System based on vehicle speed planning and energy management strategy collaborative optimization and application method

Abstract

The invention relates to the field of vehicle control, in particular to a system based on vehicle speed planning and energy management strategy collaborative optimization and an application method. The system comprises an offline optimization subsystem and an online application subsystem, wherein the offline optimization subsystem comprises a scene pre-analysis module and a multi-dimensional dynamic planning algorithm module; the online application subsystem comprises a double-LSTM (Long Short Term Memory) model module and a vehicle-mounted control module; according to the invention, the scene pre-analysis module obtains and analyzes the scene of the urban continuous intersection; then solving a vehicle speed sequence and a fuel cell control sequence with the optimal comprehensive performance through a multi-dimensional dynamic programming algorithm module, and taking the vehicle speed sequence and the fuel cell control sequence as a training data set; the economic vehicle speed and the fuel cell output power are obtained through the double-LSTM model module; and the output power of the power battery is obtained through the vehicle-mounted control module. Therefore, the optimal output power of the fuel cell and the power cell at each moment is determined by considering external interference factors from driving scenes such as traffic lights and front vehicles and economic vehicle speeds.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control, and more specifically, to a system and application method based on coordinated optimization of vehicle speed planning and energy management strategy. Background Art

[0002] Fuel cell hybrid electric vehicles have a diversity of energy sources and working modes. Therefore, it is crucial to achieve efficient energy management of fuel cell hybrid electric vehicles. However, both traditional rule-based energy management strategies and optimization-based energy management strategies are greatly affected by the actual driving environment. For example, the degree of traffic congestion, the location and time period of traffic lights, pedestrians, rainy and snowy weather, random traffic events, etc. This kind of traffic information has the characteristics of wide sources, complex data, and even incomplete traffic information. Therefore, most of the current technologies focus on energy management strategies that integrate traffic information, that is, speed planning and energy management strategy collaborative optimization. However, the diversification of fuel cell vehicle power systems has brought great challenges to the collaborative optimization of speed planning and energy management strategies, because there are two power sources in the power system, and the energy management strategy is required to determine the power distribution between the fuel cell and the power battery at each moment according to the planned speed. In the joint optimization problem, the increase in state variables and control variables leads to an increase in the difficulty of solving and the computational burden index. In addition, due to the simplification of the model, external interference from driving scenes such as traffic lights and leading vehicles is often ignored in the collaborative optimization process. Therefore, most collaborative optimization strategies are suitable for free flow driving, but are not practical for online implementation in urban driving scenarios.

[0003] The prior art discloses a power distribution method and device for a hybrid vehicle, and a vehicle controller. The power distribution method includes: obtaining the current vehicle speed and current required torque of the hybrid vehicle; when determining that the driving mode of the hybrid vehicle is the parallel mode, determining the current SOC value of the power battery and determining the current required power of the hybrid vehicle; and distributing the output power of the engine and the working power of the electric generator according to the current SOC value of the power battery, the current required power and the engine economic line power of the hybrid vehicle. However, this method still ignores external interference from driving scenes such as traffic lights and the vehicle in front. This affects the power distribution performance. Summary of the invention

[0004] The purpose of the present invention is to disclose a system and application method based on coordinated optimization of vehicle speed planning and energy management strategy with better performance and consideration of external factors.

[0005] In order to achieve the above objectives, the present invention provides a system and an application method based on coordinated optimization of vehicle speed planning and energy management strategy, including: Offline optimization subsystem and online application subsystem; the offline optimization subsystem includes: scenario pre-analysis module and multi-dimensional dynamic programming algorithm module; the online application subsystem includes: dual LSTM model module and vehicle control module; Scenario pre-analysis module: obtains the scenario of continuous urban intersections, and pre-analyzes the scenario of continuous urban intersections to obtain reasonable green wave time windows and vehicle data, and obtains the driving speed range of the green wave time windows; Multi-dimensional dynamic programming algorithm module: Constructs the coordinated optimal control problem of vehicle speed planning and energy management based on the driving speed range and vehicle data, and uses the multi-dimensional dynamic programming algorithm to solve the vehicle speed sequence and fuel cell control sequence with the best comprehensive performance as the training data set; Dual LSTM model module: obtains economic vehicle speed and fuel cell output power based on training data set and vehicle data; On-board control module: obtains the power battery output power according to the economic vehicle speed and the fuel cell output power; controls the vehicle according to the fuel cell output power and the power battery output power.

