Energy recovery method and system of hydrogen energy vehicle

Abstract

The invention relates to the technical field of new energy vehicle control, in particular to an energy recovery method and system for a hydrogen energy vehicle. Segmenting the planned driving path by taking the position of a traffic signal lamp on the planned driving path as a segmentation point to obtain a first road section to an Nth road section; the first road section to the Nth road section serve as calculation units, first optimization models to Nth optimization models are constructed, the first optimization models to the Nth optimization models are solved through a particle swarm optimization algorithm, and first optimal braking energy recovery time to Nth optimal braking energy recovery time are obtained and output to a hydrogen energy vehicle energy management system. According to the method, the hydrogen energy vehicle braking energy recovery time is optimized, so that the adverse effect of hydrogen energy vehicle braking energy recovery on the driving safety and the service life of the vehicle is overcome while the energy recovery and utilization efficiency of the hydrogen energy vehicle is improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy vehicle control technology, and in particular to an energy recovery method and system for a hydrogen energy vehicle. Background Art

[0002] With the continuous development of energy-saving technology, vehicle energy management systems have been increasingly widely used. Vehicle energy management systems achieve the goal of energy conservation and emission reduction by rationally planning vehicle energy use behavior. Traditional vehicle energy management systems can only optimize vehicle energy use, and there are natural constraints on energy utilization efficiency. In recent years, with the increasing maturity of vehicle brake energy recovery technology, it has become a current research trend to use vehicle brake energy recovery behavior as one of the control objects of vehicle energy management systems. Vehicle brake energy recovery technology can convert mechanical energy during vehicle braking into electrical energy, thereby achieving energy recovery.

[0003] However, the temperature rise and speed fluctuations generated by the vehicle's brake energy recovery system will have an adverse effect on the vehicle's service life and safety, especially for hydrogen-powered vehicles. This adverse effect often poses a serious safety hazard. The hydrogen fuel cells used in hydrogen-powered vehicles have a certain degree of instability and are easily degraded due to sudden changes in temperature rise or speed fluctuations, and may even explode in extreme cases. Therefore, the traditional vehicle brake energy recovery system has a more prominent adverse effect on the life and safety of hydrogen-powered vehicles. How to optimize the operating parameters of hydrogen-powered vehicles and reduce the adverse effects of the vehicle's brake energy recovery system on the life and safety of hydrogen-powered vehicles while taking into account the vehicle's energy utilization efficiency is a technical problem that needs to be solved urgently. Summary of the invention

[0004] (1) Technical issues to be solved

[0005] The purpose of the present invention is to provide an energy recovery method and system for hydrogen-powered vehicles, so as to overcome the adverse effects of braking energy recovery of hydrogen-powered vehicles on driving safety and vehicle service life.

[0006] (2) Technical solution

[0007] To achieve the above object, on the one hand, the present invention provides an energy recovery method for a hydrogen energy vehicle, the method comprising the following steps:

[0008] S1, obtain first data; the first data includes the initial weight of the hydrogen energy vehicle, the initial remaining power of the hydrogen fuel cell, the planned driving route, the planned driving distance, the number of traffic lights on the planned driving route, the location of the traffic lights on the planned driving route, and the average waiting time of the traffic lights on the planned driving route.

[0009] S2, taking the position of the traffic lights on the planned driving path as the segmentation point, dividing the planned driving path into N driving sections, which are respectively recorded as the first section to the Nth section; mapping the first section to the Nth section to a two-dimensional coordinate plane, and calculating the length of the first section to the Nth section.

[0010] S3, taking the first to Nth road sections as calculation units, respectively, constructing the first to Nth optimization models, the optimization objectives of the first to Nth optimization models are the first objective function to the Nth objective function respectively; taking the maximum value of the first to Nth objective functions as the goal, using the particle swarm optimization algorithm to solve the first to Nth optimization models, to obtain the optimal values ​​of the first to Nth braking energy recovery time, recorded as the first optimal braking energy recovery time to the Nth optimal braking energy recovery time.

[0011] S4, outputting the first optimal braking energy recovery time to the Nth optimal braking energy recovery time to the hydrogen energy vehicle energy management system.

[0012] Furthermore, the method of constructing first to Nth optimization models by taking the first to Nth road sections as calculation units, wherein the optimization targets of the first to Nth optimization models are first to Nth objective functions, respectively, includes:

[0013] According to the first data, the first objective function to the Nth objective function are constructed, and the first function, the second function, the third function, the fifth function, the first recovery energy to the Nth recovery energy, and the first energy consumption to the Nth energy consumption are obtained; the independent variables of the first objective function to the Nth objective function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time.

[0014] Optimization constraints are constructed based on the first data, the third function, the first to Nth recovered energies, and the first to Nth energy consumptions.

[0015] The first to Nth objective functions are respectively combined with the optimization constraints to obtain first to Nth optimization models.

[0016] Further, the method of constructing the first objective function to the Nth objective function according to the first data, and obtaining the first function, the second function, the third function, the fifth function, the first recovered energy to the Nth recovered energy, and the first energy consumption to the Nth energy consumption includes:

[0017] The functional relationship between the first recovery energy to the Nth recovery energy and the first acceleration time to the Nth acceleration time, and the first start acceleration to the Nth start acceleration is constructed by the recovery energy calculation formula, which is recorded as the first function p i =H1(a i , t ci ); the recovery energy calculation formula is:

[0018]

[0019] Among them, p i is the i-th recovered energy, in joules; a i is the i-th starting acceleration, in meters per second squared; t ci is the i-th acceleration time, in seconds; m is the initial weight of the hydrogen energy vehicle, in kilograms; η is the preset braking energy recovery efficiency; i is an integer variable with a value from 1 to N.

[0020] According to the preset first mapping relationship, a functional relationship between the first energy consumption to the Nth energy consumption and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is constructed, which is recorded as the second function r i =H2(a i , t ci , t di ), where r i is the energy consumption of the ith energy, in joules; t di is the i-th braking energy recovery time, in seconds.

[0021] According to the preset acceleration and speed mapping relationship, a functional relationship between the first speed to the Nth speed and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is constructed, which is recorded as the third function v i (t) = H3(a i , t ci , t di ), where v i (t) is the i-th speed, in meters per second; t is the driving time, in seconds; the third function is discretized with a preset first interval Δt as the time interval to obtain the first speed sequence to the N-th speed sequence; according to the first speed sequence to the N-th speed sequence, the first speed smoothing value to the N-th speed smoothing value are obtained by the speed smoothing value calculation formula; the speed smoothing value calculation formula is:

[0022]

[0023] Among them, z iis the i-th speed smoothing value; is the mean value of the ith velocity, in meters per second; n i is the number of data in the i-th speed sequence; j i The value is 1 to n i Integer variable; x i (j i ) is the jth in the i-th speed sequence i elements.

