阅读说明：本技术 一种混合动力车辆参数标定方法及装置 (Hybrid vehicle parameter calibration method and device ) 是由 陈慧勇 王印束 曾小华 王越 王兴 王富生 蒋银飞 李建锋 刘小伟 于 2020-04-28 设计创作，主要内容包括：本发明属于混合动力车辆技术领域,具体涉及一种混合动力车辆参数标定方法及装置。该方法首先根据整车实际情况搭建包括能耗模型的整车仿真模型,能耗模型包括理论油耗模型和理论电耗模型,采用电耗或者油耗表征：然后获取整车工况数据和待标定参数的可行域,将整车工况数据输入至整车仿真模型中,在待标定参数的可行域内进行遍历,找到能耗模型的能耗值最低的结果对应的待标定参数的取值作为该整车工况下的最佳标定参数。本发明将两个不同量纲的物理量处理为一种量纲的物理量,从而对具体能耗的多少有了清楚明确的判断,可准确判断不同取值的待标定参数对应的能耗值大小,提高了标定效果。(The invention belongs to the technical field of hybrid vehicles, and particularly relates to a method and a device for calibrating parameters of a hybrid vehicle. The method comprises the following steps of firstly, building a whole vehicle simulation model comprising an energy consumption model according to the actual condition of the whole vehicle, wherein the energy consumption model comprises a theoretical oil consumption model and a theoretical power consumption model, and is characterized by adopting power consumption or oil consumption: and then acquiring the whole vehicle working condition data and the feasible region of the parameter to be calibrated, inputting the whole vehicle working condition data into the whole vehicle simulation model, traversing in the feasible region of the parameter to be calibrated, and finding the value of the parameter to be calibrated corresponding to the result with the lowest energy consumption value of the energy consumption model as the optimal calibration parameter under the whole vehicle working condition. The invention processes the physical quantities of two different dimensions into one dimensional physical quantity, thereby clearly and definitely judging the specific energy consumption, accurately judging the energy consumption values corresponding to the parameters to be calibrated with different values and improving the calibration effect.)

1. A method for calibrating parameters of a hybrid vehicle is characterized by comprising the following steps:

building a whole vehicle simulation model according to the actual condition of a whole vehicle, wherein the whole vehicle simulation model comprises an energy consumption model, and the energy consumption model is determined according to the energy flow condition among all modules in a power system of the whole vehicle; the energy consumption model comprises a theoretical oil consumption model and a theoretical power consumption model, and the energy consumption model is characterized by adopting power consumption or oil consumption: if the oil consumption representation is adopted, the theoretical power consumption model is equivalent to the oil consumption model to obtain a first equivalent model, and the sum of the equivalent oil consumption value of the first equivalent model and the oil consumption value of the theoretical oil consumption model is used as the energy consumption value of the energy consumption model; if the power consumption representation is adopted, the theoretical oil consumption model is equivalent to the power consumption model to obtain a second equivalent model, and the sum of the equivalent power consumption value of the second equivalent model and the power consumption value of the theoretical power consumption model is used as the energy consumption value of the energy consumption model;

acquiring finished automobile working condition data and a feasible region of a parameter to be calibrated, inputting the finished automobile working condition data into the finished automobile simulation model, traversing in the feasible region of the parameter to be calibrated, and finding a value of the parameter to be calibrated corresponding to a result with the lowest energy consumption value of the energy consumption model as an optimal calibration parameter under the finished automobile working condition; if the number of the parameters to be calibrated is one, traversing each value in the feasible region of the parameters to be calibrated, and if the number of the parameters to be calibrated is at least two, traversing combinations of different values of each parameter to be calibrated, which are obtained by traversing in the feasible region of each parameter to be calibrated.

2. The method for calibrating the parameters of the hybrid vehicle according to claim 1, wherein if the actual running state of the whole vehicle is off-line, the working condition of the whole vehicle is a standard working condition; and if the actual running state of the whole vehicle is on-line, the working condition of the whole vehicle is a synthetic working condition obtained by synthesizing historical working conditions.

