Rail vehicle and operation optimization control method, system and storage medium thereof

Abstract

The present invention discloses a rail vehicle and an operation optimization control method, system and storage medium thereof. An evaluation objective function is constructed based on the safety, comfort and energy consumption of rail vehicle travel, and a rail vehicle optimization control strategy model is constructed with the highest evaluation as the goal to optimize the operation control of the rail vehicle. The rail vehicle optimization control strategy model determines the rail vehicle power control parameters of several sub-tracks between the current station and the next station based on the departure time of the rail vehicle and the operating environment parameters, adjusts the operating status of the rail vehicle on each sub-track, and realizes safer, more efficient, energy-saving and comfortable rail vehicle operation adjustment.

Description

Technical Field

[0001] The present invention relates to the field of rail transit technology, and in particular to a rail vehicle and an operation optimization control method, system and storage medium thereof. Background Art

[0002] The vibration caused by the formation of rail vehicles will seriously affect the comfort of passengers. As the comfort level expected by passengers increases, the comfort, safety and timeliness of rail vehicle operation become more and more important. Traditional mechanical solutions cannot effectively avoid the vibration and impact caused by track roughness, nor can they balance the comfort and timeliness of rail vehicle operation, resulting in poor passenger experience. In order to enhance the comfort and timeliness of rail vehicle operation, a method that can actively make autonomous decisions is needed to optimize rail vehicle operating parameters.

[0003] Invention patent application CN114580290A discloses a high-speed train group coordinated control method and system for joint optimization of energy consumption and comfort, obtains real-time operation information of high-speed trains, calculates the speed deviation between the actual speed and the expected speed of high-speed trains, considers the energy consumption and ride comfort of high-speed trains during operation as optimization targets, establishes a multi-objective optimization model for coordinated control of multiple trains in high-speed train group based on the speed deviation and dynamic relationship of high-speed trains, solves the multi-objective optimization model by a fast non-dominated sorting algorithm with elite retention strategy, obtains the control variables of each high-speed train, applies the control variables to the power device to generate traction or braking force; applies the traction or braking force to the high-speed train to control the speed change of the high-speed train until the high-speed train group runs in a consistent state. This method takes into account the running speed and energy consumption during the train operation and optimizes the ride comfort. However, this scheme does not take into account the safety of train operation, and cannot guarantee the running timeliness, and cannot balance the comfort and timeliness of rail vehicle driving. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a rail vehicle and its operation optimization control method, system and storage medium in view of the deficiencies in the prior art, so as to ensure the timeliness of rail vehicle travel and improve the travel comfort of rail vehicle.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a rail vehicle operation optimization control method, comprising the following steps:

[0006] S1. Divide the running track of the rail vehicle into a plurality of sub-tracks, and the track parameters of each sub-track are the same;

[0007] S2. Construct a multi-scale rail vehicle driving evaluation index; the multi-scale rail vehicle driving evaluation index includes a rail vehicle driving energy consumption index, a driving safety index and a comfort index; the driving energy consumption index indicates the proportion of residual electric energy of the rail vehicle; the driving safety index indicates the speed limit of the rail vehicle on the sub-track; the comfort index indicates the on-time rate of the rail vehicle arriving at the station and the severity of the acceleration change during driving;

[0008] S3. Determine the rail vehicle safe and comfortable operation consistency evaluation objective function according to the driving evaluation index;

[0009] S4. Taking the current operating environment parameters of the rail vehicle as the input of the neural network model, and the vehicle power control parameters of the rail vehicle in different sub-tracks as the output of the neural network model, the neural network model is trained, and the objective function of maximizing the consistency evaluation of safe and comfortable operation of the rail vehicle is used as the parameter optimization target of the trained neural network model, and the trained neural network model is optimized to obtain a vehicle optimization control strategy model.

[0010] The present invention constructs an evaluation objective function based on the safety, comfort and energy consumption of rail vehicle travel, and constructs a rail vehicle optimization control strategy model with the highest evaluation as the goal to optimize the operation control of the rail vehicle. The rail vehicle optimization control strategy model will determine the rail vehicle power control parameters of several sub-tracks between the current station and the next station based on the departure time of the rail vehicle and the operating environment parameters, adjust the operating conditions of the rail vehicle on each sub-track, and achieve safer, more efficient, energy-saving and comfortable rail vehicle operation adjustment. Therefore, the method of the present invention can improve the comfort of rail vehicle travel while ensuring the timeliness of rail vehicle travel. The present invention divides the track into multiple sub-tracks, further improving the optimization control accuracy.

[0011] In step S1, the track parameters include the slope, resistance coefficient and track type of the track; the track type includes straight type and curved type.

[0012] The track parameters of the present invention take into full account factors such as the track slope, resistance coefficient and track type, which can ensure accurate control of relevant parameters of the rail vehicle and further improve ride comfort.

