Fuel cell hybrid electric heavy truck power source optimization design method and system

Abstract

The invention discloses a fuel cell hybrid electric heavy truck power source optimization design method and system, and the method comprises the steps: building a target function of the total cost per unit travelled distance in a life cycle based on the basic parameters of a fuel cell, a power cell and a super capacitor, and the degradation parameters and hydrogen consumption of the fuel cell and the power cell; an energy management strategy of the fuel cell hybrid electric heavy truck is established based on the average required power of the truck and the SOC of the power cell and the super capacitor; and a cancer cell competition and transfer algorithm is adopted, under an energy management strategy, a minimum objective function meeting the driving working condition is searched, and energy size optimization is completed. According to the method, the power source of the fuel cell hybrid electric heavy truck is subjected to global optimization design by adopting the cancer cell competition and transfer algorithm, the service life of the power source is prolonged while the power performance of the truck is guaranteed, the manufacturing cost, degradation cost and hydrogen consumption of a hybrid power system are remarkably reduced, and the energy efficiency of the system is improved.

Description

Technical Field

[0001] The present invention relates to the field of energy management, and in particular to a method and system for optimizing the design of a power source of a fuel cell hybrid electric heavy truck. Background Art

[0002] Electrification is a necessary way to decarbonize the transportation industry. At present, the electrification technology routes for heavy-duty trucks are mainly divided into two types: lithium batteries and fuel cells (FC). In order to meet the 500km range, a 49-ton pure electric heavy truck needs to be equipped with a 1000kWh lithium battery. The battery weighs about 5 tons, which greatly affects the ability of heavy trucks to load goods. Compared with pure lithium heavy trucks, fuel cells can ensure longer endurance under the premise of similar quality to internal combustion engines, and are more suitable for long-distance transportation needs. At the same time, the hydrogen refueling time of fuel cells is relatively short, only a few minutes, while the charging time of lithium batteries is very long, which affects the logistics efficiency. Therefore, fuel cell systems are more suitable as a power source for heavy trucks. Compared with traditional fuel vehicles, the energy density of hydrogen used in fuel cells is about 3 times that of gasoline, and the emissions are water vapor, which significantly reduces the emission of room temperature gases and other pollutants and slows down climate deterioration.

[0003] Fuel cell vehicles can be divided into two categories according to the energy source: All fuel cell vehicles (AFCV) and fuel cell hybrid vehicles (FCHV). The former is powered entirely by fuel cells, while the latter is powered by fuel cells and energy storage systems. Energy storage systems generally include power batteries (B) or supercapacitors (SC), or hybrid energy storage systems (HESS) consisting of power batteries and supercapacitors. Pure fuel cell vehicles respond slowly to the instantaneous power demand of the vehicle, affecting the starting and acceleration performance of the vehicle. For heavy trucks that travel at high loads for a long time, a single fuel cell system faces severe temperature management and life degradation problems. In addition, the fuel cell system cannot recover the braking energy of the vehicle, and the overall energy efficiency of the vehicle is low. Therefore, in order to improve the instantaneous power response of fuel cell vehicles, extend their life and recover the braking energy of the vehicle, a fuel cell hybrid system is a good choice.

[0004] The size of the power source in a fuel cell hybrid vehicle has a great impact on the initial cost, operating cost, life of the power source and energy management of the vehicle. Many scholars have done a lot of research on the size design of fuel cell hybrid vehicles, which are mainly divided into FC+B and FC+B+SC. Min-Joong Kim et al. first studied the optimal size of the FC+B power source, and proposed a subsystem scaling model and a parameterizable power management strategy, and incorporated them into the system optimization as design variables. The simulation showed that this method can provide better fuel economy. In addition, other researchers have proposed an optimal size method for plug-in FC+B-driven buses. In this work, L.Xu et al. proposed a theoretical model to describe the relationship between component parameters and vehicle performance, adopted a charge depletion and charge sustaining (CDCS) strategy for energy management, and used the contour map method to optimize the energy size to reduce system cost and operating cost. The above studies on the optimization of FC+B system size did not consider the life of the system. Summary of the invention

[0005] In order to solve the technical problems in the above background, the present invention proposes a new energy size optimization method for a fuel cell hybrid heavy-duty tractor based on the characteristics of three power sources.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A fuel cell hybrid electric heavy truck power source optimization design method, the steps comprising:

[0008] Based on the basic parameters of fuel cells, power batteries and supercapacitors, the degradation parameters of fuel cells and power batteries and hydrogen consumption, an objective function of the total cost per unit mileage during the life cycle is constructed;

[0009] Based on the average required power of the vehicle, the SOC of the power battery and supercapacitor, an energy management strategy for a fuel cell hybrid electric heavy truck is established.

[0010] The cancer cell competition and transfer algorithm is used to find the minimum objective function under the driving condition under the energy management strategy to complete the energy size optimization.

[0011] Preferably, the objective function constructed includes:

[0012]

[0013] Cost in =N fc ·c fc +N bs ·Nbp ·c b +N scs ·N scp ·c sc

[0014] N cycle =(N fc ·c fc +N bs ·N bp ·c b ) / (cost fcdeg +cost bdeg )

[0015] Cost fcdeg =(D fc / 0.1)·N fc ·c fc

[0016] Cost bdeg =(Q b_loss / 0.2)·N bs ·N bp ·c b

[0017] X total =N cycle ·x cycle

[0018] Cost H2 =cost H2 ·N cycle

[0019] in, Indicates that every 10 4 km total cost consumed; Cost in Indicates initial cost; Cost H2 represents the cost of hydrogen use during the life cycle; X total Indicates the total mileage during the life cycle; N fc Indicates the number of fuel cell units; c fc Represents the single cell cost of the fuel cell; N bs and N bp Respectively represent the number of batteries in series and parallel; c b Represents the cost of a single battery; N scs and N scp Respectively represent the number of supercapacitors in series and in parallel; c sc Represents the cost of a single supercapacitor; cost fcdeg and cost bdeg Respectively represent the degradation cost of fuel cell and battery in a single cycle condition; N cycleThe number of cycles representing the total mileage; x cycle Indicates the mileage of a single cycle; cost H2 is the hydrogen consumption in a single cycle.

