A power distribution method for pure electric vehicle
By using fuzzy control theory and large sample multi-scene data acquisition methods in the energy management system of electric vehicles, a two-layer fuzzy control model was established, and the energy management control results were optimized, the power distribution problem in the existing technology was solved, and the power performance was improved and economic performance.
Patent Information
- Application Number
- CN202010661774.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-07-10
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2040-07-10
AI Technical Summary
The existing electric vehicle energy management system adopts a threshold value control method, which cannot optimally solve the power distribution problem, affecting power performance and economic performance. In addition, the confirmation and calibration of membership function of controlling model variables requires a large amount of data, and the effect is subject to sample size and scenarios. It has a great impact and is difficult to achieve working condition adaptability and global optimization.
Through large sample multi-scene data acquisition, the control model input and output membership function is formulated, and the fuzzy control theory is used to establish a two-layer fuzzy control model, define the fuzzy subset and membership function, formulate inference rules and defuzzy methods, and optimize the energy management control results.
It achieves relatively ideal power and economy, can provide higher output power at low power, reduce energy consumption at high power, and improve the power and economic performance of the vehicle for long-term cycles.
Smart Images

Figure CN111823883B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a power distribution method for a pure electric vehicle, and belongs to the technical field of energy management of pure electric vehicles. Background Art
[0002] The software and hardware system coordinates, distributes and controls the working energy of the energy conversion device of the electric vehicle power system. The hardware of the energy management system consists of a series of sensors, control units ECU and actuators, and the function of the software system is mainly to analyze and process the signals of the sensors, optimize the working energy of the energy conversion device, and issue instructions to the actuators. Therefore, it can be said that the function of the electric vehicle energy management system is to achieve the best energy flow between energy conversion devices (such as motors, energy storage devices, power conversion modules, thermal management systems, etc.) according to the characteristics of each component and the operating conditions of the vehicle, so as to maximize the energy utilization efficiency of the whole vehicle, while meeting the requirements of the basic technical performance (such as power, driving stability, etc.) and cost of the vehicle.
[0003] The current energy management of electric vehicles adopts the threshold value control method. Although the control is simple and reliable, the control rules are fixed and cannot optimally solve the power distribution problem, which affects the power performance and economic performance.
[0004] The confirmation and calibration of the membership function of the control model variables requires a large amount of data. The actual effect is greatly affected by the sample size and scenario. It is necessary to consider the global optimization of operating conditions, cycle conditions, special conditions, and dynamic programming to make the single driving cycle globally optimal. In order to make electric vehicles have good mechanical properties, electric drive performance and reasonable energy distribution, the energy management system of electric vehicles must effectively monitor and control the work of the energy system so that the energy of electric vehicles can flow optimally to maximize the use of energy and improve the economic performance of the vehicle. Summary of the invention
[0005] The present invention proposes a power distribution method for pure electric vehicles, which formulates the input-output membership function of the control model through large-sample multi-scenario data collection, and uses fuzzy control theory to optimize the energy management control results. The method has relatively ideal power and economy.
[0006] The present invention adopts the following technical solutions to solve the technical problems:
[0007] A power distribution method for a pure electric vehicle comprises the following steps:
[0008] Step 1: Build the control model
[0009] Through the analysis of the energy flow of the whole vehicle, the input and output variables and their relationships were found, and a two-layer fuzzy control model was developed;
[0010] Step 2: Define fuzzy subsets
[0011] Geometrically define and normalize input variables through practical experience and effects;
[0012] Step 3: Determine the membership function
[0013] Determine the membership function through data statistics of large samples and multiple scenarios combined with expert experience;
[0014] Step 4: Develop inference rules
[0015] Formulate reasoning rules based on project experience, usage scenarios, and subsystem requirements;
[0016] Step 5: Choose an inference method
[0017] Determine the classic Mamdani reasoning method through project experience and simulation analysis;
[0018] Step 6: Select the defuzzification method
[0019] The classic mom defuzzification method is determined through project experience and simulation analysis.
[0020] The input variables in step 1 are battery power, temperature offset, torque change rate, motor temperature, and vehicle speed, and the output variables are power distribution coefficient and motor power.
[0021] The two-layer fuzzy control model described in step 1 is a power allocation coefficient fuzzy control model and a motor output power fuzzy control model.
[0022] The fuzzy subsets in step 2 include input fuzzy subsets and output fuzzy subsets.
[0023] The input fuzzy subset is:
[0024] Fuzzy set of torque change rate (region division) {small, medium, large}; domain (value range) {0:0.1:1}
[0025] SOC fuzzy set {small, medium, large}; domain {0:0.1:1}
[0026] Temperature deviation rate fuzzy set {small, medium, large}; domain {0:0.1:1}
[0027] Motor temperature fuzzy set {small, medium, large}; domain {0:0.1:1}
[0028] Vehicle speed fuzzy set {small, medium, large}; domain {0:0.1:1}
[0029] The output fuzzy subset is:
[0030] Fuzzy set of power allocation coefficient {small, medium-small, medium, medium-large, large}; domain {0:0.1:1}.
