Optimal intelligent energy management method under alternation of multiple working conditions

Abstract

The invention discloses an optimal intelligent energy management method under alternation of multiple working conditions, relates to the field of automobile energy management, and solves the technical problem that an existing energy management strategy affects power output due to a real driving environment and working conditions. The method comprises the following steps of: 1, acquiring an automobile standard power domain, current traffic condition information and a current automobile power domain, and establishing an automobile power domain database according to the automobile standard power domain, the current traffic condition information and the current automobile power domain; 2, constructing a classification set model according to the automobile power domain database; and 3, detecting and monitoring the subsystems in the classification set model in real time and acquiring data in the subsystems. Reasonable utilization of energy of the hybrid electric vehicle is achieved, the dynamic property and economical efficiency of the vehicle are improved, meanwhile, data contained in the power domain database are various, working conditions capable of being compared and analyzed are many, and the application scene and field range is wide.

Description

Technical Field

[0001] The present invention relates to the field of automobile energy management, and more specifically, to an optimal intelligent energy management method under multiple alternating working conditions. Background Art

[0002] The function of the power domain controller is mainly to control the vehicle's powertrain, optimize the vehicle's power performance, and ensure the vehicle's power safety. Other functions of the power domain controller include but are not limited to engine management, transmission management, battery management, power distribution management, emission management, speed limit management, fuel and power saving management, etc. For hybrid vehicles, select an energy management strategy that matches the power system, and effectively coordinate various energy sources through the energy management strategy, so that the power system can work in the high-efficiency zone as much as possible, improve the efficiency of the power system and the utilization rate of various energy sources, thereby increasing the service life of the battery.

[0003] However, due to the characteristics of the current battery itself, the performance will be attenuated at high or low temperatures, and accessories such as comfort and entertainment consume a lot of energy. If the battery temperature is too high or too low, it will affect the power output, thereby affecting the vehicle's power and economy, causing more inconvenience to users. Secondly, the current energy management strategies have certain limitations, because there are many uncertainties in the real driving environment. In the actual driving process, there are many transient and steady-state multi-operating conditions, which have a great impact on the energy management strategy. Therefore, it is necessary to consider a variety of basic operating conditions to set the energy management strategy. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an optimal intelligent energy management method under multiple alternating working conditions in view of the deficiencies in the prior art, so as to solve the technical problem that the existing energy management strategy is affected by the real driving environment and working conditions and thus the power output is affected.

[0005] The optimal intelligent energy management method under multiple alternating working conditions described in the present invention comprises:

[0006] Step 1: Obtaining a standard power domain of an automobile, current traffic condition information, and a current power domain of an automobile, and establishing an automobile power domain database according to the standard power domain of the automobile, the current traffic condition information, and the current power domain of the automobile;

[0007] Step 2: constructing a classification set model based on the automobile power domain database;

[0008] Step 3: Perform real-time detection and monitoring of the subsystems in the classification set model and obtain data in the subsystems;

[0009] Step 4: Optimize the data in the subsystem to output an energy control decision signal.

[0010] As a further improvement, the method for establishing the automobile power domain database is to obtain the standard state value and standard parameters in the automobile standard power domain, obtain the current state value and current parameters of the current automobile power domain, and obtain a state difference value by subtracting the standard state value from the current state value. When the state difference value is greater than a preset state difference threshold, the standard state value is taken as the optimal state value; otherwise, the current state value is taken as the optimal state value.

[0011] Subtracting the standard parameter from the current standard parameter to obtain a parameter difference value, and when the parameter difference value is greater than a preset parameter difference threshold, the standard parameter is taken as the optimal parameter; otherwise, the current parameter is taken as the optimal parameter;

[0012] An automobile power domain database is established according to the optimal parameters, optimal state values ​​and current traffic condition information.

[0013] Furthermore, the method for constructing the classification set model is:

[0014] The first step: extracting keywords from the data in the automobile power domain database and indexing the keywords;

[0015] Step 2: Classify the keywords and generate multiple sets;

[0016] Step 3: Set up a classifier, input the set into the classifier for training to obtain a subsystem corresponding to the set, and construct the classification set model based on the subsystem.

[0017] Furthermore, the keywords include engine operating conditions, road conditions and vehicle mileage.

[0018] Furthermore, the method for optimizing the data in the subsystem is:

[0019] The data in the preset automobile standard database are integrated into standard cases, and the data in the subsystem are integrated into existing cases. The case similarity between the existing cases and the standard cases is calculated. When the case similarity is less than the preset standard similarity, the data in the subsystem corresponding to the existing case is used as an energy control decision signal; when the case similarity is greater than or equal to the standard similarity, the data in the automobile standard database corresponding to the standard case is used as an energy control decision signal.

[0020] Furthermore, the data in the automobile standard database and the data in the subsystem are integrated into the form of tables or images.

