CN101372235A - Automatic train operating device and train operating auxiliary device - Google Patents

Abstract

The invention relates to an automatic train operation device capable of reducing energy loss generated during travel of the train and realizing energy conservation of the train operation. The device is able to generate an operation mode causing the train to stop at a designated location at a designated time, and providing a thrust force command to a driving brake device of an electromotive machine to carrying the operation mode. The automatic train operation device comprises: a loss indication calculating means (16) for calculating the loss indicator representing the energy loss generated during operation of train; and an operation mode compensating measure (19) for compensating mode based on the loss indicator to reach the special value causing the train to stop at a designated location until a designated time.

Description

Automatic train operation device and train operation auxiliary device

(the application be denomination of invention be " automatic train operation device and train operation auxiliary device " the dividing an application of No. the 200610059702.2nd, first to file, should be for No. 200610059702.2 that denomination of invention is another dividing an application No. the 03133028.2nd, first to file of " automatic train operation device and train operation auxiliary device " in first to file.)

Technical field

The present invention relates to a kind ofly be used for automatic running train and make unpiloted electric car can regularly locate the automatic train operation device that stops, and will recommend the gear of thrust or master controller to indicate train operation auxiliary device to the driver.

Background technology

The purpose of automatic train operation device (below be called ATO) is to carry out automatically to move between the station of train, and train can be stopped at down on the given stop position at station.The electric car system architecture that is equipped with this kind ATO as shown in figure 47.

The restricting signal that is sent by not shown automatic train control device (ATC) will be imported in the automatic train operation device 1, except route conditions such as the gradient and curvature, also be imported in the automatic train operation device 1 from fixed canned datas such as the vehicle condition in the data bank 3, operation schedule, resistance to motions.Automatic train operation device 1 will be inferred current vehicle location according to the car speed of vehicle location that is detected by ground upper sensor 10 and speed detector 9 detections, and the thrust (size) that this moment should export is exported to driving brake equipment 2 as thrust instruction Fcmd.Braking force when the instruction of the thrust here Fcmd is defined as the pulling force instruction that comprises when vehicle quickens and car retardation in this manual instructs.If pulling force, Fcmd is just greater than 0 in the thrust instruction; If braking force, Fcmd is just less than 0 in the thrust instruction.

Driving brake equipment 2 is to be made of VVVF (variable voltage and variable frequency) current converter 4, main motor 5, braking force control system 6 and mechanical brake 8.Main motor 5 and wheel 7 mechanical connections that travel on track 11, mechanical brake 8 can carry out mechanical braking to wheel 7.

Obtain acting on of actual thrust from thrust instruction Fcmd and obtain under pulling force and the acquisition braking force both of these case to be different, to describe respectively below.

Obtain under the situation of pulling force, thrust instruction Fcmd (greater than 0) will input to current converter 4.Current converter 4 will be controlled so that obtain the corresponding to pulling force with thrust instruction Fcmd the moment of torsion of main motor 5.At this moment, braking force control system 6 and mechanical brake 8 are not worked.

Obtain under the situation of braking force, thrust instruction Fcmd (less than 0) will input to braking force control system 6 rather than current converter 4.Braking force control system 6 at first instructs thrust braking force instruction just to export to current converter 4.By the effect of current converter 4, will feed back to braking force control system 6 through the electric braking force Felec of main motor 5 outputs.Because the effect of thrust instruction Fcmd braking force instruction just, braking force control system 6 will be controlled mechanical brake 8, electric braking force Felec is played a role, and the insufficient section that can not be kept by this electric braking force is supplied by the mechanical braking power Fmech of mechanical brake 8.Like this, mechanical braking power Fmech becomes

Fmech＝Fcmd-Felec ......(1)。

Automatic train operation device 1, for example such as shown in figure 48, comprising temporarily travels plans device 12, the best is travelled and planned device 13 and thrust command generating device 14.The tentative planning device 12 that travels will work up tentative driving mode (F0 (x), V0 (x)) as the initial value of working out best driving mode.The thrust Fn (x) and the speed Vn (x) at x place, position on and the expression route corresponding of driving mode herein with a series of position.The best travel the planning device 13 will work up the best driving mode F1 (x) of train according to the savings information in tentative driving mode (F0 (x), V0 (x)) and the data bank 3.In thrust command generating device 14, the restricting signal that worked-out best driving mode F1 (x) will send according to detection speed and detection position, the ATC of train engraves the thrust (size) that should export and exports to current converter 4 as thrust instruction Fcmd during with this.

During the best driving mode of planning train, a variety of driving modes that can realize are arranged generally.Particularly with the train operation quantity of sooner or later moving overstocked asynchronism(-nization) seldom daytime or early in the morning, the late into the night, the operation of train is very long at interval, and also leaves affluence on the conceptual design, can alleviate the restriction of the scheme of travelling.

In the flat 5-193502 communique of Te Kaiping 8-216885 communique and Te Kai, be documented in the best as assessment item and travel in the scheme energy-conservation.But so-called energy-conservation in these known examples but do not relate to the viewpoint of produce power loss in the driving of train, control of braking such as actuating device and brake equipment.

Given this, in " the braking mode change causes the basis investigation that regenerated energy effectively utilizes effect " (Japanese railway technical tie-up seminar the 7th time), " pure electric braking practicability investigation " (5-244 of national congress of Japanese Electrotechnical Committee (JEC)), discussed specially at train braking control, the particularly scheme of travelling of the degradation of energy that produces on the glancing impact mechanical brake.But train drives the degradation of energy that produces in the control of braking and also can occur in drive controlling, and when control of braking, the device beyond the mechanical brake also has degradation of energy.If really be like this, just can't reduce the complex energy loss.

Summary of the invention

The degradation of energy that the intent of the present invention produces when being that train driven control of braking is made comprehensive evaluation, and the degradation of energy that produces in the driving process that reduces as far as possible to stand is realized energy-conservation travelling.The degradation of energy that will solve at the present invention simply describes below.

The loss meeting that produces in the train operation changes according to driving mode, but is taking place can be divided into two kinds of primary and secondarys on the machine.The first is as the degradation of energy on the electric machines such as the current converter 4 of actuating device and main motor 5.These losses are expressed as thrust and function of speed.It two is the degradations of energy that produce in mechanical brake when work.If observe the acceleration and deceleration action of train from the viewpoint of energy stream, under the degradation of energy in ignoring aforementioned electric machine and the situation of resistance to motion, in power travels accelerator, actuating devices such as current converter 4 and main motor 5 will be converted into the energy of vehicle movement by stringing (figure slightly) electric energy that provides; And electric brake is in moderating process, and then the transformation of energy with vehicle movement becomes electric energy, and power supply is given in regeneration.There is not degradation of energy in this perfect condition.But electric brake if the braking force that ATO or driver send instruction exceeds the braking force that electric machine can be exported, is just supplied with the braking force of 8 pairs of deficiencies of mechanical brake in moderating process, makes deceleration/decel (acceleration/accel negative value) maintain given value.When mechanical brake 8 is worked like this, the kinergety of vehicle as heat exhaustion, has just been produced degradation of energy.Among the present invention, the loss that produces because of mechanical brake work partly is defined as brake loss.

If it is the allowance of actuating device that the braking force instruction surpasses electric machine, as long as be not present in the power supply with the corresponding load of regenerated electric power, brake loss will take place.For the latter,, will control current converter 4 so that allow main motor 5 outputs and braking force instruct corresponding braking force if actuating device obtains braking force instruction.At this moment, the kinergety of vehicle will be transformed into the regenerated energy of power supply, but if there is no with power supply in load accordingly train in the accelerator just of regenerated electric power, will produce unnecessary regenerated electric power, so stringing voltage will rise.Actuating device will be controlled, and braking force is suppressed so that can suppress the rising of stringing voltage.More than be called light load regeneration control.In the action of this light load regeneration control, main motor 5 will be exported than the little braking force of braking force instruction.At this moment, Bu Zu braking force is replenished by mechanical brake 8.

Under the situation of energy-saving run, should work up best driving mode, also require simultaneously to travel by this photograph driving mode in practice.As the consistent device that moves of realization with driving mode, automatic train operation device (ATO) and ATS Automatic Train Stopper (TASC) etc., the device of driverless operation and generation thrust instruction automatically is to be known by people.Utilize these devices, realize following the tracks of travelling of the best scheme of travelling thereby can send thrust accurately.But, be directly connected to the driving brake equipment of vehicle, need the ground equipment of detection position, so the systematic comparison complexity, cost is very high.

In addition, by optimal thrust being indicated way, can expect to finish the train driving approaching with worked-out driving mode by driver's technical ability to the driver.Here it is, the operation auxiliary device.Though the energy-saving effect that obtains during the utilization and operation auxiliary device is owing to reasons such as driver's response lag are not so good as to use ATO and TASC, but owing to only need give the driver with the instruction indication, directly do not relate to the driving brake equipment of vehicle, therefore have the advantage that system is oversimplified.Finally by driver's operation, can remove the wayside equipment used the detection position etc. or system is oversimplified.By these a series of ways, just can expect cost-cutting, realize cost optimization.In addition, in recent years because ATOization gradually, driver's running technology is as downward trend gradually having occurred, and in the situation of utilization and operation auxiliary device, the driver will often carry out the thrust adjustment by judgement, therefore just can not produce the problem of running technology decline.

In addition, automatic train operation device is as a kind ofly making train defer to train speed limit, entering practicability already to the device that has a speed limit of certain allowance apart from speed limit.But, because PI is controlled the equal error follow-up control as main body, need also have much according to the situation of train and route characteristic, according to present situation, the work of its characteristic or controlled variable being adjusted at each train and each route also need spend a large amount of time and labors.

In addition, also consider to use and work up the scheme of travelling and be the automatic train operation device that originally carries out train driving with scheme.When scheme is travelled in formulation, utilize simple train driving pattern through regular meeting.The simplest way is that the train operation as object is showed according to following simple physics formula.

F-Fr＝M·α ......(7)

The F here represents tractive force or braking force, and Fr represents train resistance, and M is the weight of train, and α shows acceleration/accel (comprise negative acceleration, be deceleration/decel).Train resistance Fr is the resistance that train produces in the process of moving, for ease of calculating, often classifies in the following manner.That is,

Resistance sets out: the resistance when dispatching a car

Air resistance: air is to the resistance of the train that travels

Grade resistance: the resistance that route tilts to produce

Curve resistance: the crooked resistance that produces of route

Tunnel resistance: the resistance that train travels in the tunnel and produces

Also to consider the resistance between wheel tread and the orbital plane in the air resistance under most situation, and handle as 2 formulas of speed.

Generally speaking, train resistance Fr often considers according to grade resistance, air resistance, curve resistance, tunnel resistance, the resistance that sets out etc.When train travels, consider grade resistance, air resistance and curve resistance outside the tunnel.At this moment, grade resistance, air resistance and curve resistance will be asked (for example, reference literature " operation is theoretical " (direct current AC electric locomotive) friend-making society compiles) according to (8), (9), (10) respectively.Promptly

(a) grade resistance formula

Frg＝s ......(8)

Frg: grade resistance (kgf/ton)

S: the gradient (‰) (upwards for just, downwards for negative)

(a) air resistance formula

Fra＝A+Bv+Cv ² ......(9)

Fra: air resistance (kgf/ton)

A, B, C: coefficient

V: speed (km/h)

(b) curve resistance formula

Frc＝800/r ......(10)

Frc: curve resistance (kgf/ton)

R: radius of curvature (m)

During automatic train operation, when utilizing the model in (7), in the driverless train operation scheme based on the scheme of travelling, characteristics such as train characteristic and route characteristic will improve by bus comfort level greatly and improve and stop precision.

The purpose of this invention is to provide a kind of automatic train operation device and train operation auxiliary device, utilize this device can be between the train station in the driving process, reach regularly to locate and stop under the prerequisite of requirement the energy-saving run of the degradation of energy that the realization reduction produces in travelling.

Another object of the present invention is, a kind of automatic train operation device is provided, reduce adjust in required time or labour, and after business is travelled beginning, proceed to improve comfort level by bus, and improve and stop precision from dynamic characteristic study.

Still a further object of the present invention is, a kind of device is provided, and utilizes this device, only need allow train back and forth travel on the projected route of travelling and just can collect the collection operation of the required data of running gear.

A further object of the present invention is, a kind of automatic train operation device is provided, thereby the 1st, the swing influence got rid of when the automatic train operation improves energy-saving effect greatly; Thereby 2nd, calculating dead time improves the accuracy situation that stops on the target location greatly; 3rd, the phase change because of rate control instruction in the time of can be to gear position operation causes that abominable environment by bus improves.

A further object of the invention provides a kind of train positioning and stops automatic control device, need not carry out gear switch continually and just can guarantee to stop precision, and no longer need the very long adjustment time.

In order to achieve the above object, the present invention works up and can make train at given time and the driving mode that stops at given position, and the driving control device that is provided with the electric machine that comprises current converter and main motor sent the thrust instruction that is used to realize driving mode, it is characterized in that, comprising: the loss index arithmetic element that the loss index of the degradation of energy that is used for representing that the above-mentioned driving brake equipment of train driving process is produced is carried out computing; Reach according to this loss index and above-mentioned driving mode is carried out revisal so that reduce the 1st driving mode correcting unit of degradation of energy.

In order to achieve the above object, the automatic train operation device among the present invention also comprises: the data that obtain during to train driving are carried out the data processing unit of online processing; According to data that obtain when the train driving by this data processing unit and the data that obtain in advance, controlled variable, train characteristic and the route characteristic when train driving during to train driving learn automatically from dynamic characteristic unit; Use is carried out the automatic train operation unit of automatic train operation by train characteristic and route characteristic from the study of dynamic characteristic unit.

In order to achieve the above object, the automatic train operation device among the present invention also comprises: the train characteristic unit that is used for collecting train characteristic and route characteristic in the train driving process; Reach the driverless train operation unit that the information calculations of collecting relevant train according to above-mentioned train characteristic unit goes out best driving mode and according to this pattern train moved automatically.

For achieving the above object, automatic train operation device among the present invention, characteristic data and from the condition of service of automatic train control device during the operation of in according to train detection position, train detection speed, data bank, preserving, control by actuating device or control setup train, automatically move, comprising: circuit is implemented in computing when carrying out when stopping in the station of above-mentioned train stopping in the station of given computing; Reach computing enforcement circuit when travelling between the station of carrying out given computing or control between the station when train travels, computing is implemented circuit and is had and can work up the train that allows in stopping can stop at the best on target location the best of scheme scheme generation unit that travels that travels constantly in target in a station when stopping in the above-mentioned station, computing is implemented circuit and is comprised when travelling between above-mentioned station: travel the scheme driving process in the above-mentioned train worked-out the best of scheme generation unit of dispatching a car from an above-mentioned station and travel according to the best, travel scheme and actual travel result's error above the scheme scheme of the travelling arithmetic element again of computing again of travelling under the situation of given value in this best; From above-mentioned travel scheme again arithmetic element extract the control command extraction unit of control command again the scheme of travelling of computing; Reach the control command that the control command extraction unit is extracted and export to the control command output unit of above-mentioned actuating device or control setup.