[0024] A two-layer neural network model is established based on the first speed average to the Nth speed average and the first braking energy recovery time to the Nth braking energy recovery time. Through model training and function inverse transformation, the functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is obtained, which is recorded as the fifth function w i =H5(a i , t ci , t di ), where w i is the ith brake temperature rise, in Kelvin.

[0025] According to the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the first speed smoothing value to the Nth speed smoothing value, the first objective function to the Nth objective function are constructed through the objective function construction formula; the objective function construction formula is:

[0026]

[0027] Wherein, k1 is a preset first coefficient; k2 is a preset second coefficient; k3 is a preset third coefficient; G i (a i , t ci , t di ) is the i-th objective function.

[0028] Furthermore, a two-layer neural network model is established according to the first speed average value to the Nth speed average value and the first braking energy recovery time to the Nth braking energy recovery time, and a functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is obtained through model training and function inverse transformation, which is recorded as the fifth function w i =H5(a i , t ci , t di ), where w i The methods for calculating the i-th brake temperature rise in Kelvin include:

[0029] A two-layer neural network model is constructed, wherein the input of the two-layer neural network model is one of the first speed average to the Nth speed average and one of the first braking energy recovery time to the Nth braking energy recovery time; the output of the two-layer neural network model is the braking temperature rise obtained by subtracting the temperature at the time when the hydrogen energy vehicle completes braking from the temperature at the time when the hydrogen energy vehicle starts braking; the two-layer neural network model is trained using a pre-set sample set to obtain a first neural network model; and the functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first speed average to the Nth speed average and the first braking energy recovery time to the Nth braking energy recovery time is obtained by fitting the first neural network model, which is recorded as a fourth function.

[0030] The speed smoothing value calculation formula and the third function are respectively inversely transformed and combined with the fourth function to obtain the functional relationship between the first brake temperature rise to the Nth brake temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first brake energy recovery time to the Nth brake energy recovery time.

[0031] Furthermore, the method for constructing optimization constraints according to the first data, the third function, the first recovered energy to the Nth recovered energy, and the first energy consumption to the Nth energy consumption includes:

[0032] According to the third function, the number of traffic lights on the planned driving path, and the average waiting time of traffic lights on the planned driving path, a sixth function is obtained through a preset speed-distance mapping relationship; the independent variables of the sixth function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time; the dependent variable of the sixth function is the driving distance S e .

[0033] A first constraint condition is constructed according to the driving distance and the planned driving distance; the first constraint condition is:

[0034] S e ≥S z ;

[0035] Among them, S z It is the planned driving distance in meters.

[0036] The second constraint condition is constructed based on the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the initial remaining power of the hydrogen fuel cell; the second constraint condition is:

[0037]

[0038] Where c0 is the initial remaining power of the hydrogen fuel cell, in joules; p i is the i-th recovered energy, in joules; ri is the i-th energy consumption, in joules; i is an integer variable with values ​​from 1 to N.

[0039] The first constraint condition is combined with the second constraint condition to obtain an optimized constraint condition.

[0040] Based on the same inventive concept, on the other hand, the present invention also provides an energy recovery system for a hydrogen energy vehicle, the system comprising:

[0041] A data reading module is used to obtain first data. The first data includes the initial weight of the hydrogen energy vehicle, the initial remaining power of the hydrogen fuel cell, the planned driving route, the planned driving distance, the number of traffic lights on the planned driving route, the location of the traffic lights on the planned driving route, and the average waiting time of the traffic lights on the planned driving route.

[0042] The road segmentation module is connected to the data reading module and is used to divide the planned driving path into N driving segments using the traffic light positions on the planned driving path as segmentation points, and respectively record them as the first segment to the Nth segment; the first segment to the Nth segment are mapped to a two-dimensional coordinate plane, and the length of the first segment to the Nth segment is calculated.

[0043] The optimization solution module is connected to the road section division module, and is used to construct the first optimization model to the Nth optimization model with the first section to the Nth section as calculation units, respectively, and the optimization objectives of the first optimization model to the Nth optimization model are the first objective function to the Nth objective function respectively; with the maximum value of the first objective function to the Nth objective function as the goal, the first optimization model to the Nth optimization model are solved by using the particle swarm optimization algorithm to obtain the optimal value of the first braking energy recovery time to the Nth braking energy recovery time, which is recorded as the first optimal braking energy recovery time to the Nth optimal braking energy recovery time.

[0044] The data output module is connected to the optimization solution module and is used to output the first optimal braking energy recovery time to the Nth optimal braking energy recovery time to the hydrogen energy vehicle energy management system.

[0045] Furthermore, the optimization solution module includes:

[0046] An objective function construction module is used to construct the first objective function to the Nth objective function according to the first data, and obtain the first function, the second function, the third function, the fifth function, the first recovery energy to the Nth recovery energy, and the first energy consumption to the Nth energy consumption; the independent variables of the first objective function to the Nth objective function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time.

[0047] The constraint condition construction module is connected to the objective function construction module, and is used to construct optimization constraint conditions according to the first data, the third function, the first recovered energy to the Nth recovered energy, and the first energy consumption to the Nth energy consumption.

[0048] The optimization model construction module is connected to the constraint condition construction module, and is used to combine the first to Nth objective functions with the optimization constraint conditions respectively to obtain the first to Nth optimization models.

[0049] Furthermore, the objective function construction module includes:

[0050] The first function construction module is used to construct the functional relationship between the first recovery energy to the Nth recovery energy and the first acceleration time to the Nth acceleration time, and the first start acceleration to the Nth start acceleration through the recovery energy calculation formula, which is recorded as the first function p i =H1(a i , t ci ); the recovery energy calculation formula is:

[0051]

[0052] Among them, p i is the i-th recovered energy, in joules; a i is the i-th starting acceleration, in meters per second squared; t ci is the i-th acceleration time, in seconds; m is the initial weight of the hydrogen energy vehicle, in kilograms; η is the preset braking energy recovery efficiency; i is an integer variable with a value from 1 to N.