3. The hybrid vehicle parameter calibration method according to claim 1, wherein the first equivalent model is:

of formula (II) to'_uAs a first equivalent model, E_b,dcTotal energy of discharge, η, for the cell_b,dcFor average efficiency of cell discharge, E_b,cTotal energy, η, charged to the battery_trIs the average overall transmission efficiency, eta_b,cAverage efficiency of charging, Soc, for batteries_iniIs the initial SOC, SOC of the battery_endIs the final value SOC of the battery, C is the unit conversion coefficient of oil consumption, b_e,avgThe average fuel consumption rate of the engine is as follows:

in the formula, Me (t) is the fuel injection quantity of the engine at each point of the cycle working condition, and Pe (t) is the output power of the engine at each point of the cycle working condition.

4. The method of calibrating parameters of a hybrid vehicle according to claim 1, characterized in that the parameters that have a greater influence on the energy consumption value of the energy consumption model are used as the parameters to be calibrated.

5. The method for calibrating the parameters of the hybrid vehicle according to claim 4, characterized in that the parameters to be calibrated comprise an SOC threshold value and a power threshold value for switching between an electric-only mode and a hybrid mode.

6. The method for calibrating the parameters of the hybrid vehicle according to claim 4 or 5, characterized in that the parameters having a greater influence on the energy consumption value of the energy consumption model are obtained by analyzing the results of the energy consumption model corresponding to the selected parameters by using principal component analysis and/or pareto analysis.

7. A hybrid vehicle parameter calibration device, characterized by comprising a memory and a processor, wherein the processor is used for executing instructions stored in the memory to realize the hybrid vehicle parameter calibration method according to any one of claims 1-6.

Technical Field

The invention belongs to the technical field of hybrid vehicles, and particularly relates to a method and a device for calibrating parameters of a hybrid vehicle.

Background

In the face of the current situation of global energy shortage, the current vehicle technology is developing towards low-carbon and intelligent direction. Hybrid pneumatic vehicles are the focus of global research as one of the important energy saving driving modalities in the automotive industry.

The actual running performance of the hybrid vehicle is closely related to the type of the adopted control strategy and the selection of the control parameters, and the good parameter calibration result is finally ensured that the vehicle control unit realizes excellent control effect on the premise of a set power assembly power parameter and a complete functional vehicle energy management strategy.

The calibration of the whole hybrid electric vehicle controller is the whole process of adjusting and determining the operation parameters and the control parameters in the whole hybrid electric vehicle controller according to characteristic parameters of each part of a power assembly and various performance indexes such as the whole vehicle economy, the emission performance, the extreme condition applicability and the like. The complete vehicle controller calibration process comprises the steps of energy management strategy calibration, actuator calibration, operation condition calibration, economy calibration, emission calibration, OBD calibration, safety monitoring calibration and the like. In the calibration method of the vehicle control unit, a calibration tool is usually adopted to perform data adjustment verification, readjustment and revalidation through repeated tests, and finally a better test value is taken as a calibration value in a comprehensive consideration.

Chinese patent application publication No. CN108515962AA method for quickly calibrating a whole vehicle controller of a hybrid electric vehicle is provided, and fuel consumption and electric energy consumption are taken as calibration targets when the method is used for calibrating. The method for determining the fuel consumption and the electric energy consumption comprises the following steps: the test was performed according to the mode specified by the european economic commission in the regulation ECE R101 based on the condition a and the condition B, respectively, and the fuel consumption/electric energy consumption was obtained according to the test, the total mileage before the mode switching, and the average mileage between two charges. That is to say, the whole method is carried out under two set models, the corresponding fuel consumption and the corresponding electric energy consumption are obtained based on the test calibration results of the two models, and the dependence on the driving condition is strong. The vehicle basically cannot run according to the two set working conditions in the actual running process, the characteristics of randomness and diversity exist, and if the calibration result obtained under the test condition is applied to an actual vehicle, the calibration result is not necessarily applicable to the actual vehicle. And the method separately calculates the fuel consumption and the electric energy consumption, and the consumption units are different due to different power sources, so that the optimal result is the result which cannot be accurately measured in the face of the consumption of two different units. For example, the first result is a fuel consumption of 100L (100km)^-1The electric energy consumption is 30Wh km^-1The second result is a fuel consumption of 20L (100km)^-1The electric energy consumption is 110Wh km^-1The first result is higher than the second result in fuel consumption, but the power consumption is low, and it is very likely that the optimal calibration target is not determined accurately if no result is determined accurately, and if the calibration target is determined accurately, the calibration result is not good.