[0013] In step S3, the expression of the rail vehicle safe and comfortable operation consistency evaluation objective function is:

[0014]

[0015] Where n represents the number of sub-tracks from the current station position to the next station, a i Indicates that the rail vehicle is on subtrack x i The acceleration, xi represents any i-th sub-track, v i Indicates that the rail vehicle is on subtrack x i The speed at the end, α represents the drag coefficient of the rail vehicle itself, β i ={0,1}, Indicates that the rail vehicle is on subtrack x i The average speed, v i, Represents sub-track x i The speed limit, E represents the planned power consumption of the rail vehicle from the current station to the next station, t represents the planned travel time of the rail vehicle from the current station to the next station, t ′ Indicates the time when the rail vehicle departs from the current station. represents the actual power consumption of the rail vehicle from the current station to the next station, m represents the mass of the rail vehicle; L i Represents sub-track x i Path distance.

[0016] In order to further improve the efficiency and comfort of rail vehicle operation, in step S4, the current operating environment parameters of the rail vehicle include track slope, rail vehicle resistance coefficient, sub-track parameters, expected arrival time and track distance to the station, and the sub-track parameters include the distance, slope, track type and speed limit of different sub-tracks.

[0017] In the present invention, in order to further improve the optimization control accuracy, in step S4, the neural network model is trained to maximize the rail vehicle safe and comfortable operation consistency evaluation objective function as the trained neural network model parameter optimization target. The specific implementation process includes:

[0018] 1) Encoding the sub-orbital parameters to obtain a coded representation vector of the sub-orbital parameters;

[0019] 2) Selecting a corresponding strategy of the encoding representation vector from a strategy selection space, wherein the strategy selection space includes a plurality of different groups of accelerations, and the selected strategy is the acceleration of the rail vehicle on the corresponding sub-track;

[0020] 3) Get the i-th sub-track x i Strategy a i , based on the vehicle on the ith sub-track x i The force relationship analysis results show that the rail vehicle on the ith sub-track x i The power F(i); F(i) = ma i +F ′ (i)+mgsinθ i ; where θ iRepresents sub-track x i The slope, a i Indicates that the rail vehicle is on subtrack x i The acceleration of α represents the drag coefficient of the rail vehicle itself, β i ={0,1}, g represents the acceleration due to gravity;

[0021] 4) converting the power F(i) into a corresponding vehicle power control parameter σ(i);

[0022] 5) Repeat the above steps 3) and 4) to obtain a set of vehicle power control parameters {σ(i)|∈[1,]} of the rail vehicle on the n sub-tracks;

[0023] 6) Repeat steps 2), 3), 4), and 5) to obtain several sets of vehicle power parameter sets, and the set of vehicle power control parameter sets that maximizes the rail vehicle safety and comfort operation consistency evaluation objective function is the optimal control strategy.

[0024] In step S4, the specific implementation process of optimizing the trained neural network model includes:

[0025] A) Initialization strategy selects the parameters θ of the neural network model a ; Generate N groups of orbits with different operating environment parameters, where the generated dth group of orbits has m d sub-trajectories; generate D groups of different accelerations, and construct all the generated accelerations into a strategy selection space. Then any w-th group of strategies in the strategy selection space is a w ;

[0026] B) The track environment operating parameters and the corresponding strategy selection space are stored in the memory bank, based on the parameter θ a The strategy selects the neural network model as any sub-orbit x d, Select strategy a w , x d, represents the u-th sub-track in the d-th group of tracks generated;

[0027] C) Determine the safety and comfort operation consistency evaluation objective function of rail vehicles on any sub-track

[0028] x d, Select strategy a w The value of value(x d, ,a w )for:

[0029] Among them: a d,-1Indicates the u-1th sub-orbit x in the generated dth group of orbitals d,-1 The acceleration, ∈[1,m d ],m d represents the mass of the vehicle in the dth set of tracks;

[0030] F ′ (d,) indicates that the rail vehicle is on the sub-track x d, The resistance encountered; L d, Represents sub-track x d, The length distance, v d, Indicates that the rail vehicle is on subtrack x d, The speed of the end;

[0031] v u, Represents sub-track x d, Speed ​​limit

[0032] D) Repeat steps B) and C), select a strategy for each sub-track, and optimize the neural network model parameters θ by minimizing the following loss function a Update to obtain an updated strategy selection neural network model, that is, a vehicle optimization control strategy model; the loss function Loss(θ a )for:

[0033]

[0034] Among them, γ d,, (θ a )={0,1},γ d,, (θ a )=1 means based on parameter θ a The strategy selects the neural network model as any sub-orbit x d, Select strategy a w , γ d,, (θ a )=0 means based on parameter θ a The strategy selection neural network model does not have any sub-orbit x d, Select strategy a w ,value(x d, ,ax) represents the strategy selection space for x d, Select the maximum value generated by the corresponding strategy; N is the number of different operating environment parameter groups of the track; D is the number of different acceleration groups.