In addition, in order to achieve the above object, train positioning among the present invention stops automatic control device, comprising: being used for train car respectively brakes " the braking characteristics data store " of braking characteristics data such as the dead time of deceleration/decel on the gear, braking gear switch and response lag time; Obtain " the train current data acquiring unit " of data such as train present speed, current location, current braking gear; According to the braking characteristics data of storage in " braking characteristics data store " and the train current data that obtains by " train current data acquiring unit ", work up by using a plurality of braking gears to make train stop at " the deceleration control scheme generation unit " of the deceleration control scheme on the given position; " deceleration control instruction fetch unit " from the deceleration control instruction that engraves when extracting each " deceleration control scheme generation unit " worked-out deceleration control scheme; " the deceleration control instruction output unit " of brake equipment exported in the deceleration control instruction that " deceleration control instruction fetch unit " extracted.

The present invention can realize in the driving process, guaranteeing regularly to locate on the basis of stop condition the energy-saving run that the degradation of energy in travelling is reduced when train between the station.

According to the present invention, with online mode train characteristic and route characteristic, controlled variable are learnt automatically in travelling, realized using effective automatic train operation of learning outcome.

In addition, the present invention also provides a kind of and can just can finish the device that the collection running gear moves the data acquisition operation of desired data by allowing train come and go to travel on the predetermined route that travels.

In addition, according to the present invention, therefore the influence that swing produces when having got rid of automatic train operation greatly improves energy-saving effect greatly.According to certain embodiments, can obtain dead time, therefore the precision that stops on the target location improves greatly, and in addition in other embodiments, the discontinuous variation of rate control instruction causes the bad situation of ride comfort in the time of can also improving because of gear position operation.

According to the present invention, can be according to the deceleration/decel on each gear of train, the braking characteristics data of the dead time of braking gear switch and response lag time etc., the present speed of train, current location, the data of current braking gear etc., work up by using several braking gears to make train can stop at deceleration control scheme on the assigned address, like this, can only set the scheme that the frequent switching that also need not carry out gear under the situation of deceleration/decel just can be stopped on given position with discrete value even if just can work up, just can improve comfort level by bus if carry out deceleration control, and can guarantee the precision of stopping according to this scheme.

According to the present invention, calculating the time of respectively braking gear distributes so that utilize the combination of several braking gears to reach the purpose that the location stops, and the deceleration control scheme can be worked out according to the switching instant of employed braking gear and this braking gear, like this, even also can be under the situation that deceleration/decel changes by the way of change time distribution, need not just can adjust stop position by the lifting gear, can make comfort level raising by bus, and can guarantee to stop precision.

According to the present invention, in the deceleration control scheme, slow down with the high braking gear of deceleration/decel earlier, switch to the low braking gear of deceleration/decel then, can when stopping, stop by this way with lower braking gear, travelling comfort is improved.

According to the present invention, predetermined speed when slowing down by the deceleration control scheme on the switching instant is compared with the true train speed on the switching instant, both just do not change the deceleration control scheme simultaneously, so just easily the deceleration regime of true train is made evaluation, can carry out computing again to the deceleration control scheme at the change of deceleration/decel, make to stop the precision raising.

According to the present invention, after the deceleration control scheme working-out, under the situation that the value that deceleration/decel uses when working out a scheme begins to change, the deceleration control scheme is changed, that so just can improve opposing deceleration/decel variation interference can be handling, guarantees to stop precision.

According to the present invention, time series data according to train speed in slowing down is inferred deceleration/decel, and works out the deceleration control scheme according to the deceleration/decel of inferring, that so just can improve opposing deceleration/decel variation interference can be handling, and need not carry out complicated adjustment, guarantee to stop precision.

According to the present invention, when slowing down by the deceleration control scheme each constantly or each locational predetermined speed and true train speed compare, come the deceleration/decel that uses in the revisal deceleration control scheme at difference relatively, deceleration/decel according to this revisal changes adjustment to the deceleration control scheme, that so just can improve that the opposing deceleration/decel change to disturb can be handling, and need not carry out complicated adjustment, guarantee to stop precision.

According to the present invention, deceleration/decel, gear switch dead time and the response lag time etc. of using during according to the speed in the last time step, scheme working-out, come when slowing down by the deceleration control scheme each constantly or each locational predetermined speed calculate one by one, even like this under the situation that the memory size of control computer is restricted, also can according to circumstances improve anti-deceleration/decel change disturb can be handling, and need not carry out complicated adjustment, to guarantee to stop precision.

Description of drawings

Fig. 1 is the block diagram that the driverless train operation of the expression embodiment of the invention 1 is put.

Fig. 2 is the diagram of curves of expression power machine loss index and total loss index example when travelling.

Fig. 3 is the diagram of curves of machine loss index, brake loss index and total loss index example in the expression braking procedure.

Fig. 4 is the diagram of curves of expression power transducer loss index and motor losses index example when travelling.

Fig. 5 is the diagram of curves of expression power transducer loss and motor losses example when travelling

Fig. 6 is the diagram of curves of the driving mode example of expression embodiment 1.

Fig. 7 is the block diagram of the automatic train operation device of the expression embodiment of the invention 2.

Fig. 8 is the travel diagram of curves of the brake loss example under the limited situation of charge capacity of expression power.

Fig. 9 is the block diagram of the automatic train operation device of the expression embodiment of the invention 3.

Figure 10 is the block diagram of the train operation auxiliary device of the expression embodiment of the invention 4.

Figure 11 is the block diagram of the thrust indicating device configuration example of expression embodiment 4.

Figure 12 is the block diagram of the control system of the thrust indicating device among expression Figure 11.

Figure 13 is the block diagram of thrust indicating device configuration example of the train operation auxiliary device of the expression embodiment of the invention 5.

Figure 14 is the block diagram of the train operation auxiliary device of the expression embodiment of the invention 6.

Figure 15 is the train entire block diagram that expression is equipped with automatic train operation device of the present invention.

Figure 16 is the explanation block diagram of inner structure of the automatic train operation device of Figure 15.

Figure 17 infers the concept map that carries out the driving mode revisal at the power initial stage of travelling according to weight.

Figure 18 considers that business is preceding and the flow chart of the back characteristic learning procedure of doing business.

Figure 19 is that expression is carried out the block diagram of the correcting unit of revisal to one embodiment of the invention from the dynamic characteristic learning outcome.

Figure 20 is the constructional drawing of automatic train operation device and data preservation portion.

Figure 21 is an example of driverless train driving mode.

Figure 22 is the train composition block diagram that expression is equipped with the related automatic train operation device of various embodiments of the present invention.

Figure 23 is the structured flowchart of the related automatic train operation device 1 of the expression embodiment of the invention 13.

Figure 24 is the structured flowchart of the related automatic train operation device 1 of the expression embodiment of the invention 14.

Figure 25 is the structured flowchart of the related automatic train operation device 1 of the expression embodiment of the invention 15.

Figure 26 is the structured flowchart of the related automatic train operation device 1 of the expression embodiment of the invention 16.

Figure 27 is the structured flowchart of the related automatic train operation device 1 of expression embodiments of the invention 17.

Figure 28 is the structured flowchart of the related automatic train operation device 1 of expression embodiments of the invention 18.

Figure 29 is the structured flowchart of the related automatic train operation device 1 of the expression embodiment of the invention 19.

Figure 30 is the structured flowchart of the related automatic train operation device 1 of expression embodiments of the invention 20.

Figure 31 is the structured flowchart of the related automatic train operation device 1 of expression embodiments of the invention 21.

Figure 32 is the structured flowchart of the related automatic train operation device 1 of expression embodiments of the invention 22.

Figure 33 is the structured flowchart of the related automatic train operation device 1 of expression embodiments of the invention 23.

Figure 34 is the structured flowchart of the related automatic train operation device 1 of expression embodiments of the invention 24.

Figure 35 is the structured flowchart of the related automatic train operation device 1 of expression embodiments of the invention 25.

Figure 36 is the basic structure block diagram of the related automatic train operation device 1 of expression embodiments of the invention 26.

Figure 37 represents the travel instruction diagram of scheme characteristic example of worked-out in embodiments of the present invention the best.

Figure 38 is worked-out in embodiments of the present invention or through the instruction diagram of the scheme of the travelling characteristic example of computing again.

Figure 39 is the instruction diagram of the worked-out in embodiments of the present invention scheme characteristic example of temporarily travelling.

Figure 40 is the principle of work diagram of circuit that the scheme of travelling shown in Figure 36 adopts device 24.

Figure 41 is the summary pie graph that expression train positioning of the present invention stops the embodiment 27 of automatic control device.

Figure 42 is the skeleton diagram that train positioning involved in the present invention stops deceleration control side's one case to use in the automatic control device.

Figure 43 is illustrated in the skeleton diagram that the train positioning that the present invention relates to stops to adjust by the way that changes the handover scheme time of origin in the automatic control device example of stop position.

Figure 44 is the skeleton diagram that stops in the automatic control device example of the stop position set-up procedure of being undertaken by the way that changes the handover scheme time of origin in the train positioning that the present invention relates to.

Figure 45 is the summary pie graph that train positioning involved in the present invention stops the embodiment 28 of automatic control device.

Figure 46 is the summary pie graph that train positioning involved in the present invention stops the embodiment 29 of automatic control device.

Figure 47 is the block diagram that the general electric car system configuration example of automatic train operation device is installed.

Figure 48 is the block diagram of the automatic train operation device in the system shown in Figure 47.

The specific embodiment

Below, with reference to the accompanying drawings embodiments of the invention are elaborated

Fig. 1 represents the block diagram of the summary formation that the driverless train operation of embodiment 1 is put.The best that present embodiment relates generally to automatic train operation device planning device that travels, and omitted the diagram of other parts.

The best shown in Figure 1 Planning Department 13 that travels is made up of driving mode revisal index operational part 15, driving mode correcting section 19, operating range correcting section 20 and timing property judging part 21.Driving mode revisal index operational part 15 is made up of loss index operational part 16, overload objective operational part 17 and adder 18.Loss index operational part 16 calculates the loss index CPL (x) of train position x according to tentative driving mode (F0 (x), V0 (x)).Here, CPL represents Cost of Power Loss (loss in efficiency expends).Driving mode is represented thrust Fn (x) or the speed Vn (x) on the x of position.

Fig. 2 and Fig. 3 represent the example of various loss indexs.Loss index when Fig. 2 shows power and travels, the loss index when Fig. 3 shows braking deceleration.In more detail, Fig. 2 (a) expression machine loss index, Fig. 2 (b) expression total loss index, Fig. 3 (a) expression machine loss index, Fig. 3 (b) expression brake loss index, Fig. 3 (c) expression total loss index.The machine loss index here is meant the loss index of electric machine, just refers to the amount of changer (current converter) loss index after motor (main motor) loss index adds.

These indexs are that (v, loss F) [W] is multiplied by the amount of the back reciprocal acquisition of speed [m/s] at certain operating point as the function of speed v and thrust F.After the inverse of taking advantage of calculation speed, the speed v 1[m/s on certain operating point] subtle change Δ v[m/s takes place] time loss that produces just can access study plot and evaluate and test.

Total loss index CPL (x) carries out addition with machine loss index and brake loss index, multiply by the amount that draws behind the weight coefficient W1 on its result.Weight coefficient W1 be from for the viewpoint of effect of reduction loss is to a certain degree set, or obtain and other each indexs between balance set.Promptly

Loss index CPL (x)

=W1 * (machine loss index+brake loss index) ... (2)

Overload objective operational part 17 carries out computing according to tentative driving mode (F0 (x), V0 (x)) to the overload load index COL (x) on the train position x.COL is Cost of OverLoad (overload expends).

Machine loss is carried out computing as transducer loss and motor losses sum.Fig. 4 (a) (b) represents transducer loss [W] and motor losses [W] on each operating point.According to tentative driving mode (F0 (x), V0 (x)), corresponding transducer loss [W] and motor losses [W] are carried out integration respectively, so just can calculate the transducer loss [J] and motor losses [J] that have added running time between the station.Under the situation of these (index) overrates [W], correspondingly calculate overload objective.For example, weight coefficient is made as W2, with transducer loss index COLC (x) according to

COLC(x)

=W2 * { transducer loss [J]/(running time+parking period in standing)

-changer rated value [W] } * transducer loss index (Fig. 5 (a)) ... (3)

Calculate.

COLC is Cost of Over Load in Converter (overload in the changer expends).

Can use the motor losses index shown in Fig. 5 (b) to obtain motor losses index COLM (x) (still, weight coefficient is made as W3, can independently set) equally.Overload objective COL (x) be according to

COL(x)＝COLC(x)+COLM(x) ......(4)

Calculate.

COLM is Cost of Over Loss in Motor (loss of crossing in the motor expends).

Adder 18 will be lost index CPL (x) and overload objective COL (x) addition, and with the overall performance C (x) on the x of train present position according to

C(x)＝CPL(x)+COL(x) ......(5)

Calculate.

Driving mode correcting section 19 is added to overall performance C (x) on the thrust mode F0 (x) of tentative driving mode, and exports the 1st revisal driving mode F01 (x) (this stage goes up velocity mode V0 (x) and do not change) on this basis.

Because the 1st revisal driving mode F01 (x) revisal thrust mode, so operating range and given value are inconsistent.Be consistent apart from x and given value in order to allow travel, operating range correcting section 20 according to the condition of the route conditions of storage in the data bank 3 or vehicle condition, resistance to motion to the 1st revisal driving mode (F01 (x), V0 (x)) carries out revisal, and export the 2nd revisal driving mode (F02 (x), V02 (x)) and running time T run.Can realize by adjusting the method that waits between inertial time apart from revisal.Certainly, the method apart from revisal is not limited in this.

Judge by 21 couples of running time T of timing property judging part run, judge at the situation whether given value is in the error allowed band.Running time T run exceeds under the situation of error allowed band, the 2nd revisal driving mode (F02 (x), V02 (x)) is made as new tentative driving mode (F0 ' (x), V0 ' (x)), and calculates again.Running time T run is under the interior situation of error allowed band, and it is made as the suitableeest driving mode (F1 (x), V1 (x)) and output.