[0031] The membership function in step 3 is
[0032] P bat =P mot +K TMS P TMS-MAX +P LP (1)
[0033]
[0034] P mot =f(V,T,K TMS ) (3)
[0035] Where: P bat is the battery power, P TMS-MAX is the maximum power of the thermal management system, P LP is the low voltage system power, P LP_i is the power of the low voltage subsystem, ΔP i is the power change of the low-voltage subsystem, λ is the power adjustment coefficient of the low-voltage system, and f() represents the mapping relationship between the motor power and the vehicle speed, motor temperature, and thermal management temperature function; K TMS is the power distribution coefficient; T is the motor temperature; V is the vehicle speed; P mot is the available output power of the motor.
[0036] The inference rule in step 4 is:
[0037] (1) When the vehicle starts, if the power battery operating temperature Tbat is low, P TMS , where P TMS Indicates the thermal management system power.
[0038] (2) When the vehicle starts, if the power battery operating temperature Tbat is normal, P motor , where P motor Indicates the motor power.
[0039] (3) When the car accelerates suddenly or climbs a slope, that is, dθ / dt is greater than a certain threshold, P is activated first. motor , meet the high power demand of the drive motor, and then start P TMS .
[0040] (4) When the car is running at normal speed, that is, dθ / dt is less than a certain threshold, P is satisfied first. TMS , improve the charging and discharging efficiency, and then meet P motor , ensuring the normal power demand of the drive motor.
[0041] (5) If the battery SOC is low, P should be satisfied first motor .
[0042] (6) If the battery SOC is high, P should be satisfied first TMS .
[0043] (7) If the motor temperature is high, P should be satisfied first TMS .
[0044] (8) If the vehicle speed is high, P should be given priority. TMS .
[0045] (9) If the vehicle speed is medium, P should be given priority. motor .
[0046] The classic mom defuzzification method described in step 6 is the maximum membership average method.
[0047] The beneficial effects of the present invention are as follows:
[0048] The present invention can provide a higher output power for the drive system when the battery power is low, and can reduce the energy consumption rate when the battery power is high, thereby improving the power performance and economic performance of the entire vehicle over a long period of time without increasing or decreasing the overall energy of the power battery. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] Figure 1 It is a topological diagram of the high-voltage components of a pure electric vehicle.
[0050] Figure 2 This is the structural diagram of the high-voltage components of a pure electric vehicle.
[0051] Figure 3 It is the control strategy model diagram.
[0052] Figure 4 This is an energy consumption comparison chart.
[0053] Figure 5 This is a comparison chart of available power.
[0054] Figure 6 This is a simulation diagram of driving range. DETAILED DESCRIPTION
[0055] The present invention is further described in detail below in conjunction with the accompanying drawings.
[0056] like Figure 1As shown in the figure, the power electrical components of pure electric vehicles mainly include air conditioning, motors, thermal management, and low-voltage power systems; the energy management system adjusts the air conditioning power according to the available battery power and the motor power demand to adapt to the power demand and economic demand of power performance. The power management controller, as the central control unit of the power / energy management and distribution system, is responsible for commanding, controlling and coordinating various components. The communication between the power management controller and all power converters and power control units is to send control instructions and receive sensor signals and status reports from these devices. It also communicates with the vehicle system controller to achieve interaction with other vehicle systems. All power and energy management algorithms are implemented in the power manager.
[0057] like Figure 2 As shown, the battery provides the energy required by all components to supply the thermal management system motor, power system, and low-voltage power supply system.
[0058] P bat =P mot +K TMS P TMS-MAX +P LP (1)
[0059]
[0060] P mot =f(V,T,K TMS ) (3)
[0061] Where: P bat is the battery power, P TMS-MAX is the maximum power of the thermal management system, P LP is the low voltage system power, P LP_i is the power of the low voltage subsystem, ΔP i is the power change of the low-voltage subsystem, λ is the power adjustment coefficient of the low-voltage system, and f() represents the mapping relationship between the motor power and the vehicle speed, motor temperature, and thermal management temperature function; K TMS is the power distribution coefficient; T is the motor temperature; V is the vehicle speed; P mot is the available output power of the motor.
[0062] like Figure 3 As shown, the energy flow distribution for the electric drive system and high-voltage accessories is to coordinate the output power of the power system and high-voltage accessories in different driving scenarios under the constraints of vehicle power, economy and smoothness to reduce the energy consumption of the whole vehicle; the power distribution coefficient of the energy management system is introduced according to the characteristics of the whole vehicle and the battery, and a two-level fuzzy control model is established.
[0063] SOC represents the battery charge level, where 0 indicates the lowest remaining power of the power battery; 1 indicates the highest remaining power of the power battery. ΔT represents the battery temperature deviation rate, where 0 indicates the smallest temperature difference between the current battery temperature and the optimal battery temperature; 1 indicates the largest temperature difference between the current battery temperature and the optimal battery temperature. K TMS represents the power distribution coefficient; 0 indicates the lowest available power required for thermal management; 1 indicates the highest available power required for thermal management. T represents the motor temperature; 0 indicates the lowest motor temperature; 1 indicates the highest motor temperature. V represents the vehicle speed; 0 indicates a lower vehicle speed; 1 indicates the maximum vehicle speed. P mot represents the maximum output power of the motor. dθ / dt represents the torque change rate; 0 indicates a lower frequency of the driver stepping on the accelerator pedal, i.e., the vehicle is driving normally; 1 indicates a higher frequency of the driver stepping on the accelerator pedal, i.e., the vehicle is accelerating rapidly or climbing a slope.