[0006] Further, the scenario pre-analysis module includes: a traffic intersection queue prediction submodule and a time and speed pre-analysis submodule; Traffic intersection queue prediction submodule: obtains the length of queued vehicles at the intersection and the time it takes for the queued vehicles to dissipate based on the scene of continuous intersections in the city; Time and speed pre-analysis submodule: The green wave travel time window is obtained based on the length of the queued vehicles at the intersection and the time it takes for the queued vehicles to dissipate, and the driving speed range is calculated.

[0007] Furthermore, the traffic intersection queue prediction submodule includes: in the scene of continuous urban intersections, when the traffic flow encounters a red light at the intersection, it will form a queue behind the stop line, generating a stop wave. ; When the signal light switches to green, a start wave is generated The moment when the start wave and the stop wave meet marks the end of the queue caused by the red light phase. At this time, the distance from the stop line to the farthest waiting vehicle is the farthest distance of the queue. The time from when the green light comes on to when the last vehicle in the queue leaves the stop line is the queue dissipation time. ; The formula is as follows:

[0008]

[0009]

[0010]

[0011] In the formula, , Indicates the flow and density of traffic flow in the free driving state, , represents the flow rate and density in the saturated state, represents the density under the blocked traffic state, These are the moments when the traffic light switches to green and red respectively.

[0012] Furthermore, the time and speed pre-analysis submodule includes: The calculation formula for the travel time interval to the intersection is as follows:

[0013]

[0014] In the formula, , For the vehicle The minimum and maximum travel time for a road segment; For the Length of road section; , The upper and lower limits of the vehicle's speed. The upper limit is set according to the road speed limit, and the lower limit is set to prevent the vehicle from driving at an ultra-low speed and affecting the driving of the following vehicle. , are the upper and lower limits of vehicle acceleration; According to the traffic light phase and timing information of each intersection, the estimated road section travel time interval is intersected with the green light window to obtain the passable time interval; Arrive at The fastest and slowest times at the intersection are ; After selecting the target travel time interval, determine the upper and lower speed limits for each road section:

[0015] .

[0016] Furthermore, the multi-dimensional dynamic programming algorithm module includes: an optimal control problem submodule and a dynamic programming solution algorithm submodule; Optimal control problem submodule: used to construct the optimal control problem of coordinated control of fuel cell vehicle speed planning and energy management; Dynamic programming algorithm submodule: solves the optimal control problem and obtains the speed control sequence and fuel cell output power sequence with the optimal cost function performance index value as the training data set.

[0017] Furthermore, the optimal control problem submodule includes: Establishing the vehicle longitudinal dynamics model: =

[0018]

[0019] In the formula is the wheel end drive power, Indicates the battery output power, represents the fuel cell output power, represents the motor efficiency, Indicates the mass of the car (kg); represents the gravitational acceleration ( ); represents the rolling resistance coefficient; Indicates the road slope angle; represents the air resistance coefficient; represents the frontal area ( ); represents the air density ( ); Indicates the driving speed ( ), represents the rotational mass conversion factor, Indicates transmission efficiency; The Thevenin model is used to mathematically model the battery, and the state of charge of the power battery is calculated according to the ampere-hour integration method. SOC for:

[0020] In the formula, is the initial charge state of the battery, is the battery capacity, is the battery current; The output voltage of PEMFC monomer is:

[0021] In the formula is the activation polarization voltage loss, is the ohmic polarization voltage loss, is the concentration polarization voltage loss; The voltage of a single cell of PEMFC is about 0.7 V; The total voltage of the fuel cell stack is:

[0022] in is the number of cells in the battery stack; Fuel cell hydrogen consumption rate Expressed as:

[0023]

[0024] in is the molar mass of hydrogen, is the superposition current, F is the Faraday coefficient; The fuel cell performance attenuation rate formula is:

[0025] in is the fuel cell performance decay rate, is the correction factor, , , and Respectively represent the duration of low load operation, high load operation, rapid load change, and start-stop cycle. , , and is the corresponding attenuation coefficient; The state variable has two dimensions: speed and battery state of charge ; and the control variable is the motor power demand and fuel cell output power , the state transfer equation is:

[0026] Taking the comprehensive performance of hydrogen consumption, fuel cell loss, driving time and speed fluctuation as the optimization target, the optimal control problem of fuel cell vehicle speed planning and energy management coordinated control is constructed:

[0027]

[0028] In the formula, , and For the vehicle The hydrogen consumed during the driving phase, the fuel cell life and time, is the absolute value of the speed difference between the current stage and the previous stage, , , , is the weight factor of the corresponding cost item.