According to said structure, tentative driving mode (F0 (x), V0 (x)) will carry out revisal to the thrust on the position x with remarkable result according to loss index CPL (x) and overload objective COL (x).For example, Fig. 6 represents the result that the driving mode in the embodiment of the invention generates., do not reach overload here, overload objective can not exert an influence.What " proterotype (A) " was represented is tentative pattern.That " index is suitable for (B) " is represented is the 1st revisal driving mode (F01 (x), V01 (x)).The high speed glancing impact that the loss index is big more just needs big thrust revisal more, and braking force is died down.On the other hand, to quicken duration little though power travels, and can obtain revisal with the cooresponding pulling force of loss index." gear quantization (C) " be not though appear in the example that originally carries out an invention, but gear has only 6 grades, corresponding to the situation that can't export continuous thrust, it can select the cooresponding thrust of gear (size) with the thrust error minimum at the thrust F01 (x) of the 1st revisal driving mode." distance adjust (D) " is at " gear quantization (C) " pattern operating range to be carried out the 2nd revisal driving mode (F02 (x), V02 (x)) that revisal makes the suitable and given value 1300m of this distance.Loss in the driving mode before the revisal is 2070[KJ], the loss of the 2nd revisal driving mode is 1650[KJ by contrast], degradation of energy has lowered much.On in motion, the former is 84.5[sec], and the latter is increased to 84.9[sec].Repeat computing till running time reaches given value, not only can guarantee timing property and stopping property of location, and can also produce to make and drive the best driving mode F1 (x) that degradation of energy that control of braking relates to becomes minimum.Like this, on the basis of guaranteeing timing property and stopping property of location, can also reach best energy-saving effect.

Make in the driving mode that total energy loss reduces, the degradation of energy that is included in the electric machine generation that the current converter 4 (changer) that drives in the brake equipment 2 or main motor 5 (motor) wait might increase.Electric machine has specified operating range,, heating can occur and cause temperature to rise just under overload condition at the condition of service of over range, and protection work or fault, burn etc. will take place.Overload objective operational part 17 is judged the overload degree of each machine according to tentative driving mode.Judged result be the overload situation under, just according to overload objective revisal thrust so that the degradation of energy in the electric machine is inhibited.Because the thrust in the excessive zone of degradation of energy can obtain revisal, thereby can avoid entering overload effectively.So just, can avoid stopping to take place, the reliability of system is improved greatly with fault because of the electric machine overload causes running.

By in the driving process of train, implementing the best scheme of travelling, each instantaneous position and speed is used as initial condition (IC), guarantee timing property and stopping property of location before the train arrival, can work up optimal energy-conservation driving mode on this basis.In other words, utilize speed limit conditions such as ATC,, also can draw a kind of maximum energy-saving driving mode at this state immediately even broken away from initial driving mode temporarily.If simply according to initial driving mode, just might make loss become big, be worthless from the angle of energy.That is to say, cause still can realizing the best energy-conservation scheme of travelling constantly under the unexpected situation about taking place at this when breaking away from initial driving mode.

In the present embodiment, initial condition (IC) is used as in position and speed, guarantee train arrival timing property and stopping property of location before, and can generate the maximum energy-saving driving mode on this basis, between not only can being applied to stand, the present invention carries out in the automatic train operation device (ATO) of driverless train operation, also applicable in the braking interval, positioning in the automatic stop control of train (TASC) of the control of stopping.

The represented algorithm of present embodiment is that the revisal driving mode is so that running time reaches given value under the consistent prerequisite with given value of operating range.But under the prerequisite consistent with given value, the revisal driving mode is so that operating range reaches the algorithm of given value between also can adopting in motion.

Fig. 7 is the block diagram of the basic comprising example of driverless train rotating device among the embodiment 2, and part identical with Fig. 1 among Fig. 7 adopts identical symbol, and it illustrates omission, at this part different with Fig. 1 is illustrated.

To be input to from the operation schedule in the data bank 3 in the loss index operational part 16, the charge capacity of will travelling from the power in the data bank 36 is input in the loss index operational part 16.Power in the data bank 36 charge capacity of travelling is power in the some time is engraved between the service area electric power on the train in the accelerator that travels, and promptly the power charge capacity of travelling stores.From travelling the database information of load, operation schedule and power extracts the load (amount) of travelling of cooresponding power by loss index operational part 16.As previously mentioned, because brake loss can change according to the load that power travels, thereby can calculate and the power corresponding loss index of charge capacity of travelling.The same Fig. 1 of other parts.

Foregoing has following effect and effect.

At the power that the dopes charge capacity of travelling, will loss index CPL (x) particularly suitably be adjusted the brake loss index.For example, the expression power brake loss index under the very big situation of load of travelling among Fig. 3 (b), in current converter 4 under the restriction of electric weight, the situation of the high braking force of high speed, the loss index will be big more.Fig. 8 represents the travel brake loss index of (125KW/ master's motor) under the little situation of load of power.In the case, because the load that power travels is little, thereby the zone that can not export the braking electric power identical with thrust instruction Fcmd is arranged.That is begin to increase when, the loss index is from low speed.So just can predict degradation of energy exactly, realize energy-conservation travelling more effectively by loading condition.

Fig. 9 is the block diagram of the summary configuration example of the automatic train operation device among the embodiment 3, and the part identical with Figure 16 all adopts same-sign and save explanation among this figure, only different parts is illustrated at this.

Adopt data bank 34 and driving mode extracting part 35 in the device shown in Figure 9, replace tentative among Figure 48 Planning Department 12 and the best Planning Department 13 that travels that travels.Store the driving mode that each train travels in the data bank 34 between each station.Driving mode extracting part 35 will from the data bank 3 of storing operation schedule, extract and existing station between the corresponding driving mode F1 (x) that travels.By carrying out the way that embodiment 1 described the best is travelled scheme and the best driving mode among its result is stored in advance, the driving mode that is stored in the data bank 34 just can be accomplished.

Above-mentioned structure can reach following effect and effect.

For the generation of best driving mode, calculate and carry out preferred plan owing to require to repeat convergence, thereby the computing suitable spended time that gets up.When implementing the scheme of travelling at the next stop in the docking process, because the restriction on operation time just can't reach optimum efficiency.If carry out the restriction that these schemes just can be avoided operation time in advance, and can access best driving mode.Like this, just can better realize energy-conservation effect.And, by precomputing the way of driving mode, can also confirm driving mode exactly.Thereby get rid of unusual pattern, the confidence level of system is improved greatly.

Figure 10 is that the summary of the electric car system that train operation auxiliary device is installed among the embodiment 4 constitutes block diagram, among this figure with Figure 47 in same section all adopt identical symbol and omit its explanation, only different parts is illustrated at this.

At this, the automatic train operation device 1 that adopts train operation auxiliary device 22 to replace among the embodiment 1.Train operation auxiliary device 22 is to carry out and the identical processing of automatic train operation device 1 among the embodiment 1, and produces thrust output recommendation Frec.In other words, train operation auxiliary device 22 is equivalent to a thrust instruction Fcmd that can replace in the automatic train operation device 1, the device of thrust output recommendation Frec.This thrust recommendation Frec is imported in the thrust indicating device 24 that is attached on the master controller 23.Master controller 23 will be exported to the cooresponding thrust instruction of angle or the position F cmd of master controller and drive brake equipment 2.

The configuration example of thrust indicating device 24 as shown in figure 11.Thrust indicating device 24 is made of angle ordering calculation portion 25, impedance controller 26, servoamplifier 27, servomotor 28 and coder 29.Servomotor 28 mechanically combines with master controller 23.

To input to angle ordering calculation portion 25 from the thrust recommendation Frec of train operation auxiliary device 22 outputs.Angle ordering calculation portion 25 calculates the cooresponding master controller angle with this thrust recommendation Frec, and it is exported as angle instruction θ cmd.Impedance controller 26 instructs the angle θ of θ cmd and coder 29 detected actual master controllers to import angle, and can make the latter (angle θ) and the torque command T cmd that the former (angle instruction θ cmd) is consistent input to servoamplifier 27.27 pairs of servomotors 28 of servoamplifier drive so that allow the output torque of servomotor 28 and torque command T cmd be consistent.

Impedance controller 26 imposes on torque T ope on the master controller 23 at the driver, and servomotor 28 is controlled so that reach desirable impedance size (moment of inertia J, damping D, stiffness K), and the block diagram of control system can be as shown in Figure 12.J0 is equivalent to the moment of inertia of equal value after the rotor of servomotor 28 and master controller 23 merging, and g1 and g2 are equivalent to the cut-out frequency of noise reduction with filter.

Angle is instructed under the situation of θ cmd zero setting (0), from the required moment of torsion of exterior master controller 23, promptly, the driver is applied to torque T ope on the master controller 23 and the transfer function θ (s) between the master controller angle θ, under the situation of ignoring noise filter, can calculate according to following formula, and can obtain desirable impedance (J, D, K).

$\mathrm{\θ}\left(s\right)=\frac{1}{J\·s^2+D\·s+K}\·\mathrm{Tope}......\left(6\right)$

According to above-mentioned formation, can reach following effect and effect.

Utilize servomotor 28 to control the angle θ of master controller 23, so that the thrust recommendation Frec that thrust instruction unit 24 can be accessed calculate with train operation auxiliary device 22 is consistent thrust instruction F cmd.Therefore, the driver utilizes the control of 26 pairs of impedance of impedance controller in operation master controller 23, can allow the driver obtain own desirable impedance (J, D, K).In other words, do not contact the driver under the state of master controller 23, can obtain the thrust instruction F cmd consistent with thrust recommendation Frec.When driver operation master controller 23, will be subjected to the effect of the power sent from servomotor 28 on the direction of thrust recommendation Frec, can be set at arbitrarily angled is thrust instruction F cmd.Like this, the driver just can give train operation auxiliary device 22 with the operation Control work and do, and can operate master controller 23 as required, freely controls the thrust instruction.For the angle θ that realizes the master controller 23 that energy-saving run adopts can be used as the antagonistic force of autonomous controller 23 to detect, and can constantly move in the stop mode of the energy-conservation location of understanding.Like this, the operation by the driver just can realize energy-conservation travel and locate stop to travel, when the situation of beyong contemplation takes place, also can promptly take measures to deal with simultaneously.

The thrust instruction F cmd that supplies with driving control device 2 is not directly actuated by train operation auxiliary device 22, but from the angle θ of original master controller 23, and it can make easier that system becomes.And train operation auxiliary device finally is by driver operation, and the location that need not strictly set train operation auxiliary device 22 stops precision, can make device become very easy.Like this, the confidence level of system can improve, and cost also reduces greatly.

For the situation of train operation auxiliary device,, therefore require driver's operative technique very consummate because ultimate demand is operated by the driver.If described, utilize system that automatic train operation device is housed just the situation etc. of unforeseen circumstances generation can not occur causing inadequately because of the driver operation technology according to present embodiment.

Figure 13 is the block diagram of the summary configuration example of train operation auxiliary device among the embodiment 5.Present embodiment is compared with embodiment 4, and the structure of thrust indicating device 24 is different, is elaborated at this part below.But in the present embodiment, the thrust instruction will be configured to the gear of several master controllers, and promptly power travels and quickens 6 grades of (P1-P6), braking deceleration 6 grades of (B1-B6), neutral gear (N) and special brakings (EB).

The gear is here represented the result of speed and thrust modeization, and is used in the current electric car drive controlling.The shelves number of gear from several grades to not waiting more than 30 grades, and select variety classes for use according to the difference of system.In addition, master controller 23 shown in Figure 13 is the summary pie graphs when watching from above.

Thrust indicating device 24 is by recommending gear display control unit 30 and lamp group 31 to constitute.In illustrated embodiment, lamp group 31 is made up of 14 lamps altogether, and they are respectively: travel with power and quicken corresponding 6 lamps of gear P1-P6, with corresponding 6 lamps of braking deceleration gear B1-B6, with corresponding 1 lamp of neutral gear N, and with cooresponding 1 lamp of unusual braking gear EB.Recommend gear display control unit 30 to receive the recommendation gear command N rec that train operation auxiliary device 22 sends, and light cooresponding lamp.

According to said structure, this example has following effect and effect.

The driver confirms the gear that will set by lighting of lamp, so that carry out energy-conservation the travelling (scheme) that guarantees timing property and stopping property of location.For example, recommend the content of gear command N rec to be in power and travel and quicken a shelves P6, lamp correspondingly will be lighted, if braking deceleration shelves B3, cooresponding with it lamp also can be lighted.The driver carries out gear position operation according to the situation of lighting of lamp to master controller 23, so just can suppress degradation of energy, thereby reach purpose of energy saving.

There is no directly electric or mechanically contact between thrust indicating device 24 and the driving braking control system, owing to also need between this to finish by driver's operation, thereby thrust indicating device 24 can earlier be handled it than driver's judgement when unexpected situation takes place, and the confidence level of system is improved greatly.Lamp or LED read out instruments such as (light-emitting diodes) is simpler than the servo structure of the master controller among the embodiment 4 23, easier realization, and the confidence level of system improves greatly, and reaches the purpose that reduces the constituent apparatus cost.

Figure 14 is the summary configuration example block diagram of the train operation auxiliary device among the embodiment 6.Present embodiment and embodiment 5 are different, and the structure of thrust indicating device 24 changes, and is elaborated with regard to difference below.

Thrust indicating device 24 in the present embodiment is by recommending gear display control unit 32 and audio output unit 33 to be constituted.Recommend gear display control unit 32 to receive the recommendation gear command N rec that is sent by train operation auxiliary device 22, control audio output unit 33 makes the corresponding newspaper of its output sound.For example, when recommending gear to be B3, will export similar newspaper sounds such as " braking 3 grades ".

Said structure has following effects and effect.

The driver can be known the situation of set position by the newspaper sound, so that guarantee timing property and stopping property of location, reaches energy-conservation purpose of travelling, and so also can reach same effect and effect among the embodiment 5.But embodiment 5 is the states by lamp shows the recommendation gear, and driver's attention all concentrates on gear and shows, causes the place ahead absent minded, probably causes accident to take place.Relative therewith, if instruct reception and registration, just there is not this problem with sound, the confidence level of simultaneity factor also improves greatly.

Figure 15 and Figure 16 represent an embodiment of the automatic train operation device among the present invention.The automatic train operation device (ATO) 108 that carries on the train 0 among the figure obtains the speed limit data from the automatic train control device (ATC) 102 as overground system, and from the data bank (DB) 103 of train 0, obtain route conditions (angle of inclination and curve curvature radius etc.), vehicle condition (train marshalling list, weight etc.), data such as condition of service, and obtain departure signal from operating console 104 respectively, from answer loading device 105, obtain and answer the loading signal, from speed detector 106, obtain the train speed signal, and from making the position signal that obtains train the ground sensor detector 107 of response to suitably being configured in ground upper sensor on the route.The ground upper sensor that is provided with on the route is used for confirming the position of train.Herein, DB103 is installed in the train 0, but the outside that also can according to circumstances be installed in train 0 in addition, according to circumstances can also be arranged in the train 0 and the ground as ground system dispersedly.