[0053] The second function construction module is connected to the first function construction module and is used to construct a functional relationship between the first energy consumption to the Nth energy consumption and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time according to a preset first mapping relationship, which is recorded as the second function r i =H2(a i , t ci , t di ), where r i is the energy consumption of the ith energy, in joules; tdi is the i-th braking energy recovery time, in seconds.

[0054] The speed smoothing value calculation module is connected to the second function construction module and is used to construct a functional relationship between the first speed to the Nth speed and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time according to a preset acceleration and speed mapping relationship, which is recorded as a third function v i (t) = H3(a i , t ci , t di ), where v i (t) is the i-th speed, in meters per second; t is the driving time, in seconds; the third function is discretized with a preset first interval Δt as the time interval to obtain the first speed sequence to the N-th speed sequence; according to the first speed sequence to the N-th speed sequence, the first speed smoothing value to the N-th speed smoothing value are obtained by the speed smoothing value calculation formula; the speed smoothing value calculation formula is:

[0055]

[0056] Among them, z i is the i-th speed smoothing value; is the mean value of the ith velocity, in meters per second; n i is the number of data in the i-th speed sequence; j i The value is 1 to n i Integer variable; x i (j i ) is the jth in the i-th speed sequence i elements.

[0057] The fifth function construction module is connected to the speed smoothing value calculation module, and is used to establish a two-layer neural network model according to the first speed average value to the Nth speed average value and the first braking energy recovery time to the Nth braking energy recovery time, and obtain the functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time through model training and function inverse transformation, which is recorded as the fifth function w i =H5(a i , t ci , t di ), where w i is the ith brake temperature rise, in Kelvin.

[0058] The objective function construction module is connected to the fifth function construction module, and is used to construct the first objective function to the Nth objective function according to the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the first speed smoothing value to the Nth speed smoothing value through the objective function construction formula; the objective function construction formula is:

[0059]

[0060] Wherein, k1 is a preset first coefficient; k2 is a preset second coefficient; k3 is a preset third coefficient; G i (a i , t ci , t di ) is the i-th objective function.

[0061] Furthermore, the fifth function construction module includes:

[0062] The fourth function construction module is used to construct a two-layer neural network model, the input of the two-layer neural network model is one of the first speed average to the Nth speed average and one of the first braking energy recovery time to the Nth braking energy recovery time; the output of the two-layer neural network model is the braking temperature rise obtained by subtracting the temperature at the time when the hydrogen energy vehicle completes braking from the temperature at the time when the hydrogen energy vehicle starts braking; the two-layer neural network model is trained using a pre-set sample set to obtain a first neural network model; according to the first neural network model, the functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first speed average to the Nth speed average and the first braking energy recovery time to the Nth braking energy recovery time is fitted, which is recorded as the fourth function.

[0063] The inverse transformation joint module is connected to the fourth function construction module, and is used to respectively perform inverse transformation on the speed smoothing value calculation formula and the third function, and to combine them with the fourth function to obtain the functional relationship between the first brake temperature rise to the Nth brake temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first brake energy recovery time to the Nth brake energy recovery time.

[0064] Furthermore, the constraint condition construction module includes:

[0065] The first constraint condition construction module is used to obtain a sixth function through a preset speed-distance mapping relationship according to the third function, the number of traffic lights on the planned driving path, and the average waiting time of the traffic lights on the planned driving path; the independent variables of the sixth function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time; the dependent variable of the sixth function is the driving distance S e.

[0066] A first constraint condition is constructed according to the driving distance and the planned driving distance; the first constraint condition is:

[0067] S e ≥S z ;

[0068] Among them, S z It is the planned driving distance in meters.

[0069] The second constraint condition construction module is connected to the first constraint condition construction module and is used to construct a second constraint condition according to the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the initial remaining power of the hydrogen fuel cell; the second constraint condition is:

[0070]

[0071] Where c0 is the initial remaining power of the hydrogen fuel cell, in joules; p i is the i-th recovered energy, in joules; r i is the i-th energy consumption in joules; i is an integer variable ranging from 1 to N.

[0072] The constraint condition combination module is connected to the second constraint condition construction module and is used to combine the first constraint condition with the second constraint condition to obtain an optimized constraint condition.

[0073] (3) Beneficial effects

[0074] Compared with the prior art, the present invention has the following beneficial effects:

[0075] 1. By considering the brake temperature rise, speed smoothing value, recovered energy and energy consumption factors in the objective function, the brake energy recovery time of hydrogen-powered vehicles is optimized, thereby improving the energy recovery efficiency of hydrogen-powered vehicles and overcoming the adverse effects of brake energy recovery of hydrogen-powered vehicles on driving safety and vehicle service life.

[0076] 2. The present invention divides the planned driving path into N driving sections, and uses the first section to the Nth section as calculation units to construct the first optimization model to the Nth optimization model, and then uses the particle swarm optimization algorithm to solve them respectively. Compared with the traditional method of optimizing and solving the entire planned driving path as a whole, the calculation complexity is reduced and the calculation efficiency is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 This is a flowchart of an energy recovery method for a hydrogen energy vehicle according to Embodiment 1 of the present invention;

[0078] Figure 2This is a schematic diagram of the module composition of an energy recovery system for a hydrogen-powered vehicle according to Example 2 of the present invention. DETAILED DESCRIPTION

[0079] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0080] Before giving examples, it is necessary to explain the application scenarios conceived by the present invention. The present invention is used to optimize the operating parameters of autonomous driving hydrogen-powered vehicles when the roads in the enterprise's internal park are unobstructed, thereby improving the energy recovery efficiency of hydrogen-powered vehicles while overcoming the adverse effects of braking energy recovery of hydrogen-powered vehicles on driving safety and vehicle service life.

[0081] Example 1: Figure 1 As shown, this embodiment provides an energy recovery method for a hydrogen energy vehicle, the method comprising the following steps:

[0082] S1, obtain first data; the first data includes the initial weight of the hydrogen energy vehicle, the initial remaining power of the hydrogen fuel cell, the planned driving route, the planned driving distance, the number of traffic lights on the planned driving route, the location of the traffic lights on the planned driving route, and the average waiting time of the traffic lights on the planned driving route.

[0083] S2, taking the position of the traffic lights on the planned driving path as the segmentation point, dividing the planned driving path into N driving sections, which are respectively recorded as the first section to the Nth section; mapping the first section to the Nth section to a two-dimensional coordinate plane, and calculating the length of the first section to the Nth section.