[0029] Furthermore, the dynamic programming solution algorithm submodule includes: The optimal control problem is transformed into a distance-based discrete system, by dividing the total road length into The road is divided into Equal division, number of stages ; Use multi-dimensional dynamic programming algorithm to solve: discretize the state variables to get The state variable matrix of indivual as well as indivual , respectively denoted as and and , the control variable is recorded as as well as ; Dynamic programming works backwards from the final state to the initial state, and gradually calculates the minimum cost of reaching the final state at each stage, and determines the best decision path based on the minimum cost. The cost function expression of dynamic programming solution at each stage is as follows:

[0030] is the cumulative optimal cost of the subsequent sub-routes, is the cost function value of the state variables and control variables at the current stage; The multi-dimensional dynamic programming algorithm starts from the last stage N and works backwards to solve the problem. First, the transfer cost to the feasible state point of the N-1th stage is calculated. The feasible state point is determined by the maximum charge and discharge power of the power battery and the speed range obtained by pre-analysis. The decision of the optimal cost of each state point from the N-1th stage to the Nth stage is recorded, and the solution is solved in this way. During the whole planning process, the optimal vehicle speed and fuel cell power sequence were generated by backtracking method; Starting from the initial value, the optimal control variable in the current state is forward indexed, and the state value of the next stage is solved through the state transfer equation; and so on, finally the speed sequence and fuel cell output power sequence with the optimal cost function are obtained as the training data set.

[0031] Furthermore, the dual LSTM model module includes: an LSTM1 submodule and an LSTM2 submodule; LSTM1 submodule: obtains the solution rules for the vehicle speed planning problem based on the multi-dimensional dynamic programming algorithm according to the training data set, and generates the economic vehicle speed; The network structure of LSTM1 includes: input layer, hidden layer and output layer; the input layer has four variables, namely, vehicle speed V, distance L from the vehicle to the next intersection, time T from the current moment to the next intersection when the traffic light switches to green, and phase length of the green light at the next intersection. , all four variables are input in the form of sequences; the hidden layer includes LSTM layer and fully connected layer, and LSTM layer can capture the long-term dependencies in time series data; the output layer is the vehicle speed at the next moment ; LSTM2 submodule: used to generate fuel cell output power; The network structure of LSTM2 includes: input layer, hidden layer and output layer; the input of LSTM2 network is vehicle speed V, power battery charge state S, acceleration A, where the acceleration is obtained according to the economic vehicle speed generated by LSTM1, and finally outputs the fuel cell power at the current moment .

[0032] Furthermore, according to the training data set, the solution rule for the vehicle speed planning problem based on the multi-dimensional dynamic programming algorithm includes: all the time series information in the training data set is calculated according to the time interval Interpolation and resampling are performed, and the resampled time series information is reorganized in order according to each group of n time series, and the reorganized training data set is used as the input of the LSTM1 submodule.

[0033] In addition, the present invention also provides an application method of a system based on coordinated optimization of vehicle speed planning and energy management strategy, including: S1: Obtain the scene of the continuous urban intersection through the scene pre-analysis module, and pre-analyze the scene of the continuous urban intersection to obtain a reasonable green wave passage time window and vehicle data, and obtain the driving speed range of the green wave passage time window; S2: construct the coordinated optimal control problem of vehicle speed planning and energy management according to the driving speed range and vehicle data through the multi-dimensional dynamic programming algorithm module, and use the multi-dimensional dynamic programming algorithm to solve the vehicle speed sequence and fuel cell control sequence with the best comprehensive performance as the training data set; S3: Obtaining the economic vehicle speed and fuel cell output power according to the training data set and vehicle data through the dual LSTM model module; S4: Obtain the power battery output power according to the economic vehicle speed and the fuel cell output power through the vehicle control module; and control the vehicle according to the fuel cell output power and the power battery output power.