ATO108 is used for handling the data processing unit 180 and automatic train operation unit 181 of online data except that being installed with, and also being equipped with characteristic judging unit 124 before doing business and business back characteristic unit 134 is the deduction unit and the unit of representative.Data processing unit 180 carries out the train speed Signal Processing, except train speed, also to carry out continuous computing to train position (time integral of speed), train acceleration/accel (differential of speed) and train driving distance (time integral of speed absolute value).Train position and train driving distance will suitably be carried out revisal by the train position signal from ground upper sensor 107.Digital processing element 180 carries out given computing according to each incoming signal, for study described later or train operation provide required instrumentation data.The necessary instrumentation data of automatic train operation are to be provided by automatic train operation unit 181.Actuating device 9 is issued in the instruction that the result that automatic train operation unit 181 goes out according to each input data operation travels power, and deceleration instruction is issued speed reduction gearing 110.Actuating device 109 comprises the main motor of tractor-trailer train and the power converter of the main motor of control.Speed reduction gearing 110 generally all has mechanical brake and power brake simultaneously.

Carrying ATO108 in the train 0, as Figure 16 in detail shown in like that, wherein relate to the business of learning functionality of the present invention before characteristic infer that unit 124 and business back characteristic unit 134 are made of following each device institute: judging unit 120 travels before the business, characteristic initial value setup unit 121 before the business, test is travelled with automatic train operation unit 122 before the business, the memory cell 123 as a result of travelling, characteristic is inferred unit 124 before the business, inferred results correcting unit 125, characteristic inferred value memory cell 126, learning characteristic data bank (learning characteristic DB) 130, characteristic initial setting unit 131, automatic train operation unit 132, the memory cell 133 as a result of travelling after the business, business back characteristic unit 134, and learning outcome correcting unit 135.Unit 121-126 belongs to the preceding processing equipment used when travelling of testing that travels of doing business, unit 131-135 belongs to the used processing equipment in beginning back that travels of doing business, do not concern that both are all general before and after judging unit 120 and learning characteristic DB130 and the business of travelling before the business.

Among Figure 16, ATO108 will be used as automatic train operation device and use, data processing unit 180 that omission is installed originally and automatic train operation unit 181 etc.

The following describes the effect of installing among Figure 15 and Figure 16.

Among Figure 15, ATO108 will obtain the speed limit data from ATC102, from DB103, obtain available in advance various information such as route conditions or vehicle condition, condition of service etc., and acquisition speed, carry out given computing as standard, send control commands such as power driving instruction or deceleration instruction, realize the automatic operation of train 0 as described above like that.

ATO108 will accept the departure signal sent by operating console 104, allow the automatic train operation unit begin to carry out automatic run action.After dispatching a car, will use from the speed data of answering loading information, obtaining from speed detector 106 of answering that loading device 105 obtains and from ground sensor detection information that ground sensor detector 107 obtains.Answer loading information to be used as the information relevant with the weight of train, sensor detection information is used for the revisal location information on the ground.Utilize these information, ATO108 will work up the instruction (power driving instruction/deceleration instruction) of control train.The power driving instruction is as under the worked-out situation of control command, the outputting power driving instruction, and drive train drivings by actuating device 109.The power driving instruction can also adopt the power gear instruction or the like of travelling under the situation that gear travels except dynamic torque (power travel tractive force) instruction.Under the worked-out situation, export deceleration instruction as control command at deceleration instruction, and make train deceleration by speed reduction gearing 110.As deceleration instruction, can adopt the braking force instruction, under the situation that gear power travels, can adopt drg gear instruction or the like.

Below, describe the effect of ATO108 in detail with reference to Figure 16.

The departure signal of sending according to the operating console 104 that receives at first allows the judging unit 120 that travels before the business judge that (current) belongs to travelling travelling after the still business beginning of test before the business.The method of judging can adopt the method that works up soft sign, " does not travel for test during day-mark will " such as picture, and such methods of utilizing soft sign such as " travelling for business during day-mark will " perhaps adopts hard switching to set method of result or the like.

The judged result of judging unit 120 is that the preceding test of business is travelled if travel before the business, just sets the initial characteristic parameter when test is travelled before doing business before business in the characteristic initial value setup unit 121.The method of setting can adopt man-machine interface, the method for utilizing manual mode to set before the beginning of travelling.The content of setting value can be the specification of train or route characteristic etc., extracts characteristic parameter and input from the information that in advance can set about.

Then, use the characteristic parameter that characteristic initial value setup unit 121 configures before the business, the train that allows test before the business travel to carry out automatic operation scheme with automatic train operation unit 122 is tested and is travelled.Method as the driverless train operation, being to work up the best scheme of travelling when stopping in the station, is that benchmark moves automatically with this scheme, if travel under the situation of scheme generation than large deviation with the best, just work out the scheme of travelling again, carry out revisal by the method for Error Feedback at control command.And, just in time belong to travelling before the business this moment, for example, the train that travels for gear can travel being used for measuring testing under the gear of characteristic, and can carry out to measure characteristic is travelling of purpose.

Then, test before the business is travelled and moved the result who obtains automatically with automatic train operation unit 122, store in the memory cell 123 of travelling as a result.Storing process be with target travel scheme and when travelling the speed data of instrumentation or position data or the like store in the media of hard disk (HD) etc. with the form of electronic file.

Then, the test that stores according to the memory cell 123 as a result of the travelling result that travels allows before the business characteristic infer that unit 124 infers characteristic parameter.Should infer before the business that the characteristic parameter of getting well can select weight, accelerating performance, decelerating behavior etc.

About marshaling single-piece weight, the test travel phase before business, because the passenger do not get on the bus, acceleration/accel in the time of can be according to inertia traveling or deceleration/decel (acceleration/accel negative value) and train resistance are inferred.At this, can consider situation about showing as the train of determination object according to the such simple physical formula of formula (7).

When calculating train resistance, can carry out computing according to the formula of considering route characteristic such as the gradient, curvature or air resistance, friction drag.In addition, the computing about train resistance please refer to " operation theoretical (the direct current AC electric locomotive) " that document friend-making society compiles.Generally speaking, train resistance Fr can be expressed as following formula

Fr＝Frg+Fra+Frc

＝s+(A+Bv+Cv2)+800/r ......(11)

Wherein, Fr is train resistance [kgf/ton], and Frg is grade resistance [kgf/ton] (upwards for just, downwards for negative), Fra is resistance to motion [kgf/ton], and Frc is curve resistance [kgf/ton], and s is the gradient [‰], A, B, C are coefficient, and v is a train speed, and r is a radius of curvature.

Consider these parameters, (7) are deformed into

M＝(F-Fr)/α ......(12)，

Weight is inferred according to formula (12).In the formula (12), under inertia traveling, tractive force can be 0.For acceleration/accel (deceleration/decel) α, can adopt method of least squares, by carrying out obtaining after the computing from instrumentation result (train running speed).Utilize above-mentioned each processing just can infer weight M.

The deduction computing of weight M just can utilize this to be worth and infer power ride characteristic and braking characteristics after finishing.

At first, working weight inferred value Mest, the acceleration alpha acc when power travels and train resistance Fr carry out the deduction of power ride characteristic (power travel relation of gear and tractive force etc.).Acceleration alpha ccc when travelling about power and train resistance Fr can be by obtaining with the same processing of aforementioned weight computing.Use this tittle and weight inferred value, infer tractive force F according to following formula

F＝Mestαacc+Fr ......(13)。

In the time of can inferring that according to formula (13) train carries out the power mobility operation of gear, the tractive force size of travelling of the power on each gear.So just, can infer travel relation between gear and the tractive force of power.

Working weight inferred value, deceleration/decel and train resistance when slowing down can be inferred the braking force characteristic.Deceleration/decel during deceleration and train resistance can obtain in the same processing of computing by with above-mentioned weight the time.Use this tittle and weight inferred value, infer braking force F according to following formula

F＝＝Mestαdec+Fr ......(14)

Wherein α dec is deceleration/decel (an acceleration/accel negative value).

According to formula (14), when train carries out the gear brake operating, can infer the braking force on each gear.Just can infer the relation of braking between gear and the braking force according to inferred results.

For above-mentioned inferred value, carry out computing in the time of preferably can between the station, travelling back or parking, but also can in the train driving process, carry out computing, and in the process of moving operation result is confirmed.Like this, by the way that weight, power ride characteristic, braking characteristics are inferred, just can before travelling, business use the deviation of adjusting each train marshalling list than the shorter time of the past.

Then, allow inferred results correcting unit 125 infer that at the preceding characteristic of business the unit good characteristic inferred value of 124 deductions carries out revisal.Carry out before the revisal, answer the allowed band of the pre-set characteristic parameter that can realize in theory, the purpose of revisal is in order to guarantee not exceed this allowed band.For example, exceed at the characteristic inferred value under the situation of allowed band, consider to use the setting value that calculates in advance, perhaps use critical value in the allowed band etc.Under the situation of this allowed band of substantial deviation, also need to test once more the operation of travelling etc.

Then, the characteristic inferred value that characteristic inferred value memory cell 126 is good with 125 revisals of inferred results correcting unit stores among the learning characteristic DB130.The method of storage can adopt and the identical method of the aforementioned memory cell as a result 123 of travelling.The test of learning characteristic DB130 before storage business is travelled travelled the characteristic inferred results that obtains, and also stores the characteristic learning outcome that the study after business described later is travelled obtains.

Next the situation of result that the judging unit 120 that travels before the explanation business is judged for travelling after doing business.

Under the situation that business is travelled, set by the initial value of 131 pairs of characteristic parameters of characteristic initial value setup unit earlier.In initial business was travelled, the characteristic parameter (characteristic inferred results) that takes out from learning characteristic DB130 by 126 storages of characteristic inferred value memory cell used.Travel carrying on the business, and when learning, will use the characteristic parameter (characteristic learning outcome) that from learning outcome, obtains.

Then, use the characteristic parameter that configures by characteristic initial value setup unit 131, travel automatically by automatic train operation unit 132 control trains.The automatic operation of train is travelled with the preceding test of business basically and is used automatic train operation unit 122 identical, but because the ridership of riding after the business is indefinite, weight will change.Therefore, when train departures have just begun to enter power and travel, must infer the weight when travelling between departures.For the deduction method of weight,, just use and answer loading if can obtain answering loading.Answer loading if can not utilize, just infer weight by the way of the identical effect of characteristic deduction unit 124 before performance when just the entering power and travel after departures and the business and inferred results correcting unit 125.If inferred results is different with the value that characteristic initial value setup unit 131 is set, just need work out processing such as the scheme of travelling again.After departures, just carried out carrying out when power travels the basic condition of weight deduction referring to situation shown in Figure 17.

In Figure 17, transverse axis represents to arrive from departures the distance of the next stop, i.e. position, and the longitudinal axis is represented with the situation of each locational speed as velocity mode.Under the dead ship condition of departures, to follow begin to travel according to the worked-out best driving mode 131 of characteristic inferred value (fine dotted line among the figure) after, according to travel actual travel result in interval 130 of initial stage power, be that actual travel pattern 132 (heavy line among the figure) is carried out the weight deduction, by computing again based on the weight inferred value, work up the driving mode 132 (thick dashed line among the figure) after the revisal, and carry out the actual travel operation as standard.

Then, travel after business memory cell 33 as a result stores the automatic operation result of automatic train operation unit 32.The method of storage can be taked and the identical method of the aforementioned memory cell as a result 23 of travelling.

Then, use the result that travels who is being stored by the memory cell 133 as a result of travelling after the business, back characteristic unit 134 is carried out the study of characteristic by doing business.About the regular learning content of this characteristic, be divided into following several situation:

(1) learns according to the result that travels between the station

(2) learn according to the system-wide line result that travels

(3) learn according to 1 day the result that travels

(4) result that travels according to a few days learns

(5) result that travels according to several months learns

Below respectively (1)-(5) are described respectively.

(1) learns according to the result that travels between the station

The result that travels between the station that obtains after travelling between standing learns as benchmark, learning outcome is reflected to travels between next secondary.For example, the characteristics that Sometimes When It Rains reduce at braking force are learnt.As judging the example that the result that is necessary to travel between slave station learns, the situation that braking force reduces during the rainy day is exactly an example.Rainy day if will use the power-absorption fan of train, will consider that the frictional behaviour that makes brake plate owing to raining reduces the situation generation that causes braking force (deceleration performance) to reduce when driving a vehicle.In case rain, deceleration descends possibly.The study of carrying out the braking force characteristic on this result's basis is relatively good.Here the learning outcome that obtains mostly is temporary, also can other approach, keep utilizing as temporary transient characteristic parameter.

(2) result that travels according to whole process learns

Learn according to the result who from start to finish travels on the route, when next bar route begins to travel, learning outcome is reflected.For example, travelling on the route that is through with must arrive each station, under the situation of the stop position that departs from objectives when stopping (having produced bias), in order to reduce this bias, preferably learns the braking force characteristic at bias.For example, surpass under the situation of target stop position, we can expect that the setting value of braking force characteristic will be suitably greater than actual value.Want big because braking-force ratio is actual, therefore the deceleration/decel that can't obtain envisioning.In this case, preferably learn for the purpose that reaches the setting value that suitably reduces the braking force characteristic.

(3) learn according to one day result that travels

The result that travels according to one day learns, and learning outcome is reflected in the travelling of next day.For example, when observing one day the result that travels (for example 1 route on the through running result that travels repeatedly), can be described as on certain necessary station for the target stop position the phenomenon that walks far with degree takes place, might be because produced deviation in the setting of route characteristic parameters such as the gradient or curve between the station.At this moment, the destination of study is exactly in order step by step route characteristic parameters such as the gradient or curve to be adjusted slightly according to the result that travels.

(4) result that travels according to a few days learns

The result that travels of a few days is stored, learn according to this event memory.For example, observation a few days travel as a result the time, on the scheme of in the identical time period, travelling, produced deviation, whether so just can consider because some influence, made power in this time period travel tractive force characteristic or the final and actual deviation that takes place of braking force characteristic.In other time period, do not observe under the situation of deviation, can think in fact and, therefore can only carry out revisal time period as object not exclusively because deviation takes place characteristic parameter itself, and then by learning that this compensating value is carried out revisal.

