AU2011229236A1

AU2011229236A1 - Method and device for train length detection

Info

Publication number: AU2011229236A1
Application number: AU2011229236A
Authority: AU
Inventors: Walter Schlosser; Christoph Strasser; Gotz Wiedmann
Original assignee: Knorr Bremse Systeme fuer Nutzfahrzeuge GmbH; Knorr Bremse Systeme fuer Schienenfahrzeuge GmbH
Current assignee: Knorr Bremse Systeme fuer Schienenfahrzeuge GmbH
Priority date: 2010-03-18
Filing date: 2011-03-16
Publication date: 2012-10-11
Anticipated expiration: 2031-03-16
Also published as: RU2561481C2; EP2547568A1; ES2586578T3; AU2011229236B2; CN102822032B; EP2547568B1; RU2012144281A; WO2011113856A1; DE102010011949A1; CN102822032A

Abstract

The invention relates to a method and a device for train length detection in a train set which is composed of a plurality of carriages (1a - 1c) and which, by means of a pneumatic braking system, is braked in a plurality of braking stages according to the pressure in a main air line (HL) coupled from carriage (1a) to carriage (1c), the pressure (p) and throughflow (formula (I)) of said main air line and the ambient temperature (T) being detected along the time axis by sensors, and from these the train length (L) being calculated by means of an electronic evaluation unit (4), wherein the measurement variables are detected by sensors, proceeding from the steady state of a present braking stage (I.), during the execution of the next braking stage (II.) until a steady state is again reached. The volume (V) of the main air line (HL) is calculated by additive integration of the throughflow (formula (I)) during the ventilation of the main air line (HL) for the execution of the next braking state (II.), taking into consideration the pressure (p) prevailing in the initial state and the end state and the ambient temperature (T), in order to determine the train length (L) corresponding to the main air line length at a known line cross section (Q) using said volume.

Description

WO 2011/113856 PCT/EP2011/053950 Method and Device for Train Length Detection 5 The present invention relates to a method and a device for train length detection in a train set which is composed of a plurality of cars and which, by means of a pneumatic brake system, is braked in a plurality of 10 braking stages according to the pressure in a main air line HL coupled from car to car, the pressure PHL and throughflow 9 of said main air line and the ambient temperature T being detected by sensors along the time axis, on the basis of which the train length L is 15 finally calculated by means of an electronic evaluation unit. Furthermore, the invention also relates to a device which implements the method, and to a train set in which such a device is installed. 20 The main air line HL is used in train sets primarily to trigger the pneumatically operated brakes which, in terms of signal transmission, move into the braking position by reducing the pressure and are released when the pressure rises. The main air line HL, which runs 25 along all the cars of a train set, can also be used to acquire information about train-specific properties. It is therefore possible to monitor the main air line HL in terms of a train separation. In this context, the subsequently fed-in volume flow and the pressure 30 conditions are monitored during travel with the brakes released, during braking and during the releasing process. The basis for the detection of train separations or for detecting the length of the main air line HL and therefore of the entire train set are 35 characteristic properties of the brake system, such as, for example, the maximum subsequent supply to the main air line HL on the basis of the maximum leakage of the system and the typical length-dependent lag time in WO 2011/113856 - 2 - PCT/EP2011/053950 which a change in the pressure conditions can be detected. This and other characteristic properties of a pneumatic 5 brake system are preferably based on standardized definitions of the main air line HL in order to permit generally valid applicability. Through derivation from these properties, threshold values and gradients are defined for the throughflow values and pressure values 10 which permit conclusions about the train length or the continuity of the main air line HL to be drawn through signal processing. If it is established, for example, that the main air line HL is not continuous, a disruptive cause of this may be determined to be a 15 closed check valve within the main air line HL between two cars. DE 199 02 777 Al discloses a technical solution for -monitoring the completeness of a train, in which 20 solution a message about the state of the train set is output by means of a compressed air sensor and a throughflow meter in order to determine the volume flow in the main air line HL. The main air line of the train set usually runs through all the connected cars and can 25 be monitored by sensor, for example at the relay valve on the tractive unit, wherein the direction and quantity of the volume flow of compressed air are measured by sensors which are known per se. Overall, in the steady state of the brake system there is an 30 equilibrium between the inflow and outflow of the quantity of air. The inflowing compressed air replaces here merely the air which flows out of the brake system due to leaks, which air escapes over the entire length of the main air line HL. If the brakes are applied, the 35 air pressure in the main air line HL drops in a defined fashion in usually a plurality of braking stages.