[0034] Compared with the prior art, the technical solution of the present invention has the following beneficial effects: The present invention obtains and analyzes the scene of continuous urban intersections through the scene pre-analysis module, thereby taking into account external interference factors from driving scenes such as traffic lights and the front car; then solves the vehicle speed sequence and fuel cell control sequence with the best comprehensive performance through the multi-dimensional dynamic programming algorithm module as a training data set; obtains the economic vehicle speed and fuel cell output power through the dual LSTM model module; obtains the power battery output power according to the economic vehicle speed and the fuel cell output power through the vehicle control module. In this way, the optimal output power of the fuel cell and the power battery at each moment is determined by combining the driving scene and the economic vehicle speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A system block diagram of the coordinated optimization of vehicle speed planning and energy management strategy according to the first embodiment; Figure 2 This is the car-cloud layered architecture diagram described in Example 2; Figure 3 is a schematic diagram of the scene pre-analysis described in Example 2; Figure 4 is a configuration diagram of a fuel cell vehicle according to the second embodiment; Figure 5 is a schematic diagram of the multi-dimensional dynamic programming algorithm solution process described in Example 2; Figure 6 It is the LSTM1 model structure described in Example 2; Figure 7 It is the LSTM2 model structure described in Example 2; Figure 8 A flow chart of an application method of the system based on coordinated optimization of vehicle speed planning and energy management strategy according to the first embodiment; DETAILED DESCRIPTION The drawings are for illustrative purposes only and are not to be construed as limiting the present patent; The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0036] Embodiment 1: This embodiment provides Figure 1 The system and application method and system based on coordinated optimization of vehicle speed planning and energy management strategy include: In order to achieve the above objectives, the present invention provides a system and an application method based on coordinated optimization of vehicle speed planning and energy management strategy, including: Offline optimization subsystem and online application subsystem; the offline optimization subsystem includes: scenario pre-analysis module and multi-dimensional dynamic programming algorithm module; the online application subsystem includes: dual LSTM model module and vehicle control module; Scenario pre-analysis module: obtains the scenario of continuous urban intersections, and pre-analyzes the scenario of continuous urban intersections to obtain reasonable green wave time windows and vehicle data, and obtains the driving speed range of the green wave time windows; Multi-dimensional dynamic programming algorithm module: Constructs the coordinated optimal control problem of vehicle speed planning and energy management based on the driving speed range and vehicle data, and uses the multi-dimensional dynamic programming algorithm to solve the vehicle speed sequence and fuel cell control sequence with the best comprehensive performance as the training data set; Dual LSTM model module: obtains economic vehicle speed and fuel cell output power based on training data set and vehicle data; On-board control module: obtains the power battery output power according to the economic vehicle speed and the fuel cell output power; controls the vehicle according to the fuel cell output power and the power battery output power.

[0037] This embodiment acquires and analyzes the scene of continuous urban intersections through the scene pre-analysis module, thereby taking into account external interference factors from driving scenes such as traffic lights and the vehicle in front; then the multi-dimensional dynamic programming algorithm module solves the vehicle speed sequence and fuel cell control sequence with the best comprehensive performance as the training data set; the dual LSTM model module obtains the economic vehicle speed and fuel cell output power; the vehicle control module calculates the power battery output power based on the economic vehicle speed and the fuel cell output power. In this way, the optimal output power of the fuel cell and power battery at each moment can be determined by combining the driving scene and the economic vehicle speed.

[0038] Embodiment 2: This embodiment further discloses on the basis of the first embodiment: The system vehicle-cloud layered architecture based on the coordinated optimization of vehicle speed planning and energy management strategy in this embodiment is shown in the figure below: Figure 2 As shown; The scenario pre-analysis module includes: a traffic intersection queue prediction submodule and a time and speed pre-analysis submodule; Traffic intersection queue prediction submodule: obtains the length of queued vehicles at the intersection and the time it takes for the queued vehicles to dissipate based on the scene of continuous intersections in the city; Time and speed pre-analysis submodule: The green wave travel time window is obtained based on the length of the queued vehicles at the intersection and the time it takes for the queued vehicles to dissipate, and the driving speed range is calculated.

[0039] The traffic intersection queue prediction submodule includes: in the scene of continuous urban intersections, when the traffic flow encounters a red light at the intersection, it will gradually form a queue behind the stop line, generating a stop wave ; When the signal light switches to green, a start wave is generated The moment when the start wave and the stop wave meet marks the end of the queue caused by the red light phase. At this time, the distance from the stop line to the farthest waiting vehicle is the farthest distance of the queue. The time from when the green light comes on to when the last vehicle in the queue leaves the stop line is the queue dissipation time. ; The formula is as follows:

[0040]

[0041]

[0042]

[0043] In the formula, , Indicates the flow and density of traffic flow in the free driving state, , represents the flow rate and density in the saturated state, represents the density under the blocked traffic state, These are the moments when the traffic light switches to green and red respectively.

[0044] The time and speed pre-analysis submodules include: The calculation formula for the travel time interval to the intersection is as follows:

[0045]

[0046] In the formula, , For the vehicle The minimum and maximum travel time for a road segment; For the Length of road section; , The upper and lower limits of the vehicle's speed. The upper limit is set according to the road speed limit, and the lower limit is set to prevent the vehicle from driving at an ultra-low speed and affecting the driving of the following vehicle. , are the upper and lower limits of vehicle acceleration; like Figure 3 As shown, according to the phase and timing information of traffic lights at each intersection, the estimated road section travel time interval is intersected with the green light window to obtain the passable time interval; Arrive at The fastest and slowest times at the intersection are ; After selecting the target travel time interval, determine the upper and lower speed limits for each road section:

[0047] .