[0084] S3, taking the first to Nth road sections as calculation units, respectively, constructing the first to Nth optimization models, the optimization objectives of the first to Nth optimization models are the first objective function to the Nth objective function respectively; taking the maximum value of the first to Nth objective functions as the goal, using the particle swarm optimization algorithm to solve the first to Nth optimization models, to obtain the optimal values ​​of the first to Nth braking energy recovery time, recorded as the first optimal braking energy recovery time to the Nth optimal braking energy recovery time.

[0085] S4, outputting the first optimal braking energy recovery time to the Nth optimal braking energy recovery time to the hydrogen energy vehicle energy management system.

[0086] Exemplarily, the initial weight of the hydrogen energy vehicle can be obtained by querying the vehicle parameter information, the initial remaining power of the hydrogen fuel cell can be obtained by querying the battery management system of the hydrogen energy vehicle, and the planned driving path, the planned driving distance, the number of traffic lights on the planned driving path, the location of the traffic lights on the planned driving path, and the average waiting time of the traffic lights on the planned driving path can be pre-set by the user or obtained by big data analysis of the user's historical driving records. It is worth noting that the present invention is used for planning the situation where there are no other driving obstacles except traffic lights on the driving path, and is suitable for driving planning of internal roads in industrial parks. Therefore, traffic lights are the only factor that causes hydrogen energy vehicles to brake and stop. According to the location of the traffic lights on the planned driving path, the planned driving path is divided into N driving sections. In each driving section, the hydrogen energy vehicle only needs to perform one acceleration movement, one uniform motion and one braking at most. The optimization model is constructed one by one in the N driving sections, and the particle swarm optimization algorithm is used for optimization and solution, and the first optimal braking energy recovery time to the Nth optimal braking energy recovery time can be obtained.

[0087] Exemplarily, the initial weight of the hydrogen energy vehicle is 1437 kg, the initial remaining power of the hydrogen fuel cell is 4 kWh, i.e. 14400000 joules, the planned driving route is from Office Building 1 to Warehouse 3 in the industrial park, the planned driving distance is 4700 meters, the number of traffic lights on the planned driving route is 2, the positions of the traffic lights on the planned driving route are respectively 1500 meters and 3200 meters from the starting point, and the average waiting time of the traffic lights on the planned driving route is 32 seconds. Therefore, the planned driving route is divided into 3 driving sections, N=3. The length of the first section is 1500 meters, the length of the second section is 1700 meters, and the length of the third section is 1500 meters.

[0088] Furthermore, the method of constructing first to Nth optimization models by taking the first to Nth road sections as calculation units, wherein the optimization targets of the first to Nth optimization models are first to Nth objective functions, respectively, includes:

[0089] According to the first data, the first objective function to the Nth objective function are constructed, and the first function, the second function, the third function, the fifth function, the first recovery energy to the Nth recovery energy, and the first energy consumption to the Nth energy consumption are obtained; the independent variables of the first objective function to the Nth objective function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time.

[0090] Optimization constraints are constructed based on the first data, the third function, the first to Nth recovered energies, and the first to Nth energy consumptions.

[0091] The first to Nth objective functions are respectively combined with the optimization constraints to obtain first to Nth optimization models.

[0092] Further, the method of constructing the first objective function to the Nth objective function according to the first data, and obtaining the first function, the second function, the third function, the fifth function, the first recovered energy to the Nth recovered energy, and the first energy consumption to the Nth energy consumption includes:

[0093] The functional relationship between the first recovery energy to the Nth recovery energy and the first acceleration time to the Nth acceleration time, and the first start acceleration to the Nth start acceleration is constructed by the recovery energy calculation formula, which is recorded as the first function p i =H1(a i , t ci ); the recovery energy calculation formula is:

[0094]

[0095] Among them, p i is the i-th recovered energy, in joules; a i is the i-th starting acceleration, in meters per second squared; t ci is the i-th acceleration time, in seconds; m is the initial weight of the hydrogen energy vehicle, in kilograms; η is the preset braking energy recovery efficiency; i is an integer variable with a value from 1 to N.

[0096] For example, the initial weight of a hydrogen-powered vehicle is 1437 kg, and the preset braking energy recovery efficiency is 0.17. Substituting it into the recovery energy calculation formula, we get:

[0097] p1=H1(a1,t c1 )=122.145(a1t c1 ) 2 ;

[0098] p2=H1(a2,t c2 )=122.145(a2t c2 ) 2 ;

[0099] p3=H1(a3,t c3 )=122.145(a3t c3 ) 2 .

[0100] According to the preset first mapping relationship, a functional relationship between the first energy consumption to the Nth energy consumption and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is constructed, which is recorded as the second function r i =H2(a i , t ci , t di ), where r i is the energy consumption of the ith energy, in joules; t di is the i-th braking energy recovery time, in seconds.

[0101] Exemplarily, the first mapping relationship can be obtained by fitting a neural network algorithm. The specific steps are: the operation data of new energy vehicles of the same model can be collected to obtain data samples of energy consumption, acceleration time, acceleration, and braking time. The sample size of the data samples should be greater than 500. Through the neural network algorithm fitting, the first mapping relationship is established, and finally the second function is obtained:

[0102] r1=H2(a1,t c1 , t d1 );

[0103] r2=H2(a2,t c2 , t d2 );

[0104] r3=H2(a3,t c3 , t d3 ).

[0105] According to the preset acceleration and speed mapping relationship, a functional relationship between the first speed to the Nth speed and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is constructed, which is recorded as the third function v i (t) = H3(a i , t ci , t di ), where v i (t) is the i-th speed, in meters per second; t is the driving time, in seconds; the third function is discretized with a preset first interval Δt as the time interval to obtain the first speed sequence to the N-th speed sequence; according to the first speed sequence to the N-th speed sequence, the first speed smoothing value to the N-th speed smoothing value are obtained by the speed smoothing value calculation formula; the speed smoothing value calculation formula is:

[0106]

[0107] Among them, z i is the i-th speed smoothing value; is the mean value of the ith velocity, in meters per second; n i is the number of data in the i-th speed sequence; j i The value is 1 to n i Integer variable; x i (j i ) is the jth in the i-th speed sequence i elements.

[0108] Exemplarily, the first speed to the third speed are variables that change with time. In order to realize data processing, they need to be discretized with a preset first interval Δt as the time interval, and finally obtain the first speed smoothing value z1, the second speed smoothing value z2, and the third speed smoothing value z3.