[0020] Preferably, the energy management strategy established includes the required power of fuel cells and hybrid energy storage:

[0021] P fc =αP avg +βP fc_Eff +(1-α-β)P fc_Max

[0022]

[0023] Among them, P avg It represents the average power demand during the total running time from vehicle startup to the current moment; P fc_Eff Indicates the power at the highest efficiency point of the fuel cell; P fx_Max represents the rated power of the fuel cell, α represents P avg The pre-dynamic coefficient of β is determined by the SOC of the battery and supercapacitor; P HESS Represents the required power of the hybrid energy storage system; P fc Represents the output power of the fuel cell; η DC / DC Indicates the efficiency of the DC converter, P DC_dem Indicates the DC bus demand power.

[0024] Preferably, the cancer cell competition and metastasis algorithm comprises two stages: early stage of cancer and late stage of cancer;

[0025] Early signs of cancer include:

[0026]

[0027] D=ln(n)+b

[0028]

[0029] w i ~N(0,UL)

[0030] In the formula, a represents the pre-coefficient that decreases from 0.1 to 0 as the number of iterations increases; b represents the variable that decreases from 1 to 0 as the number of iterations increases; and They represent the positions of the ith healthy cell at iterations it and it+1, and their values ​​decrease gradually with the increase of iteration times; k represents a random number between [-1, 1]; meancell_H represents the average value of the healthy cell positions; D represents the spatial dimension influence coefficient; w iThe zero-mean Gaussian noise vector representing the i-th healthy cell; the covariance is represented by U-L, where U = (U1, U2,...U n ) represents the peak value of all healthy cells; L = (L1, L2,...L n ) represents the minimum boundary of all healthy cells.

[0031] The expression for the advanced stage of cancer includes:

[0032] R1 = rand, H = 0.5

[0033]

[0034] In the formula, and respectively represent the positions of the i-th central cancer cell at the it-th and it+1-th iterations; represents the division and mutation behavior of cancer cells; H represents the metastasis probability; R1 is a random number on [0, 1]; when R1 < H, the cancer cells continue to proliferate and divide, otherwise the cancer cells spread to other places through blood vessels; represents the moving speed of the i-th central cancer cell at the it+1-th iteration; c represents a pre-factor that decreases from 0.3 to 0 as the number of iterations increases; rand is a random number on [0, 1].

[0035] The present invention also provides a fuel cell hybrid electric heavy truck power source optimization design system, which is used to implement the above method and includes: an objective function construction module, a vehicle energy management module, and a solution module;

[0036] The objective function construction module is used to construct an objective function of the total cost per unit driving mileage within the life cycle based on the basic parameters of the fuel cell, power battery, and supercapacitor, the degradation parameters of the fuel cell and power battery, and the hydrogen consumption;

[0037] The energy management module is used to allocate the power of the fuel cell and the energy storage system based on the average demand power of the vehicle, the SOC of the power battery and the supercapacitor, and perform charging and discharging according to the SOC, maximum charging power, and maximum discharging power of the power battery and the supercapacitor;

[0038] The solution module is used to adopt the cancer cell competition and metastasis algorithm to find the minimum objective function that meets the driving conditions under the energy management strategy and complete the energy size optimization.

[0039] Preferably, the constructed objective function includes:

[0040]

[0041] Cost in = Nfc ·c fc +N bs ·N bp ·c b +N scs ·N scp ·c sc

[0042] N cycle =(N fc ·c fc +N bs ·N bp ·c b ) / (cost fcdeg +cost bdeg )

[0043] cost fcdeg =(D fc / 0.1)·N fc ·c fc

[0044] cost bdeg =(Q b_loss / 0.2)·N bs ·N bp ·c b

[0045] X total =N cycle ·x cycle

[0046] Cost H2 =cost H2 ·N cycle

[0047] in, Indicates that every 10 4 km total cost consumed; Cost in Indicates initial cost; Cost H2 represents the cost of hydrogen use during the life cycle; X total Indicates the total mileage during the life cycle; N fc Indicates the number of fuel cell units; c fc Represents the single cell cost of the fuel cell; N bs and N bp Respectively represent the number of batteries in series and parallel; c b Represents the cost of a single battery; N scs and N scp Respectively represent the number of supercapacitors in series and in parallel; c sc Represents the cost of a single supercapacitor; cost fcdeg and cost bdegRespectively represent the degradation cost of fuel cell and battery in a single cycle condition; N cycle The number of cycles representing the total mileage; x cycle Indicates the mileage of a single cycle; cost H2 is the hydrogen consumption in a single cycle.

[0048] Preferably, the energy management strategy established includes the required power of fuel cells and hybrid energy storage:

[0049] P fc =αP avg +βP fc_Eff +(1-α-β)P fc_Max

[0050]

[0051] Among them, P avg It represents the average power demand during the total running time from vehicle startup to the current moment; P fc_Eff Indicates the power at the highest efficiency point of the fuel cell; P fx_Max represents the rated power of the fuel cell, α represents P avg The pre-dynamic coefficient of β is determined by the SOC of the battery and supercapacitor; P HEsS Represents the required power of the hybrid energy storage system; P fc Represents the output power of the fuel cell; η DC / DC Indicates the efficiency of the DC converter, P DC_dem Indicates the DC bus demand power.