Disclosure of Invention

The invention provides a method and a device for calibrating parameters of a hybrid vehicle, which are used for solving the problem of poor calibration result caused by separate calculation of fuel consumption and electric energy consumption in the prior art.

In order to solve the technical problem, the technical scheme of the invention comprises the following steps:

the invention provides a method for calibrating parameters of a hybrid vehicle, which comprises the following steps:

building a whole vehicle simulation model according to the actual condition of a whole vehicle, wherein the whole vehicle simulation model comprises an energy consumption model, and the energy consumption model is determined according to the energy flow condition among all modules in a power system of the whole vehicle; the energy consumption model comprises a theoretical oil consumption model and a theoretical power consumption model, and the energy consumption model is characterized by adopting power consumption or oil consumption: if the oil consumption representation is adopted, the theoretical power consumption model is equivalent to the oil consumption model to obtain a first equivalent model, and the sum of the equivalent oil consumption value of the first equivalent model and the oil consumption value of the theoretical oil consumption model is used as the energy consumption value of the energy consumption model; if the power consumption representation is adopted, the theoretical oil consumption model is equivalent to the power consumption model to obtain a second equivalent model, and the sum of the equivalent power consumption value of the second equivalent model and the power consumption value of the theoretical power consumption model is used as the energy consumption value of the energy consumption model;

acquiring finished automobile working condition data and a feasible region of a parameter to be calibrated, inputting the finished automobile working condition data into the finished automobile simulation model, traversing in the feasible region of the parameter to be calibrated, and finding a value of the parameter to be calibrated corresponding to a result with the lowest energy consumption value of the energy consumption model as an optimal calibration parameter under the finished automobile working condition; if the number of the parameters to be calibrated is one, traversing each value in the feasible region of the parameters to be calibrated, and if the number of the parameters to be calibrated is at least two, traversing combinations of different values of each parameter to be calibrated, which are obtained by traversing in the feasible region of each parameter to be calibrated.

The beneficial effects of the above technical scheme are: the method comprises the steps of determining an energy consumption model based on the energy flow condition among all modules in the whole vehicle power system, using the energy consumption model as a system response, traversing parameters to be calibrated, and using the value of the parameters to be calibrated with the best system response (namely the lowest energy consumption and the best economy) as the best calibration parameter. The energy consumption model is characterized by power consumption or oil consumption, the physical quantities of two different dimensions are finally processed into the physical quantity of one dimension, the physical quantity can be characterized by the oil consumption and the power consumption, so that the specific energy consumption can be clearly and definitely judged, the energy consumption values corresponding to the parameters to be calibrated with different values can be accurately judged, and the calibration effect is improved.

As a further improvement of the method, in order to ensure that the calibration method can ensure the calibration effect when the whole vehicle is in an off-line state and adapt to various working conditions with strong randomness and diversity when the whole vehicle is in an on-line state, if the actual running state of the whole vehicle is off-line, the working condition of the whole vehicle is a standard working condition; and if the actual running state of the whole vehicle is on-line, the working condition of the whole vehicle is a synthetic working condition obtained by synthesizing historical working conditions.