(5) result that travels according to the several months learns

Store several months travel as a result the time, according to the storage the result learn.The result who stores during for example, according to maintenance examination learns.For example, observe 3 months travel as a result the time, sometimes may be according to before 3 months, before 2 months, before 1 month, observe the situation that reduces gradually through braking force along with the time.These situations are if be difficult to judge according to travel result's study of a few days.Under the situation of using air brake, the abrasion of the brake plate that might produce rubs causes.Therefore, just be necessary characteristic parameter to be changed (study), and take to change measures such as brake plate according to degree according to this result.In addition, also to take to change radius of wheel, perhaps take measures to wear out preventing.

Above-mentioned study scheme as an example, can be learnt by program as shown in figure 18, realize selectively.In Figure 18, the judging unit 120 of travelling before the business is judged the test that belongs to before the business the travel result of (step 151) of business after the still business that travels, if the former (business before test travel), test before just carrying on the business travel (step 152), and initial parameter inferred (step 153), finish then.If the judged result in the step 151 is travelled for business, just carry out and travel any in the corresponding 5 kinds of study of content.That is to say, state when travelling of travelling of business finished is judged (step 154), if travel end between the station, just carry out " learn according to the result that travels between the station (1) " (step 155), if through running finishes, just carry out " learn according to the through running result (2) " (step 156).In the step 154, if 1 day the end of travelling, just judge the data whether (step 157) of having stored several days, according to this result, if 1 day data of storage are just carried out " learn according to 1 day the result that travels (3) " (step 158), if stored the data of a few days, just carry out " learn according to a few days result that travels (4) " (step 159),, just carry out " learn according to the result that travels of several months (5) " (step 160) if stored the data of several months.

But each learning procedure 155,156,158,159,160 that thick line is represented among Figure 18 only is to have under the situation of following study tendency and just can carry out observing the result that travels.Promptly

(a) deviation of same tendency situation about always continuing (for example,, between all stations, all observing with the situation of missing the target stop position of degree etc.) for the through running result, and

(b) the tangible situation of deviation.

About study, can consider to adopt the method that relevant characteristic parameter is increased and decreased according to certain ratio.For example, as described above, for the through running result, if between all stations, all observe with degree miss the target stop position time, just can consider to surpass the actual too much situation about causing of braking force, thereby can learn so that the setting value of braking force characteristic suitably can be reduced certain proportion because of the setting value of braking force.

Particularly for being the situation that benchmark is learnt with the result that travels between the station, can observe several situations with deviation of same tendency is seldom.Therefore, in this case, can consider the study of following manner.Promptly

Become the driverless train operation scheme of object:

This mode is to have taken place under the situation of bigger deviation on scheme and the actual measured value when travelling, the driverless train operation scheme of control command (power travel gear instruction, the instruction of braking gear etc.) being carried out revisal at this deviation.

Learning method:

Control command revisal situation when deviation takes place between scheme of travelling and the actual measured value is learnt.With the characteristic of braking force,,, just can't obtain the deceleration/decel of being envisioned if the situation of the control command revisal of the braking force gear more eager to excel in whatever one does occurs than plan for example at glancing impact.At this moment, because the setting value of braking force characteristic is more a lot of greatly than actual braking force value, therefore can learn the predetermined value of braking force characteristic is suitably reduced certain proportion.If the situation of the control command revisal of the braking force gear more weak occurs than plan, then opposite, learn, braking force property settings value is increased certain ratio.

For inferring the characteristic judgement different with actual value, it is the acceleration/accel that obtains according to as the instrumentation data, use related characteristics, route style characteristic (gradient, curve etc.), weight, tractive force or braking force, judge whether to satisfy formula (7) as the train driving of hypothesis characteristic.

As mentioned above, learning outcome correcting unit 135 will carry out revisal to the learning outcome that is obtained by business back characteristic unit 134.The method of revisal can adopt the processing identical with aforementioned inferred results correcting unit 125 to implement.Revisal result herein will be used as the characteristic learning outcome and store among the learning characteristic DB130.

The above also will be learnt after entering business operation, on one side the operation of carrying on the business the regulating characteristic parameter time.

Be in the on-line automatic study in the docking process when study of above-mentioned most of kinds it is envisaged for train arrival etc.But the weight when power travels is inferred the on-line automatic deduction that then it is envisaged in the driving process.

By learning on one side and carry out when inferring the method for automatic train operation, can be effectively move automatically at factors such as the deviation of train marshalling list and timeliness variations.

As mentioned above, if adopt the automatic train operation device among the embodiment 7, by before business is travelled to the travel deduction of tractive force, braking force of weight, power, just can be enough adjust at the deviation of every row train marshalling list than the shorter in the past time, by carrying out the way of characteristic parameter study after the business, even the automatic operation that also can realize under the situation about changing on the characteristic parameter satisfying ride comfort and stop precision.In the study after business,, can realize more study near actual conditions by partly waiting and divide enforcement according to being divided into the partial sum route running that travels between the station during the data of utilizing study.In addition, in the study after deduction before the business and the business, by to the way of inferring and learning outcome carries out revisal, even extremely uncommon afterclap occurred, also, just can infer and learn, and needn't use those uncommon characteristic parameters as long as by revisal.

As mentioned above, just might make actv. the best scheme of travelling by the study of constantly carrying out characteristic.If fairly large study occurs in train driving, just taking this as an opportunity works out the scheme of travelling again, so just can realize satisfying comfort level, target stop position precision, the operation of the driverless train on the opportunity of travelling by bus.

Among the embodiment 7, most study is contemplated when all being train arrival etc. is in the on-line automatic study of carrying out in the docking process, and the weight in the power driving process infers that contemplated is the on-line automatic deduction of carrying out in the driving process.But,, just can be implemented in the system that carries out on-line automatic study in the driving process and utilize learning outcome by driver's judgement if the man-machine interface that can confirm to learn to carry out situation in the train driving process has been installed.In the case, can be with learning device with the auxiliary device of single assembly as the driverless train operation.

Figure 19 represents the structure of the main portion of the automatic train operation device among the embodiment 9.In the present embodiment, as business back characteristic learning device, comprise in the claim from dynamic characteristic unit 1341, from dynamic characteristic unit 1342, from dynamic characteristic unit 1343, from dynamic characteristic unit 1344 and from dynamic characteristic unit 1345, also comprise the learning outcome comparing unit 136 that is used for importing these learning outcomes of obtaining from dynamic characteristic unit in addition, also comprise the learning outcome correcting unit 137 that learning outcome is carried out revisal according to the comparative result in the learning outcome comparing unit 136.

As explanation among the embodiment 7, carry out characteristic study respectively from dynamic characteristic unit 1341-1345.Learning outcome comparing unit 136 is accepted from the learning outcome that obtains in dynamic characteristic unit 1341-1345 these learning outcomes to be compared, and whether check has big contradiction each other.Between dynamic characteristic unit 1341-1345,, therefore can the result between the long learning period be checked by the result that obtain between the short learning period because between the learning period, promptly the interval of learning differs bigger.For example, learning outcome in dynamic characteristic unit 1345 and the learning outcome in dynamic characteristic unit 1344 at the same time between section differ n doubly, such as differing under 10 times the situation, be judged as and occur obviously unusually, and can think that very big contradiction appears in the learning outcome from dynamic characteristic unit 1345.In addition, utilize a plurality of results in dynamic characteristic unit 1341-1345 to check, the precision of inspection is improved greatly.

Then, learning outcome correcting unit 137 carries out revisal at the comparative result of the big contradiction of appearance of learning outcome comparing unit 136.The simplest correction method is exactly directly to utilize the learning outcome of (learning at interval) between the short learning period from dynamic characteristic unit.But, when using a plurality of learning outcome in dynamic characteristic unit 1341-1345, can consider to use the aviation value of these learning outcomes.The situation of contradiction appears in the major part in the learning outcome in dynamic characteristic unit 1341-1345, and exist each other under the situation of big deviation from the learning outcome of dynamic characteristic unit 1341-1345, can consider that also these learning outcomes are taken out aviation value to be used.

In dynamic characteristic unit 134, use suitable observer just can carry out the study of characteristic.Suitable observer is that object apparatus is being carried out under the situation of numerical modeling according to formula (7), according to observation (mensuration) probable value its parameter is discerned.Also can consider to carry out system identification, constantly allow automatic train operation unit 181 utilize the recognition result of suitable observer, constitute a kind of suitable control system according to observation procedure.According to formula (7), use suitable observer, can utilize the power of the acceleration-deceleration (detection speed by speed detector 106 is calculated) of observed value and control command value travel tractive force or braking force, constantly weight, train resistance are discerned.Algorithm as suitable observer, ((repair, the U.S. male work of Jin Jing happiness by trailing vortex mirror in temple referring to " strong adaptation control is crossed the threshold " for details can to adopt extendible method of least square or extended Kalman filter, suitable filter etc., オ-system society periodical) the 2nd chapter " deduction of unknown devices and adaptation observer " P.47-87,3.3 joints " adaptive filter " of perhaps " system controls serial 6 optimum filterings " (Western Hills are outstanding clearly, training wind shop) P.50-57).

As mentioned above,, the learning outcome that contradiction occurs is excluded, just can obtain more high-precision characteristic learning outcome by (study at interval) during the comparative learning different several from the dynamic characteristic learning device.

Among the embodiment 11, automated characterization unit 134 is carried out characteristic study by interference observer.Interference observer is used action control fully and is waited in the work, be that the device that is used for discerning interference ((write referring to " MATLAB Control System Design " in detail by strong Tibetan of wild ripple, west village's show and, flat Tian Guangnan is outstanding altogether, Tokyo motor university press) 4.4 joints " interference observer in the action control " P.99-102).Regard the train resistance in the formula (1) as move in controlling exciting force, just can utilize interference observer constantly train driving external force to be inferred.Utilize inferred results to learn, can improve the precision of study greatly.

Below, describe embodiments of the invention 12 with reference to the accompanying drawings in detail.Figure 20 is the constructional drawing that expression automatic train operation device 1 and data are preserved unit 201.

Automatic train operation device 1 is by being constituted as the train characteristic learning device 207 of train characteristic unit with as the automatic operation control part 208 of driverless train operation unit.Train characteristic learning device 207 can obtain train characteristic data (train resistance, dead time etc. (aftermentioned)) and route data etc. in the train driving process.The data that train characteristic learning device 207 gets access to will be stored in the data preservation portion 201.To be output in the automatic operation control part 208 by train characteristic learning device 207 data that get access to and that be kept in the data preservation portion 201.By automatic operation control part 208,, set up the scheme of travelling according to the data that obtain and be stored in by train characteristic learning device 207 in the data preservation portion 201.Allow train move automatically according to this scheme.

Train characteristic learning device 207 is made of following each several part: the data preservation portion 201 that preserves the unit as data, as the travel weight of train calculating part 209 of tractive force deviation detecting unit of weight of train calculating unit and power, train resistance calculating part 210 as the train resistance calculating unit, braking force calculating part 211 as braking force calculating unit and braking force deviation detecting unit, dead time calculating part 212 as the dead time calculating unit, riding rate calculating part 213 as the riding rate calculating unit, and the device that detects train speed.

The content of data preservation portion 201 outputs will be input to weight of train calculating part 209, train resistance calculating part 210, braking force calculating part 211, riding rate calculating part 213, move in the control part 208 automatically.The content of weight of train calculating part 209 outputs will be input in the data preservation portion 201.The output content of train resistance calculating part 210 will be input in the data preservation portion 201.The output content of braking force calculating part 211 will be input in the data preservation portion 201.

The output content of dead time calculating part 212 will be input in the data preservation portion 201.The output content of riding rate calculating part 213 will be input in the data preservation portion 201.The output content of operation control part 8 will be input in weight of train calculating part 209, braking force calculating part 211, dead time calculating part 212, the riding rate calculating part 213.

When the power that weight of train calculating part 209 gives it the gun at train travels, with input from the train resistance value in the data preservation portion 201, the train speed V that engraves when travelling value of thrust F and this from the power of automatic operation control part 208.Weight of train calculating part 209 calculates according to formula (15) according to train resistance value Fr, value of thrust F and train speed V, draws weight of train M.The weight of train M that weight of train calculating part 209 is tried to achieve will be kept in the data preservation portion.In formula (15), M is that the weight of train, F are that travel value of thrust, Fr of power is that train resistance, α are the train acceleration/accel.The train acceleration alpha can be obtained by train speed V.

M＝(F-Fr)/α ......(15)

Weight of train calculating part 209 is as the power tractive force deviation test section that travels, about the power value of thrust F that travels, weight of train calculating part 209 also uses the weight of train value M that calculates, when the V1 value of using when calculating weight of train M when the value of speed V is different, the value substitution formula (15) of M just can be obtained the correct power value of thrust F that travels.And this power travel deviation of tractive effort command value Fk of power that value of thrust F and automatic operation control part 208 send of travelling also can be measured by weight of train calculating part 209.Power travel the deviation between the value of thrust F of tractive effort command value Fk and power of travelling will be output in the data preservation portion 201 and preserve.In order to determine power travel the deviation between the value of thrust F of tractive effort command value Fk and power of travelling, power travel the deviation between the value of thrust F of tractive effort command value Fk and power of travelling is added to and surveys timer-operated power and travel on the tractive effort command value Fk, the value that obtains is calculated as the new power tractive effort command value Fk that travels, can more correctly carry out automatic train operation by such processing.

Train resistance calculating part 210 is imported weight of train M, speed V by data preservation portion 201 when train is done inertia and slided.Weight of train M, speed V by 201 inputs of data preservation portion will calculate train Resistance Value Fr according to formula (16).During the train coast, because the power tractive force that travels is inoperative, can establish the power value of thrust F that travels is 0.Value of thrust F is 0 because power travels, and formula (15) is deformed into formula (16).The train resistance value Fr that through type (16) is calculated will export to data preservation portion and store there.In the formula (16), M is the weight of train, and F is the power value of thrust that travels, and Fr is the train resistance value, and α is the train acceleration/accel.The train acceleration alpha can be obtained by train speed V.

Fr＝F-Mα＝0-Mα ......(16)

In the ordinary train shown in " operation theoretical (direct current AC electric locomotive) friend-making society compile " (hot-short is more or less different), train resistance value Fr will by formula (17) be expressed as like that grade resistance value Frg, curve resistance value Frc and resistance to motion value Fra three's and.In addition, know that easily grade resistance value Frg, resistance to motion value Fra, curve resistance value Frc are respectively suc as formula shown in (18), formula (19), the formula (20).