[0048] The multi-dimensional dynamic programming algorithm module includes: an optimal control problem submodule and a dynamic programming solution algorithm submodule; Optimal control problem submodule: used to construct the optimal control problem of coordinated control of fuel cell vehicle speed planning and energy management; Dynamic programming algorithm submodule: solves the optimal control problem and obtains the speed control sequence and fuel cell output power sequence with the optimal cost function performance index value as the training data set.

[0049] The optimal control problem submodules include: The research object of the embodiment is a fuel cell hybrid vehicle of the "fuel cell + power battery" configuration, such as Figure 4 As shown. In this configuration, the fuel cell mainly meets the continuous power demand of the vehicle, while the battery supplements the peak and transient power demand, as well as recovers regenerative energy during braking. This configuration can reduce the power fluctuation of the fuel cell and reduce hydrogen consumption, thereby improving the life of the fuel cell and the economy of the whole vehicle.

[0050] Establishing the vehicle longitudinal dynamics model: =

[0051]

[0052] In the formula is the wheel end drive power, Indicates the battery output power, represents the fuel cell output power, represents the motor efficiency, Indicates the mass of the car (kg); represents the gravitational acceleration ( ); represents the rolling resistance coefficient; Indicates the road slope angle; represents the air resistance coefficient; represents the frontal area ( ); represents the air density ( ); Indicates the driving speed ( ), represents the rotational mass conversion factor, Indicates transmission efficiency; The Thevenin model is used to mathematically model the battery, and the state of charge of the power battery is calculated according to the ampere-hour integration method. SOC for:

[0053] In the formula, is the initial charge state of the battery, is the battery capacity, is the battery current; The output voltage of PEMFC monomer is:

[0054] In the formula is the activation polarization voltage loss, is the ohmic polarization voltage loss, is the concentration polarization voltage loss; The voltage of a single cell of PEMFC is about 0.7 V; The total voltage of the fuel cell stack is:

[0055] in is the number of cells in the battery stack; Fuel cell hydrogen consumption rate Expressed as:

[0056]

[0057] in is the molar mass of hydrogen, is the superposition current, F is the Faraday coefficient; The fuel cell performance attenuation rate formula is:

[0058] in is the fuel cell performance decay rate, is the correction factor, , , and Respectively represent the duration of low load operation, high load operation, rapid load change, and start-stop cycle. , , and is the corresponding attenuation coefficient; The state variable has two dimensions: speed and battery state of charge ; and the control variable is the motor power demand and fuel cell output power , the state transfer equation is:

[0059] Taking the comprehensive performance of hydrogen consumption, fuel cell loss, driving time and speed fluctuation as the optimization target, the optimal control problem of fuel cell vehicle speed planning and energy management coordinated control is constructed:

[0060]

[0061] In the formula, , and For the vehicle The hydrogen consumed during the driving phase, the fuel cell life and time, is the absolute value of the speed difference between the current stage and the previous stage, , , , is the weight factor of the corresponding cost item.

[0062] The dynamic programming solution algorithm submodules include: The optimal control problem is transformed into a distance-based discrete system, by dividing the total road length into The road is divided into Equal division, number of stages ;like Figure 5 The multidimensional dynamic programming algorithm is used to solve the problem: the state variables are discretized to obtain The state variable matrix of indivual as well as indivual , respectively denoted as and and , the control variable is recorded as as well as ; Dynamic programming works backwards from the final state to the initial state, and gradually calculates the minimum cost of reaching the final state at each stage, and determines the best decision path based on the minimum cost. The cost function expression of dynamic programming solution at each stage is as follows:

[0063] is the cumulative optimal cost of the subsequent sub-routes, is the cost function value of the state variables and control variables at the current stage; The multi-dimensional dynamic programming algorithm starts from the last stage N and works backwards to solve the problem. First, the transfer cost to the feasible state point of the N-1th stage is calculated. The feasible state point is determined by the maximum charge and discharge power of the power battery and the speed range obtained by pre-analysis. The decision of the optimal cost of each state point from the N-1th stage to the Nth stage is recorded, and the solution is solved in this way. During the whole planning process, the optimal vehicle speed and fuel cell power sequence were generated by backtracking method; Starting from the initial value, the optimal control variable in the current state is forward indexed, and the state value of the next stage is solved through the state transfer equation; and so on, finally the speed sequence and fuel cell output power sequence with the optimal cost function are obtained as the training data set.