[0035] In order to continuously optimize the vehicle optimization control strategy model, the method of the present invention also includes:

[0036] E) The final control strategy output by the vehicle optimization control strategy model is retained in the memory library.

[0037] In the present invention, after the vehicle optimization control strategy model is obtained, the method further includes: S5, using the vehicle optimization control strategy model to obtain power control parameters of the rail vehicle on the subsequent rail road.

[0038] In the present invention, the specific implementation process of step S5 includes: when the rail vehicle arrives at the current station to stop, the current operating environment parameters of the rail vehicle are input into the vehicle optimization control strategy model, the vehicle optimization control strategy model outputs the power control parameters of the rail vehicle on the subsequent rail road, and the real-time control and adjustment of the rail vehicle operation status is performed based on the power control parameters (i.e., adjusting the speed and acceleration of the rail vehicle), wherein the rail vehicle operation status includes the speed and acceleration of the rail vehicle.

[0039] The present invention optimizes and controls the speed and acceleration of rail vehicles, thereby ensuring the efficiency of vehicle operation and improving the riding comfort of passengers, thereby ensuring the comfort, safety and timeliness of rail vehicle operation. The deep learning method is used to actively adjust the operating speed of rail vehicles, thereby minimizing the vibration and impact caused by the roughness of the track.

[0040] As an inventive concept, the present invention also provides a rail transit vehicle operation optimization control system, including a memory, a processor and a computer program stored in the memory; characterized in that the processor executes the computer program to implement the steps of the above-mentioned method of the present invention.

[0041] As an inventive concept, the present invention also provides a rail vehicle, which adopts the above-mentioned operation optimization control system of the present invention.

[0042] As an inventive concept, the present invention also provides a computer-readable storage medium on which a computer program / instruction is stored; when the computer program / instruction is executed by a processor, the steps of the above method of the present invention are implemented.

[0043] Compared with the prior art, the beneficial effects of the present invention are as follows: based on the departure time of the rail vehicle and the operating environment parameters, the present invention determines the rail vehicle power control parameters of several sub-tracks between the current station and the next station, adjusts the operating conditions of the rail vehicle on each sub-track, and realizes safer, more efficient, energy-saving and comfortable rail vehicle operation adjustment, which can actively avoid passenger vibration and impact caused by track roughness, and balance the comfort and timeliness of rail vehicle travel. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of the method of Example 1 of the present invention;

[0045] Figure 2This is a functional module diagram of a rail vehicle operation optimization control system according to Embodiment 2 of the present invention;

[0046] Figure 3 This is a structural diagram of the rail vehicle operation optimization control system according to Example 2 of the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] In this article, the terms "include", "comprises" and other similar words are intended to indicate logical relationships, and should not be considered to indicate spatial structural relationships. For example, "A includes B" is intended to indicate that B logically belongs to A, but does not mean that B is spatially located inside A. In addition, the meanings of the terms "include", "comprises" and other similar words should be considered open, not closed. For example, "A includes B" is intended to indicate that B belongs to A, but B does not necessarily constitute the whole of A, and A may also include other elements such as C, D, and E.

[0049] The execution subject of the rail vehicle operation optimization control method of the embodiment of the present invention includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the rail vehicle operation optimization control method can be executed by software or hardware installed in a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0050] Example 1

[0051] Embodiment 1 of the present invention provides a method for optimizing the operation control of a rail vehicle, such as Figure 1 As shown, the method comprises the following steps:

[0052] S1: Divide the rail vehicle's running track into several sub-tracks, construct a multi-scale rail vehicle running evaluation index, and determine the rail vehicle safe and comfortable operation consistency evaluation objective function based on the constructed evaluation index.

[0053] In step S1, the travel track of the rail vehicle is divided into several sub-tracks, including:

[0054] Rail vehicles refer to vehicles that can travel on rails, mainly including subways, light rails, trains and high-speed rails. Rail vehicles have several stops during their travel, and they will stop when they arrive at a stop.

[0055] The running track of the rail vehicle is divided into several sub-tracks. The track parameters of each sub-track are the same. The track parameters include the slope, resistance coefficient and track type of the track. The track type includes straight and curved types, and each sub-track is limited to a maximum of 2 kilometers.

[0056] In an embodiment of the present invention, when a rail vehicle arrives at the current station to stop, the rail vehicle optimization control strategy model will determine the rail vehicle power control parameters of several sub-tracks between the current station and the next station based on the departure time of the rail vehicle and the operating environment parameters, and adjust the operating conditions of the rail vehicle on each sub-track to achieve safer, more efficient, energy-saving and comfortable rail vehicle operation adjustment.