[0109] A two-layer neural network model is established based on the first speed average to the Nth speed average and the first braking energy recovery time to the Nth braking energy recovery time. Through model training and function inverse transformation, the functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is obtained, which is recorded as the fifth function w i =H5(a i , t ci , t di ), where w i is the ith brake temperature rise, in Kelvin.

[0110] Exemplarily, since N=3, the fifth function obtained is:

[0111] w1=H5(a1,t c1 , t d1 );

[0112] w2=H5(a2,t c2 , t d2 );

[0113] w3=H5(a3,t c3 , t d3 ).

[0114] According to the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the first speed smoothing value to the Nth speed smoothing value, the first objective function to the Nth objective function are constructed through the objective function construction formula; the objective function construction formula is:

[0115]

[0116] Wherein, k1 is a preset first coefficient; k2 is a preset second coefficient; k3 is a preset third coefficient; Gi (a i , t ci , t di ) is the i-th objective function.

[0117] For example, the values ​​of the first coefficient, the second coefficient, and the third coefficient comprehensively reflect the dimensions and emphasis of the recovered energy, energy consumption, and speed smoothing value, and can be preset by the user. Take k1=3270000, k2=23, and k3=107. Therefore, the first objective function to the third objective function are respectively:

[0118]

[0119] Furthermore, a two-layer neural network model is established according to the first speed average value to the Nth speed average value and the first braking energy recovery time to the Nth braking energy recovery time, and a functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is obtained through model training and function inverse transformation, which is recorded as the fifth function w i =H5(a i , t ci , t di ), where w i The methods for calculating the i-th brake temperature rise in Kelvin include:

[0120] A two-layer neural network model is constructed, wherein the input of the two-layer neural network model is one of the first speed average to the Nth speed average and one of the first braking energy recovery time to the Nth braking energy recovery time; the output of the two-layer neural network model is the braking temperature rise obtained by subtracting the temperature at the time when the hydrogen energy vehicle completes braking from the temperature at the time when the hydrogen energy vehicle starts braking; the two-layer neural network model is trained using a pre-set sample set to obtain a first neural network model; and the functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first speed average to the Nth speed average and the first braking energy recovery time to the Nth braking energy recovery time is obtained by fitting the first neural network model, which is recorded as a fourth function.

[0121] The speed smoothing value calculation formula and the third function are respectively inversely transformed and combined with the fourth function to obtain the functional relationship between the first brake temperature rise to the Nth brake temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first brake energy recovery time to the Nth brake energy recovery time.

[0122] Exemplarily, the inverse transformation of the third function can be achieved by inverse function transformation. The inverse transformation of the speed smoothing value calculation formula needs to be firstly numerically fitted and then achieved by inverse function transformation.

[0123] Furthermore, the method for constructing optimization constraints according to the first data, the third function, the first recovered energy to the Nth recovered energy, and the first energy consumption to the Nth energy consumption includes:

[0124] According to the third function, the number of traffic lights on the planned driving path, and the average waiting time of traffic lights on the planned driving path, a sixth function is obtained through a preset speed-distance mapping relationship; the independent variables of the sixth function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time; the dependent variable of the sixth function is the driving distance S e .

[0125] A first constraint condition is constructed according to the driving distance and the planned driving distance; the first constraint condition is:

[0126] S e ≥S z ;

[0127] Among them, S z It is the planned driving distance in meters.

[0128] For example, since the planned driving distance is 4700 meters, the first constraint is S e ≥4700.

[0129] The second constraint condition is constructed based on the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the initial remaining power of the hydrogen fuel cell; the second constraint condition is:

[0130]

[0131] Where c0 is the initial remaining power of the hydrogen fuel cell, in joules; p i is the i-th recovered energy, in joules; r i is the i-th energy consumption in joules; i is an integer variable ranging from 1 to N.

[0132] Exemplarily, the second constraint condition is obtained as follows:

[0133]

[0134] The first constraint condition is combined with the second constraint condition to obtain an optimized constraint condition.

[0135] Embodiment 2: Based on the same inventive concept, Figure 2 As shown, this embodiment also provides an energy recovery system for a hydrogen energy vehicle, the system comprising:

[0136] The data reading module is used to obtain the first data; the first data includes the initial weight of the hydrogen energy vehicle, the initial remaining power of the hydrogen fuel cell, the planned driving route, the planned driving distance, the number of traffic lights on the planned driving route, the location of the traffic lights on the planned driving route, and the average waiting time of the traffic lights on the planned driving route.

[0137] The road segmentation module is connected to the data reading module and is used to divide the planned driving path into N driving segments using the traffic light positions on the planned driving path as segmentation points, and respectively record them as the first segment to the Nth segment; the first segment to the Nth segment are mapped to a two-dimensional coordinate plane, and the length of the first segment to the Nth segment is calculated.

[0138] The optimization solution module is connected to the road section division module, and is used to construct the first optimization model to the Nth optimization model with the first section to the Nth section as calculation units, respectively, and the optimization objectives of the first optimization model to the Nth optimization model are the first objective function to the Nth objective function respectively; with the maximum value of the first objective function to the Nth objective function as the goal, the first optimization model to the Nth optimization model are solved by using the particle swarm optimization algorithm to obtain the optimal value of the first braking energy recovery time to the Nth braking energy recovery time, which is recorded as the first optimal braking energy recovery time to the Nth optimal braking energy recovery time.

[0139] The data output module is connected to the optimization solution module and is used to output the first optimal braking energy recovery time to the Nth optimal braking energy recovery time to the hydrogen energy vehicle energy management system.

[0140] Furthermore, the optimization solution module includes:

[0141] An objective function construction module is used to construct the first objective function to the Nth objective function according to the first data, and obtain the first function, the second function, the third function, the fifth function, the first recovery energy to the Nth recovery energy, and the first energy consumption to the Nth energy consumption; the independent variables of the first objective function to the Nth objective function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time.

[0142] The constraint condition construction module is connected to the objective function construction module, and is used to construct optimization constraint conditions according to the first data, the third function, the first recovered energy to the Nth recovered energy, and the first energy consumption to the Nth energy consumption.

[0143] The optimization model construction module is connected to the constraint condition construction module, and is used to combine the first to Nth objective functions with the optimization constraint conditions respectively to obtain the first to Nth optimization models.