[0021] Furthermore, the expression for calculating the case similarity between the existing case and the standard case is:

[0022]

[0023] Where n is the number of case attributes, w j is the weight of the i-th attribute, p is a direct match, q is an indirect match, C i For the existing case, C r For the standard case, S(C i , C r ) is the case similarity.

[0024] Furthermore, the method for real-time detection and monitoring of the subsystems in the classification set model is:

[0025] Obtain the standard parameters in the preset automobile standard database, and subtract the standard parameters from the parameters in the subsystem to obtain a data difference. When the data difference is greater than or equal to the preset data alarm difference, a fault alarm is issued; when the data difference is less than the preset data alarm difference, no fault alarm is issued.

[0026] Furthermore, after the energy control decision signal is output, the energy control decision signal is recorded in a vehicle standard database.

[0027] Furthermore, a decision priority is set, and the data in the subsystem is optimized according to the decision priority and an energy control decision signal is output.

[0028] Beneficial Effects

[0029] The advantages of the present invention are:

[0030] The present invention establishes an automobile power domain database according to the acquired automobile standard power domain, current traffic condition information and current automobile power domain, constructs a classification set model according to the automobile power domain database, performs real-time detection and monitoring of subsystems in the classification set model and obtains data in the subsystems, optimizes the data in the subsystems to output energy control decision signals, realizes the rational use of hybrid automobile energy, and improves the power and economy of the vehicle. At the same time, the power domain database contains a wide range of data, and can perform comparative analysis on a large number of working conditions, and has a wide range of applicable scenarios and fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a flow chart of the energy management method of the present invention;

[0032] Figure 2 It is a schematic diagram of the energy management strategy flow based on fuzzy rules in the energy management method of the present invention;

[0033] Figure 3It is a torque curve diagram of the energy management strategy based on the Pontrya minimum in the energy management method of the present invention. DETAILED DESCRIPTION

[0034] The present invention will be further described below in conjunction with the embodiments, but it does not constitute any limitation to the present invention. Any limited number of modifications made by anyone within the scope of the claims of the present invention are still within the scope of the claims of the present invention.

[0035] See also Figure 1-Figure 3 , an optimal intelligent energy management method under multiple alternating working conditions of the present invention, the method comprises:

[0036] Step 1: Obtain the standard power domain of the vehicle, current traffic condition information and the current vehicle power domain, and establish a vehicle power domain database based on the standard power domain of the vehicle, the current traffic condition information and the current vehicle power domain.

[0037] The method for establishing the automobile power domain database is to obtain the standard state value and standard parameters in the automobile standard power domain, obtain the current state value and current parameters of the current automobile power domain, and obtain the state difference value by subtracting the standard state value from the current state value. When the state difference value is greater than a preset state difference threshold, the standard state value is taken as the optimal state value; otherwise, the current state value is taken as the optimal state value.

[0038] The standard parameter is subtracted from the current standard parameter to obtain a parameter difference. When the parameter difference is greater than a preset parameter difference threshold, the standard parameter is taken as the optimal parameter; otherwise, the current parameter is taken as the optimal parameter.

[0039] The current vehicle power domain database is established based on the optimal parameters, optimal state values ​​and current traffic conditions information.

[0040] The present invention also provides a method for establishing a vehicle longitudinal dynamics model in a vehicle power domain:

[0041] In the time domain model, the state quantity is x = [sv] T , where s is the vehicle displacement over time, and v is the vehicle speed. In summary, the vehicle longitudinal dynamics state expression is:

[0042]

[0043] Where u = F t , F t is the vehicle driving force in the current automotive power domain, is the vehicle longitudinal dynamic state quantity.

[0044] The general driving equation of a car is F t＝ F f+ F w+ Fi+ F j ;

[0045] Among them, F f is the rolling resistance of the current vehicle power domain, F w is the air resistance of the current vehicle power domain, F i is the slope resistance of the current vehicle power domain, F j It is the acceleration resistance in the current vehicle power domain.

[0046] The expression to obtain the vehicle power balance value is:

[0047] Create a vehicle longitudinal dynamics model based on vehicle power balance values, vehicle driving force, rolling resistance, air resistance, slope resistance, and acceleration resistance.

[0048] Step 2: Construct a classification set model based on the automobile power domain database. The method for constructing the classification set model is:

[0049] Step 1: Extract keywords from the data in the automotive power domain database and index the keywords;

[0050] Step 2: Classify keywords and generate multiple sets;

[0051] Step 3: Set up a classifier, input the set into the classifier for training to obtain the subsystem corresponding to the set, and build a classification set model based on the subsystem.

[0052] Keywords include engine operating conditions, road conditions and vehicle mileage.

[0053] Step 3: Perform real-time detection and monitoring of the subsystems in the classification set model and obtain data from the subsystems.