[0064] When Topi_min ≤ Tbat ≤ Topti_max, ΔT = 0;
[0065] When Tbat ≥ Topti_max, ΔT = (Tbat - Topti_max) / (Tmax - Topti_max);
[0066] When Tbat < Topti_min, ΔT = (Topti_min - Tbat) / (Topti_min - Tmin).
[0067] Among them, Topti_max is the upper limit value of the optimal working temperature range of the battery; Topti_min is the lower limit value of the optimal working temperature range of the battery; Tbat is the median value of the battery temperature.
[0068] Figure 4 Compared with the AND gate threshold control method and the no-thermal management control method, the energy consumption of the fuzzy control has better improvement, and the system economy is improved to a certain extent. Figure 5 Compared with the AND gate threshold control method and the no-thermal management control method, the available power of the battery of the fuzzy control has better improvement when meeting the power demand, and the system power performance is improved to a certain extent. Figure 6 Regarding the relationship between the cruising range, vehicle speed, and air conditioner, the strategy function reflecting the influence of the vehicle speed change and the change in air conditioner power on the cruising range is effective.
[0069] The vehicle energy consumption of the fuzzy control method is lower than that of the threshold control method and the no-thermal management control method, and it has better economy compared with the two. In the low SOC range, the output power that the energy system using this method can provide is higher than that of the threshold control method and the no-thermal management control method, indicating that it can provide a higher output power for the drive system in the later stage of vehicle driving, so that the whole vehicle can have an ideal power performance for a long time without increasing the overall energy of the power battery.
Claims
1. A power distribution method for a pure electric vehicle, characterized in that: The steps include: Step 1: Build the control model Through the analysis of the energy flow of the whole vehicle, the input and output variables and their relationships were determined, and a two-layer fuzzy control model was developed; The input variables are battery power, temperature offset, torque change rate, motor temperature and vehicle speed, and the output variables are power distribution coefficient and motor power; The two-layer fuzzy control model is a power allocation coefficient fuzzy control model and a motor output power fuzzy control model; Step 2: Define fuzzy subsets Geometrically define and normalize input variables; The fuzzy subsets include an input fuzzy subset and an output fuzzy subset. The input fuzzy subset is: The torque change rate fuzzy set {small, medium, large}, i.e., the region division; the domain {0:0.1:1}, i.e., the value range; SOC fuzzy set {small, medium, large}; domain {0:0.1:1}; Temperature deviation rate fuzzy set {small, medium, large}; domain {0:0.1:1}; Motor temperature fuzzy set {small, medium, large}; domain {0:0.1:1}; Vehicle speed fuzzy set {small, medium, large}; domain {0:0.1:1}; The output fuzzy subset is: Power allocation coefficient fuzzy set {small, medium-small, medium, medium-large, large}; domain {0:0.1:1}; Step 3: Determine the membership function The membership function is determined by statistically analyzing large sample and multiple scenarios; multiple scenarios refer to different driving scenarios; The membership function is: P bat =P mot +K TMS P TMS-MAX +P LP Formula 1 Formula 2 P mot =f(V,T,K TMS ) Formula 3 Where: P bat is the battery power, P TMS-MAX is the maximum power of the thermal management system, P LP is the low voltage system power, P LP_i is the power of the low voltage subsystem, ΔP i is the power change of the low-voltage subsystem, λ is the power adjustment coefficient of the low-voltage system, and f() represents the mapping relationship between the motor power and the vehicle speed, motor temperature, and thermal management temperature function; K TMS is the power distribution coefficient; T is the motor temperature; V is the vehicle speed; P mot is the available output power of the motor; Step 4: Develop inference rules Formulate reasoning rules by using scenarios and subsystem requirements; The inference rule is: When the vehicle starts, if the power battery operating temperature Tbat is low, P TMS , where P TMS Indicates the thermal management system power; When the vehicle starts, if the power battery operating temperature Tbat is normal, P motor , where P motor Indicates motor power; When the car accelerates suddenly or climbs a slope, that is, the torque change rate dθ / dt is greater than a certain set threshold, P is started first. motor , to meet the high power demand of the drive motor, and then start P TMS ; When the car is running at normal speed, that is, the torque change rate dθ / dt is less than a certain threshold, P is satisfied first. TMS , improve the charging and discharging efficiency, and then meet P motor , ensure the normal power demand of the drive motor; Step 5: Choose an inference method Determine the classic Mamdani reasoning method through simulation analysis; Step 6: Select the defuzzification method Determine the classic mom defuzzification method through simulation analysis; The classical mom defuzzification method is the maximum membership average method; The input-output membership function of the control model is formulated through large-sample multi-scenario data collection, and the fuzzy control theory is used to optimize the energy management control results.