[0057] The sub-track set for a rail vehicle to reach the next station from the current station is {x i |∈[1,]}, where n represents the number of sub-tracks from the current station position to the next station, and x i represents any i-th sub-track, and the rail vehicle adjusts the power control parameters at the intersection of any two sub-tracks to control the running status of the rail vehicle, wherein the running status of the rail vehicle includes the speed and acceleration of the rail vehicle.

[0058] In step S1, a multi-scale rail vehicle driving evaluation index is constructed, including:

[0059] The rail vehicle driving evaluation index includes the rail vehicle driving energy consumption index, driving safety index and comfort index;

[0060] The driving energy consumption index indicates the proportion of the remaining electric energy of the rail vehicle. The higher the proportion of the remaining electric energy, the less electric energy the rail vehicle consumes during driving. In the embodiment of the present invention, the rail vehicle converts electric energy into driving kinetic energy during driving to provide power for the rail vehicle.

[0061] The driving safety index indicates the speed limit of the rail vehicle on the sub-track. Different sub-tracks have different speed limits. If the speed of the rail vehicle on the sub-track exceeds the corresponding speed limit, it means that danger may occur.

[0062] The comfort index indicates the punctuality of the rail vehicle arriving at the station and the severity of the acceleration change during the driving process. The higher the punctuality rate and the lower the severity of the acceleration change, the higher the comfort index.

[0063] In step S1, the objective function of the safety and comfort operation consistency evaluation of rail vehicles is determined according to the constructed evaluation index, including:

[0064] The objective function for evaluating the consistency of safe and comfortable operation of rail vehicles is determined based on the constructed evaluation index. The constructed objective function is:

[0065]

[0066] F(i)-F′(i)-mgsinθ i =ma i

[0067]

[0068]

[0069]

[0070] t=t″-t′

[0071] v i =v i-1 +a i t i '

[0072] in:

[0073] F(i) represents the number of vehicles on the track x i The power includes traction and braking force, which is provided by the rail vehicle itself;

[0074] a i Indicates that the rail vehicle is on subtrack x i acceleration;

[0075] v i Indicates that the rail vehicle is on subtrack x i The speed at the end, Indicates that the rail vehicle is on subtrack x i The average speed of the track vehicle on subtrack x i The velocity at the starting point is v i-1 ;

[0076] v i,limit Represents sub-track x i Speed ​​limit

[0077] θ i Represents sub-track x i The slope of

[0078] L i Represents sub-track x i The path distance, t iIndicates that the rail vehicle is moving from sub-track x i Start from the starting point and drive to sub-track x i The time required to reach the end, t″ represents the planned arrival time of the rail vehicle at the next station, t′ represents the departure time of the rail vehicle from the current station, and t represents the planned travel time of the rail vehicle from the current station to the next station. Indicates the actual travel time of a rail vehicle from the current station to the next station;

[0079] E represents the planned electric energy consumption of the rail vehicle from the current station to the next station. Indicates the actual power consumption of a rail vehicle from the current station to the next station;

[0080] F ′ (i) indicates that the rail vehicle is on sub-track x i resistance, g represents the acceleration of gravity, α represents the resistance coefficient of the rail vehicle itself, β i ={0,1}, if β i =0, indicating sub-track x i is a straight line, if β i =1, indicating sub-track x i is the curve, r i Then it represents the corresponding curve radius;

[0081] Corresponding to the comfort index, |a i -a i-1 The smaller the value, the smaller the change in acceleration during the running of the rail vehicle, and the lower the frequency of body vibration caused by sudden acceleration or deceleration of the vehicle. The higher the value of; Corresponding to the comfort index, Indicates the time that the rail vehicle arrives at the next station early or late. The early or late arrival of the rail vehicle will affect the operation of other vehicles on the track. Indicates that rail vehicles arrive on time;

[0082] Corresponding driving energy consumption index; Corresponding driving safety indicators;

[0083] In this embodiment, the parameter to be solved of the objective function is the power of the rail vehicle on different sub-tracks. Based on the solved rail vehicle power, the power parameters of the rail vehicle on different sub-tracks are controlled so that the operation of the rail vehicle meets the index conditions of safety, high efficiency, energy saving and comfort.

[0084] The constraints of the objective function for evaluating the consistency of safe and comfortable operation of rail vehicles in step S1 include:

[0085] The constraints of the objective function are:

[0086]

[0087] in:

[0088] v max represents the maximum speed of the rail vehicle, v represents the speed of the rail vehicle at any time, if the rail vehicle reaches the speed v by accelerating max , with speed v max Drive at a constant speed;

[0089] a represents the acceleration threshold;

[0090] v1 represents the speed of the rail vehicle starting from the current station, v n Indicates the speed at which the rail vehicle reaches the next station.