Train resistance value Fr calculates by weight of train M and speed V during inertia traveling, therefore also can calculate grade resistance value Frg and resistance to motion value Fra by train resistance calculating part 210.Resistance to motion value Fra calculates by speed V.Curve resistance value Frc then uses the data that are kept in advance in the data preservation portion 1.For train resistance value Fr, resistance to motion value Fr, curve resistance value Frc being used as numeric data, train resistance calculating part 210 will calculate grade resistance value Frg according to the deformation type of formula (17).The grade resistance value Frg that is calculated by train resistance calculating part 210 will be output in the data preservation portion 201 and remain there.In the formula (18), s represents the gradient (%) (upwards for just, downwards for negative).In the formula (19), A, B, C are coefficient, V represents speed (km/h).In the formula (20), r represents radius of curve (m).The train resistance calculating part can be measured grade resistance value or train resistance value when train driving, therefore is not difficult to obtain correct data.In addition, can determination data because just back and forth travel once according to predetermined route, so can shorten greatly on the time.In formula (17), formula (18), formula (19), the formula (20), train resistance is Fr, and the train resistance value is Fr, and the resistance to motion value is Fr, and the curve resistance value is Frc.A, B, C is a coefficient, r represents radius of curvature.

Fr＝Frg+Fra+Frc ......(17)

Frg＝s ......(18)

Fra＝A+Bv+Cv ² ......(19)

Frc＝800/r ......(20)

Braking force calculating part 211 is respectively by moving control part 208 input train speed V, braking instruction Fs automatically, by the 201 input weight of train M of data preservation portion, train resistance value Fr.Braking force calculating part 211 will utilize train speed V, weight of train M, train resistance value Fr according to formula (21) thus carrying out computing draws braking force value Fb.The braking force value Fb that is calculated by braking force calculating part 211 will be output in the data preservation portion 201 and be stored in the there.

Use aforesaid formula to describe once more.In the formula (21), braking force is Fb, and weight is M, and acceleration/accel is α, and the train resistance value representation is Fr.

Fb＝Mα+Fr ......(21)

Braking force calculating part 211 will calculate (with reference to formula (22)) to braking force value Fb and the deviation Fh between the braking force command value Fs that braking force calculating part 211 is calculated as the braking force error measuring means.Deviation Fh between braking force value Fb that braking force calculating part 211 is calculated and the braking force command value Fs will be output to preservation portion 201 and be kept at the there.Braking force value Fb that braking force calculating part 211 is calculated and the deviation Fh between the braking force command value Fs are added to and carry out deviation Fh and survey timer-operated braking instruction value Fs and go up and just can obtain new braking force command value Fs, and braking force value Fb carries out work just can to allow train choose more according to the method.In the formula (22), the braking force value is Fb, and the braking force command value is Fs, and deviation card is shown as Fh.

Fh＝Fs-Fb ......(22)

Carry out glancing impact, the quantitative data input of the moment T2 by automatic operation control part 208 during with the data of the moment T1 of braking instruction value Fs output and train deceleration is in the dead time calculating part.Deviation Th (with reference to formula (23)) between the data in the T2 moment when dead time calculating part 211 will calculate T1 data constantly that braking force command value Fs is output with train deceleration.The deviation Th that dead time calculating part 211 calculates will be output in the data preservation portion 201 and be kept at the there.Dead time Th represents that actual braking instruction instructs the back to arrive the time of the whole process need of actuating device 205 or brake equipment 206 up to braking instruction by automatic operation control part 208.By way to the mensuration of dead time Th, on the basis of considering dead time Th, work out the scheme of travelling, just can set up the more correct and safe scheme of travelling.In formula (23), by automatic operation control part 208 moment after the braking instruction value F output is called T1, the moment of train deceleration is T2, dead time is expressed as Th.

Th＝T2-T1 ......(23)

The weight of train M, at full load passenger's number N, people's the average weight Mc that engrave when the weight of train Mk when riding rate calculating part 213 is imported empty wagons respectively by data store 201 and this.Weight of train M, the patronage N of at full load that engraves when the weight of train Mk when riding rate calculating part 213 will use empty wagons and this, these data of average weight Mc of people are carried out computing according to formula (24), and Mrate calculates with the riding rate inferred value.The riding rate inferred value Mrate that riding rate calculating part 213 calculates will be output in the data preservation portion 201, and be kept in the data preservation portion 201.In formula (24), the weight of train during empty wagons is Mk, this time weight of train that engraves be M, train at full load patronage is N, people's average weight is Mc, the riding rate inferred value is expressed as Mrate.

$\mathrm{Mrate}=\frac{\frac{M-\mathrm{Mk}}{\mathrm{Mc}}}{N}......\left(24\right)$

In the aforesaid train characteristic learning device 202, weight of train calculating part 209 can carry out power at train and weight of train M be calculated when travelling, and can the weight of train M on the current time be outputed in the riding rate computer device by data preservation portion 201.Therefore, can estimate each the station between riding rate Mrate.Because can between the station, infer riding rate Mrate, therefore, just can analyze the variation of every station riding rate and cause the variation of riding rate because of asynchronism(-nization).And, because weight of train calculating part 9 can calculate the weight of train M on the current time, therefore can calculate correct data with regard to train resistance value Fr or grade resistance value Frg.Automatically move control part 208, as as opening flat 5-193502 the spy and the flat 6-284519 of Te Kai records and narrates, by ground upper sensor and train speed and elapsed time the train current location is detected, ((longitudinal axis is a speed with reference to Figure 21 according to the driverless train operation mode, transverse axis is a distance (position))), the decision target velocity.Allow train follow the tracks of this target velocity by automatic operation control.In addition, also have the method for coming the detection position by operating range and ground upper sensor, therefore the mode of operation control part is hard-core automatically.

Operation control part 208 in the present embodiment comprises the dead time correcting unit, the power that are not installed in traditional automatic operation control part travel tractive force deviation correcting unit, braking force deviation correcting unit.Dead time correcting section (figure part omitted) as the dead time correcting unit is imported dead times by dead time calculating part 212.Dead time correcting section (figure part omitted) calculates braking force on the basis that has added dead time or power travels the tractive force time opening, and control power travels the tractive force time opening.Power weight of train calculating part 209 imput powers of tractive force deviation correcting unit (figure part omitted) by the tractive force deviation detecting unit that travels as the power tractive force deviation of travelling of travelling.Travel tractive force deviation correcting unit (figure part omitted) of power travels and recalculates the power tractive effort command value of travelling on the basis of tractive force deviation having added power, and the control power tractive force that travels.Braking force deviation correcting unit (figure part omitted) is by braking force computer device input brake power deviation compensating value.Braking force deviation correcting unit (figure part omitted) recalculates the braking force command value, control brake power on the basis of additional brake power deviation compensating value.

Because train characteristic learning device 207 can obtain picture riding rate, the weight of train, train resistance, braking force or the like data in the train driving process, therefore the automatic train operation device among the embodiment 12 among the present invention not only is applicable to the safety situation of operation data acquisition before automatically, also be applicable to the situation that in fact has the passenger to take operation, or work out again in the vehicle of the scheme of travelling by the data that when operation gathers.In the present embodiment, take to be undertaken by train characteristic learning device 207 mode of data handing in the train driving process, certainly, it is also no problem that data handing carries out behind train driving.In the present embodiment, the situation of braking force has been described, has also comprised the braking gear certainly, the method for braking is hard-core.In addition, the train characteristic learning device in the present embodiment is also considered the data acquisition of rainy day, the data of each Various Seasonal, the data of each different route, the data acquisition at each different station or the like, so route data is not limited to only gather once.

Figure 22 represents to be equipped with the train composition block diagram of the related automatic train operation device of various embodiments of the present invention.Be equipped with by the speed detector 302 of the formations such as (PG) of the impulse deviser in the rotating shaft that is installed in wheel on the train 0 and be used for detecting the ground sensor detector 303 of setting ground upper sensor (transmitter responder) in orbit, also be equipped with in addition these train detection speed of input put the automatic train operation device 1 of signal and train detection position signal, by the actuating device 305 and the brake equipment 306 of this automatic train operation device 1 control.What the automatic train control device of not expressing among the figure (ATC) was sent will be imported in the automatic train operation device 4 about the ATC signal of speed limit etc. or condition of service etc.

Be provided with data bank 300 in the automatic train operation device 1, circuit 304B was implemented in computing when computing was implemented to travel between circuit 304A and station when stopping in standing, above-mentioned train detection speed signal and train detection position signal will be imported into when travelling between this station computing and implement among the circuit 304B.Computing enforcement circuit 304A is used for carrying out following given computing when train 0 stops when stopping in standing, and computing enforcement circuit 304B carries out following given computing or control when travelling between standing in the driving process between train 0 station.And, characteristic data and operation schedule various data such as (tables) when storing operation such as route conditions (gradient, curvature etc.) and vehicle condition (train characteristic such as speed limit, vehicle weight and acceleration and deceleration performance etc.) in the data bank 300.Can adopt as hard disk that carries in the automatic train operation device 1 or the like as data bank 300, but along with the continuous development of magnetic card technique, the driver also can use portable IC-card.

Figure 23 is the block diagram that expression embodiments of the invention 13 related automatic train operation devices 1 constitute.Computing is implemented circuit 304A and is provided with the best scheme generation unit 307 that travels when stopping in standing, and computing is implemented circuit 304B and is provided with the scheme of travelling arithmetic element 308, control command extraction unit 309 again when travelling between standing, and control command output unit 310.And canned data will be transfused to when computing is implemented to travel between circuit 304A and station when stopping in arriving at a station arithmetical circuit and implement 304B in the two in the data bank 300.From speed detector 302 and on the ground the various detection signals of sensor detector 303 and ATC signal then only are transfused to when travelling between arriving at a station computing and implement among the circuit 304B.

The best scheme generation unit 307 that travels will generate the bests scheme of travelling according to the various data of storage in the data bank 300, so that train 0 can be parked on the target location of next halting point in target constantly from a certain station." the best " condition here can be made as various conditions.For example, can be made as the running time override, improve the energy-saving efficiency override, thereby perhaps override is considered to reach good environment by bus for acceleration and deceleration suddenly not etc.In addition, for the best travel maintenance method of related data in the scheme of the best of scheme generation unit 307 of travelling, can consider and to treat as control command with the time or the corresponding speed target value of distance etc. as an example.

The best scheme generation unit 307 that travels is made the travel method of scheme of the best, has and for example uses the mechanics train model, the method (for example, the spy opens flat 5-193502 number) of prediction train driving movement etc.As shown in figure 37, this method is to obtain power travel curve, inertia traveling curve and retrograde braking curve earlier by the way of prediction, and the intersection point of inertia curve and retrograde braking curve is made as the braking starting point.

Travelled by the best again the scheme of travelling scheme generation unit 307 is worked-out to travel scheme the time by arithmetic element 308 input, respectively by speed detector 302 and sensor detector 303 input train detection speed, train detection position and from the ATC signal of ATC on the ground, when the scheme of travelling that generates and the error between the actual travel result exceed given value, the computing once more of the scheme of travelling.

Control command extraction unit 309 is according to from the scheme of the travelling scheme of travelling of arithmetic element 308 inputs again, extract with actuating device 305 and brake equipment 306 corresponding this time assisted instruction and the deceleration instruction that engrave, and export to control command output unit 310.Control command output unit 310 will be exported to actuating device 305 and brake equipment 306 from the assisted instruction and the deceleration instruction of 9 inputs of control command extraction element.

Below, the action among Figure 22 of structure as mentioned above is described.Now, suppose that train 0 is parked in certain station, the best canned data that travels in the scheme generation unit 307 comparable data storehouses 300 generates the best that arrives next halting point scheme of travelling.Then, train 0 begins to travel, the scheme of travelling arithmetic element 308 again will be to the best travel scheme and according to speed detector 302, computing is carried out and the actual travel result that obtains compares in the detection speed that obtains of sensor detector 303 and train detection position on the ground of the best that scheme generation unit 307 makes of travelling, when both poor (for example, the best travel poor between speed target value in the scheme and the speed actual value, i.e. speed course latitude error) computing again of the scheme of just travelling when surpassing predefined critical value.

At this, both differences exceed the state of critical value, except that wave the phenomenon initiation because of aforesaid train, also might be other car because the place ahead of working direction is at a stop, and send the situation that the speed limit alteration command causes from ATC.Carry out computing again on the basis that the scheme of travelling is travelled between arithmetic element 308 actual speed that can engrave and actual distance (train position) or station excess time of being allowed again when considering by computing again.

And, control command extraction unit 309 from the scheme of travelling again arithmetic element 308 through the scheme of travelling that computing draws again, extract control commands such as assisted instruction or deceleration instruction, these control commands that control command output unit 310 will extract are exported to actuating device 305 or brake equipment 306.According to the computing and the control of automatic train operation device 304, train 0 will reach the target location of next halting point constantly in target.Afterwards, train 0 is at next halting point parking period, and the best scheme generation unit 307 that travels also will work up the best that arrives next stop scheme of travelling, and unit 308-310 will carry out identical action.In addition, the scheme of travelling arithmetic element 308 again must arrive the best that the best scheme generation unit 307 that travels works out and travels and just carry out computing again when error between scheme and the actual travel result exceeds given value, will directly export to control command extraction unit 309 by the best scheme generation unit 307 worked-out the bests scheme of travelling of travelling.

According to the embodiment among above-mentioned Figure 23 13, after train 0 beginning is travelled according to the scheme generation unit 307 worked-out the best schemes of travelling of being travelled by the best, actual travel result and this scheme of travelling produce under the situation of certain deviation, the scheme of travelling arithmetic element 308 again can be carried out the computing again of form scheme immediately, so just, the bigger train that once took place before can suppressing waves phenomenon, reaches energy-conservation effect.

Figure 24 is the structured flowchart of the related automatic train operation device 1 of expression embodiments of the invention 14.The difference of Figure 24 and Figure 23 is, used on the arithmetic element 308 cumulative errors with reference to the type scheme arithmetic element 311 again of travelling again in the scheme of travelling of Figure 23.Therefore the scheme of travelling among Figure 23 arithmetic element 308 again might be subjected to bigger noise effect, thereby cause unsettled computing again owing to engrave error when computing constantly all must be to this again at each and whether surpass critical value and make judgement.Therefore, in the present embodiment, cumulative errors with reference to type travel scheme again arithmetic element 311 will judge the error that is accumulated to a certain degree (for example, accumulative total reaches error between time of 5 minutes).Do like this and just can prevent to be subjected to above-mentioned big noise effect and unsettled computing again generation.

Figure 25 represents the structured flowchart of the automatic train operation device 1 that embodiments of the invention 15 are related.The difference of Figure 25 and Figure 24 is, is provided with control command correcting unit 312 between control command extraction unit 309 and the control command output unit 310.This control command correcting unit 312 has judges the scheme of the travelling scheme of travelling of arithmetic element 308 outputs and the function whether error between the actual travel result exceeds critical value again, be judged as under the situation that exceeds critical value, will carrying out revisal the control command that extracts by control command extraction unit 309.Owing to designed above-mentioned control command correcting unit 312, therefore had the function of supporting automatic train operation device 1.