As a further improvement of the method, in order to obtain an accurate energy consumption model, the first equivalent model is:

in the formula (f)_u' is a first equivalent model, E_b,dcTotal energy of discharge, η, for the cell_b,dcFor average efficiency of cell discharge, E_b,cTotal energy, η, charged to the battery_trIs the average overall transmission efficiency, eta_b,cAverage efficiency of charging, Soc, for batteries_iniIs the initial SOC, SOC of the battery_endIs the final value SOC of the battery, C is the unit conversion coefficient of oil consumption, b_e,avgThe average fuel consumption rate of the engine is as follows:

in the formula, Me (t) is the fuel injection quantity of the engine at each point of the cycle working condition, and Pe (t) is the output power of the engine at each point of the cycle working condition.

As a further improvement of the method, in order to improve the calibration efficiency, a parameter that has a large influence on the energy consumption value of the energy consumption model is used as a parameter to be calibrated.

As a further improvement of the method, in order to improve the calibration efficiency, the parameters to be calibrated comprise an SOC threshold value and a power threshold value for switching between the pure electric mode and the hybrid power mode.

As a further improvement of the method, in order to obtain accurate parameters having a large influence on the energy consumption value of the energy consumption model, the results of the energy consumption model corresponding to the selected parameters are analyzed by using principal component analysis and/or pareto analysis to obtain the parameters having a large influence on the energy consumption value of the energy consumption model.

The invention also provides a hybrid vehicle parameter calibration device which comprises a memory and a processor, wherein the processor is used for executing instructions stored in the memory to realize the hybrid vehicle parameter calibration method introduced above and achieve the same effect as the method.

Drawings

FIG. 1 is an overall framework flow diagram of a method embodiment of the present invention;

FIG. 2 is a flow chart of co-simulation of Matlab and Isight software platforms in an embodiment of the method of the present invention;

FIG. 3 is a block diagram of a hybrid powertrain in a method embodiment of the present invention;

FIG. 4 is a schematic illustration of the division of portions of a hybrid powertrain in an embodiment of the method of the present invention;

FIG. 5 is a flow chart of a method embodiment of the present invention for constructing a synthetic regime;

FIG. 6 is a flow chart of an optimized calibration algorithm for a hybrid vehicle in an embodiment of the method of the present invention;

FIG. 7 is a pareto chart for each calibration parameter of an embodiment of a method of the present invention;

FIG. 8 is a master effect plot for each calibration parameter of a method embodiment of the present invention;

fig. 9 is a connection block diagram of a hybrid vehicle parameter calibration apparatus of the apparatus embodiment of the present invention.

Detailed Description

The method comprises the following steps:

the embodiment provides a method for calibrating parameters of a hybrid vehicle, wherein an overall framework flowchart of the method is shown in fig. 1, and the method aims to achieve the best overall vehicle economy (the overall vehicle energy consumption is low) on the premise of meeting the dynamic performance.

The method adopts different processing modes for the whole vehicle to be on-line or not, and a detailed description is firstly made on a calibration method of the whole vehicle in an on-line state.

Step one, as shown in fig. 2, a test platform is built. The test platform is realized by joint simulation of a Matlab software platform and an Isigh software platform. The Isigh software platform is used for defining each parameter to be calibrated and determining a feasible domain of the parameter to be calibrated, finally generating a design matrix, and sequentially writing each group of data of the design matrix into a parameter input m file of the Matlab software platform. And the Matlab software platform is used for building a finished automobile simulation model according to the actual condition of the finished automobile, simulating according to the content in the m file, feeding the simulation result back to the Isight software platform, and calculating each group of data in the design matrix to obtain the corresponding simulation result. After all the group data in the design matrix are calculated, the Isight software platform can analyze to determine the contribution (influence) of each parameter to be calibrated to the system response (simulation result) or the optimal calibration parameter.