[0052] Preferably, the cancer cell competition and metastasis algorithm comprises two stages: early stage of cancer and late stage of cancer;

[0053] Early signs of cancer include:

[0054]

[0055]

[0056] D=ln(n)+b

[0057]

[0058] w i ~N(0,UL)

[0059] In the formula, a represents the pre-coefficient that decreases from 0.1 to 0 as the number of iterations increases; b represents the variable that decreases from 1 to 0 as the number of iterations increases; and respectively represent the positions of the i-th healthy cell at the it-th and it+1-th iterations, and their values gradually decrease as the number of iterations increases; k represents a random number between [-1, 1]; meancell_H represents the average value of the positions of healthy cells; D represents the spatial dimension influence coefficient; w i represents the zero-mean Gaussian noise vector of the i-th healthy cell; the covariance represents U-L, where U = (U1, U2,...U n ) represents the peak values of all healthy cells; L = (L1, L2,...L n ) represents the minimum boundaries of all healthy cells.

[0060] The expressions in the late stage of cancer include:

[0061] R1 = rand, H = 0.5

[0062]

[0063] In the formula, and respectively represent the positions of the i-th central cancer cell at the it-th and it+1-th iterations; represents the division and mutation behavior of cancer cells; H represents the metastasis probability; R1 is a random number on [0, 1]; when R1 < H, the cancer cells continue to proliferate and divide, otherwise the cancer cells spread to other places through blood vessels; represents the moving speed of the i-th central cancer cell at the it+1-th iteration; c represents a pre-factor that decreases from 0.3 to 0 as the number of iterations increases; rand is a random number on [0, 1].

[0064] Compared with the existing technology, the beneficial effects of the present invention are as follows:

[0065] The present invention globally optimizes the design of the power source of a fuel cell hybrid electric heavy truck by adopting a cancer cell competition and metastasis algorithm. While ensuring the vehicle's power performance, it extends the service life of the power source, significantly reduces the manufacturing cost, degradation cost, and hydrogen consumption of the hybrid power system, improves the energy efficiency of the system, and provides an economical, environmentally friendly, and efficient solution for the electrification of heavy trucks. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0067] Figure 1 is a schematic flow chart of the method of the present invention;

[0068] Figure 2 A schematic diagram of a CHTC-TT-400 operating curve according to an embodiment of the present invention;

[0069] Figure 3 Schematic diagram of the power requirements of a vehicle under the cycle conditions of an embodiment of the present invention; wherein (a) represents the CHTC-TT-400 cycle condition; (b) represents the power requirements of a fully loaded vehicle; (c) represents the power requirements of an unloaded vehicle;

[0070] Figure 4 A schematic diagram of an energy management strategy according to an embodiment of the present invention;

[0071] Figure 5 A pseudo code diagram of an algorithm according to an embodiment of the present invention;

[0072] Figure 6 Schematic diagram of simulation results of an embodiment of the present invention; wherein (a) represents 10 4 km represents the comprehensive cost of system consumption; (b) represents the total mileage during the life cycle. DETAILED DESCRIPTION

[0073] 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.

[0074] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0075] Before the description is given, the technical background of the present invention is briefly described.

[0076] Currently, the commercial fuel cell traction heavy trucks are still mainly FC+B (i.e. fuel cell + battery). However, in order to meet the high power energy demand of the vehicle and the demand for instantaneous high power braking energy recovery, the FC+B solution requires more lithium batteries to compensate, which increases the weight of the power transmission, affects the dynamic performance of the vehicle, increases the cost of the system, and limits the performance of the vehicle. Although supercapacitors have low energy density, they have high power density and are a good choice as a supplement to FC+B. This embodiment adopts a hybrid solution of FC+B+SC (i.e. fuel cell + battery + supercapacitor).

[0077] Fuel cell hybrid systems are divided into active and passive solutions. Compared with the passive solution, the DC / DC converter of the active solution increases the system cost, but it can handle complex and dynamic driving conditions well, maintain the stability and performance of the system, and can achieve dynamic energy management, automatically adjust the output of each energy source, and optimize energy distribution to cope with changing load requirements.

[0078] Embodiment 1

[0079] like Figure 1 FIG. 1 is a schematic diagram of the method flow of this embodiment, and the steps include:

[0080] S1. Based on the basic parameters of fuel cells, power batteries and supercapacitors, the degradation parameters of fuel cells and power batteries and hydrogen consumption, an objective function of the total cost per unit mileage during the life cycle is constructed.

[0081] In this embodiment, the basic parameters of the hybrid power system include: unit manufacturing costs of fuel cells, power batteries and supercapacitors, life costs of fuel cells and power batteries, hydrogen consumption costs and single-cycle driving mileage.

[0082] Since heavy trucks are often used for long-distance transportation, in order to simulate the long-distance transportation of heavy trucks, this embodiment is based on CHTC-TT (China Semi-trailer Tractor Cycle Condition), repeating the part > 80km / h to construct a total length of 400km cycle condition, called CHTC-TT-400, and the total working time is 17750s. Among them, CHTC-TT contains 2 speed intervals, and the total working time is 1800s. The low-speed interval is 473s, accounting for 26.3% of the total time, and the high-speed interval is 1327s, accounting for 73.7% of the total time. The average speed is 46.6km / h, the maximum speed is 88.0km / h, and the idling ratio is 8.6%. Figure 2 The CHTC-TT operating curve is shown.

[0083] At the same time, in order to indicate the running status of the tractor when fully loaded and unloaded, the tractor will run CHTC-TT-400 in fully loaded and unloaded states respectively, with a full-load gross mass of 49,000 kg and an unloaded gross mass of 11,500 kg. Figure 3 (a) shows the speed-time image of CHTC-TT-400. Figure 3 (b) and (c) show the variation of DC bus power and average required power over time when the vehicle is fully loaded and unloaded, respectively.