In other words, train 0 according to the best travel scheme generation unit 307 or travel scheme again the scheme of travelling that calculates of arithmetic element 308 carry out actual travel and just do not have any problem, travel but also must break away from the scheme of travelling sometimes.For example, taken place on one of them in a plurality of drgs when unusual etc.But in the present embodiment, even this situation, braking instruction correcting unit 312 also can reach the auxiliary effect of supporting, can carry out suitable revisal to control command, and the stop position that therefore can prevent train 0 position that departs from objectives is excessive.In addition, structure according to Figure 25, between control command extraction unit 309 among Figure 23 and the control command output unit 310 control command correcting unit 312 has been installed, but, this control command correcting unit 312 also can be arranged on the control command extraction unit 309 among Figure 24 and accuse between the instruction output unit 310.

Figure 26 represents the structured flowchart of the automatic train operation device 1 that the embodiment of the invention 16 is related.Figure 26 and Figure 25 difference are, have used cumulative errors with reference to type control command correcting unit 313 in the control command correcting unit 312 shown in Figure 25.For the control command correcting unit 312 among Figure 25, as long as when once judging error between the scheme of travelling and the actual travel scheme and exceeding critical value, control command correcting unit 312 will carry out revisal at the control command from control command extraction unit 309 immediately, do the influence that is easy to be subjected to noise like this, cause unsettled control easily.Therefore, in the present embodiment, cumulative errors will be differentiated around the error (for example, accumulative total reaches 5 fens errors between the clock time) that adds up to a certain degree with reference to type control command correcting unit 313.Mention noise effect and unsettled control above can preventing like this.

Figure 27 represents the structured flowchart of the automatic train operation device 1 that embodiments of the invention 17 are related.Difference among Figure 27 and Figure 26 is, the scheme of travelling arithmetic element 308 again becomes cumulative errors with reference to the type scheme arithmetic element 311 again of travelling.The structure of other parts is identical with Figure 26, here no longer elaborates.In addition, among this embodiment, cumulative errors between the scheme of travelling and the actual travel result will judge that the critical value that these 2 unit use can be set different values for according to the difference of condition by 2 unit 311,313 when cumulative errors are judged.

Figure 28 represents the structured flowchart of the automatic train operation device 1 that embodiments of the invention 18 are related.The difference of Figure 28 and Figure 27 is, the best that circuit 304A is implemented in computing when stopping in standing scheme generation unit 307 that travels will become dead time consideration type the best scheme generation unit 314 that travels, and comprise " dead time " data in the train characteristic data of storage in the data bank 300 in addition.

Generate the dead time of the train response of the control command when the computing at the scheme of travelling, travelling during this period of time of true train given in the influence from control command output back to control command just effect really, for its solution procedure, it is excessive that the load of mathematical algorithm can become, and is not very practical with regard to arithmetic speed.Therefore, in the present embodiment, be stored in and comprised the dead time of obtaining in advance in the data bank 300 in the train characteristic data, what travel the scheme generation unit as the best is the best of " dead time consideration type " the scheme generation unit 314 that travels, and can take into account this dead time when the best scheme of travelling generates.Do like this, can when arriving next halting point, improve the target location greatly and stop precision.

Figure 29 represents the structured flowchart of the automatic train operation device 1 that the embodiment of the invention 19 is related.The difference of Figure 29 and Figure 28 is, with the cumulative errors among Figure 28 with reference to type travel scheme again arithmetic element 311 be re-established as the dead time consideration type scheme arithmetic element 315 again of travelling.This dead time consideration type travels, and arithmetic element 315 and dead time consideration type the best scheme generation unit 314 that travels is identical again for scheme, and the time data with reference in the train characteristic data that are stored in the data bank 300 carries out the again computing relevant with the scheme of travelling.Do like this, can when arriving next halting point, improve the target location greatly and stop precision.

Expression is the scheme of travelling of " dead time consideration type " travel the structure after 314 combinations of scheme generation unit of unit 315 and " dead time consideration type " the best of reruning among the embodiment 19.But, the travel structure of scheme generation unit 307 combinations of common the best with non-" dead time consideration type " also can be arranged, just with the scheme of travelling among Figure 23 to Figure 27 again arithmetic element 308,311 be replaced as the dead time consideration type scheme structure of arithmetic element 315 again of travelling.

Figure 30 represents the structured flowchart of the automatic train operation device 1 that embodiments of the invention 20 are related.The difference of Figure 30 and Figure 29 is, the scheme generation unit 314 that travels of the dead time consideration type the best among Figure 29 is configured to be the place ahead forecasting type the best scheme generation unit 316 that travels.This place ahead forecasting type the best scheme generation unit 316 that travels also belongs to a kind of of " dead time consideration type ", but it is to work out the scheme of travelling according to the prediction on train 0 working direction, so that train 0 can be parked on the target location in next halting point.

As shown in figure 38, along train exercise direction to the train motivation prediction carry out computing, and restrain computing so as to allow train with target velocity by objective (restraining computing in the time of the deceleration starting point that perhaps staggers bit by bit), do not use the curve that drives in the wrong direction just can work up the best scheme of travelling like this.If do not consider under the yet passable situation of dead time, owing to the place that counter-rotating target location in the reference braking characteristics is obtained can be made as the braking starting point, therefore computing can be said to be easily and carries out, but must consider that the words mathematical algorithm that calculates according to this reversible type can become very complicated under the situation of dead time.Therefore, asking braking to need expensive operation time on the starting point, obtain braking starting point moment of operation result target location of having passed by.In addition, method shown in Figure 38 will be obtained the braking starting point by the prediction computing on the travel direction repeatedly, even of this sort computing has many times, also can carry out in each given sample period, and end with the short time.

Figure 31 represents the basic structure block diagram of the automatic train operation device 1 that embodiments of the invention 21 are related.The difference of Figure 31 and Figure 29 is, with the dead time consideration type among Figure 29 travel scheme again arithmetic element 315 set the place ahead forecasting type scheme arithmetic element 317 again of travelling for.This place ahead forecasting type travels, and arithmetic element 317 and the place ahead forecasting type the best scheme generation unit 316 that travels is the same again for scheme, carries out computing according to the prediction on the working direction of train 0 during computing again so that train 0 can be parked on the target location in next halting point in the scheme of travelling.Therefore, only just can finish the scheme of the travelling computing again of having considered dead time with very short time.In addition, this the place ahead forecasting type travel scheme again arithmetic element 317 not only can travel with the dead time consideration type among Figure 29 scheme again arithmetic element 315 replace, can also with the scheme of travelling among Figure 23 to Figure 27 and Figure 30 again arithmetic element 308,311,315 replace.

Figure 32 represents the basic structure block diagram of the automatic train operation device 1 that embodiments of the invention 22 are related.The difference of Figure 32 and Figure 31 is, with the place ahead forecasting type among Figure 31 travel scheme again arithmetic element 317 set one by one the place ahead forecasting type scheme arithmetic element 318 again of travelling for.The place ahead forecasting type among Figure 31 travel scheme again arithmetic element 317 in predefined each given control cycle, carry out the scheme of the travelling computing again of the place ahead prediction computing, but, if utilize the forecasting type of the place ahead one by one in the present embodiment travel scheme again arithmetic element 318 just not necessarily must in each control cycle, carry out computing again.For example, the sampling control cycle is 0.3 second, and so per 1 second, carrying out once-through operation in perhaps per 10 seconds could.Changed the cycle of computing more like this, computational load is diminished.Can on the basis in the abrupt change of considering line slope and variation place of speed limit or the like, suitably make decision when in addition, determining execution cycle.

Figure 33 represents the basic structure block diagram of the automatic train operation device 1 that embodiments of the invention 23 are related.The difference of Figure 33 and Figure 32 is, with the forecasting type of the place ahead one by one among Figure 32 travel scheme again arithmetic element 318 set the driving the place ahead one by one of measurement of velocity forecasting type scheme arithmetic element 319 again of travelling for.Promptly, if the detection sample period of speed detector 302 is set for example 1 (msec) for, computing is implemented circuit 304B and is directly used the speed detection signal of importing with this cycle when exercising between standing, but the speed detection signal of input in during 5-10 (msec) degree is carried out processing such as filtering, new data more then.And, the driving the place ahead one by one of measurement of velocity forecasting type travel scheme again arithmetic element 319 will carry out the scheme of the travelling computing again of the place ahead forecasting type according to this Data Update cycle.So just can suppress the influence of noise etc., the operational precision when improving computing again.

Figure 34 represents the structured flowchart of the automatic train operation device 1 that embodiments of the invention 24 are related.This embodiment is that computing is implemented to add on the circuit 304B and exercised the result between the station and preserve unit 320 when travelling between station in the structure of Figure 31, the device that forms behind the computing enforcement circuit 304A when while stops in the additional station on dead time deduction unit 321 can be inferred dead time according to up-to-date enforcement result.Like this, not necessarily store the dead time data in the data bank 300 among this embodiment.

In other words, train 0 is in case send from the station, arrive before next halting point during the station between exercise result data, as train position, train speed, ATC signal etc. just will be stored arrive at a station between enforcement as a result in the memory cell 320.And, when in case train 0 arrives next stop parking, in this docking process, dead time infers that unit 321 just preserves the data of storing in the unit 320 with reference to the result that travels between the station and carry out the time deduction, and inferred results is exported to dead time consideration type the best scheme generation unit 314 and the place ahead forecasting type scheme arithmetic element 317 again of travelling of travelling.Travel scheme generation unit 314 and the place ahead forecasting type of dead time consideration type the best travels that arithmetic element 317 will be when considering the dead time of having inferred again for scheme, works up the scheme of travelling that arrives the interval before next halting point and carries out computing again.

At this, will illustrate dead time infer the dead time deduction method that unit 321 is carried out, this method does not use the complex calculations algorithm, employing be that signal value according to the instrumentation data changes the straightforward procedure of inferring.For example, be example with the glancing impact, with control of braking instruction output and carry out gear position operation, through the phenomenon that rank rear vehicle speed after a while will occur descending, can infer the time that drops to before the pre-set critical value this moment as dead time.In addition, being stored in dead time in the data bank 300 among Figure 28 to Figure 33 there is no need especially restriction in the time and arrives under certain state and obtain, therefore the result who infers with the complex calculations algorithm also can preserve, allowing train 0 test travels, use the dead time among this embodiment to infer unit 321, can obtain data more simply.

According to this embodiment, therefore can obtain to reflect the dead time of up-to-date train characteristic, travel through travel scheme generation unit 314 and the place ahead forecasting type of dead time consideration type the best that arithmetic element 317 is worked-out respectively and can improve greatly by the reliability of the enforcement scheme of calculation process more again for scheme.

Figure 35 represents the structured flowchart of the automatic train operation device 1 that embodiments of the invention 25 are related.The difference of Figure 35 and Figure 34 is, computing is implemented to have added the online deduction of dead time unit 322 on the circuit 304B when exercising between standing, and the place ahead forecasting type enforcement scheme arithmetic element 317 again can be carried out computing again according to the dead time of being inferred by the online deduction of this dead time unit 322.

In other words, structure shown in Figure 34 will be inferred dead time according to exercising the result between certain interval interior station, and this inferred results is applied in the computing again of the scheme of travelling in the next interval, and among the embodiment shown in Figure 35, just can infer dead time in the driving process in same interval, and inferred results can be used in calculating process again according to exercising the result between the station.Like this, the place ahead forecasting type in the present embodiment exercise scheme again the operation result again of arithmetic element 317 just can reflect up-to-date train characteristic better than Figure 34.

Figure 36 represents the structured flowchart of the automatic train operation device 1 that embodiments of the invention 26 are related.Among this embodiment, computing is implemented to have added on the circuit 304B the place ahead forecasting type parking and is adopted unit 324 with temporarily travel the scheme arithmetic element 323 and the scheme of travelling when travelling between station shown in Figure 35.And, in the present embodiment, the scheme of travelling is divided into P1, P2, these 3 kinds of P3 according to the enforcement time of train, train 0 arrive given place before the target location the time engrave, the scheme of travelling adopts unit 324 will begin to adopt the scheme of the travelling P3 that is stopped and calculated with the scheme arithmetic element 323 of temporarily travelling by the place ahead forecasting type.Next will be elaborated to embodiment 26.

At first, like that travel scheme P1, P2, P3 are defined according to following.

P1: train 1 in the station when parking by the scheme of the travelling worked-out the best of arithmetic element 314 (perhaps 307,316) scheme of travelling again.

P2: in travelling between train 1 station by the scheme of travelling arithmetic element 317 (perhaps 308,311,315,318, the 319) computing again of carrying out drawing after the computing again scheme of travelling again.

P3: in travelling between train 0 station and train 0 arrive after the moment that N rice (for example N=300 (m)) is located before the target location by the parking of the place ahead forecasting type with the scheme arithmetic element 323 worked-out parkings of temporarily travelling with the scheme of temporarily travelling.

The scheme of temporarily travelling arithmetic element 323 arrives before the target location behind N rice at train 0, works up according to period demand (for example, the detection sample period of speed detector 2) and stops with the scheme P3 that temporarily travels.Work out this parking when temporarily travelling scheme P3, the train detection speed that has engraved when having utilized this, train detection position (etc. condition) are Yi Bian and the parking movement of prediction train when considering dead time on the train working direction.As the parking movement here, for example, begin promptly to plan to work up to take to brake the basic movement of the parking of stopping in this case from current time, and can utilize in given braking gear positions.And the prediction of train driving movement can be adopted the physical model of following formula (25).

F-Fr＝M·α ......(25)

F: power travel tractive force or braking force

Fr: train resistance (resistance to motion, grade resistance, curve resistance, tunnel resistance etc.)

M: train quality

α: acceleration/accel or deceleration/decel

The resistance that train impedance Fr produces when being train driving for convenience of calculation, often is considered to it by resistance to motion, grade resistance, curve resistance, tunnel resistance etc. as described above and forms.Like this, train resistance Fr can obtain according to following formula (26).

Fr＝Frg+Fra+Frc+Frt ......(26)

(26) each Resistance Value in the formula then uses the data that are kept in the data bank 300, can ask (with reference to " running theoretical (direct current AC electric locomotive) ", friend-making society compiles) according to following resistance type (27)-(30).