The finished automobile simulation model built in the Matlab software platform is shown in FIG. 3, two motors are arranged in the model, namely CL1 and CL2, and the finished automobile simulation model comprises an energy consumption model. According to the energy flow condition of the power system of the whole vehicle, the energy flow among different modules (end points) in the power system can be determined, and an energy balance equation is established for the whole vehicle according to the energy conservation law, so that the energy consumption model can be deduced. The energy consumption model comprises a theoretical oil consumption model and a theoretical power consumption model, the theoretical power consumption model needs to be equivalent to the oil consumption model to obtain a first equivalent model, and the sum of the oil consumption value of the theoretical oil consumption model and the equivalent oil consumption value of the equivalent model is used as the energy consumption value (also system response) of the energy consumption model. The specific process is as follows:

as shown in fig. 4, the entire system is divided into a power source, a transmission system, and a vehicle body. The power source consists of an engine and a power battery (hereinafter referred to as a battery), the transmission system consists of a motor MG1, a motor MG2 and a two-gear gearbox, and the vehicle body is simplified into a whole vehicle longitudinal dynamics model according to the automobile theory, as shown in formula (1):

F_t＝F_f+F_w+F_i+F_j (1)

in the formula, F_tThe unit is Nm of the whole vehicle driving force at the wheel; f_fIs rolling resistance in Nm; f_wIs the air resistance in Nm; f_iIs ramp resistance in Nm; f_jFor acceleration resistance, the unit is Nm.

The definitions of the energy flows at different end positions in the hybrid system in fig. 4 are shown as equations (2) to (7), respectively:

E_fuel＝(f_e·Ca)/(3600·C) (3)

E_ice＝f_e/(C·b_e,avg) (4)

E_b,dc＝max{0,(Soc_ini-Soc_end)·BE·3600} (6)

E_b,c＝min{0,(Soc_ini-Soc_end)·BE·3600} (7)

in the formula, E_whThe unit is KJ which is the theoretical total driving energy of the cycle condition at the wheel; e_fuelThe unit is KJ for the total energy of fuel consumed by the engine; e_iceThe unit of the energy actually provided by the engine is KJ; e_rgbThe total recovered energy is regenerative braking at the battery end, and the unit is KJ; e_b,dcThe total discharge energy of the battery is KJ; e_b,cThe total charging energy of the battery is KJ; e.g. of the type_rgb(t) the energy recovered by each point regenerative braking under the condition of the braking under the circulating working condition, and the unit is KJ; f_t(t) is the driving force required by each point under the circulating working condition, and the unit is Nm; f_t' (t) is the whole vehicle driving force under the driving condition in the circulating working condition, and the unit is Nm; v (t) is the speed of each point under the cycle condition, and the unit is m/s; n is the total time of the cycle condition and the unit is s (the calculation step is 1)s)；f_eThe unit is L/100km for the fuel consumption of hundred kilometers of the whole vehicle; ca is the heat value of fuel oil, and the unit is KJ/g; soc (t) is the system SOC change corresponding to each point of the cycle condition; soc_iniAn initial SOC of the battery; soc_endIs the battery final value SOC; BrkP (t) is the opening degree of the brake pedal at each point of the cycle working condition; BE is the battery energy in KWh; b_e,avgThe average fuel consumption rate of the engine is g/KWh; c is the conversion coefficient of the unit of oil consumption. Wherein the average fuel consumption b of the engine_e,avgThe calculation formula of the conversion coefficient C of the unit of oil consumption is respectively shown as a formula (8) and a formula (9):

C＝1/(36000·ρ_fuel·x_tot) (8)

in the formula, ρ_fuelThe unit is the density of fuel oil and is Kg/L; x is the number of_totThe unit is Km, which is the total driving mileage under the circulating working condition; me (t) is the oil injection quantity of each point of the engine under the circulating working condition, and the unit is g/h; pe (t) is the output power of the engine at each point of the cycle working condition, and the unit is KW.

According to the energy flow at each endpoint in the hybrid power system, the total input energy of the transmission system comprises: energy E actually supplied by the engine_iceTotal recovered energy E of regenerative braking at battery end_rgbAnd total energy E of battery discharge_b,dc(ii) a The total output energy of the transmission system comprises: theoretical total driving energy E of cycle condition at wheel_whAnd total energy E of battery charge_b,c. Defining the average integrated transmission efficiency eta of the system by combining the law of conservation of energy_trAs shown in formula (10):

based on the definition of average comprehensive transmission efficiency, a theoretical oil consumption model is obtained, and the theoretical oil consumption model is shown as a formula (11):

in the formula eta_b,cAverage efficiency of charging the battery; eta_b,dcThe average efficiency of cell discharge.