[0091] S2: Construct a rail vehicle optimization control strategy model, wherein the deep neural network model is the main model structure of the rail vehicle optimization control strategy model, and the input of the model is the current operating environment parameters of the rail vehicle, including the rail vehicle resistance coefficient, sub-track parameters, expected arrival time and track distance to the station. The sub-track parameters include the distance, slope, track type and speed limit of different sub-tracks. The output of the model is the vehicle power control parameters of the rail vehicle on different sub-tracks.

[0092] In step S2, a rail vehicle optimization control strategy model is constructed, wherein the deep neural network model is the main model structure of the rail vehicle optimization control strategy model, including:

[0093] Constructing a rail vehicle optimization control strategy model, wherein the deep neural network model is the main model structure of the rail vehicle optimization control strategy model, and the rail vehicle optimization control strategy model includes an input layer, a coding layer, a strategy selection layer, and an output layer;

[0094] The strategy selection layer includes a strategy selection neural network and a strategy evaluation neural network. The strategy selection layer receives the encoding representation vector input by the encoding layer. The strategy selection neural network is used to select a strategy from the strategy selection space. The strategy evaluation neural network is used to evaluate the selected strategy set and take the strategy set with the best evaluation result as the final control strategy.

[0095] The operation process of the rail vehicle optimization control strategy model is as follows:

[0096] S21: The input layer receives the parameter values ​​of the input model, wherein the input of the model is the current operating environment parameters of the rail vehicle, including the track slope, the rail vehicle resistance coefficient, the sub-track parameters, the estimated arrival time and the track distance to the station, and the sub-track parameters include the distance, slope, track type and speed limit of different sub-tracks;

[0097] S22: the coding layer encodes the received sub-track parameters to obtain a coding representation vector of the sub-track parameters, and inputs the coding representation vector of the sub-track parameters and the remaining received parameters to the strategy selection layer;

[0098] S23: the strategy selection layer selects a corresponding strategy of the encoding representation vector from the strategy selection space, wherein the strategy selection space includes a plurality of different acceleration groups, and the selected strategy is the acceleration of the rail vehicle on the sub-track. In the embodiment of the present invention, the acceleration value range satisfies the constraint condition of the objective function;

[0099] S24: The output layer receives the selected arbitrary sub-track x i Strategy a i , and based on the track vehicle on any sub-track x i The force relationship analysis results show that the rail vehicle on any sub-track x i The power F(i):

[0100] F(i)=ma i +F ′ (i)+mgsinθ i ;

[0101] and converting the power F(i) into a corresponding vehicle power control parameter σ(i) for output;

[0102] The rail vehicle moves on the sub-track x based on the vehicle dynamics control parameter σ(i). i If the rail vehicle is traveling on the sub-track x i The acceleration is a i ;

[0103] Repeat step S24 to obtain a set of vehicle power control parameters {σ(i)|∈[1,]} for the rail vehicle on n sub-tracks; repeat this step to obtain several sets of vehicle power parameter sets, and use the strategy evaluation neural network in the strategy selection layer to evaluate each set of vehicle power control parameters. The evaluation basis is that a set of vehicle power control parameters that maximizes the consistency evaluation objective function of the safe and comfortable operation of the rail vehicle is the optimal control strategy.

[0104] S3: Parameter optimization is performed on the rail vehicle optimization control strategy model based on the offline optimization strategy to obtain the optimized rail vehicle optimization control strategy model, wherein the deep deterministic policy gradient optimization algorithm is the main method of the offline optimization strategy, and the parameter optimization goal is to maximize the consistency evaluation objective function of the safe and comfortable operation of the rail vehicle.

[0105] In step S3, the rail vehicle optimization control strategy model is optimized based on the offline optimization strategy, including:

[0106] The model parameters to be optimized of the rail vehicle optimization control strategy model include a strategy selection neural network, wherein a deep deterministic policy gradient optimization algorithm is the main method of the offline optimization strategy;

[0107] The optimization process of the rail vehicle optimization control strategy model is:

[0108] S31: Initialization strategy selects the parameters θ of the neural network a ;

[0109] S32: Generate N groups of orbits with different operating environment parameters, where the generated dth group of orbits has m d sub-tracks;

[0110] S33: Generate D groups of different accelerations, and construct all the generated accelerations into a strategy selection space. Then any w-th group of strategies in the strategy selection space is a w ;

[0111] S34: storing the generated track environment operating parameters and the corresponding strategy selection space in the memory bank, based on the parameter θ a The strategy of selecting a neural network as any sub-orbit x d, Select strategy a w , x d, represents the u-th sub-track in the d-th group of tracks generated;

[0112] S35: Determine the rail vehicle on any sub-track x based on the rail vehicle safety and comfort operation consistency evaluation objective function d, Select strategy a w The value is:

[0113]

[0114] in:

[0115] a d,-1 Represents sub-track x d,-1 The acceleration, x d, represents the u-th sub-orbit in the d-th group of generated orbitals, u∈[1,m d ]; m drepresents the mass of the vehicle in the dth track

