TW201418087A - An operation control method of rail train for saving electricity - Google Patents

Abstract

An operation control method of rail rain for saving electricity includes following steps: Predicting a time - speed curve between the train stations of a rail rain, based on the operating data of rail train; Using the time - speed curve to control a operating mode of the rail trains, the rail train operation providing a voltage of traction power supply, a linear parameter of a track, a resistance of the rail train and an adhesion. Adjusting the operating mode of the rail train the in the next moment according to the voltage of traction power supply, the linear parameter of the track, the resistance of the rail train and the adhesion, wherein the operating mode includes a value, a speed and an acceleration of the train.

Description

Railway train energy saving operation control method

The invention relates to a rail transit operation control technology, in particular to a rail train energy saving operation control method.

Nowadays, the cost of rail transit in each city is very high, not only the construction cost is very high, but also the operation and maintenance cost of the line is very large every year after the official operation and opening, especially the line power consumption cost is the most serious, resulting in the urban rail transit operating cost. High, this has become the most prominent problem in urban rail transit. Therefore, reducing the energy consumption of urban rail transit and reducing the total amount of electricity consumption has become the most effective way to reduce the operating cost of urban rail transit.

The main form of energy consumption in the operation of urban rail transit systems is energy consumption. According to the statistical analysis of the urban rail transit power load, energy consumption is mainly used in the following aspects: traction power supply, ventilation and air conditioning, escalators, lighting, water supply and drainage, weak current systems, etc., especially with traction power supply, the largest energy consumption, Nearly 50% of energy consumption comes from train traction. Therefore, one of the important ways to reduce the energy consumption of urban rail transit systems is to reduce the energy consumption of train traction. Train traction energy consumption is mainly used for train operation. Therefore, the goal of energy-saving operation of trains has become the most effective means to reduce train traction energy consumption.

The existing methods for energy-saving control of rail trains are to adjust the train speed in an off-line manner, but fail to propose an effective method for online implementation for saving energy, and also do not consider the problem of timetable adjustment and train adhesion.

An object of the present invention is to provide a method for saving operation control of a railway train.

In order to achieve the above object, the present invention provides a rail train energy saving operation control method, which comprises the following steps: predicting a train station and a train based on operational data of a train a time-speed curve running between stations; and controlling a running mode of the train by using a time-speed curve, having a traction power supply network voltage, a track linear parameter, a train resistance and a resistance during train operation Adhesion, according to the voltage of the traction power supply network, the linear parameters of the track, the resistance of the train and the adhesion force, adjust the running mode of the train in the next moment, wherein the operation mode includes the calculation value of a value, a speed and an acceleration of the train.

In one embodiment, the voltage of the traction power supply network is the voltage of a third rail.

In an embodiment, the operating mode of the control train utilizes a three-layer reverse transfer-like neural network to adjust the operating mode of the train in the next moment, wherein the three-layer inverted transmission-like neural network has an input layer and an output layer. And a hidden layer.

In an embodiment, the rail train saves power operation control method, and further comprises the steps of: taking the voltage of the traction power supply network, the track linear parameter, the resistance of the train and the adhesion force as an input layer of the three-layer inverted transmission type neural network. The value, and preset an output target value, calculate the hidden layer value of the three-layer inverse transfer type neural network according to the value of the input layer; calculate the output layer of the three-layer inverted transfer type neural network by using the value of the hidden layer The value of the output layer is the value, speed and acceleration of the operating mode; the value of the output layer is compared with the preset output target value, and the value of the output layer is adjusted so that the value of the output layer and the preset output An error value between the target values is the smallest.

In one embodiment, the operating mode is selected from the group consisting of an acceleration mode, a deceleration mode, a constant speed mode, and a idle mode.

In one embodiment, the track linear parameters include the slope of a track, the curvature of a track, and the speed limit of a track.

1‧‧‧ train

100‧‧‧ Train Traction System

110‧‧‧ traction power supply network

120‧‧‧Power-receiving device

130‧‧‧Direct AC converter

140‧‧‧Motor

150‧‧‧Transmission

160‧‧‧ Wheels

Figure 1 is a schematic diagram of the energy flow of the train traction system.