Wherein, the wheel is provided with the theoretical total driving energy E of the cycle working condition_whDepending on the cycle condition requirements, the energy is constant at the selected cycle condition; charge and discharge energy E of battery_b,c、E_b,dcDepending on the final value of the battery SOC after the end of the cycle condition simulation: if the SOC is completely balanced after the working condition operation is finished, the total charging and discharging energy of the battery is 0; and it is difficult to achieve complete balancing of the SOC during actual control. Thus, the theoretical power consumption model is converted into a fuel consumption model to obtain a first equivalent model, a first equivalent model f_u' As shown in formula (12):

therefore, the energy consumption model is as follows:

further derivation of formula (13) yields formula (14):

in the formula (f)_e,uThe unit is L/100km for energy consumption model.

And step two, acquiring the whole vehicle working condition data (including data such as corresponding relation between vehicle speed and time), wherein the vehicle can acquire useful information in real time through a vehicle network cradle head because the actual running state of the whole vehicle is on-line, and the synthetic working condition obtained by synthesizing historical working conditions is adopted in the whole vehicle working condition in order to enable the finally obtained optimal calibration parameter to be suitable for various actual working conditions. As shown in fig. 5, the specific process of synthesizing the synthetic working condition is as follows:

the car networking cloud deck acquires the running condition information (the change relation curve of the speed and the time) of all the front cars in front of the car in real time, and the Markov algorithm is utilized to construct the representative typical working condition of the front cars. Firstly, carrying out motion segment classification (for example, classification is carried out by adopting a clustering algorithm, the classification result can be a working condition based on traffic conditions, including a working condition under road congestion, a working condition under road smoothness and the like), calculating a state transition probability matrix among all the classes, then realizing random recombination of the kinematics segments by utilizing a Markov method, considering that the synthesis of a single working condition is finished when the recombination segments reach the boundary of the previously calculated driving mileage, and when a statistical characteristic vector theta of the synthetic working condition is_iStatistical feature vector theta with the original data_tWhen the error between the two is smaller than the given threshold value, the two are considered to have approximate statistical characteristics, the working condition is recorded as a candidate working condition, and the next cycle is continued. After all the cycles are completed, the variance sum of the vehicle speed-acceleration joint probability density of each candidate working condition and the vehicle speed-acceleration joint probability density of the original working condition data is calculated, and the candidate working condition with the minimum variance sum is used as the final synthetic working condition. This part of the content is prior art and will not be described here.

And step three, preliminarily selecting calibration parameters. According to experience, the SOC threshold value SOC for switching the pure electric mode and the hybrid power mode_oPower threshold value P_oIs two important parameters influencing mode switching, and when the SOC of the whole vehicle is lower than the SOC threshold value_oTime, charge coefficient C_chgBased on the influence on the charging speed of the engine on the power battery and the distribution of the working points of the engine, the switching threshold value SOC of the pure electric mode and the hybrid power mode is adopted in the embodiment_oPower threshold P_oAnd coefficient of charge C_chgAs a preliminary selection of calibration parameters.

And step four, because the preliminarily selected calibration parameters are only determined according to experience, but actually have small influence on the system response, the sensitivity analysis of the calibration parameters is needed after the preliminary selection of the calibration parameters, simulation and result analysis are carried out in the platform shown in fig. 2, and the calibration parameters with the maximum influence on the system response are screened out from the sensitivity analysis, so that the final parameters to be calibrated are obtained. As shown in fig. 6, the specific process is as follows:

1) the SOC threshold value SOC needs to be determined first_oPower threshold P_oAnd coefficient of charge C_chgThe feasible region has more values in the feasible region, so that in an Iight software platform, a Latin hypercube method with less test times and uniform distribution is firstly utilized to sample the sample. The method ensures that the fitting of the calibration parameters and the system response is more accurate and real, and has very good space filling property and equilibrium.