[0084] In order to solve the problem of inconsistent scales and orders of magnitude of different optimization objectives (including initial cost, fuel cell degradation, battery degradation, and hydrogen consumption, etc.), the embodiment analyzes the optimal energy size Pareto frontier obtained under the same energy management and optimization algorithm, and finds that as the initial cost of the system increases, the system life will be correspondingly extended and the cycle hydrogen consumption will be correspondingly reduced. In order to consider these optimization objectives at the same time, this embodiment unifies them into unit mileage (10 4 km), as shown in formula (1), avoiding setting weights. Every 10 4 km total cost consumed; Cost in Cost is the initial cost; H2 is the total cost of hydrogen consumed during the life cycle; X total is the total mileage during the life cycle. in Calculated from the energy size, as shown in formula (2), X total The number of cycles N in the life cycle cycle and the mileage of a single cycle x cycle Determine and calculate as shown in formula (3) to (6). Formula (3) calculates the number of cycles N of the system in its life cycle based on the degradation costs of FC and B under each cycle condition (calculated by formula (4) and (5)). cycle , x cycle =800km. Cost H2 By N cycle and the hydrogen consumption cost of a single cycle H2 Determine and calculate as formula (7).

[0085]

[0086] Cost in =N fc ·c fc +N bs ·N bp ·c b +N scs ·N scp ·c sc (2)

[0087] N cycle =(N fc ·c fc +N bs ·N bp ·c b ) / (cost fcdeg +cost bdeg ) (3)

[0088] cost fcdeg=(D fc / 0.1)·N fc ·c fc (4)

[0089] cost bdeg =(Q b_loss / 0.2)·N bs ·N bp ·c b (5)

[0090] X total =N cycle ·x cycle (6)

[0091] Cost H2 =cost H2 ·N cycle (7)

[0092] Among them, N fc Indicates the number of fuel cell units; c fc Represents the single cell cost of the fuel cell; N bs and N bp Respectively represent the number of batteries in series and parallel; c b Represents the cost of a single battery; N scs and N scp Respectively represent the number of supercapacitors in series and in parallel; c sc Represents the cost of a single supercapacitor; cost fcdeg and cost bdeg Represent the degradation costs of the fuel cell and battery in a single cycle condition respectively.

[0093] S2. Based on the average power demand of the vehicle, the SOC of the power battery and the supercapacitor, an energy management strategy (Average Power and SOC, APS) for the fuel cell hybrid electric heavy truck is established.

[0094] The APS proposed in this embodiment not only takes into account the dynamic changes and change rates of the output power of the fuel cell and the power battery, but also avoids the waste of fuel cell power generation and reduces the number of fuel cell start / stop times.

[0095] Specifically, this embodiment proposes a method for calculating the fuel cell power by using the average power demand and the SOC state of the energy storage system, as shown in formula (8). This method introduces the vehicle average power demand P avg 、The fuel cell's highest efficiency point power P fc_Eff and the maximum power P fc_Max , which can avoid the drastic change of fuel cell output power and make P fc_EffThe coefficient β increases with the increase of the SOC of the hybrid energy storage system. When the SOC of the energy storage system increases, the output power of the FC decreases and moves to the maximum efficiency point. When the system SOC decreases, the output power of the FC (fuel cell) increases and moves to P fc_Max Mobile, ensuring that the energy storage system does not need additional charging to meet the high power requirements of the vehicle. The change in fuel cell output power not only avoids the inability to fully utilize the fuel cell power when the fuel cell works at a constant power, but also ensures the slow change of the fuel cell output power to reduce the degradation of the fuel cell.

[0096] The output power of the fuel cell P fc The calculation is as follows:

[0097] P fc =aP avg +βP fc_Eff +(1-α-β)P fc_Max (8)

[0098] Among them, P avg represents the average power demand of the vehicle; Δt represents the total running time from the start of the vehicle to the present; P fc_Eff and P fc_Max They represent the power and maximum power of the fuel cell at maximum efficiency, respectively; α represents P avg The value of β is determined by the SOC of the battery and supercapacitor and is calculated as follows:

[0099]

[0100] Among them, E b and E sc Represents the maximum available energy that can be stored by batteries and supercapacitors, respectively; and Respectively represent the maximum and minimum SOC values ​​of B (battery) and SC (supercapacitor); σ represents the pre-correction factor to ensure α+β∈[0,1], which is calculated as follows:

[0101]

[0102] Therefore, the required power P of the vehicle's hybrid energy storage system (HESS) is HESS for:

[0103]

[0104] Among them, η DC / DC Indicates the efficiency of the DC converter, set to 0.95; if P HESS >0, the HESS is discharged, otherwise the HESS is charged; P DC_demIndicates the DC bus demand power.

[0105] When HESS is discharged, the supercapacitor has a high power density and a low energy density, so its main purpose is to provide peak power. sc When >0.4, the supercapacitor outputs electrical energy first, and the maximum output power is P sc_disMax The part exceeding the maximum output power is provided by the supercapacitor, and vice versa, the battery outputs the power first. The maximum output power is P b_disMax The part exceeding the maximum output power is provided by the supercapacitor, so that the SOC value of the supercapacitor is maintained at 0.4 to ensure that it has sufficient remaining space to absorb the high power energy generated by braking. When charging HESS, since the battery can store more energy than the supercapacitor, the battery is charged first within the range allowed by the battery SOC value, and the maximum charging power is P b_chaMax The part that exceeds the maximum charging power of the battery is allocated to the supercapacitor to ensure that the power of the system is not wasted.

[0106] In APS, the SOC value of the supercapacitor SOC sc The range is set to [0.05, 1] ​​to avoid low voltage drop of the DC converter. Overcharging or overdischarging as well as fast charging often cause irreversible damage to the battery and serious life degradation. Therefore, the SOC value of the battery SOC b The range is set to [0.2, 0.9] and the maximum charge rate of the battery is set to 0.1C. Figure 4 Energy management strategies are demonstrated.

[0107] Therefore, during the operation of APS, the system needs to meet the following constraints:

[0108] P xmin ≤P x ≤P xmax (13)

[0109] V xmin ≤V x ≤V xmax (14)

[0110] I xmin ≤I x ≤I xmax (15)

[0111] SOC ymin ≤SOC y ≤SOC ymax (16)

[0112] ΔI fcmin ≤ΔI fc ≤ΔIfcmax (17)

[0113] V DCmin ≤V DC ≤V DCmax (18)

[0114] Where x represents any one of FC, B or SC; y represents any one of B and SC; V DC Indicates the voltage of the DC converter; ΔI fc Indicates the current change rate of FC. In the embodiment, ΔI is specified fcmin =-3A / s, ΔI fcmax =3A / s.