[0064] The dual LSTM model module includes: LSTM1 submodule and LSTM2 submodule; LSTM1 submodule: obtains the solution rules for the vehicle speed planning problem based on the multi-dimensional dynamic programming algorithm according to the training data set, and generates the economic vehicle speed; The network structure of LSTM1 is as follows Figure 6 The figure includes: input layer, hidden layer and output layer; the input layer has four variables, namely, vehicle speed V, distance L from the vehicle to the next intersection, time T from the current moment to the next intersection signal light switching to green light, and phase length of the next intersection green light , all four variables are input in the form of sequences; the hidden layer includes LSTM layer and fully connected layer, and LSTM layer can capture the long-term dependencies in time series data; the output layer is the vehicle speed at the next moment ; LSTM2 submodule: used to generate fuel cell output power; The network structure of LSTM2 is as follows Figure 7 The figure shows: input layer, hidden layer and output layer; the input of LSTM2 network is vehicle speed V, power battery charge state S, acceleration A, where the acceleration is obtained according to the economic vehicle speed generated by LSTM1, and finally outputs the fuel cell power at the current moment .

[0065] The solution rules for the vehicle speed planning problem based on the multidimensional dynamic programming algorithm obtained according to the training data set include: Interpolation and resampling are performed, and the resampled time series information is reorganized in order according to each group of n time series, and the reorganized training data set is used as the input of the LSTM1 submodule.

[0066] This embodiment acquires and analyzes the scene of continuous urban intersections through the scene pre-analysis module, thereby taking into account external interference factors from driving scenes such as traffic lights and the vehicle in front; then the multi-dimensional dynamic programming algorithm module solves the vehicle speed sequence and fuel cell control sequence with the best comprehensive performance as the training data set; the dual LSTM model module obtains the economic vehicle speed and fuel cell output power; the vehicle control module calculates the power battery output power based on the economic vehicle speed and the fuel cell output power. In this way, the optimal output power of the fuel cell and power battery at each moment can be determined by combining the driving scene and the economic vehicle speed.

[0067] Embodiment three: This embodiment provides a Figure 8 The application method of the system based on the coordinated optimization of vehicle speed planning and energy management strategy includes: S1: Obtain the scene of the continuous urban intersection through the scene pre-analysis module, and pre-analyze the scene of the continuous urban intersection to obtain a reasonable green wave passage time window and vehicle data, and obtain the driving speed range of the green wave passage time window; S2: construct the coordinated optimal control problem of vehicle speed planning and energy management according to the driving speed range and vehicle data through the multi-dimensional dynamic programming algorithm module, and use the multi-dimensional dynamic programming algorithm to solve the vehicle speed sequence and fuel cell control sequence with the best comprehensive performance as the training data set; S3: Obtaining the economic vehicle speed and fuel cell output power according to the training data set and vehicle data through the dual LSTM model module; S4: Obtain the power battery output power according to the economic vehicle speed and the fuel cell output power through the vehicle control module; and control the vehicle according to the fuel cell output power and the power battery output power.

[0068] This embodiment acquires and analyzes the scene of continuous urban intersections through the scene pre-analysis module, thereby taking into account external interference factors from driving scenes such as traffic lights and the vehicle in front; then the multi-dimensional dynamic programming algorithm module solves the vehicle speed sequence and fuel cell control sequence with the best comprehensive performance as the training data set; the dual LSTM model module obtains the economic vehicle speed and fuel cell output power; the vehicle control module calculates the power battery output power based on the economic vehicle speed and the fuel cell output power. In this way, the optimal output power of the fuel cell and power battery at each moment can be determined by combining the driving scene and the economic vehicle speed.

[0069] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A system and application method based on coordinated optimization of vehicle speed planning and energy management strategy, characterized in that: include: Offline optimization subsystem and online application subsystem; The offline optimization subsystem includes: scenario pre-analysis module and multi-dimensional dynamic programming algorithm module; the online application subsystem includes: dual LSTM model module and vehicle control module; Scenario pre-analysis module: obtains the scenario of continuous urban intersections, and pre-analyzes the scenario of continuous urban intersections to obtain reasonable green wave time windows and vehicle data, and obtains the driving speed range of the green wave time windows; Multi-dimensional dynamic programming algorithm module: Constructs the coordinated optimal control problem of vehicle speed planning and energy management based on the driving speed range and vehicle data, and uses the multi-dimensional dynamic programming algorithm to solve the vehicle speed sequence and fuel cell control sequence with the best comprehensive performance as the training data set; Dual LSTM model module: obtains economic vehicle speed and fuel cell output power based on training data set and vehicle data; On-board control module: obtains the power battery output power according to the economic vehicle speed and the fuel cell output power; controls the vehicle according to the fuel cell output power and the power battery output power.