[0116] F ′ (d,) indicates that the rail vehicle is on the sub-track x d, the resistance encountered;

[0117] L d, Represents sub-track x d, The length distance, v d, Indicates that the rail vehicle is on subtrack x d, The speed of the end;

[0118] S36: Repeat the above steps S34-S35, use the strategy selection neural network to select a strategy for each sub-track, and minimize the following loss function to optimize the model parameter θ a Update to obtain the updated strategy selection neural network, the loss function Loss(θ a )for:

[0119]

[0120] in:

[0121] γ d,, (θ a )={0,1},γ d,, (θ a )=1 means based on parameter θ a The strategy of selecting a neural network as any sub-orbit x d, Select strategy a w , γ d,, (θ a )=0 means based on parameter θ a The strategy selection neural network does not have any sub-orbit x d, Select strategy a w ;

[0122] value(x d, ,ax) represents the strategy selection space for x d, Select the corresponding strategy to generate the maximum value;

[0123] The final control strategy output by the optimized rail vehicle optimization control strategy model is retained in the memory library.

[0124] S4: using the optimized rail vehicle optimization control strategy model to output the power control parameters of the rail vehicle on the subsequent rail road, and performing real-time control and adjustment of the rail vehicle operation status based on the real-time power control parameters of the rail vehicle, wherein the rail vehicle operation status includes the speed and acceleration of the rail vehicle.

[0125] In step S4, the optimized rail vehicle optimization control strategy model is used to output the power control parameters of the rail vehicle on the subsequent rail road, including: inputting the current operating environment parameters of the rail vehicle into the optimized rail vehicle optimization control strategy model, the model outputs the power control parameters of the rail vehicle on the subsequent rail road, and the rail vehicle performs real-time control and adjustment of the rail vehicle operating status based on the power control parameters, wherein the rail vehicle operating status includes the speed and acceleration of the rail vehicle.

[0126] Example 2

[0127] like Figure 2 As shown, this embodiment provides a functional module diagram of a rail vehicle operation optimization control system, which can implement the rail vehicle operation optimization control method in Example 1.

[0128] The rail vehicle operation optimization control system 100 can be installed in an electronic device. According to the functions to be implemented, the rail vehicle operation optimization control system can include a data acquisition module 101, a model building device 102 and a model optimization module 103. The module described in the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0129] The data acquisition module 101 is used to divide the running track of the rail vehicle into several sub-tracks, and collect the current running environment parameters of the rail vehicle, including the rail vehicle resistance coefficient, sub-track parameters, expected arrival time and track distance to the station. The sub-track parameters include the distance, slope, track type and speed limit of different sub-tracks;

[0130] The model building device 102 is used to build a multi-scale rail vehicle driving evaluation index, determine the rail vehicle safe and comfortable operation consistency evaluation objective function according to the built evaluation index, and build a rail vehicle optimization control strategy model;

[0131] The model optimization module 103 is used to optimize the parameters of the rail vehicle optimization control strategy model based on the offline optimization strategy, use the optimized rail vehicle optimization control strategy model to output the power control parameters of the rail vehicle on the subsequent track road, and perform real-time control and adjustment of the rail vehicle operation status based on the real-time power control parameters of the rail vehicle.

[0132] In detail, the modules in the rail vehicle operation optimization control system 100 in the embodiment of the present invention are used in the same manner as described above. Figure 1 The technical means are the same as the rail vehicle operation optimization control method described in, and can produce the same technical effects, so they will not be repeated here.

[0133] Example 3

[0134] The rail vehicle operation optimization control system provided in this embodiment is shown in Figure 3 The system includes a processor 10, a memory 11 and a bus, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as program 12. The processor executes the computer program in the memory to implement the steps of the method in the above-mentioned embodiment 1.

[0135] In some implementations, the memory may be a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.

[0136] In some other implementations, the processor may be a central processing unit (CPU), a digital signal processor (DSP), or other general-purpose processors of various types, which are not limited herein.

[0137] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes conventional removable storage media, such as: USB flash drive, SD card, solid state drive, etc., and can also be a new type of storage medium, such as: ferroelectric memory FeRAM, magnetic random access memory MRAM, phase change memory OUM, PCM phase change memory, ReRAM impedance random access memory, NRAM carbon nanotube random access memory, etc. The memory 11 can be an internal storage unit of the electronic device 1 in some embodiments, and can also be an external storage device of the electronic device 1 in other embodiments, such as an internal flash storage space integrated in the CPU processor, or various types of on-board storage chips mounted separately on the PCBA circuit board inside the electronic device. Further, the memory 11 can also include both an internal storage unit of the electronic device 1 and an external storage device extended by an external communication bus such as I2C, SPI, USB, MVB, TRDP, etc. The memory 11 is not only used to store the operating system for the normal operation of the electronic device 1, the application programs required for acquisition and control, and the big data storage required for algorithm and strategy optimization, such as the code of the program 12.