WO 2011/113856 - 3 - PCT/EP2011/053950 In order to monitor the completeness of a train, the measured values of the sensors which monitor the main air line HL are fed to an electronic evaluation unit which compares the detected measured values with 5 predetermined values of the respective operating variables for a corresponding operating state of the train set. The completeness of the train is determined as a function of the comparison result. In this context, the evaluation and acquisition of the measured 10 values for determining the train completeness information takes place only at a single location in the train set, preferably in the tractive unit of the train, with the result that further devices for detecting operating variables of the main air line HL 15 are not necessary at other locations of the train set, in particular at the rear of the train. However, this monitoring of the completeness of a train has the disadvantage that in this way it is not 20 simultaneously possible to determine precisely the length of the train. The knowledge of the length of the train is useful, for example, for detecting what are referred to as "black cars". The arrangement of the cars in series and the properties are generally known 25 by reference to a car list. The significant information for the tractive unit driver, such as the braking properties, are compiled on what is referred to as a brake slip, derived from the car list. In addition, the train length on frequently travelled route sections 30 during the driving mode is important, for example in order to be able to comply with safety distances. DE 199 33 798 Al discloses a method for train length detection in which the length of the train is measured 35 directly and transmitted to the tractive unit. For this purpose, volumes and pressure signals in the main air line HL are determined by sensors, wherein, in particular, transfer of information in close to real- WO 2011/113856 - 4 - PCT/EP2011/053950 time conditions occurs to the tractive unit via the last car of the train set. Subsequently, an evaluation device checks whether the volume signals and pressure signals and physical variables derived therefrom 5 correspond to a known setpoint value range, stored in the evaluation unit, for the train length. As a function of this, a signal is output which supplies the information as to whether the measured values are within the stored setpoint value ranges. In addition it 10 is proposed to determine the length, to be stored in the evaluation unit of the tractive unit, of the train set to be measured by a train length measuring device and to transfer said length to the tractive unit. In addition, the train length can also be measured by an 15 axle counter during starting or when a station is exited, and transferred to the tractive unit. As a result, a fixed measuring device on the track for measuring the train length is activated here. 20 All these measures appear to be extremely costly since sensors located outside the tractive unit, specifically in the last car or even outside the train set, are used to acquire measured values in order to detect the length of the train. 25 DE 100 09 324 Al discloses, in contrast, a method for power-train-based determination of the train length of a train set, in which solely the physical state variables of the pressure, throughflow and temperature 30 of the air in the main air line HL in the region of the tractive unit are measured, and wherein changes in pressure in the main air line HL are generated from a defined sequence of via the driver's brake valve in the tractive unit or other suitable actuators, the 35 associated flows are integrated over time and the leakage rates are determined during pressure which is kept constant, that is to say the fixed state thereof, and the volume of the main air line HL is calculated WO 2011/113856 - 5 - PCT/EP2011/053950 from these variables, on which basis the train length can be inferred. Although this calculation method takes into account the 5 leakage of the brake system which is present owing to the system, other interference variables such as, for example, local venting in the region of the control valves assigned to the individual brake cylinders of the cars are not taken into account during the 10 acceleration of said cars. This is because the control valves ensure that there is temporary additional venting of the main air line HL in the first braking stage for the purpose of braking acceleration. 