2 is a method for controlling energy saving operation of a railway train according to the present invention.

3 is a flow chart of a method for controlling power running operation of a railway train according to the present invention.

Figure 4 is a schematic diagram of a three-layer inverted transfer neural network.

The foregoing and other technical contents, features and effects of the present invention are as follows A detailed description of one of the preferred embodiments will be apparent from the detailed description. The directional terms mentioned in the following embodiments, such as upper, lower, left, right, front or rear, etc., are only used to refer to the directions of the accompanying drawings. Therefore, the directional terms are used for illustration only and are not intended to limit the invention.

First introduce several train models.

Train dynamics model

In urban rail transit, the train will be subjected to many forces with different directions and sizes during the operation. The force situation is more complicated. However, for the train energy-saving operation control, the longitudinal motion of the train on the track is mainly considered, so only Study the forces in the longitudinal direction of train operation. In this way, the external forces that the train receives during operation are mainly: train traction, train braking force, and train running resistance.

Train traction

Because the trains in urban rail transit are grouped by trains and trailers, the traction of trains comes from various trains. The essence of the motor is an energy conversion mechanism, which converts the electric energy provided by the traction net or the third rail into mechanical energy by the traction motor, and then transmits it to the moving wheel of the moving car, and the moving wheel passes the contact and friction with the rail. The force is generated on the rail, and the rail has a reaction force, that is, traction force, for the moving train to the train running direction, so that the train can run forward.

2. Train braking force

The train braking force in urban rail transit is mainly generated by the braking device, which is opposite to the running direction of the train on the track and can hinder the train movement, and belongs to the external force that the driver can control and adjust according to needs or by the automatic driving equipment. . Nowadays, most urban rail transit vehicle traction electric drive systems adopt advanced frequency-adjusting AC induction motor drive system. This system has good electric braking performance at high speed, but when the train is at low speed, the electric braking efficiency is better. Low, the braking effect is not good, for this reason, after the train speed is reduced to a certain extent, the air brake system must be used to brake the train. Therefore, the brakes of general trains have two types of electric brakes and air (friction) brakes.

3. Train running resistance

One of the trains produced during operation is opposite to its running direction, preventing it from The external force that runs and cannot be controlled by the driver is the resistance of the train running. It can be divided into basic resistance and additional resistance according to the formation reasons.

(1) Basic resistance: The resistance that exists when the train is in any operation (including the start-up phase).

There are many factors that cause the basic resistance of the train, mainly due to the friction and impact between the various parts of the vehicle, between the surface of the vehicle and the air, and between the wheel and the rail. However, in practical application, these factors are difficult to calculate by theoretical formula. Therefore, in order to simplify the calculation method, the empirical formula derived from a large number of experiments is usually used for calculation, which can be generally described as the unit basic resistance equal to the train running speed. The form of the quadratic formula, ie: w ₀ = a + bv + cv ² (N/kN) (1)

Where a, b, c are empirical constants related to the vehicle type; v is the train speed in m/s.

(2) Additional resistance: the resistance that exists when the train is operated under individual conditions.

The additional resistance is different from the basic resistance, which is less affected by the difference in vehicle type. It depends on the line conditions, including the additional resistance of the slope, the additional resistance of the curve, and the air resistance.

Additional resistance to the ramp: refers to the component of the train's gravity along the track when the train is running on the ramp.

When the train is on the uphill, the additional resistance of the ramp hinders the train from moving forward; otherwise, it helps the train to move forward. Unit trains ramp w _i is equal to the resistance of the train in which the value of the ramp slope mille i, _{i.e.: w i = i (N /} kN) (2)

Additional resistance to the curve: refers to the additional resistance of the track to the train when the train is in a curved orbit. The empirical formula for calculating the additional resistance w _r of the unit curve is: w _r = A / R (N/kN) (3)

In the formula, A is the constant determined by the test method, usually 450-800. According to China's Train Traction Calculation Procedure, A takes 600; R is the curve radius, the unit is m.

Total additional resistance: refers to the train running at the same time on the ramp, curve and air resistance. Usually the total unit additional resistance w _j is the sum of the above three.