[0138] In some embodiments, the processor 10 may be composed of an MCU microprocessor, such as C51, AVR, CORTEX-M, DSP and other chips, or an MPU microprocessor with more powerful performance and functions, such as Intel X86, Freescale I.MX6, Allwinner A20, TI AM335X, etc. Further, the processor 10 may also include the minimum control core system components necessary for the electronic device 1 other than the MCU and MPU, such as: a crystal oscillator circuit, a power supply system, a code download interface, necessary operation and status indication circuits, etc. The processor 10 is the control core of the electronic device, and executes or executes the program or module stored in the memory 11 (such as the program 12 for optimizing the operation of rail vehicles) and calls the data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

[0139] The bus is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc. The bus can be a data bus, an address bus and a control bus, or any combination thereof according to the application scenario. The electronic device bus can be based on a standard serial communication protocol, such as RS232, RS485, CAN, etc., or can be some specific bus protocols, such as advanced eXtensible interface, Profinet, EtherCAT, etc. Furthermore, it can also be a self-defined dedicated protocol.

[0140] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0141] For example, although not described, the electronic device 1 may also include redundant components that are mutually cold and hot backup, such as a CPU board, a power board, etc. Through one or more redundant backup components, the reliability and availability of the system can be greatly improved, and the functions and performance of the system can be further improved. The electronic device 1 may also include various external communication and positioning function modules, such as: 4G, 5G interface modules, Zigbee communication modules, GPS, Beidou positioning modules, etc., which will not be repeated here.

[0142] Example 4

[0143] Embodiment 4 of the present invention provides a computer-readable storage medium corresponding to the above embodiment 1, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the steps of the method of the above embodiment 1 are implemented.

[0144] Computer readable storage media can be tangible devices that hold and store instructions used by instruction execution devices. Computer readable storage media can be, for example, but not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any combination thereof.

[0145] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0146] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0147] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0148] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0149] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A rail vehicle operation optimization control method, characterized in that: The following steps are involved: S1. Divide the running track of the rail vehicle into a plurality of sub-tracks, and the track parameters of each sub-track are the same; S2. Construct a multi-scale rail vehicle driving evaluation index; the multi-scale rail vehicle driving evaluation index includes a rail vehicle driving energy consumption index, a driving safety index and a comfort index; the driving energy consumption index indicates the proportion of residual electric energy of the rail vehicle; the driving safety index indicates the speed limit of the rail vehicle on the sub-track; the comfort index indicates the on-time rate of the rail vehicle arriving at the station and the severity of the acceleration change during driving; S3. Determine the rail vehicle safe and comfortable operation consistency evaluation objective function according to the driving evaluation index; The expression of the rail vehicle safety and comfort operation consistency evaluation objective function is: Where n represents the number of sub-tracks from the current station position to the next station, a i Indicates that the rail vehicle is on subtrack x i The acceleration, x i represents any i-th sub-track, v i Indicates that the rail vehicle is on subtrack x i The speed at the end, F′(i) represents the number of vehicles on the subtrack x. i resistance, g represents the acceleration of gravity, α represents the resistance coefficient of the rail vehicle itself, β i ={0,1}, if β i =0, indicating sub-track x i is a straight line, if β i =1, indicating sub-track x i is the curve, r i Indicates the corresponding curve radius; Indicates that the rail vehicle is on subtrack x i The average speed, v i，limit Represents sub-track x i The speed limit, E represents the planned power consumption of the rail vehicle from the current station to the next station, t represents the planned travel time of the rail vehicle from the current station to the next station, t i Indicates the time when the rail vehicle departs from the current station. represents the actual power consumption of the rail vehicle from the current station to the next station, m represents the mass of the rail vehicle, L i Represents sub-track x i Path distance; S4. Taking the current operating environment parameters of the rail vehicle as the input of the neural network model, and the vehicle power control parameters of the rail vehicle in different sub-tracks as the output of the neural network model, the neural network model is trained, and the objective function of maximizing the consistency evaluation of safe and comfortable operation of the rail vehicle is used as the parameter optimization target of the trained neural network model, and the trained neural network model is optimized to obtain a vehicle optimization control strategy model.

2. The rail vehicle operation optimization control method according to claim 1, characterized in that: In step S1, the track parameters include the slope, resistance coefficient and track type of the track; the track type includes straight type and curved type.

3. The rail vehicle operation optimization control method according to claim 1, characterized in that: In step S4, the current operating environment parameters of the rail vehicle include track slope, rail vehicle resistance coefficient, sub-track parameters, estimated arrival time and track distance to the station, and the sub-track parameters include distance, slope, track type and speed limit of different sub-tracks.