15 However, this measure gives rise to inaccurate measurement results during the determination of the train length. The object of the present invention is therefore to 20 provide a method and a device for train length detection in which precise determination of length is possible solely with a sensor system internal to the train unit. 25 The object is achieved on the basis of a method according to the preamble of claim 1 in conjunction with the characterizing features thereof. The following dependent claims present advantageous developments of the invention. Reference is made to claim 9 for a 30 device which corresponds to the method. Claim 11 specifies a train set which contains this device. The invention includes the solution that the detection of a measured variable by sensors is only carried out 35 starting from the steady state of an existing braking stage I. during the execution of the following braking stage II. until a steady state is reached again within this braking stage II. Subsequent integration of the WO 2011/113856 - 6 - PCT/EP2011/053950 throughflow during the venting of the main air line HL in order to execute the following braking stage II., the volume of the main air line HL is calculated taking into account the pressure prevailing in the initial 5 state and the final state, as well as the ambient temperature. Finally, from the volume calculated in this way it is possible, given a known line cross section of the main air line HL, to determine the length of said main air line HL in a manner known per 10 se, and therefore to determine the train length L. The volume can be determined concretely by means of the following formulaic relationship: f*dt y V,- *dt " I dp* dr| . |p(t2 )- P( 1 )| 15 Finally, the line length and therefore the train length L of the train set is obtained from L = . The cross section of the main air line HL and of the couplings is 20 known generally. With the method according to the invention it is therefore possible to check the line length in the case of any braking request which does not occur from the 25 released state. As a result, the continuity of the main air line HL can be checked and a closed cut-off faucet can be detected. If the determination of the line length is integrated into the braking test before the start of the journey, the system can output a warning 30 about a deviating train length compared to the specifications on the brake slip. If, for example, after the braking test, further cars are coupled to the train set with a closed cut-off faucet, this fault is WO 2011/113856 - 7 - PCT/EP2011/053950 detected at the other end with respect to the change in direction when the tractive unit is connected. In order to detect the correct volume flow, the leakage 5 must also be considered in the method presented above. When the brakes are applied starting from an existing braking stage, the main air line HL is vented completely via the driver's brake system with the exception of the leak, and is therefore detected by the 10 throughflow measurement. The leakage rate must additionally be added to the volume flow. The following relationship applies: VNoverall : VNMeas + VNLeakage 15 The leakage rate is dependent on the pressure level in the main air line HL. For the purpose of calculation, the leakage is considered as a nozzle in the main air line HL with a constant nozzle cross section which 20 vents to the atmosphere. From the volume flow Y = A*(2*R*T) 0 o 5 *Y and the relationship VN = r, the following expression is obtained with the throughflow coefficient Y: 25 [III] VN = A* *(2*287*T)0-5*60*10- 3 *Y*pl [-] where A is in mm2 where TN = 293.15K and pN = 1.013 barA, where the 30 following applies: = *, if 2- > 0.528 where k=1.402, otherwise Y=0.484. k--1 p pA WO 2011/113856 - 8 - PCT/EP2011/053950 pl corresponds here to the absolute pressure upstream of the nozzle, P2 corresponds to the absolute pressure downstream of the nozzle, and T corresponds to the temperature. R = 287 1 is the universal gas constant. kg ocK 5 After the leakage has been determined for a constant pressure level, the constant nozzle cross section A can therefore be determined as a function of the temperature by means of the formal relationship [III] . 10 VNLeakage can therefore be approximately calculated from the measured pressure profile in the main air line HL. The advantage of the solution according to the invention results, in particular, from the measure that 15 the air flowing into the main air line HL during the initial charging of the brake system can be disregarded. This is because during the initial charging the air flows not only into the main air line HL but also into the working chambers of the control 20 valves and into various reservoir containers of the cars. In this context, the volume of the reservoir containers of the individual cars can vary, with the result that in practice this interference variable cannot be corrected by computational means. In 25 addition, the initial state of the working chamber of the control valves and of the reservoir containers is usually not known. The solution according to the invention completely excludes the measuring errors which result from this. In order to solve this problem, 30 the solution according to the invention provides in principle for the throughflow rate of the air to be detected only in the charged state, for example after the initial charging during travel or in any desired steady state of the main air line HL in which the 35 pressure PHL is constant. By excluding the acceleration effect, the solution according to the invention avoids an unknown throughflow variable, which gives rise to a more accurate measurement result.