[0115] S3. Using the cancer cell competition and transfer algorithm, under the energy management strategy, we find the minimum objective function that meets the driving conditions and complete the energy size optimization.

[0116] Optimization algorithms are crucial in engineering problems that require finding optimal solutions, and directly affect the quality of the results. Researchers usually use particle swarm optimization (PSO), grey wolf optimizer (GWO), and genetic algorithm (GA) to find the optimal solution. Particle swarm optimization simulates the predation behavior of bird flocks. Although it is fast and easy to implement, it is easy to fall into the local optimal solution, and is not suitable for high-dimensional problems, and is prone to premature convergence. The grey wolf algorithm simulates the hunting behavior of gray wolf groups. The search ability for the global optimal solution is relatively weak, and it is easy to fall into the local optimal solution when dealing with complex problems. Although the genetic algorithm has good search ability for high-dimensional problems, it has a large amount of calculation, requires parameter adjustment, and has a slow convergence speed. In response to these problems, this embodiment proposes a new population position update strategy to improve the ability to search for the optimal solution. The position update strategy is very similar to the competition between cancer cells and surrounding healthy cells and the metastasis and split mutation behavior of cancer cells during cancer evolution, so the algorithm is named Cancer Cell Competition and Transfer Algorithm (C3TA).

[0117] The algorithm mainly includes two stages: (1) In the early stage of cancer, the cancer cells at the primary site are in the process of self-division and proliferation and compete with the surrounding healthy cells for nutritional resources. Due to insufficient nutrition, the healthy cells gradually fall into disadvantage and are devoured by the cancer cells. The cancer cells take the place of the healthy cells, thus further expanding the cancer tumor. (2) In the late stage of cancer, the number of cancer cells at the primary site has accumulated greatly. As the cancer cells grow, some of them begin to metastasize to other places in the body through blood vessels, thus spreading the cancer.

[0118] (1) Initialize population position and displacement speed

[0119] For n-dimensional optimization problems, the position of individuals in the population cell = (x1, x2, ..., x n ), in the optimization process, the total number of individuals in the population is set to N, then the population C = {cell1, cell2, ..., cell N}, firstly use a random method to initialize the individual positions in the population:

[0120] cell i = l b +rand(1,n)·(u b -l b )(i=1, 2, 3,...,N) (19)

[0121] Among them, l b Indicates the lower boundary; u b Indicates the upper boundary; rand(1,n)[0,1] indicates a random number within 1 row and n columns.

[0122] For the results of random initialization, an elite reverse learning strategy is used to improve population diversity:

[0123] cell i =δ·(u b +l b )-cell i (20)

[0124]

[0125] Among them, δ represents a random number in [0, 1]; fit(cell i ) and fit(cell i ) represent cells respectively i and cell i The fitness is the calculation result of the cost function. Since the goal of this algorithm is the minimum fitness, the cell positions with small fitness are retained.

[0126] The range of individual movement speeds in the population is set according to the upper boundary of the position:

[0127] u v =0.3·u b (twenty two)

[0128] l v =-0.3·u b (twenty three)

[0129] In the formula, l v and u v They represent the lower and upper boundaries of the individual movement speed respectively.

[0130] The individual speed is initialized as:

[0131] V i = l u +rand(1,n)·(u v -l v )(i=1, 2, 3,...,N) (24)

[0132] (2) Stage 1: Early stage of cancer

[0133] At this stage, since the number of cancer cells is relatively small, it is assumed that the top 20% of individuals in terms of fitness are healthy cells, 20% to 40% are cancer cells in contact with healthy cells, called border cancer cells, and the rest are cancer cells inside the tumor, called central cancer cells. To determine the number of the three types of cells, let S1 = 0.2, then the number of healthy cells is:

[0134] S 1_Num =round(S1·N) (25)

[0135] In the formula, round represents the rounding operation.

[0136] At this stage, healthy cells (1≤i≤S 1_Num ) is less threatened by cancer cells, and its position update is mainly based on the average position of healthy cells, and has a smaller impact. The position update formula is:

[0137]

[0138]

[0139] D=ln(n)+b (29)

[0140]

[0141] w i ~N(0,UL) (31)

[0142] In the formula, a is the pre-coefficient that decreases from 0.1 to 0 as the number of iterations increases; b represents the variable that decreases from 1 to 0 as the number of iterations increases; and They represent the positions of the ith healthy cell at iterations it and it+1, and their values ​​decrease gradually with the increase of iteration times; k represents a random number between [-1, 1]; meancell_H represents the average value of the healthy cell positions; D represents the spatial dimension influence coefficient; w i represents the zero-mean Gaussian noise vector of the i-th healthy cell; the covariance represents UL, where U = (U1, U2, ...U n) represents the peak value of all healthy cells; L=(L1,L2,...L n ) represents the minimum boundary of all healthy cells.

[0143] For borderline cancer cells (S1 Num <i≤2·S1 Num ), in the process of it and it+1 nearby healthy cells competing for nutrition resources, it gradually engulfs healthy cells and occupies the position of healthy cells. Therefore, its position is mainly affected by nearby healthy cells. The position update formula is:

[0144]

[0145] In the formula, and They represent the positions of the i-th border cancer cell at and iterations, respectively, and their values ​​gradually decrease with the increase of iteration times; represents the position of the healthy cell with the smallest Euclidean distance to the i-th border cancer cell.