2. The system based on coordinated optimization of vehicle speed planning and energy management strategy according to claim 1 is characterized in that: The scenario pre-analysis module includes: a traffic intersection queue prediction submodule and a time and speed pre-analysis submodule; Traffic intersection queue prediction submodule: obtains the length of queued vehicles at the intersection and the time it takes for the queued vehicles to dissipate based on the scene of continuous intersections in the city; Time and speed pre-analysis submodule: The green wave travel time window is obtained based on the length of the queued vehicles at the intersection and the time it takes for the queued vehicles to dissipate, and the driving speed range is calculated.

3. The system based on coordinated optimization of vehicle speed planning and energy management strategy according to claim 2 is characterized in that: The traffic intersection queue prediction submodule includes: in the scene of continuous urban intersections, when the traffic flow encounters a red light at the intersection, it will gradually form a queue behind the stop line, generating a stop wave ; When the signal light switches to green, a start wave is generated The moment when the start wave and the stop wave meet marks the end of the queue caused by the red light phase. At this time, the distance from the stop line to the farthest waiting vehicle is the farthest distance of the queue. The time from when the green light comes on to when the last vehicle in the queue leaves the stop line is the queue dissipation time. ; The formula is as follows: In the formula, , Indicates the flow and density of traffic flow in the free driving state, , represents the flow rate and density in the saturated state, represents the density under the blocked traffic state, These are the moments when the traffic light switches to green and red respectively.

4. The system based on coordinated optimization of vehicle speed planning and energy management strategy according to claim 2 is characterized in that: The time and speed pre-analysis submodules include: The calculation formula for the travel time interval to the intersection is as follows: In the formula, , For the vehicle The minimum and maximum travel time for a road segment; For the Length of road section; , The upper and lower limits of the vehicle's speed. The upper limit is set according to the road speed limit, and the lower limit is set to prevent the vehicle from driving at an ultra-low speed and affecting the driving of the following vehicle. , are the upper and lower limits of vehicle acceleration; According to the traffic light phase and timing information of each intersection, the estimated road section travel time interval is intersected with the green light window to obtain the passable time interval; Arrive at The fastest and slowest times at the intersection are ; After selecting the target travel time interval, determine the upper and lower speed limits for each road section: 。 5. The system based on coordinated optimization of vehicle speed planning and energy management strategy according to claim 1 is characterized in that: The multi-dimensional dynamic programming algorithm module includes: an optimal control problem submodule and a dynamic programming solution algorithm submodule; Optimal control problem submodule: used to construct the optimal control problem of coordinated control of fuel cell vehicle speed planning and energy management; Dynamic programming algorithm submodule: solves the optimal control problem and obtains the speed control sequence and fuel cell output power sequence with the optimal cost function performance index value as the training data set.

6. The system based on coordinated optimization of vehicle speed planning and energy management strategy according to claim 5 is characterized in that: The optimal control problem submodules include: Establishing the vehicle longitudinal dynamics model: = In the formula is the wheel end drive power, Indicates the battery output power, represents the fuel cell output power, represents the motor efficiency, Indicates the mass of the car (kg); represents the gravitational acceleration ( ); represents the rolling resistance coefficient; Indicates the road slope angle; represents the air resistance coefficient; represents the frontal area ( ); represents the air density ( ); Indicates the driving speed ( ), represents the rotational mass conversion factor, Indicates transmission efficiency; The Thevenin model is used to mathematically model the battery, and the state of charge of the power battery is calculated according to the ampere-hour integration method. SOC for: In the formula, is the initial charge state of the battery, is the battery capacity, is the battery current; The output voltage of PEMFC monomer is: In the formula is the activation polarization voltage loss, is the ohmic polarization voltage loss, is the concentration polarization voltage loss; The total voltage of the fuel cell stack is: in is the number of cells in the battery stack; Fuel cell hydrogen consumption rate Expressed as: in is the molar mass of hydrogen, is the superposition current, F is the Faraday coefficient; The fuel cell performance attenuation rate formula is: in is the fuel cell performance decay rate, is the correction factor, , , and Respectively represent the duration of low load operation, high load operation, rapid load change, and start-stop cycle. , , and is the corresponding attenuation coefficient; The state variable has two dimensions: speed and battery state of charge ; and the control variable is the motor power demand and fuel cell output power , the state transfer equation is: Taking the comprehensive performance of hydrogen consumption, fuel cell loss, driving time and speed fluctuation as the optimization target, the optimal control problem of fuel cell vehicle speed planning and energy management coordinated control is constructed: In the formula, , and For the vehicle The hydrogen consumed during the driving phase, the fuel cell life and time, is the absolute value of the speed difference between the current stage and the previous stage, , , , is the weight factor of the corresponding cost item.