The grade resistance formula

Frg＝s ......(27)

Frg: grade resistance (kgf/ton)

S: the gradient (‰) (upwards for just, downwards for negative)

The resistance to motion formula

Fra=A+Bv+Cv ²(quadratic power of v) ... (28)

Fra: resistance to motion (kgf/ton)

A, B, C: coefficient

V: speed (km/h)

The curve resistance formula

Frc＝800/r ......(29)

Frc: curve resistance (kgf/ton)

R: radius of curve (m)

The tunnel resistance type (because the tunnel resistance can cause significantly variation because of tunnel cross sectional shape, size and train speed etc., so, for simplicity, will use following value)

Frt=2 (single track tunnel)

Perhaps

=1 (multiple track tunnel) ... (30)

Frt: tunnel resistance (kgf/ton)

According to the physical model in the above-mentioned formula (25), the scheme of temporarily travelling arithmetic element 323 temporarily the travel formulation of scheme P3 of usefulness that before (train) arrives the distance objective position, begins after N rice to repeat to stop.By the formulation that repeats of this scheme, stopping will be approaching to the target location gradually with the stop position among the scheme P3 that temporarily travels.Figure 39 just represents this situation.In addition, using the value apart from N of temporarily travelling between the scheme computing starting position from the target location to the train is to determine according to " operating range " ± formula such as " surplus distances ".

Next, the principle of work of the scheme of the travelling employing unit 324 among Figure 36 is described with reference to the flow chart among Figure 40.In this flow chart, expression is worked out P1, P2, the P3 scheme of travelling in any one according to period demand or when carrying out computing again, the processing sequence in certain one-period.

At first, whether the scheme of travelling adopts unit 324 will judge current train 0 motoring condition or travel to be in when stopping in the station constantly or between just having dispatched a car soon or stood in the station the driving process or being in target parking position puts near in any situation (step 1).If judged result is " just dispatching a car soon when stopping in standing or from the station ", just adopt dead time consideration type the best scheme generation unit 314 worked-out the bests scheme P1 (step 2) that travels that travels.Afterwards, the scheme of travelling adopts unit 324 that this best scheme P1 that travels is outputed to control command extraction unit 309.The action that control command extraction unit 309 is imported after the scheme of travelling illustrated in the above-described embodiments, was not therefore giving unnecessary details here.

If the judged result in the step 1 is " when travelling between standing ", the scheme of travelling adopt unit 324 will judge this week the interim scheme of travelling again computing whether implement (step 3).If computing is implemented again, the place ahead forecasting type travel scheme again arithmetic element 317 will adopt the scheme P2 (step 4) of travelling of computing again.

On the other hand, if in the step 3, this week is interim travel scheme again computing be not implemented, just judge 1 hour before, be whether to have adopted the best scheme P1 (step 5) of travelling in the previous cycle.If adopted the best scheme P1 that travels before 1 hour, the scheme of travelling adopts unit 324 will adopt the best scheme P1 (step 2) that travels.But, if do not adopt before 1 hour under the situation of the best scheme P1 that travels, illustrate that then current time is the moment after the best is travelled that scheme P1 has been used in the past and also carried out computing soon after that, therefore the scheme that was adopted before 1 hour is exactly the scheme of travelling after the computing again.Therefore, the scheme of travelling adopts unit 324 will adopt the scheme of the travelling (step 6) that adopted before 1 hour.

If, judged result is " target parking position put near " in the step 1, promptly being positioned at target parking position puts in the N rice, the parking that the scheme of travelling adopts unit 324 import to be worked out by the scheme arithmetic element 323 of temporarily travelling is with the scheme P3 that temporarily travels, and whether judgement stop position at this moment is in (step 7) within the scope of (target parking position is put) ± (permissible error).If stop position is within this scope, just adopts and stop with the scheme P3 (step 8) of temporarily travelling.But if not in this scope, then return step 5, adopted before 1 hour (perhaps from more early the moment) scheme of travelling of computing again, pass through step 1 once more after, the judgement in the repeating step 7 is till entry range.

As implied above, according to embodiment 26, can work up can be near target parking position be put (target parking position is put) ± (permissible error) parking of stopping with the scheme of temporarily travelling, therefore can parking more exactly on the target location.In addition, on the train working direction train worked out when movement is predicted and stop, can make the processing of dead time become simpler, and can work up the very simple automatic train operation device of mathematical algorithm at an easy rate with the scheme of temporarily travelling.In addition, in the present embodiment, stopping with the scheme arithmetic element 323 of temporarily travelling is that example illustrates with " the place ahead forecasting type ", and scheme arithmetic element 323 is not limited in the form of " the place ahead forecasting type " but this temporarily travels.

Yet up to the present the automatic train operation device of described each embodiment all is an imagination by adopt power travel gear and braking gear in common train, makes control command stage of development formula (discontinuous formula) variation pattern.But, might allow actuating device and brake equipment carry out work in the near future according to the successive control signal.Therefore, make control command when quickening become continuous tractive effort command or the power torque command that travels, carry out the best the travel formulation of scheme or the computing again of the scheme of travelling again, automatic running is reached improve comfort level and energy-conservation effect by bus greatly.In addition, make the control command when slowing down also become continuous braking force instruction, carry out the best the travel formulation of scheme or the computing again of the scheme of travelling again, automatic running is reached improve comfort level and energy-conservation effect by bus greatly.If can both realize above-mentioned successive control instruction when quickening and when slowing down, running will realize improving ride comfort level and energy-conservation effect biglyyer automatically.

Next embodiment 27 is described with reference to the accompanying drawings.Figure 41 represents the summary pie graph of the embodiment of the invention.

The speed of the train 0 in travelling and position will be according to the information of tachogenerator constant velocity test section 403 and the information of the ground sensor portion 404 that is used for ground sensor signals such as detected transmission device, carry out computing by velocity location operational part 405, and utilize train current data acquiring unit 412 to put it into train positioning to stop in the automatic control unit 410.In addition, though not shown in the figures,, with current braking gear or stop information such as target location and also put into train positioning and stop automatic control unit 410 by train current data acquiring unit 412.Stop the effect of automatic control unit 410 by train positioning, according to data such as the present speed that obtains through train current data acquiring unit 412, current location, current braking gear and be kept at the dead time of the deceleration/decel of respectively braking gear in the braking characteristics data store 411, braking gear switch and the braking characteristics data of response lag time etc., work up and to realize the deceleration control scheme that the target location stops by the combination between several gears by deceleration control scheme generation unit 413.

So-called deceleration control scheme is meant to be calculated so that can stop on the given position by the gear combination the time distribution of each braking gear, for example under the situation of 2 gear combinations, at first keep the 1st braking gear at the specified time of obtaining according to time Distribution Calculation method, switch to the 2nd braking gear then, and keep this braking gear till stopping.The simple example of deceleration control scheme as shown in figure 42.This example is represented the deceleration control scheme that begins from 10 meters of distance objectives, if Distance Remaining be 6 meters near the switching gear, just can stop at the target stop position.The time distribution can be obtained by following method: for example at present speed and Distance Remaining, temporarily make 2 braking gears that in scheme, use, and the holding time of the 1st braking gear is set as variable, obtain 1st holding time of braking gear operating range and that with Distance Remaining equate of operating range when under the 2nd braking gear, slowing down when under the 1st braking gear, slowing down.What if there is no satisfy condition separates, and just changes the combination of 2 gears, repeats same calculating.When asking the integrating of operating range, can work up the scheme of braking response characteristic when having considered gear switch by following method: in the gear switch dead time behind braking gear output command, slow down with the deceleration/decel on the braking gear before switching, response lag behind the process dead time is in the time, from switching the deceleration/decel variation of the deceleration/decel on the braking gear after the switching gradually on the braking gear before, just slow down after the time through response lag, and on the basis that this a series of actions changes, operating range is calculated with the deceleration/decel on the braking gear after switching.Each deceleration value of braking on the gear reaches when normally stablizing, and carries out gear switch according to above-mentioned such worked-out scheme, carries out gear switch continually but also can stop on the given position with regard to not only avoiding.In addition, when working out a scheme, set the 1st braking gear for high deceleration/decel shelves, set the 2nd braking gear for hang down deceleration/decel shelves, so just can stop, thereby improve comfort level by bus with low grade.

When each brakes the deceleration/decel change of gear, for example, through the 1st the braking gear (gear that deceleration/decel is high) holding time in (moment of handover scheme), the speed that dopes when relatively the deceleration/decel that uses by scheme slows down and the speed of true train, if actual speed is little, be that deceleration/decel is littler than what suppose, do not switch to the 2nd braking gear (gear that deceleration/decel is low) so immediately, but prolong the 1st holding time of braking gear, to prevent to surpass the target stop position.The change of Figure 43 representation scheme switching instant is to the example of the adjustment of stop position.In this legend, because actual deceleration/decel is littler than what suppose, making slows down slows down, and therefore will switch to the situation of the low gear of deceleration/decel in the original scheme in the 5m vicinity, changes at the 3.2m place and switching, and adjust stop position.This by switching instant change to the adjustment process of stop position with flowcharting shown in Figure 44.

Prolong under the situation of holding time, for example, the true train speed that engraves during according to handover scheme is inferred actual train deceleration degree, to slow down with the deceleration/decel of inferring, try to achieve by the way of computing again since the 1st braking gear instruction deceleration control scheme constantly, perhaps, the scheme that the scheme switching instant begins is carried out computing again according to the deceleration/decel of inferring.When working out initial deceleration control scheme, use predefined maximum deceleration, so actual deceleration degree situation to situation bigger than normal less than normal can be adjusted stop position by postponing the gear switch time method.

Figure 45 represents the basic block diagram in the embodiment of the invention 28.The deceleration/decel deduction unit 416 of inferring deceleration/decel according to train speed time series data in slowing down has been installed in this example, and other structures are identical with embodiment 27, and basic function is also identical.

Utilizing deceleration/decel to infer when deceleration/decels are inferred in unit 416 can try to achieve because the effect of the given deceleration/decel on this gear causes the situation of speed reduction in the specified time by after the dead time of for example gear switch, response lag effluxion.If the data of train speed have very big deviation, preferably can perhaps take the way of suitable filtering to remove noise by obtaining the moving average of speed, infer deceleration/decel according to the data of handling again.Utilize deceleration/decel to infer the deceleration/decel that unit 416 engraves when inferring this, and deceleration control scheme is one by one carried out revisal according to the deceleration/decel of inferring, the deceleration/decel of so respectively braking on the gear is just corresponding with the variation that produces because of time or speed during 1 time is travelled, and guarantees to stop precision.

Figure 46 represents the basic block diagram of embodiments of the invention 29.Be equipped with in this example by to undertaken by the deceleration control scheme under the deceleration situation each constantly or each locational predetermined speed compare with actual train speed, the deceleration/decel that uses in the deceleration control scheme is carried out the deceleration/decel scheme correcting unit 417 of revisal according to difference.Structure in addition is identical with embodiment 27, and basic function is also identical.

When slowing down by the deceleration control scheme each constantly or the predetermined speed in each position be distribute the braking gear that in scheme, uses and time calculate respectively after, calculate again according to the dead time of the deceleration/decel on the braking gear that uses in current train speed, the scheme, gear switch and response lag time etc.Predetermined speed will keep the value in the stopped process of beginning from scheme with arrangement mode, and can carry out reference to each value of arranging one by one, but when the calculator memory capacity of control usefulness is restricted, also can calculate one by one since the deceleration/decel on train speed in the time last time step and the braking gear at that time.This time predetermined speed of engraving compare with actual train speed, if train speed is low, actual deceleration/decel will be bigger than the deceleration value of using in the scheme, therefore deceleration/decel should be improved, and the deceleration control scheme is carried out computing again.On the contrary, if the train speed height, will be little compared with the deceleration/decel of the deceleration value reality in the scheme, therefore deceleration/decel should be reduced, the deceleration control scheme is carried out computing again.When changing deceleration/decel, for example to set the error permissible value of the train speed of predetermined speed and reality, according to the change amount that reaches the required time quantum decision deceleration/decel of error permissible value.Utilize deceleration/decel scheme correcting unit 417 one by one predetermined speed and true train speed to be compared also revisal deceleration/decel, so just can change and often suitably upgrade the deceleration control scheme at the timeliness of deceleration/decel.Owing to have deviation in the actual train speed data, preferably use filtered data, set bound of deceleration/decel variable quantity or the like, take to prevent the processing of dispersing.

Claims (8)

1. a train positioning stops automatic control device, and train is stopped on given position automatically, it is characterized in that comprising:

Being used for train car respectively brakes the braking characteristics data store of braking characteristics data such as the dead time of deceleration/decel on the gear, braking gear switch and response lag time;

Obtain the train current data acquiring unit of data such as train present speed, current location, current braking gear;

According to braking characteristics data of storing in the braking characteristics data store and the train current data that obtains by train current data acquiring unit, work up by using a plurality of braking gears to make train stop at the deceleration control scheme generation unit of the deceleration control scheme on the given position;

Deceleration control instruction fetch unit from the deceleration control instruction that engraves when extracting each the worked-out deceleration control scheme of deceleration control scheme generation unit;

The deceleration control instruction output unit of brake equipment is exported in the deceleration control instruction that the deceleration control instruction fetch unit is extracted.

2. train positioning according to claim 1 stops automatic control device, it is characterized in that, computing is carried out in time distribution on each braking gear, so that can stop at by the combination of a plurality of braking gears on the given position, constitute the deceleration control scheme by braking gear that uses and the switching instant of braking on the gear.

3. train positioning according to claim 2 stops automatic control device, it is characterized in that, in the deceleration control scheme, slows down with the high braking gear of deceleration/decel earlier, and then switches to the low braking gear of deceleration/decel.

4. train positioning according to claim 2 stops automatic control device, it is characterized in that, the predetermined speed of switching instant and the true train speed of switching instant compare when slowing down by the deceleration control scheme, simultaneously the deceleration control scheme are not changed at both.

5. train positioning according to claim 1 stops automatic control device, it is characterized in that, after the deceleration control scheme working-out goes out, when the value of using when working out a scheme when deceleration value has changed, the deceleration control scheme is changed.

6. train positioning according to claim 1 stops automatic control device, it is characterized in that, also comprise the deceleration/decel deduction unit of inferring deceleration/decel according to the time series data of the train speed in slowing down, and work out the deceleration control scheme according to the deceleration/decel of inferring.

7. train positioning according to claim 1 stops automatic control device, it is characterized in that, also comprise when slowing down by the deceleration control scheme each constantly or each locational predetermined speed and true train speed compare, and, the deceleration control scheme is changed according to the deceleration/decel compensating value that calculates by deceleration/decel scheme correcting unit at the deceleration/decel scheme correcting unit of the deceleration value of using in its difference revisal deceleration control scheme.

8. train positioning according to claim 4 stops automatic control device, it is characterized in that, deceleration/decel, gear switch dead time and the response lag time of having used according to the rapid speed of a preceding time step with when working out a scheme, when slowing down according to the deceleration control scheme each constantly or each locational predetermined speed calculate one by one.

Images