WO 2011/113856 - 9 - PCT/EP2011/053950 According to a measure which improves the invention it is proposed that in order to determine the train length during the ventilation of the main air line HL, the 5 detection of the measured variable by sensors is carried out only until the pressure of the reservoir air container connected to the main air line is reached. According to the inventive method it is therefore possible, during such a single-line operating 10 mode with contemporary brake configurations, also to carry out the evaluation of the charging process of the main air line up to the point when the subsequent supply to the reservoir container starts. The throughflow is evaluated here up to a pressure value 15 below the reservoir container pressure. During this process, the unknown size of the reservoir container is advantageously excluded. According to a measure which further improves the 20 invention in terms of an accurate measurement result, it is proposed that in the case of the line cross section which is used with the determined volume of the main air line HL to calculate the train length, both the cross section of the main air line HL which runs 25 through the individual cars and the cross section of the line couplings arranged therebetween are taken into account. The train length L is obtained, as stated above, by dividing the determined volume of the main air line HL by the line cross section Q. 30 In order to compensate for the leakage as a further interference variable within the brake system by computational means, it is proposed that in addition the volume flow which is brought about within the main 35 air line HL in the steady state as a result is measured, with the result that this measured variable can be used to eliminate the interference variable as a correction value by computational means during the WO 2011/113856 - 10 - PCT/EP2011/053950 determination of the train length. The throughflow coefficient Y which is necessary for the calculation can be defined in a simplified fashion in the range 0.45 to 0.5 if the pressure ratio p2 to pl is greater 5 than the value 0.528 +/- 10%. Even in the case of a ratio which is greater than this value, the error remains relatively low since as the pressure in the main air line HL decreases, the leakage also decreases. If the leakage of the pneumatic brake system is 10 calculated or defined, it is possible to achieve a qualitatively better result by including it in the calculation of the train length. According to another measure which improves the 15 invention it is proposed that when the volume of the main air line HL has previously been determined at least once as a correction value, the air volume which is lost as a result of braking acceleration losses of the individual control valves assigned in the brake 20 cylinders from the released state of the brake system is eliminated by computational means during the determination of the train length. The method according to the invention is based on the fact that the acceleration effect of the control valves of the brake 25 system is eliminated from the determination of the length of the main air line HL. However, if the line volume is determined once, for example in the course of a braking test before the train set departs, the resulting error can be determined for a volume which is 30 now known. The background for this is that during the travel of the train it is also possible to check the train length during braking, even in the case of braking requests from the released state, in order to detect the tearing off of part of the train or a closed 35 cut-off faucet. In order to carry out this measure, at least a pressure of approximately 0.1 bar should preferably be vented via a nozzle until the acceleration effect is triggered in the individual WO 2011/113856 - 11 - PCT/EP2011/053950 control valves. The acceleration effect then locally extracts a pressure of approximately 0.3 bar from the main air line HL. The effect is then terminated and the control valves are absolutely sensitive. By means of 5 this relationship, an approximate calculation of the quantity of air which is lost as a result is then possible by means of the ideal Glass equation. The following applies: 10 Pbef ore * Vbefore "" Pafter * Vafter According to another measure which develops the invention, the determination of the volume of the main air line HL can also be carried out in what is referred 15 to as the two-line operating mode. In the two-line operating mode, reservoir containers which can vary in their size and further compressed air consumers are charged via a separate compressed air line, the main container line HB. The main container line HB runs 20 along the train set parallel to the main air line HL. As a result the method which is the subject matter of the invention can be applied in the two-line operating mode even in the case of ventilation after initial charging because there are no unknown volume sizes. In 25 other words, the volume of the main air line HL is therefore determined during the release of the brakes as a result of ventilation of the main air line HL. The charging process of the main air line HL can be 30 evaluated between any two steady states by means of the method according to the invention in order to determine the train length. Since no acceleration effect occurs during the ventilation process, all that is necessary is to also take into account the leakage as an 35 interference factor. In contrast to the determination of the volume by means of the venting, in the two-line operating mode the leakage has to be subtracted from WO 2011/113856 - 12 - PCT/EP2011/053950 the measured volume flow as a function of the pressure. The following is therefore obtained: VNtotal = VNmeas ~ VNleakage 5 The determination of the volume by means of the releasing process and charging process in the two-line operating mode is possible without faults only when the air has been extracted from the main air line HL only 10 via the driver's brake system and not by other devices. Further measures which improve the invention are presented below in more detail together with the description of a preferred exemplary embodiment of the 15 invention with reference to the figures, of which: Figure 1 is a schematic illustration of a train set which is composed of a plurality of cars, with a device for train length detection 20 via the main air line, and Figure 2 shows a flowchart illustrating the individual method steps for train line detection. 25 According to figure 1, a train set is composed of a large number of cars la to lc which are connected in series. A pneumatic brake system brakes the train set according to the pressure in a main air line HL, 30 coupled from car la to car lb and finally car lc, in one or more braking stages as far as the stationary state. In the process, sensors 2a to 2c monitor the pressure PHL, the throughflow 1 and the ambient temperature T within the main air line HL along the 35 time axis. These sensors 2a to 2c are arranged here in a tractive unit 3 which is connected in front of the cars la to 1c. An electronic evaluation unit 4, which collects the measured sensor signals and ultimately WO 2011/113856 - 13 - PCT/EP2011/053950 calculates the train length, is also positioned in the tractive unit 3. In this exemplary embodiment, two separate sensors 2b 5 and 2b' for determining the throughflow V are provided within the scope of the sensor equipment. While the first sensor 2b is used at the changeover between the steady states, that is to say at the transition from one braking stage to the next higher braking stage, the 10 second sensor 2b' is used only for the purpose of leakage measurement in the steady state. Since the changeover between the steady states generates a substantially higher throughflow V, the first sensor 2b is given larger dimensions than the second sensor 2b', 15 which has to determine only very small throughflows f' in comparison. As a result of the different measuring ranges which are used due to this, the accuracy of the determination of the throughflows 19 is improved overall. However, in this context it is necessary to 20 distinguish between a single-line operating mode and a two-line operating mode. In the two-line operating mode it is possible for one sensor to be completely sufficient in the case of ventilation, which sensor measures both leakage and ventilation processes, or two 25 sensors with the same cross section in series. The leakage sensor requires here a smaller measuring range and can therefore achieve a higher level of accuracy. In the single-line operating mode with leakage 30 measurement, it is absolutely necessary to have two sensors or a device which permits bidirectional measurement since the throughflow when the brakes are applied is opposed to the throughflow when the leakage is measured. It is also the case here that the cross 35 section of the main air line HL must not be constricted and the leakage sensor requires a smaller measuring range, and therefore a higher level of accuracy can be achieved.