2) And inputting the sampling results into a finished automobile simulation model of the Matlab software platform one by one to obtain a simulation result, and feeding the simulation result back to the Isight software platform. In the Isight software platform, a pareto chart and a main effect chart are made according to the influence degree of the three calibration parameters on the system response based on the simulation result, and the pareto chart and the main effect chart are respectively shown in FIG. 7 and FIG. 8.

As shown in FIG. 7, PbatEngOn²Indicating a power threshold value P_oThe second-order main effect of (B), PbaEngOn, represents the power threshold value P_oThe first-order dominant effect of PbaEngOn-SOC _ threshold represents the power threshold value P_oSOC threshold value SOC for mode switching_oInteraction effect of, SOC _ threshold represents SOC threshold value SOC_oFirst order dominant effect of (SOC-threshold)²SOC threshold value SOC for indicating mode switching_oSecond order main effect, Charge _ radio²Represents a charging coefficient C_chgThe Charge _ radio-PbatengOn represents the Charge coefficient C_chgAnd a power threshold value P_oThe Charge _ radio-SOC _ threshold represents the charging coefficient C_chgSOC threshold value SOC for mode switching_oThe Charge _ radio represents the charging coefficient C_chgThe first-order main effect of (2) clearly shows the contribution rate of each calibration parameter per se and the interaction effect to the system response, wherein blue represents a positive effect, red represents a negative effect, and the bar length represents the percentage of the contribution rate of each calibration parameter to the response. From pareto maps, pairs of calibration parametersThe energy consumption model presents strong interactivity and nonlinearity, and the power threshold value P_oSecond order main effect, power threshold value P_oFirst order dominant effect, power threshold value P_oSOC threshold value SOC for mode switching_oThe influence of the interaction effect on the system response is arranged in front of the influence of the interaction effect on the system response, and the charging coefficient SOC is arranged in front of the interaction effect_oThe interaction effect of (2) also has a certain influence on the response.

As shown in fig. 8, it can be seen from the graph that the slope of each curve obtains the influence degree of each calibration parameter on the system response, and the main effect of each calibration parameter on the system response can be evaluated separately. In the figure, PbatEngOn represents a power threshold value, SOC _ threshold represents an SOC threshold, and charge _ ratio represents a charge coefficient; in the main effect diagram, the larger the slope of the calibration parameter is, the larger the influence on the system response is, and the positive slope represents the positive effect, and the negative slope represents the negative effect. It can be seen that the power threshold value P_oAbsolute value of the slope of (1) and SOC threshold value SOC_oThe absolute value of the slope of (1) is large, so that the power threshold value SOC_oAnd SOC threshold value SOC_oThe impact on the system response is greatest.

According to fig. 7 and 8, the final parameter to be calibrated can be determined as the power threshold value P_oAnd SOC threshold value SOC_o。

Step five, determining the parameter to be calibrated as a power threshold value P_oAnd SOC threshold value SOC_oThen, the feasible fields of the two calibration parameters need to be determined. Moreover, because the value in the feasible region of each parameter to be calibrated is possible to be more, the value in the feasible region of the calibrated parameter is optimized by the self-adaptive simulation annealing algorithm, and the value in the feasible region of each parameter to be calibrated can be reduced, so that the simulation speed is accelerated. The adaptive simulated annealing algorithm has wide application range and can realize better optimization on nonlinear, multimodal and discontinuous target functions. And (4) inputting the whole vehicle working condition data acquired in the step two into the whole vehicle simulation model built in the step one so that the whole vehicle runs under the synthetic working condition. Since the number of the parameters to be calibrated is two, traversal is needed hereThe combination of different values of each parameter to be calibrated is obtained by traversing the feasible regions of the two parameters to be calibrated. For example, SOC_oThe optimized feasible region is [0.2,0.3,0.4,0.5 ]]，P_oThe optimized feasible region is [30,31,32 ]]Then the final combination is: (0.2,30), (0.3,30), (0.4,30), (0.5,30), (0.2,31), (0.3,31), (0.4,31), (0.5,31), (0.2,32), (0.3,32), (0.4,32), (0.5, 32). And traversing the obtained combination, inputting each group of data in the combination into a whole vehicle simulation model of the Matlab software platform to obtain a corresponding simulation result, namely outputting the energy consumption value of the energy consumption model, and finding the value of the result with the lowest energy consumption on the corresponding parameter to be calibrated, namely the optimal calibration parameter.