It is known from physics that the state of an object is determined by the resultant force of the forces acting on it. For the convenience of calculation, the train is regarded as the translational movement of a mass point along the rail operation, that is, the train traction force F , the train running resistance W , and the train braking force B all act on the center of gravity of the train, and the direction is parallel to the rail, and the train receives the resultant force. F _{total is} : F _total = F - W - B (N/kN) (4)

It can be concluded from the train _total force F _total that there are four kinds of running states of the trains in the urban rail transit, namely the traction state, the cruise state, the idle state, and the braking state. The following conditions are as follows: Traction status: This state is used when the train is in the start-up phase and acceleration phase. It includes traction and running resistance, ie: F _total = F - W (N/kN) (5)

Cruise status: This state is used when the train is in the middle of the running line. At this time, the train force is 0, and the train maintains a constant speed, ie: F _total =0 (N/kN) (6)

Skip state: This state is used when the train is in the middle of the running line. It only includes the running resistance, ie: F _total =- W (N/kN) (7)'

Braking state: This state is used when the train is in the deceleration phase or ready to stop. It includes braking force and running resistance, ie: F _total =- W - B (N/kN) (8)

Train kinematics model

Under the action of different external forces, the train can accelerate, equal speed, decelerate, etc. on the track, which is in line with the motion law of Newtonian mechanics. Assuming that the train runs the distance S after the acceleration a _{total has} passed the time t , then the train kinematics model from the Newtonian mechanical motion law is: V _t = V ₀ + a _total t (9)

Where v ₀ and v _t are the initial and final speeds of the train operation, respectively.

Energy consumption assessment model

As shown in FIG. 1 , it is a schematic diagram of the energy flow of the train traction system 100 (taking the traction power supply network as an example). Train traction system 100 includes train 1, traction power supply network 110, and The electric device 120 includes a direct AC conversion device 130, a motor 140, a transmission 150, and wheels 160. The traction power supply network 110 supplies power to the train 1 through the power receiving device 120. The direct AC conversion device 130 of the train 1 converts the input DC power into AC power, and sends it to the motor 140 to drive the motor 140 to operate. The output energy of the motor 140 drives the wheel 160 to rotate through the transmission 150 to drive the train. 1 run. The energy output by the motor 140 loses a small amount of energy during the transmission of the transmission 150, and most of the remaining energy is used for train traction. During the train operation, most of the energy consumption is used for the train's own traction. At the same time, due to the existence of running resistance during the train operation, energy is also consumed in the form of heat energy.

In order to achieve the goal of energy saving for trains, regenerative braking technology is widely used in urban rail transit. Regenerative braking means that when the train brakes, the motor 140 will reverse its rotation from the electric motor to the generator, thereby converting the running energy of the train braking into usable electric energy, and then feeding it back to the traction power supply network 110 or the three rails, so that Not only can the brake effect be generated on the train, but also the inherent energy of the train can be recycled, thereby avoiding waste of energy and achieving the goal of energy saving.

In order to meet the needs of the train traction energy assessment of the present invention, the train traction energy assessment model is described below. Part of this model is energy consumption modeling from the perspective of mechanical energy, and part of the energy consumption modeling from the perspective of electrical energy. Finally, the entire energy assessment model will be able to evaluate and calculate the energy consumption of each part of the train traction process.

The specific train traction energy assessment model is as follows:

1. Traction net/third rail supply train energy model

After the traction substation supplies power to the traction network or the third rail, the power receiving device of the train absorbs its electric energy as energy input to the train traction system. After receiving the train actual measured voltage U _i (t) at time t of the electrical terminals and the current I _i (t), can be derived time t ₁ to time t _2, the input energy model train is: P _i (t) = U _i ( t ) I _i ( t ) (10)

Where P _i ( t ) is the power input to the train; E _i is the energy input to the train.

2. Drive motor energy model

After entering the train, the energy of the input train is converted into alternating current by a direct AC conversion device as input energy for driving the traction motor. There is a partial loss in input energy during the conversion process. From the measured traction motor input voltage U _mi ( t ), current I _mi ( t ), and power factor cos The energy model that can be used to drive the motor is:

Where P _mi ( t ) is the power that drives the motor to run; E _mi is the energy that drives the motor to operate.