WO 2011/113856 - 14 - PCT/EP2011/053950 The electronic evaluation unit 4 takes into account, with respect to the line cross section during the determination of the train length, both the cross 5 section of the main air line HL running through the individual cars la to ic and the cross section of the line couplings 5 arranged therebetween, in order to achieve more accurate computational results. 10 At least one control valve 6 with a pneumatic brake cylinder 7 connected thereto for activating the brakes is arranged in each individual car la to 1c. For the purpose of braking acceleration, air volume 15 also escapes from the control valves 6, and can be detected as a correction value, in order to take this into account computationally in the determination of the train length. 20 According to figure 2, the train length detection is preferably carried out by starting from a steady state of the brake system, which occurs as a result of the braking stage I. being present. At first, a measured variable detection by sensors of the physical values 25 pressure PHL, throughflow V of the main air line HL and of the ambient temperature T takes place, specifically during the execution of the following braking stage II. until a steady state occurs again. 30 Subsequently, integration of the throughflow 9 which is obtained in this way takes place by taking into account the initial state and end state of the prevailing pressure PHL as well as of the ambient temperature T according to the equation [I] given above. The volume V 35 of the main air line HL is obtained as the computation result. Given a known line section Q of the main air line HL, the length of said airline HL, which corresponds to the train length L, is calculated from WO 2011/113856 - 15 - PCT/EP2011/053950 this volume V by the computational relationship also given above. The invention is not restricted to the preferred 5 exemplary embodiment described above. Instead, refinements thereof, which are also included in the scope of protection of the following claims, are also conceivable. It is also possible to determine further influencing interference variables and take them into 10 account computationally as correction values, so that a precise train length detection can be implemented.