[0146] For the central cancer cells, the behavior at this stage is mainly proliferation, division and gene mutation. The adhesion between cancer cells is weakened and metastasis occurs. The position update formula is:

[0147]

[0148]

[0149] In the formula, and Respectively represent the position of the i-th central cancer cell at it and it+1 iterations; and represents the moving speed of the i-th central cancer cell at it and it+1 iterations respectively; c represents the pre-factor that decreases from 0.3 to 0 as the number of iterations increases; rand is a random number on [0, 1]; Bestpos represents the position of the healthy cell with the minimum fitness; Represents the division and mutation behavior of cancer cells, as shown below:

[0150] idx = find(rand(1,n) <mu) (36)

[0151] newcell=1 b +rand(1,n)·(u b -l b ) (37)

[0152]

[0153] Where mu = 0.5 is the mutation factor, and the indices of the numbers in rand(1, n) that are less than mu are found through the find function and stored in idx; newcell represents the position of the newly generated cell, and the number at the idx position in newcell is replaced and saved the number at the idx position in, and saved in in.

[0154] (3) Stage 2 advanced cancer

[0155] A large number of cancer cells have accumulated at the primary tumor site at this stage. Therefore, the proportion of cancer cells in contact with healthy cells in the total number decreases. It is assumed that the individuals in the top 10% of fitness are healthy cells, 10% - 20% of the cancer cells are boundary cancer cells, and the remaining cancer cells are central cancer cells. To determine the number of the three types of cells, let S2 = 0.1, then the number of healthy cells is:[[]]

[0156] S 2_Num = found(S2·N) (40)

[0157] Due to the expansion of the number of cancer cells, the positions of the healthy cells (1 ≤ i ≤ S 2_Nu m) at this stage are near the positions of the healthy cells with the highest fitness in the initial stage. Therefore, the position update formula for healthy cells is:[[]]

[0158]

[0159] The position update of the boundary cancer cells (S 2_Num < i ≤ 2·S 2_Num ) is the same as in the initial stage of cancer:[[]]

[0160]

[0161] The central cancer cells may continue to proliferate and divide at this stage, or may metastasize to other parts of the body through blood vessels, causing the spread of cancer. Its position update formula is:[[]]

[0162] R1 = rand, H = 0.5 (43)

[0163]

[0164] Where H is the metastasis probability; R1 is a random number on [0, 1]. When R1 < H, the cancer cells continue to proliferate and divide, otherwise the cancer cells spread to other places through blood vessels. The pseudocode is as Figure 5 shown.

[0165] Example 2

[0166] The present embodiment also provides a fuel cell hybrid electric heavy-duty truck power source optimization design system, including: an objective function construction module, an energy management module and a solution module; the objective function construction module is used to construct an objective function of the total cost per unit mileage within the life cycle based on the basic parameters of the fuel cell, power battery and supercapacitor, the degradation parameters of the fuel cell and power battery and the hydrogen consumption; the energy management module is used to allocate the power of the fuel cell and the energy storage system based on the average required power of the vehicle, the SOC of the power battery and the supercapacitor, and to charge and discharge according to the SOC of the power battery and the supercapacitor, the maximum charging power and the maximum discharging power; the solution module is used to adopt the cancer cell competition and transfer algorithm, and under the energy management strategy, find the minimum objective function that meets the driving conditions to complete the energy size optimization.

[0167] Among them, the constructed objective functions include:

[0168]

[0169] Cost in =N fc ·c fc +N bs ·N bp ·c b +N scs ·N scp ·c sc (46)

[0170] N cycle =(N fc ·c fc +N bs ·N bp ·c b ) / (cost fcdeg +cost bdeg ) (47)

[0171] cost fcdeg =(D fc / 0.1)·N fc ·c fc (48)

[0172] cost bdeg =(Q b_loss / 0.2)·N bs ·N bp ·c b (49)

[0173] X total =N cycle ·x cycle (50)

[0174] CostH2 =cost H2 ·N cycle (51)

[0175] in, Indicates that every 10 4 km total cost consumed; Cost in Indicates initial cost; Cost H2 represents the total cost of hydrogen consumed during the life cycle; X total Indicates the total mileage during the life cycle; N fc Indicates the number of fuel cell units; c fc Represents the single cell cost of the fuel cell; N bs and N bp Respectively represent the number of batteries in series and parallel; c b Represents the cost of a single battery; N scs and N scp Respectively represent the number of supercapacitors in series and in parallel; c sc Represents the cost of a single supercapacitor; cost fcdeg and cost bdeg Respectively represent the degradation cost of the fuel cell and battery in a single cycle condition; cost H2 is the hydrogen consumption for a single cycle. The power demand relationships established include:

[0176]

[0177] Among them, P HESS Represents the required power of the vehicle's hybrid energy storage system; P fc Represents the output power of the fuel cell; η DC / DC Indicates the efficiency of the DC converter, set to 0.95; if P HESS >0, the HESS is discharged, otherwise the HESS is charged; P DC_dem Indicates the rated required power of the DC bus.

[0178] The cancer cell competition and metastasis algorithm consists of two stages: early cancer stage and late cancer stage;

[0179] Early signs of cancer include:

[0180]

[0181] D=ln(n)+b(56)

[0182]

[0183] w i ~N(0,UL)(58)

[0184] Wherein, a represents a pre-factor that decreases from 0.1 to 0 as the number of iterations increases; b represents a variable that decreases from 1 to 0 as the number of iterations increases; and respectively represent the positions of the i-th healthy cell at the it-th and it+1-th iterations, and their values gradually decrease as the number of iterations increases; k represents a random number between [-1, 1]; meancell_H represents the average value of the positions of healthy cells; D represents the spatial dimension influence coefficient; w i represents the zero-mean Gaussian noise vector of the i-th healthy cell; the covariance represents U-L, where U = (U1, U2,...U n ) represents the peak values of all healthy cells; L = (L1, L2,...L n ) represents the minimum boundaries of all healthy cells.

[0185] The expressions in the late stage of cancer include:

[0186] R1 = rand, H = 0.5 (59)

[0187]

[0188] Wherein, and respectively represent the positions of the i-th central cancer cell at the it-th and it+1-th iterations; represents the division and mutation behavior of cancer cells; H represents the metastasis probability; R1 is a random number on [0, 1]; when R1 < H, the cancer cells continue to proliferate and divide, otherwise the cancer cells spread to other places through blood vessels; represents the moving speed of the i-th central cancer cell at the it+1-th iteration; c represents a pre-factor that decreases from 0.3 to 0 as the number of iterations increases; rand is a random number on [0, 1].