3. Traction motor supply train running energy model

After the train traction motor is running, it will output the energy supplied to the train. This part of the energy can directly drive the train wheels after partial loss through the gear transmission, so that the train runs forward along the track. According to the energy conversion relationship, the input voltage U _mi ( t ), the input current I _mi ( t ), the motor efficiency η _m and the gear transmission efficiency η _tr that drive the traction motor can be used to obtain the energy model of the traction motor feeding train operation. :

Where F _mo ( t ) is the force output by the traction motor; v _i ( t ) is the ideal operating speed of the following vehicles acting on the traction motor output force; m is the number of train motors; P _mo ( t ) is the traction motor supply train Operating power; E _mo supplies the traction motor with energy for train operation.

4. Train operation energy consumption model

The train input energy will eventually be converted into the mechanical energy of the train running on the track, that is, the actual running energy consumption of the train. From the train actual running speed v _o ( t ) and the train quality M , the train running energy consumption model can be obtained as follows:

Where, P _t ( t ) is the train running power; E _t is the train running energy consumption.

5. Train running resistance energy consumption model

During the train operation, there will be train running resistance. Therefore, part of the train traction energy consumption will be lost by the train running resistance and will be dissipated to the surrounding environment in the form of heat. Train running resistance includes basic resistance and additional resistance, and the corresponding model has been previously described. From this, we can get the train running resistance energy consumption model as: P _r ( t )= F _r ( t ) v _a ( t )=[ Mg ( w ₀ + w _j )] v _a ( t ) (18)

Where F _r ( t ) is the train running resistance; P _r ( t ) is the train running resistance power; E _r is the train running resistance energy consumption.

Considering that the current regenerative braking energy-saving technology has been gradually adopted in urban rail transit, the regenerative braking energy model of the train is as follows:

6. Train regenerative braking energy model

When the train is braked with regenerative braking technology, the traction motor of the train reverses and turns into a generator to realize regenerative braking. At this point, the train operating energy will be converted to electrical energy and fed back to the power grid or third rail for reuse. Since the regenerative braking causes the motor to reverse, the motor output voltage U _mbo ( t ), the output current I _mbo ( t ) and the power factor cos The train regenerative braking energy model can be obtained as follows:

Where, P _rbo ( t ) is the power generated by the train regenerative braking; E _rbo is the energy generated by the train regenerative braking.

7. Regenerative braking feedback grid or third rail energy model

This part of the feedback energy comes from the regenerative braking energy generated after the motor is reversed. It is converted to AC power and then fed back to the power supply network (or the third track) for continuous use. According to the input voltage U _i ( t ) of the power supply network and the reverse current I _re ( t ), the energy model of the regenerative braking feedback grid is: P _rbb ( t )= U _i ( t ) I _re ( t ) (22)

Where, P _rbb ( t ) is the power that the train regenerative brake feeds back to the grid; E _rbb is the energy that the train regenerative brake feeds back to the grid.

The principles of the invention are described below.

The invention provides a train operation control method, wherein the input variables include: a pre-planned inter-station speed curve, an instantaneous traction power supply network (third rail) voltage, an orbital line shape (slope, curvature, speed limit) and related resistance ( Air resistance), instant train running adhesion estimation, train arrival Station time estimate. The output variables are the instantaneous train operation mode (acceleration, deceleration, constant speed, idle line four operating modes), instant acceleration and deceleration indication, and instant speed indication.