WO 2011/113856 PCT/EP2011/053950 List of Reference Symbols 1 Car 2 Sensor 3 Tractive unit 4 Evaluation unit 5 Line coupling 6 Control valve 7 Brake cylinder 8 Reservoir air container HL Main air line HB Main container line V Volume of the main air line Q Line cross section of the main air line PHL Pressure in the main air line V Throughflow through the main air line T Ambient temperature L Train length

Claims

1. A method for train length detection in a train set which is composed of a plurality of cars (la-ic) and which, by means of a pneumatic brake system, is braked in a plurality of braking stages according to the 10 pressure in a main air line (HL) coupled from car (la) to car (1c), the pressure (PHL) and throughflow (P) of said main air line and the ambient temperature (T) being detected by sensors along the time axis, on the basis of which the train length (L) is calculated by 15 means of an electronic evaluation unit (4), characterized in that the detection of a measured variable by sensors is carried out starting from the steady state of an existing braking stage (I.) during the execution of the following braking stage (II.) 20 until a steady state is reached again, after which the volume (V) of the main air line (HL) is calculated by integrating the throughflow (V) during the venting of the main air line (HL) in order to execute the following braking stage (II.) while taking into account 25 the pressure (PHL) prevailing in the initial state and final state as well as the ambient temperature (T) , in order to determine from this the train length (L) corresponding to the main air line length, given a known line cross section (Q). 30

2. The method as claimed in claim 1, characterized in that, in order to determine the train length (L) during the ventilation of the main air line (HL), the detection of the measured variables by 35 sensors is carried out only until the pressure (PHL) Of the reservoir air container (8) connected to the main air line (HL) is reached. WO 2011/113856 - 18 - PCT/EP2011/053950

3. The method as claimed in claim 1, characterized in that in the case of the line cross section (Q) both the cross section of the main air line (HL) which runs through the individual cars (la-1c) and 5 the cross section of the line couplings (5) arranged therebetween are taken into account.

4. The method as claimed in claim 1, characterized in that in the steady state the volume 10 flow (VNleck) which is brought about through leakage of the pneumatic brake system is measured in order to use this measured variable to eliminate, by computational means, the interference valve as a correction valve during the determination of the train length (L). 15

5. The method as claimed in claim 1, characterized in that, when the volume (V) of the main air line (HL) has previously been determined at least once as a correction value, the air volume (V) which is 20 lost as a result of braking acceleration losses of the individual control valves (6) assigned to the brake cylinders (7) from the released state of the brake system is eliminated by computational means during the determination of the train length (L). 25

6. The method as claimed in claim 1, characterized in that in the case of a two-line operating mode in which the compressed air consumers are filled via a separate main container line (HB), 30 while the main air line (HL) serves exclusively for braking, the volume (V) of the main air line (HL) is determined during the releasing of the brakes as a result of ventilation of the main air line (HL). 35

7. The method as claimed in claim 6, characterized in that the correction value which represents the leakage of the brake system is WO 2011/113856 - 19 - PCT/EP2011/053950 subtracted from the measured throughflow (V) as a function of the pressure (PHL)

8. A device for train length detection in a train set 5 which is composed of a plurality of cars (la-1c) and whose pneumatic brake system brakes in a plurality of braking stages according to the pressure in a main air line (HL) coupled from car (la) to car (lb) , wherein sensors (2a-2c) detect the pressure (PHL) and the 10 throughflow (97) as well as the ambient temperature (T) along the time axis, on the basis of which an electronic evaluation unit (4) calculates the train length (L), characterized in that the evaluation unit (4) carries 15 out the detection of the measured variables by sensors starting from the steady state of an existing braking stage (I.) during the execution of the following braking stage (II.) until a steady state is reached again, in order to calculate the volume (V) of the main 20 air line (HL) by integrating the throughflow (V) during the venting of the main air line (HL) in order to execute the following braking stage (II.) while taking into account the pressure (PHL) prevailing in the initial state and the final state as well as the 25 ambient temperature (T), in order to determine on this basis the train length (L) corresponding to the main air line length, given a known line cross section (Q).

9. The device as claimed in claim 8, 30 characterized in that the sensor (2b) is used to measure the throughflow (19) of the main air line (HL) during the changeover between the steady states, while a second sensor (2b'), which has smaller dimensions compared to the sensor (2b), is used to measure leakage 35 in a steady state.

10. A train set having a plurality of cars (la-ic), which can each be braked by a pneumatic brake system WO 2011/113856 - 20 - PCT/EP2011/053950 according to a coupled-through main air line (HL), comprising a device for train length detection as claimed in one of the preceding claims 8 and 9.