[0189] Example 3

[0190] This example provides a simulation experiment to verify the advancement of the present invention.

[0191] In this example, fuel cells, batteries, and supercapacitors are all taken as discrete units with monomers. Referring to the power of currently commercial fuel cell tractors, the minimum value of the fuel cell unit in this example is set to 30 (180 kW), and due to the constraint of the internal space of the tractor, the upper limit of the number of fuel cell units is 60 (360 kW). In order to avoid too low voltage at the low-voltage end of the DC converter, the minimum values of the series numbers of the batteries and supercapacitors are both set to 10, the maximum value of the battery series number is set to 200, and the range of the battery parallel number is set the same as the series number. Considering the power output requirements of the supercapacitor, the range of its parallel number is set to [4, 40]. Table 1 lists the settings of the simulation parameters in this example.

[0192] Table 1

[0193]

[0194] During the simulation process, the optimization algorithm C3TA provided by the present invention first generates a certain energy size, and calculates the SOC according to the obtained energy size. b , SOC sc , P DC_dem and P avg The calculated parameters are used to calculate the output power of FC, B and SC using the constructed APS to calculate the degradation and hydrogen consumption of the vehicle energy supply system. Finally, C3TA is used to optimize the target based on the above calculation results. The C3TA algorithm updates the energy size according to the position update strategy and enters the next cycle until the minimum value is retained after the iteration.

[0195] The optimal capacity configuration problem of the fuel cell tractor is studied under the APS strategy for the CHTC-TT-400 cycle, and the optimal solution is obtained through the proposed C3TA. In order to prove the performance of the proposed C3TA algorithm, we use the PSO algorithm and the GWO algorithm for comparative analysis. We also compare the performance of the APS and Discrete Wavelet Transform (DWT) energy management strategies in optimizing the optimal capacity configuration of hybrid power under the same parameters and constraint ranges. Among them, the DWT strategy allocates low-frequency power, medium-frequency power, and high-frequency power to the fuel cell, battery, and supercapacitor respectively according to the dynamic characteristics of the three power sources.

[0196] Table 2 shows the comprehensive optimization results of APS and DWT under C3TA, PSO and GWO algorithms respectively. Figure 6 In the life cycle 10 4 km comprehensive consumption cost and the total mileage during the life cycle are more intuitively compared. Figure 6 As shown in the figure, it can be seen that the life of the hybrid system using the APS strategy is increased by 77% compared with the DWT, and its life cycle is improved every 10 4 km total consumption cost is reduced by 16% compared with DWT, which means that under the same working conditions, the APS energy management strategy will reduce the decline of life and reduce the comprehensive cost of unit mileage consumption. At the same time, by comparing the optimization results of different optimization algorithms under the same energy management strategy, it is found that the C3TA algorithm proposed in this invention is better than the APS algorithm. Under DWT, we get Comparing the results obtained by PSO and GWO algorithms, it can be clearly seen that C3TA can obtain the minimum optimization target value under both energy management strategies.

[0197] Table 2

[0198]

[0199] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A fuel cell hybrid electric heavy truck power source optimization design method, characterized in that the steps include: Based on the basic parameters of fuel cells, power batteries and supercapacitors, the degradation parameters of fuel cells and power batteries and hydrogen consumption, an objective function of the total cost per unit mileage during the life cycle is constructed; Establish the energy management strategy of fuel cell hybrid electric heavy truck based on the average required power of the vehicle, the SOC of the power battery and supercapacitor; The cancer cell competition and transfer algorithm is used to find the minimum objective function under the driving condition under the energy management strategy to complete the energy size optimization.

2. The method for optimizing the design of a power source for a fuel cell hybrid electric heavy truck according to claim 1, characterized in that: The objective function constructed includes: Cost in =N fc ·c fc +N bs ·N bp ·c b +N scs ·N scp ·c sc N cycle =(N fc ·c fc +N bs ·N bp ·c b ) / (cost fcdeg +cost bdeg ) Cost fcdeg =(D fc / 0.1)·N fc ·c fc cost bdeg =(Q b_loss / 0.2)·N bs ·N bp ·c b X total =N cycle ·x cycle Cost H2 =cost H2 ·N cycle in, Indicates that every 10 4 km total cost consumed; Cost in Indicates initial cost; Cost H2 Indicates the cost of using hydrogen during the life cycle; X total Indicates the total mileage during the life cycle; N fc Indicates the number of fuel cell units; c fc Represents the single cell cost of the fuel cell; N bs and N bp Respectively represent the number of batteries in series and parallel; c b Represents the cost of a single battery; N scs and N scp Respectively represent the number of supercapacitors in series and in parallel; c sc Represents the cost of a single supercapacitor; cost fcdeg and cost bdeg Respectively represent the degradation cost of fuel cell and battery in a single cycle condition; N cycle The number of cycles representing the total mileage; x cycle Indicates the mileage of a single cycle; cos H2 is the hydrogen consumption in a single cycle.

3. The method for optimizing the design of a power source for a fuel cell hybrid electric heavy truck according to claim 1, characterized in that: The energy management strategy established includes the required power of fuel cells and hybrid energy storage: P fc =αP avg +βP fc_Eff +(1-α-β)P fc_Max Among them, P avg It represents the average power demand during the total running time from vehicle startup to the current moment; P fc_Eff Indicates the power at the highest efficiency point of the fuel cell; P fx_Max represents the rated power of the fuel cell, α represents P avg The pre-dynamic coefficient of β is determined by the SOC of the battery and supercapacitor; P HESS Represents the required power of the hybrid energy storage system; P fc Represents the output power of the fuel cell; η DC / DC Indicates the efficiency of the DC converter, P DC_dem Indicates the DC bus demand power.