[0144] Furthermore, the objective function construction module includes:

[0145] The first function construction module is used to construct the functional relationship between the first recovery energy to the Nth recovery energy and the first acceleration time to the Nth acceleration time, and the first start acceleration to the Nth start acceleration through the recovery energy calculation formula, which is recorded as the first function p i =H1(a i , t ci ); the recovery energy calculation formula is:

[0146]

[0147] Among them, p i is the i-th recovered energy, in joules; a i is the i-th starting acceleration, in meters per second squared; t ci is the i-th acceleration time, in seconds; m is the initial weight of the hydrogen energy vehicle, in kilograms; η is the preset braking energy recovery efficiency; i is an integer variable with a value from 1 to N.

[0148] The second function construction module is connected to the first function construction module and is used to construct a functional relationship between the first energy consumption to the Nth energy consumption and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time according to a preset first mapping relationship, which is recorded as the second function r i =H2(a i , t ci , t di ), where r i is the energy consumption of the ith energy, in joules; t di is the i-th braking energy recovery time, in seconds.

[0149] The speed smoothing value calculation module is connected to the second function construction module and is used to construct a functional relationship between the first speed to the Nth speed and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time according to a preset acceleration and speed mapping relationship, which is recorded as a third function v i (t) = H3(a i , t ci , t di ), where v i(t) is the i-th speed, in meters per second; t is the driving time, in seconds; the third function is discretized with a preset first interval Δt as the time interval to obtain the first speed sequence to the N-th speed sequence; according to the first speed sequence to the N-th speed sequence, the first speed smoothing value to the N-th speed smoothing value are obtained by the speed smoothing value calculation formula; the speed smoothing value calculation formula is:

[0150]

[0151] Among them, z i is the i-th speed smoothing value; is the mean value of the ith velocity, in meters per second; n i is the number of data in the i-th speed sequence; j i The value is 1 to n i Integer variable; x i (j i ) is the jth in the i-th speed sequence i elements.

[0152] The fifth function construction module is connected to the speed smoothing value calculation module, and is used to establish a two-layer neural network model according to the first speed average value to the Nth speed average value and the first braking energy recovery time to the Nth braking energy recovery time, and obtain the functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time through model training and function inverse transformation, which is recorded as the fifth function w i =H5(a i , t ci , t di ), where w i is the ith brake temperature rise, in Kelvin.

[0153] The objective function construction module is connected to the fifth function construction module, and is used to construct the first objective function to the Nth objective function according to the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the first speed smoothing value to the Nth speed smoothing value through the objective function construction formula; the objective function construction formula is:

[0154]

[0155] Wherein, k1 is a preset first coefficient; k2 is a preset second coefficient; k3 is a preset third coefficient; G i (a i , t ci , t di ) is the i-th objective function.

[0156] Furthermore, the fifth function construction module includes:

[0157] The fourth function construction module is used to construct a two-layer neural network model, the input of the two-layer neural network model is one of the first speed average to the Nth speed average and one of the first braking energy recovery time to the Nth braking energy recovery time; the output of the two-layer neural network model is the braking temperature rise obtained by subtracting the temperature at the time when the hydrogen energy vehicle completes braking from the temperature at the time when the hydrogen energy vehicle starts braking; the two-layer neural network model is trained using a pre-set sample set to obtain a first neural network model; according to the first neural network model, the functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first speed average to the Nth speed average and the first braking energy recovery time to the Nth braking energy recovery time is fitted, which is recorded as the fourth function.

[0158] The inverse transformation joint module is connected to the fourth function construction module, and is used to respectively perform inverse transformation on the speed smoothing value calculation formula and the third function, and to combine them with the fourth function to obtain the functional relationship between the first brake temperature rise to the Nth brake temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first brake energy recovery time to the Nth brake energy recovery time.

[0159] Furthermore, the constraint condition construction module includes:

[0160] The first constraint condition construction module is used to obtain a sixth function through a preset speed-distance mapping relationship according to the third function, the number of traffic lights on the planned driving path, and the average waiting time of the traffic lights on the planned driving path; the independent variables of the sixth function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time; the dependent variable of the sixth function is the driving distance S e .

[0161] A first constraint condition is constructed according to the driving distance and the planned driving distance; the first constraint condition is:

[0162] S e ≥S z ;

[0163] Among them, S z It is the planned driving distance in meters.

[0164] The second constraint condition construction module is connected to the first constraint condition construction module and is used to construct a second constraint condition according to the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the initial remaining power of the hydrogen fuel cell; the second constraint condition is:

[0165]

[0166] Where c0 is the initial remaining power of the hydrogen fuel cell, in joules; p i is the i-th recovered energy, in joules; ri is the i-th energy consumption, in joules; i is an integer variable with values ​​from 1 to N.

[0167] The constraint condition combination module is connected to the second constraint condition construction module and is used to combine the first constraint condition with the second constraint condition to obtain an optimized constraint condition.

[0168] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0169] Finally, it should be noted that: Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for recovering energy from a hydrogen-powered vehicle, characterized in that: The method comprises the following steps: S1, obtaining first data; the first data includes the initial weight of the hydrogen energy vehicle, the initial remaining power of the hydrogen fuel cell, the planned driving route, the planned driving distance, the number of traffic lights on the planned driving route, the location of the traffic lights on the planned driving route, and the average waiting time of the traffic lights on the planned driving route; S2, using the traffic light positions on the planned driving path as segmentation points, segmenting the planned driving path to obtain N driving sections, which are respectively recorded as the first section to the Nth section; mapping the first section to the Nth section to a two-dimensional coordinate plane, and calculating the length of the first section to the Nth section; S3, respectively using the first to Nth road sections as calculation units, constructing first to Nth optimization models, wherein the optimization objectives of the first to Nth optimization models are the first objective function to the Nth objective function respectively; taking the maximum value of the first to Nth objective functions as the goal, solving the first to Nth optimization models by using a particle swarm optimization algorithm, and obtaining the optimal values ​​of the first to Nth braking energy recovery time, which are recorded as the first optimal braking energy recovery time to the Nth optimal braking energy recovery time; S4, outputting the first optimal braking energy recovery time to the Nth optimal braking energy recovery time to the hydrogen energy vehicle energy management system.