Therefore, parameter calibration of the whole vehicle in an online state can be completed. The method provides a theoretical basis for the exploration of the optimal calibration parameters, so that the calibration period is greatly reduced, and the effective improvement of the calibration efficiency is realized. Moreover, the energy consumption model in the method is not limited under any working condition, has strong adaptability to the working condition and ensures the optimal calibration effect.

In addition, if the whole vehicle is in an off-line state, data interaction can not be carried out with the vehicle networking platform, the whole calibration process is the same as that of the whole vehicle in an on-line state, and only the used working condition is not a synthetic working condition any more but a standard working condition. The standard working condition is a universal working condition.

In this embodiment, the calibration parameter initially selected includes the SOC threshold value SOC_oPower threshold value P_oAnd the charging coefficient, and finally determining the SOC threshold value SOC of the parameter to be calibrated by adopting a principal component analysis method and a pareto analysis method_oPower threshold value P_o. As another embodiment, other existing methods may be used to find the parameters that have a large influence on the energy consumption value of the energy consumption model as the calibration parameters, for example, under the condition of rich experience, it is not necessary to use complicated mathematical calculation to determine which parameters to calibrate.

In this embodiment, in order to perform comprehensive analysis on the results of oil consumption and power consumption, the theoretical power consumption model is equivalent to the oil consumption model, that is, the energy consumption model is represented by oil consumption. As other implementation modes, the theoretical oil consumption model can be converted into a power consumption model, and the final energy consumption model is represented by power consumption.

The embodiment of the device is as follows:

the embodiment provides a hybrid vehicle parameter calibration device, as shown in fig. 9, the device includes a memory and a processor, as well as a communication bus and a communication interface, and the processor, the communication interface and the memory complete mutual communication through the communication bus. The processor may be a microprocessor MCU, a programmable logic device FPGA, or the like, and the memory may be a high speed random access memory, or may be a non-volatile memory, such as one or more magnetic storage devices, flash memory, or the like. The processor may call logic instructions in memory to implement the following method:

building a whole vehicle simulation model according to the actual condition of a whole vehicle, wherein the whole vehicle simulation model comprises an energy consumption model, and the energy consumption model is determined according to the energy flow condition among all modules in a power system of the whole vehicle; the energy consumption model comprises a theoretical oil consumption model and a theoretical power consumption model, and the energy consumption model is characterized by adopting power consumption or oil consumption: if the oil consumption representation is adopted, the theoretical power consumption model is equivalent to the oil consumption model to obtain a first equivalent model, and the sum of the equivalent oil consumption value of the first equivalent model and the oil consumption value of the theoretical oil consumption model is used as the energy consumption value of the energy consumption model; if the power consumption representation is adopted, the theoretical oil consumption model is equivalent to the power consumption model to obtain a second equivalent model, and the sum of the equivalent power consumption value of the second equivalent model and the power consumption value of the theoretical power consumption model is used as the energy consumption value of the energy consumption model;

acquiring finished automobile working condition data and a feasible region of a parameter to be calibrated, inputting the finished automobile working condition data into the finished automobile simulation model, traversing in the feasible region of the parameter to be calibrated, and finding a value of the parameter to be calibrated corresponding to a result with the lowest energy consumption value of the energy consumption model as an optimal calibration parameter under the finished automobile working condition; if the number of the parameters to be calibrated is one, traversing each value in the feasible region of the parameters to be calibrated, and if the number of the parameters to be calibrated is at least two, traversing combinations of different values of each parameter to be calibrated, which are obtained by traversing in the feasible region of each parameter to be calibrated.

The logic instructions in the memory may be implemented in software functional units and sold or used as independent products, or may be stored in a computer readable storage medium.