4. The rail vehicle operation optimization control method according to claim 3, characterized in that: In step S4, the neural network model is trained to maximize the rail vehicle safety and comfort operation consistency evaluation objective function as the parameter optimization target of the trained neural network model. The specific implementation process includes: 1) Encoding the sub-orbital parameters to obtain a coded representation vector of the sub-orbital parameters; 2) Selecting a corresponding strategy of the encoding representation vector from a strategy selection space, wherein the strategy selection space includes a plurality of different groups of accelerations, and the selected strategy is the acceleration of the rail vehicle on the corresponding sub-track; 3) Get the i-th sub-track x i Strategy a i , based on the vehicle on the ith sub-track x i The force relationship analysis results show that the rail vehicle on the ith sub-track x i The power F(i); F(i) = ma i +F′(i)+mgsinθ i ; where θ i Represents sub-track x i The slope, a i Indicates that the rail vehicle is on subtrack x i The acceleration of α represents the drag coefficient of the rail vehicle itself, β i ={0, 1}, g represents the acceleration due to gravity; 4) converting the power F(i) into a corresponding vehicle power control parameter σ(i); 5) Repeat the above steps 3) and 4) to obtain a set of vehicle power control parameters {σ(i)|i∈[1, n]} of the rail vehicle on n sub-tracks; 6) Repeat the above steps 2) to 5) to obtain several sets of vehicle power parameter sets, and the set of vehicle power control parameter sets that maximizes the rail vehicle safe and comfortable operation consistency evaluation objective function is the optimal control strategy.

5. The rail vehicle operation optimization control method according to claim 4, characterized in that: In step S4, the specific implementation process of optimizing the trained neural network model includes: A) Initialization strategy selects the parameters θ of the neural network model a ; Generate N groups of orbits with different operating environment parameters, where the generated dth group of orbits has m d sub-trajectories; generate D groups of different accelerations, and construct all the generated accelerations into a strategy selection space. Then any w-th group of strategies in the strategy selection space is a w ; B) The track environment operating parameters and the corresponding strategy selection space are stored in the memory bank, based on the parameter θ a The strategy of selecting a neural network model for any sub-orbit x d,u Select strategy a w , x d,u represents the u-th sub-track in the d-th group of tracks generated; C) Determine the rail vehicle on any sub-track x based on the rail vehicle safety and comfort operation consistency evaluation objective function d，u Select strategy a w The value of value(x d,u , a w )for: Among them: a d，u-1 Indicates the u-1th sub-orbit x in the generated dth group of orbitals d，u-1 The acceleration, u∈[1,m d ],m d represents the mass of the vehicle in the dth track; F′(d,u) represents the mass of the vehicle in the subtrack x d,u The resistance encountered; L d,u Represents sub-track x d，u The length distance, v d,u Indicates that the rail vehicle is on subtrack x d，u The speed of the end; v u,limit Represents sub-track x d,u Speed ​​limit D) Repeat steps B) and C), select a strategy for each sub-track, and optimize the neural network model parameters θ by minimizing the following loss function a Update to obtain an updated strategy selection neural network model, that is, a vehicle optimization control strategy model; the loss function Loss(θ a )for: Among them, γ d,u，w (θ a ) = {0, 1}, γ d,u,w (θa)=1 means that the a The strategy of selecting a neural network model for any sub-orbit x d，u Select strategy a w , γ d,u,w (θ a )=0 means based on parameter θ a The strategy selection neural network model does not have any sub-orbit x d，u Select strategy a w ,value(x d,u , max) represents the strategy selection space for x d，u Select the maximum value generated by the corresponding strategy; N is the number of different operating environment parameter groups of the track; D is the number of different acceleration groups.

6. The rail vehicle operation optimization control method according to claim 5, characterized in that: Also includes: E) The final control strategy output by the vehicle optimization control strategy model is retained in the memory library.

7. The rail vehicle operation optimization control method according to any one of claims 1 to 6, characterized in that: Also includes: S5. Utilize the vehicle optimization control strategy model to obtain the power control parameters of the rail vehicle on the subsequent rail road.

8. The rail vehicle operation optimization control method according to claim 7, characterized in that: The specific implementation process of step S5 includes: when the rail vehicle arrives at the current station to stop, the current operating environment parameters of the rail vehicle are input into the vehicle optimization control strategy model, the vehicle optimization control strategy model outputs the power control parameters of the rail vehicle on the subsequent rail road, and the real-time control and adjustment of the rail vehicle operation status are performed based on the power control parameters, wherein the rail vehicle operation status includes the speed and acceleration of the rail vehicle.

9. A rail transit vehicle operation optimization control system, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory; wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.

10. A rail vehicle, characterized in that: It adopts the control system described in claim 9.

11. A computer-readable storage medium having a computer program / instruction stored thereon; characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images