Regarding the input variables, the pre-planned inter-station speed curve is mainly planned in advance to provide the train control device as the value of the speed indication of the train traveling between the stations, regardless of whether the acquisition of the speed curve is optimized, the present invention proposes The methods are applicable. The instantaneous traction power supply network (third rail) voltage is obtained by collecting the power from the pantograph (collector shoe) on the vehicle into the vicinity of the on-board substation when the train is running, and judging the current electric energy of the traction power supply system. Usage. When the traction power supply network/third rail voltage is excessively lower than the traction power substation power supply voltage, it means that other trains on the same route are consuming electric energy, thus causing the line voltage drop to cause the traction power supply network (third rail) voltage to drop. When the traction power supply network (third rail) voltage is higher than the traction power substation power supply voltage, it means that other trains on the same route are releasing electric energy due to electric braking, thereby causing the line pressure to rise and causing the traction power supply network/third rail voltage to rise. Because the track shape (slope, curvature, speed limit) and related resistance (air resistance) will affect the speed curve and energy consumption of the train operation, the same speed curve will produce different energy consumption in different linear environments. The purpose of the instantaneous train running adhesion estimate is to ensure that the operating mode, acceleration and deceleration and speed indications produced by the method of the present invention do not cause the train to slip. The purpose of train arrival time estimation is to meet the arrival time of the train schedule. According to the pre-planned inter-station speed curve, you can know the remaining time of any mileage after the train exits to the next stop. Although the method of the present invention makes full use of the electric energy returned by the electric brakes of all the operating trains on the same route, the train does not completely travel according to the originally planned inter-station speed curve, but the originally planned inter-station speed curve still has The reference value of the train's on-time operation.

As shown in FIG. 2 and FIG. 3, it is a method for controlling power running operation of a railway train according to the present invention, which comprises the following steps:

S1. Predict a time-speed curve of the train running between the station and the station based on the running data of the train.

S2. Using the time-speed curve to control an operation mode of the train, providing a traction power supply network voltage, a track linear parameter, a train resistance and an adhesive force during the train operation, wherein the traction power supply network voltage system For the voltage of a third rail, according to the voltage of the traction power supply network, the linear parameters of the track, the resistance of the train and the adhesion force, the train is adjusted. A momentary mode of operation in which the mode of operation includes a value of the train, a speed, and an acceleration.

In step S2, a three-layer reverse transfer-like neural network is used to adjust the operation mode of the train in the next moment. The three-layer reverse transfer-type neural network has an input layer, an output layer and a hidden layer. Step S2 further includes:

S21. The voltage of the traction power supply network, the linear shape of the track, the resistance of the train, and the adhesion force are used as the input layer values of the three-layer inverted transmission type neural network, and an output target value is preset, and is calculated according to the value of the input layer. The value of the hidden layer of the three-layer inverted transfer neural network.

S22. Calculate the value of the output layer of the three-layer inverse transfer neural network by using the value of the hidden layer, wherein the value of the output layer is the value, velocity, and acceleration of the operating mode.

S23. Comparing the value of the output layer with a preset output target value, and adjusting the value of the output layer such that an error value between the value of the output layer and the preset output target value is minimized.

The trains run in the four modes of acceleration, deceleration, constant speed and idle. The acceleration and deceleration modes are given the acceleration and deceleration values. The constant speed mode maintains the original speed, and the idle mode stops supplying power. The kinetic energy maintains speed. But no matter what mode, you need to have a target speed. Therefore, the specific method proposed by the present invention for saving train energy energy is presented in the output variable, that is, the instant train operation mode, the instantaneous acceleration and deceleration indication, and the instantaneous speed indication.

In one embodiment, assume that train A will depart to the next stop, and train A will receive operational data from the control center. The train control center determines the running time between the trains A and the station according to the time schedule and system traffic, and then plans the execution time-speed curve between the train stations. The inter-station execution time-speed curve of train A can be converted into each mileage distance arrival time and each mileage train speed, and each mileage indication interval is determined according to the system reaction time and related requirements. After starting, Train A will adjust the speed according to the running speed curve to facilitate the control of the travel mode.

When train A travels between stations, in addition to adjusting the speed according to the time-speed curve of the train center, the method proposed by the present invention will use the height of the traction power supply network or the third rail voltage to determine the energy of the current traction power supply system. Use, and consider the current situation of various resistance and adhesion, according to the decision of the train A in the next moment of operation mode, speed and acceleration and deceleration adjustment.