2. The energy recovery method for a hydrogen-powered vehicle according to claim 1, characterized in that: The method of constructing first to Nth optimization models by taking the first to Nth road sections as calculation units respectively, wherein the optimization targets of the first to Nth optimization models are first to Nth objective functions respectively, comprises: Constructing the first objective function to the Nth objective function according to the first data, and obtaining the first function, the second function, the third function, the fifth function, the first recovery energy to the Nth recovery energy, and the first energy consumption to the Nth energy consumption; the independent variables of the first objective function to the Nth objective function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time; Constructing optimization constraints according to the first data, the third function, the first recovered energy to the Nth recovered energy, and the first energy consumption to the Nth energy consumption; The first to Nth objective functions are respectively combined with the optimization constraints to obtain first to Nth optimization models.

3. The energy recovery method for a hydrogen-powered vehicle according to claim 2, characterized in that: The method of constructing the first objective function to the Nth objective function according to the first data and obtaining the first function, the second function, the third function, the fifth function, the first recovered energy to the Nth recovered energy, and the first energy consumption to the Nth energy consumption includes: The functional relationship between the first recovery energy to the Nth recovery energy and the first acceleration time to the Nth acceleration time, and the first start acceleration to the Nth start acceleration is constructed by the recovery energy calculation formula, which is recorded as the first function p i =H1(a i , t ci ); the recovery energy calculation formula is: Among them, p i is the i-th recovered energy, in joules; a i is the i-th starting acceleration, in meters per second squared; t ci is the i-th acceleration time, in seconds; m is the initial weight of the hydrogen energy vehicle, in kilograms; η is the preset braking energy recovery efficiency; i is an integer variable with a value from 1 to N; According to the preset first mapping relationship, a functional relationship between the first energy consumption to the Nth energy consumption and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is constructed, which is recorded as the second function r i =H2(a i , t ci , t di ), where r i is the energy consumption of the ith energy, in joules; t di is the i-th braking energy recovery time, in seconds; According to the preset acceleration and speed mapping relationship, a functional relationship between the first speed to the Nth speed and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is constructed, which is recorded as the third function v i (t) = H3(a i , t ci , t di ), where v i (t) is the i-th speed, in meters per second; t is the driving time, in seconds; the third function is discretized with a preset first interval Δt as the time interval to obtain the first speed sequence to the N-th speed sequence; according to the first speed sequence to the N-th speed sequence, the first speed smoothing value to the N-th speed smoothing value are obtained by the speed smoothing value calculation formula; the speed smoothing value calculation formula is: Among them, z i is the i-th speed smoothing value; is the mean value of the ith velocity, in meters per second; n i is the number of data in the i-th speed sequence; j i The value is 1 to n i Integer variable; x i (j i ) is the jth in the i-th speed sequence i elements; A two-layer neural network model is established based on the first speed average to the Nth speed average and the first braking energy recovery time to the Nth braking energy recovery time. Through model training and function inverse transformation, the functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is obtained, which is recorded as the fifth function w i =H5(a i , t ci , t di ), where w i is the i-th brake temperature rise, in Kelvin; According to the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the first speed smoothing value to the Nth speed smoothing value, the first objective function to the Nth objective function are constructed through the objective function construction formula; the objective function construction formula is: Wherein, k1 is a preset first coefficient; k2 is a preset second coefficient; k3 is a preset third coefficient; G i (a i , t ci , t di ) is the i-th objective function.

4. The energy recovery method for a hydrogen-powered vehicle according to claim 3, characterized in that: A two-layer neural network model is established according to the first speed average to the Nth speed average and the first braking energy recovery time to the Nth braking energy recovery time, and a functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time is obtained through model training and function inverse transformation, which is recorded as the fifth function w i =H5(a i , t ci , t di ), where w i The methods for calculating the i-th brake temperature rise in Kelvin include: Construct a two-layer neural network model, wherein the input of the two-layer neural network model is one of the first speed average to the Nth speed average and one of the first braking energy recovery time to the Nth braking energy recovery time; the output of the two-layer neural network model is the braking temperature rise obtained by subtracting the temperature at the time when the hydrogen energy vehicle completes braking from the temperature at the time when the hydrogen energy vehicle starts braking; the two-layer neural network model is trained using a preset sample set to obtain a first neural network model; according to the first neural network model, a functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first speed average to the Nth speed average and the first braking energy recovery time to the Nth braking energy recovery time is obtained by fitting, which is recorded as a fourth function; The speed smoothing value calculation formula and the third function are respectively inversely transformed and combined with the fourth function to obtain the functional relationship between the first brake temperature rise to the Nth brake temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first brake energy recovery time to the Nth brake energy recovery time.

5. The energy recovery method for a hydrogen-powered vehicle according to claim 2, characterized in that: The method for constructing optimization constraints according to the first data, the third function, the first to Nth recovered energies, and the first to Nth energy consumptions comprises: According to the third function, the number of traffic lights on the planned driving path, and the average waiting time of traffic lights on the planned driving path, a sixth function is obtained through a preset speed-distance mapping relationship; the independent variables of the sixth function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time; the dependent variable of the sixth function is the driving distance S e ; A first constraint condition is constructed according to the driving distance and the planned driving distance; the first constraint condition is: S e ≥S z ; Among them, S z is the planned driving distance in meters; The second constraint condition is constructed based on the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the initial remaining power of the hydrogen fuel cell; the second constraint condition is: Where c0 is the initial remaining power of the hydrogen fuel cell, in joules; p i is the i-th recovered energy, in joules; r i is the energy consumption of the ith energy in joules; i is an integer variable ranging from 1 to N; The first constraint condition is combined with the second constraint condition to obtain an optimized constraint condition.

6. An energy recovery system for a hydrogen-powered vehicle, characterized in that: The system comprises: A data reading module, used to obtain first data; the first data includes the initial weight of the hydrogen energy vehicle, the initial remaining power of the hydrogen fuel cell, the planned driving route, the planned driving distance, the number of traffic lights on the planned driving route, the location of the traffic lights on the planned driving route, and the average waiting time of the traffic lights on the planned driving route; A road segmentation module is connected to the data reading module and is used to segment the planned driving path using the traffic light position on the planned driving path as a segmentation point to obtain N driving segments, which are respectively recorded as the first segment to the Nth segment; the first segment to the Nth segment are mapped to a two-dimensional coordinate plane, and the length of the first segment to the Nth segment is calculated; an optimization solution module, connected to the road section division module, for constructing a first optimization model to an Nth optimization model with the first to Nth road sections as calculation units, respectively, wherein the optimization targets of the first to Nth optimization models are the first objective function to the Nth objective function, respectively; taking the maximum value of the first to Nth objective functions as the target, solving the first to Nth optimization models by using a particle swarm optimization algorithm, and obtaining the optimal values ​​of the first to Nth braking energy recovery time, which are recorded as the first optimal braking energy recovery time to the Nth optimal braking energy recovery time; The data output module is connected to the optimization solution module and is used to output the first optimal braking energy recovery time to the Nth optimal braking energy recovery time to the hydrogen energy vehicle energy management system.