4. The method for optimizing the design of a power source for a fuel cell hybrid electric heavy truck according to claim 1, characterized in that: The cancer cell competition and migration algorithm includes two stages: early stage of cancer and late stage of cancer; Early signs of cancer include: D=ln(n)+b w i ~N(0,U-L) In the formula, a represents the pre-coefficient that decreases from 0.1 to 0 as the number of iterations increases; b represents the variable that decreases from 1 to 0 as the number of iterations increases; and They represent the positions of the ith healthy cell at iterations it and it+1, and their values ​​decrease gradually with the increase of iteration times; k represents a random number between [-1, 1]; meancell_H represents the average value of the healthy cell positions; D represents the spatial dimension influence coefficient; w i represents the zero-mean Gaussian noise vector of the i-th healthy cell; the covariance represents UL, where U = (U1, U2, ...U n ) represents the peak value of all healthy cells; L=(L1,L2,...L n ) represents the minimum boundary of all healthy cells; Expressions in later stages of cancer include: R1=rand,H=0.5 In the formula, and respectively represent the positions of the i-th central cancer cell at the it-th and it+1-th iterations; represents the division and mutation behavior of cancer cells; H represents the metastasis probability; R1 is a random number on [0, 1]; when R1 < H, the cancer cells continue to proliferate and divide, otherwise the cancer cells spread to other places through blood vessels; represents the moving speed of the i-th central cancer cell at the it+1-th iteration; c represents a pre-factor that decreases from 0.3 to 0 as the number of iterations increases; rand is a random number on [0, 1].

5. A fuel cell hybrid electric heavy truck power source optimization design system, the system is used to implement the method according to any one of claims 1 to 4, characterized in that: include: Objective function building module, vehicle energy management module and solution module; The objective function building module is used to build an objective function of the total cost per unit mileage within the life cycle based on the basic parameters of the fuel cell, the power battery and the supercapacitor, the degradation parameters of the fuel cell and the power battery and the hydrogen consumption; The energy management module is used to allocate the power of the fuel cell and the energy storage system based on the average required power of the vehicle, the SOC of the power battery and the supercapacitor, and to charge and discharge according to the SOC, maximum charging power and maximum discharging power of the power battery and the supercapacitor; The solution module is used to adopt the cancer cell competition and transfer algorithm, and under the energy management strategy, to find the minimum objective function that meets the driving conditions and complete the energy size optimization.

6. The fuel cell hybrid electric heavy truck power source optimization design system according to claim 5 is characterized in that: The objective function constructed includes: Cost in =N fc ·c fc +N bs ·N bp ·c b +N scs ·N scp ·c sc N cycle =(N fc ·c fc +N bs ·N bp ·c b ) / (cost fcdeg +cost bdeg ) cost fcdeg =(D fc / 0.1)·N fc ·c fc cost bdeg =(Q b_loss / 0.2)·N bs ·N bp ·c b X total =N cycle ·x cycle Cost H2 =cost H2 ·N cycle in, Indicates that every 10 4 km total cost consumed; Cost in Indicates initial cost; Cost H2 Indicates the cost of using hydrogen during the life cycle; X total Indicates the total mileage during the life cycle; N fc Indicates the number of fuel cell units; c fc Represents the single cell cost of the fuel cell; N bs and N bp Respectively represent the number of batteries in series and parallel; c b Represents the cost of a single battery; N scs and N scp Respectively represent the number of supercapacitors in series and in parallel; c sc Represents the cost of a single supercapacitor; cost fcdeg and cost bdeg Respectively represent the degradation cost of fuel cell and battery in a single cycle condition; N cycle The number of cycles representing the total mileage; x cycle Indicates the mileage of a single cycle; cost H2 is the hydrogen consumption in a single cycle.

7. The fuel cell hybrid electric heavy truck power source optimization design system according to claim 5, characterized in that: The energy management strategy established includes the required power of fuel cells and hybrid energy storage: P fc =αP avg +βP fc_Eff +(1-α-β)P fc_Max Among them, P avg It represents the average power demand during the total running time from vehicle startup to the current moment; P fc_Eff Indicates the power at the highest efficiency point of the fuel cell; P fx_Max represents the rated power of the fuel cell, α represents P avg The pre-dynamic coefficient of β is determined by the SOC of the battery and supercapacitor; P HESS Represents the required power of the hybrid energy storage system; P fc Represents the output power of the fuel cell; η DC / DC Indicates the efficiency of the DC converter, P DC_dem Indicates the DC bus demand power.

8. The fuel cell hybrid electric heavy truck power source optimization design system according to claim 5, characterized in that: The cancer cell competition and migration algorithm includes two stages: early stage of cancer and late stage of cancer; Early signs of cancer include: D=ln(n)+b w i ~N(0,U-L) In the formula, a represents the pre-coefficient that decreases from 0.1 to 0 as the number of iterations increases; b represents the variable that decreases from 1 to 0 as the number of iterations increases; and They represent the positions of the ith healthy cell at iterations it and it+1, and their values ​​decrease gradually with the increase of iteration times; k represents a random number between [-1, 1]; meancell_H represents the average value of the healthy cell positions; D represents the spatial dimension influence coefficient; w i represents the zero-mean Gaussian noise vector of the i-th healthy cell; the covariance represents UL, where U = (U1, U2, ...U n ) represents the peak value of all healthy cells; L=(L1,L2,...L n ) represents the minimum boundary of all healthy cells; Expressions in later stages of cancer include: R1=rand,H=0.5 In the formula, and respectively represent the positions of the i-th central cancer cell at the it-th and it+1-th iterations; represents the division and mutation behavior of cancer cells; H represents the metastasis probability; R1 is a random number on [0, 1]; when R1 < H, the cancer cells continue to proliferate and divide, otherwise the cancer cells spread to other places through blood vessels; represents the moving speed of the i-th central cancer cell at the it+1-th iteration; c represents a pre-factor that decreases from 0.3 to 0 as the number of iterations increases; rand is a random number on [0, 1].

Images