Adjustment methods can use artificial intelligence or other optimization algorithms to achieve the goal. The three-layer inverted-transfer neural network of FIG. 4 is taken as an example for illustration, and includes a plurality of neurons and a connection therebetween. In the input layer, 8 neurons are the input vector X { x ₁ , x ₂ ,..., x ₈ }, including the position of the train, the speed limit, the resistance of the train, the speed, the adhesion of the train, the traction power supply network or the third rail voltage, each mileage speed , each mileage distance arrival time (can also be the traction power supply network or the third rail voltage, track linear parameters, the resistance of the train and the adhesion of the four values). The hidden layer has 5 neurons Y { y ₁ , y ₂ ,..., y ₅ }. In the output layer, 3 neurons are the output vector Z { z ₁ , z ₂ , z ₃ }, including the train's operating mode, acceleration and deceleration, speed and other indications. The following two equations illustrate how the values of the hidden and output layer neurons are calculated.

Where u _ik is the weight value of the kth input layer neuron to the ith hidden layer neuron, and v _jk is the weight value of the kth hidden layer neuron to the jth hidden layer neuron. f is a sigmoid function with a value range between 0 and 1.

The inverse transfer-like neural network multiplies the input layer value ( x _k ) by the weight value ( u _ik ), converts it by the sigmoid function, and subtracts the threshold value of the neuron (preset threshold) to become The value of the hidden layer ( y _k ). Similarly, the value of the hidden layer ( y _k ) is multiplied by the weight value ( v _jk ) and added by the sigmoid function, and then the threshold value of the neuron (preset threshold) is subtracted to become the output layer. The value ( z _k ).

When the value of the output layer ( z _k ) is calculated, compared with the output target value ( t _k ), the energy function is used to estimate the difference between the two, and the weight values and threshold values are adjusted to minimize the energy function value as much as possible. The energy function E can be defined as

Where t _k is the output target value. z _k is the output calculated value.

The adjustment of the method of the present invention will cause the train A to fail to meet the time-speed curve provided by the original control center, but the factor of the station time will be taken into account in the decision-making, so that the purpose of operating energy saving and punctuality will be achieved at the same time.

The above is only the preferred embodiment of the present invention, and the scope of the invention is not limited thereto, that is, the simple equivalent changes and modifications made by the scope of the invention and the description of the invention are All remain within the scope of the invention patent. In addition, any of the objects or advantages or features of the present invention are not required to be achieved by any embodiment or application of the invention. In addition, the abstract sections and headings are only used to assist in the search of patent documents and are not intended to limit the scope of the invention.

Claims (6)

A rail train energy saving operation control method, comprising the steps of: predicting a time-speed curve of a train running between a station and a station according to operation data of a train; and controlling the train by using the time-speed curve The operation mode provides a traction power supply network voltage, a track linear parameter, a train resistance and an adhesive force during the running of the train, according to the voltage of the traction power supply network, the track linear parameter, and the resistance of the train. And the adhesion to adjust the running mode of the train at the next moment, wherein the operating mode includes a numerical relationship of the train, a speed, and an acceleration. The rail train energy saving operation control method according to claim 1, wherein the voltage of the traction power supply network is a voltage of a third rail. The rail train energy-saving operation control method according to claim 1, wherein the operation mode of controlling the train uses a three-layer reverse transmission-like neural network to adjust the operation mode of the train in the next moment, wherein The three-layer inverted transfer neural network has an input layer, an output layer and a hidden layer. The rail train energy-saving operation control method as described in claim 3, further comprising the steps of: the voltage of the traction power supply network, the track linear parameter, the resistance of the train, and the adhesive force as the three layers. Inverting the value of the input layer of the neural network, and presetting an output target value, and calculating a value of the hidden layer of the three-layer inverse transfer type neural network according to the value of the input layer; using the hidden layer Calculating a value of the output layer of the three-layer inverse transfer type neural network, wherein the value of the output layer includes the value, the speed, and the acceleration in the operating mode; and the value of the output layer and the preset The output target value is compared and the value of the output layer is adjusted such that an error value between the value of the output layer and the preset output target value is minimized. The rail train energy-saving operation control method according to claim 1, wherein the operation mode is selected from an acceleration mode, a deceleration mode, and a constant speed. A group of modes and a lazy mode. The method for saving power operation control of a rail train according to any one of claims 1 to 5, wherein the track linear parameter comprises a slope of a track, a curvature of a track, and a speed limit of a track.