7. The system based on coordinated optimization of vehicle speed planning and energy management strategy according to claim 5 is characterized in that: The dynamic programming solution algorithm submodules include: The optimal control problem is transformed into a distance-based discrete system, by dividing the total road length into The road is divided into Equal division, number of stages ; Use multi-dimensional dynamic programming algorithm to solve: discretize the state variables to get The state variable matrix of indivual as well as indivual , respectively denoted as and and , the control variable is recorded as as well as ; Dynamic programming works backwards from the final state to the initial state, and gradually calculates the minimum cost of reaching the final state at each stage, and determines the best decision path based on the minimum cost. The cost function expression of dynamic programming solution at each stage is as follows: is the cumulative optimal cost of the subsequent sub-routes, is the cost function value of the state variables and control variables at the current stage; The multi-dimensional dynamic programming algorithm starts from the last stage N and works backwards to solve the problem. First, the transfer cost to the feasible state point of the N-1th stage is calculated. The feasible state point is determined by the maximum charge and discharge power of the power battery and the speed range obtained by pre-analysis. The decision of the optimal cost of each state point from the N-1th stage to the Nth stage is recorded, and the solution is solved in this way. During the whole planning process, the optimal vehicle speed and fuel cell power sequence were generated by backtracking method; Starting from the initial value, the optimal control variable in the current state is forward indexed, and the state value of the next stage is solved through the state transfer equation; and so on, finally the speed sequence and fuel cell output power sequence with the optimal cost function are obtained as the training data set.

8. The system based on coordinated optimization of vehicle speed planning and energy management strategy according to claim 1 is characterized in that: The dual LSTM model module includes: LSTM1 submodule and LSTM2 submodule; LSTM1 submodule: obtains the solution rules for the vehicle speed planning problem based on the multi-dimensional dynamic programming algorithm according to the training data set, and generates the economic vehicle speed; The network structure of LSTM1 includes: input layer, hidden layer and output layer; the input layer has four variables, namely, vehicle speed V, distance L from the vehicle to the next intersection, time T from the current moment to the next intersection when the traffic light switches to green, and phase length of the green light at the next intersection. , all four variables are input in the form of sequences; the hidden layer includes LSTM layer and fully connected layer, and LSTM layer can capture the long-term dependencies in time series data; the output layer is the vehicle speed at the next moment ; LSTM2 submodule: used to generate fuel cell output power; The network structure of LSTM2 includes: input layer, hidden layer and output layer; the input of LSTM2 network is vehicle speed V, power battery charge state S, acceleration A, where the acceleration is obtained according to the economic vehicle speed generated by LSTM1, and finally outputs the fuel cell power at the current moment .

9. The system based on coordinated optimization of vehicle speed planning and energy management strategy according to claim 8 is characterized in that: The solution rules for the vehicle speed planning problem based on the multidimensional dynamic programming algorithm obtained according to the training data set include: all the temporal information in the training data set is calculated according to the time interval Interpolation and resampling are performed, and the resampled time series information is reorganized in order according to each group of n time series, and the reorganized training data set is used as the input of the LSTM1 submodule.

10. An application method of the system based on coordinated optimization of vehicle speed planning and energy management strategy according to claim 1, characterized in that: include: S1: Obtain the scene of the continuous urban intersection through the scene pre-analysis module, and pre-analyze the scene of the continuous urban intersection to obtain a reasonable green wave passage time window and vehicle data, and obtain the driving speed range of the green wave passage time window; S2: construct the coordinated optimal control problem of vehicle speed planning and energy management according to the driving speed range and vehicle data through the multi-dimensional dynamic programming algorithm module, and use the multi-dimensional dynamic programming algorithm to solve the vehicle speed sequence and fuel cell control sequence with the best comprehensive performance as the training data set; S3: Obtaining the economic vehicle speed and fuel cell output power according to the training data set and vehicle data through the dual LSTM model module; S4: Obtain the power battery output power according to the economic vehicle speed and the fuel cell output power through the vehicle control module; and control the vehicle according to the fuel cell output power and the power battery output power.

Images