7. The energy recovery system for a hydrogen-powered vehicle as claimed in claim 6, characterized in that: The optimization solution module includes: an objective function construction module, for constructing the first objective function to the Nth objective function according to the first data, and obtaining the first function, the second function, the third function, the fifth function, the first recovery energy to the Nth recovery energy, and the first energy consumption to the Nth energy consumption; the independent variables of the first objective function to the Nth objective function are respectively the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time; A constraint condition construction module, connected to the objective function construction module, for constructing an optimization constraint condition according to the first data, the third function, the first recovered energy to the Nth recovered energy, and the first energy consumption to the Nth energy consumption; The optimization model construction module is connected to the constraint condition construction module, and is used to combine the first to Nth objective functions with the optimization constraint conditions respectively to obtain the first to Nth optimization models.

8. The energy recovery system for a hydrogen-powered vehicle as claimed in claim 7, characterized in that: The objective function construction module comprises: The first function construction module is used to construct the functional relationship between the first recovery energy to the Nth recovery energy and the first acceleration time to the Nth acceleration time, and the first start acceleration to the Nth start acceleration through the recovery energy calculation formula, which is recorded as the first function p i =H1(a i , t ci ); the recovery energy calculation formula is: Among them, p i is the i-th recovered energy, in joules; a i is the i-th starting acceleration, in meters per second squared; t ci is the i-th acceleration time, in seconds; m is the initial weight of the hydrogen energy vehicle, in kilograms; η is the preset braking energy recovery efficiency; i is an integer variable with a value from 1 to N; The second function construction module is connected to the first function construction module and is used to construct a functional relationship between the first energy consumption to the Nth energy consumption and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time according to a preset first mapping relationship, which is recorded as the second function r i =H2(a i , t ci , t di ), where r i is the energy consumption of the ith energy, in joules; t di is the i-th braking energy recovery time, in seconds; The speed smoothing value calculation module is connected to the second function construction module and is used to construct a functional relationship between the first speed to the Nth speed and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time according to a preset acceleration and speed mapping relationship, which is recorded as a third function v i (t) = H3(a i , t ci , t di ), where v i (t) is the i-th speed, in meters per second; t is the driving time, in seconds; the third function is discretized with a preset first interval Δt as the time interval to obtain the first speed sequence to the N-th speed sequence; according to the first speed sequence to the N-th speed sequence, the first speed smoothing value to the N-th speed smoothing value are obtained by the speed smoothing value calculation formula; the speed smoothing value calculation formula is: Among them, z i is the i-th speed smoothing value; is the mean value of the ith velocity, in meters per second; n i is the number of data in the i-th speed sequence; j i The value is 1 to n i Integer variable; x i (j i ) is the jth in the i-th speed sequence i elements; The fifth function construction module is connected to the speed smoothing value calculation module, and is used to establish a two-layer neural network model according to the first speed average value to the Nth speed average value and the first braking energy recovery time to the Nth braking energy recovery time, and obtain the functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time through model training and function inverse transformation, which is recorded as the fifth function w i =H5(a i , t ci , t di ), where w i is the i-th brake temperature rise, in Kelvin; The objective function construction module is connected to the fifth function construction module, and is used to construct the first objective function to the Nth objective function according to the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the first speed smoothing value to the Nth speed smoothing value through the objective function construction formula; the objective function construction formula is: Wherein, k1 is a preset first coefficient; k2 is a preset second coefficient; k3 is a preset third coefficient; G i (a i , t ci , t di ) is the i-th objective function.

9. The energy recovery system for a hydrogen-powered vehicle as claimed in claim 8, characterized in that: The fifth function construction module comprises: The fourth function construction module is used to construct a two-layer neural network model, wherein the input of the two-layer neural network model is one of the first speed mean to the Nth speed mean and one of the first braking energy recovery time to the Nth braking energy recovery time; the output of the two-layer neural network model is the braking temperature rise obtained by subtracting the temperature at the time when the hydrogen energy vehicle completes braking from the temperature at the time when the hydrogen energy vehicle starts braking; the two-layer neural network model is trained using a preset sample set to obtain a first neural network model; according to the first neural network model, a functional relationship between the first braking temperature rise to the Nth braking temperature rise and the first speed mean to the Nth speed mean and the first braking energy recovery time to the Nth braking energy recovery time is obtained by fitting, which is recorded as a fourth function; The inverse transformation joint module is connected to the fourth function construction module, and is used to respectively perform inverse transformation on the speed smoothing value calculation formula and the third function, and to combine them with the fourth function to obtain the functional relationship between the first brake temperature rise to the Nth brake temperature rise and the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first brake energy recovery time to the Nth brake energy recovery time.

10. The energy recovery system for a hydrogen-powered vehicle according to claim 7, characterized in that: The constraint condition construction module includes: The first constraint condition construction module is used to obtain a sixth function through a preset speed-distance mapping relationship according to the third function, the number of traffic lights on the planned driving path, and the average waiting time of the traffic lights on the planned driving path; the independent variables of the sixth function are the first acceleration time to the Nth acceleration time, the first starting acceleration to the Nth starting acceleration, and the first braking energy recovery time to the Nth braking energy recovery time; the dependent variable of the sixth function is the driving distance S e ; A first constraint condition is constructed according to the driving distance and the planned driving distance; the first constraint condition is: S e ≥S z ; Among them, S z is the planned driving distance in meters; The second constraint condition construction module is connected to the first constraint condition construction module and is used to construct a second constraint condition according to the first recovered energy to the Nth recovered energy, the first energy consumption to the Nth energy consumption, and the initial remaining power of the hydrogen fuel cell; the second constraint condition is: Where c0 is the initial remaining power of the hydrogen fuel cell, in joules; p i is the i-th recovered energy, in joules; r i is the energy consumption of the ith energy in joules; i is an integer variable ranging from 1 to N; The constraint condition combination module is connected to the second constraint condition construction module and is used to combine the first constraint condition with the second constraint condition to obtain an optimized constraint condition.

Images