[0054] Summarize and classify based on keywords such as working conditions, road conditions, and vehicle mileage. For example, if the engine temperature is too high, the current detection will give priority to entering this set for search and analysis, that is, dividing the cluster set.

[0055] The method for real-time detection and monitoring of subsystems in the classification set model is:

[0056] Obtain the standard parameters in the preset automobile standard database, and subtract the standard data from the parameters in the subsystem to obtain the data difference. When the data difference is greater than or equal to the preset data alarm difference, a fault alarm is issued; when the data difference is less than the preset data alarm difference, no fault alarm is issued.

[0057] Step 4: Optimize the data in the subsystem to output the energy control decision signal. After the energy control decision signal is output, the energy control decision signal is recorded in the automobile standard database.

[0058] The method to optimize the data in the subsystem is:

[0059] Case-based reasoning method: Integrate the data in the preset automobile standard database into standard cases, integrate the data in the subsystem into existing cases, calculate the case similarity between the existing cases and the standard cases, and when the case similarity is less than the preset standard similarity, use the data in the subsystem corresponding to the existing case as the energy control decision signal; when the case similarity is greater than or equal to the standard similarity, use the data in the automobile standard database corresponding to the standard case as the energy control decision signal.

[0060] Integrate the data in the automotive standard database and the subsystem data into tables or images.

[0061] The expression for calculating the case similarity between the existing case and the standard case is:

[0062]

[0063] Where n is the number of case attributes, w j is the weight of the i-th attribute, p is a direct match, q is an indirect match, C i For the existing case, C r is a standard case, case set: C = {C1, C2, LC i}; The attribute set of the i-th case is

[0064] Set decision priorities, optimize the data in the subsystem according to the decision priorities, and output energy control decision signals. For example, for the problem of weak acceleration of the car, according to the preset decision priorities, first consider the key component of the internal combustion engine, obtain relevant data for analysis and optimization, and make management decisions, adjust the fuel ratio to improve combustion efficiency, adjust the gasoline pump or intake and exhaust valve power to improve, etc.; then enter the optimization of the transmission system, adjust the transmission ratio to reduce energy loss during the transmission process; finally, analyze whether the tire pressure is abnormal, and finally complete the system's optimization of the order of a certain problem.

[0065] The present invention provides three standard cases of energy control decisions:

[0066] like Figure 2 As shown, the first standard case is a control strategy based on fuzzy rules:

[0067] The control of the hybrid power system can be completed without an accurate mathematical standard model. It is more suitable for multi-variable, nonlinear, and time-varying hybrid power systems and has strong stability and robustness.

[0068] For the reasonable distribution of vehicle power of multiple power sources, when the power battery SOC (remaining power) is detected to be in a medium-high state, the ECU controls the vehicle to use the power battery energy first while meeting other needs; when the power battery SOC is in a low state, the vehicle economy is prioritized, and the fuel cell is controlled to output most of the power to meet the energy required for normal driving. The power battery SOC is maintained within a reasonable charge state range through energy management. The entire process is mainly controlled by fuzzy logic with small variables using the power battery SOC and operating power as the main input parameters.

[0069] like Figure 3 As shown in Figure 2, the second standard case is an energy management strategy based on the Pontrya minimum:

[0070] The Pontryagin Minimum (PMP) principle exists in the form of a set of optimal necessary conditions, which redefines the global optimal control problem based on local conditions that satisfy system states, co-state changes, and instantaneous minimization.

[0071] Assume that the state equation of the control system is:

[0072] where x (t) is the state quantity, u (t) For the control quantity, is the state value of the control system.

[0073] The goal of the optimization problem is to minimize the total cost consumed at the end of the system operation, which can be represented by the augmented cost function, as shown in the formula:

[0074] Among them J (tf) is the total cost of the system, tf represents the end time of the system, L(x (t) ,u (t) ) is the instantaneous cost function, Φ(x (tf) ) is the system termination state condition.

[0075] The motor-assisted energy control strategy uses the engine as the main power source, and the motor and battery assist in providing peak power. This control strategy is easy to optimize the engine operating conditions. Compared with the engine, the motor has a fast response and sensitive control, and it is easy to implement different control methods. This control strategy is adopted in most parallel hybrid systems. First, the power demand of the system is obtained according to the driver's instructions (accelerator pedal and brake pedal); according to the power demand, the controller controls the energy flow in the hybrid system; then, according to the vehicle speed, load and battery SOC, the SOC is used to divide the operating status of the engine and motor.

[0076] like Figure 3 As shown in the figure, when the SOC is greater than the lower limit cs_lo_soc, the engine efficiency is too low when the vehicle speed is low, the engine speed is lower than the starting speed, or the system torque requirement is too low. At this time, the engine is turned off and the electric motor provides all the driving torque ( Figure 3 When the system torque requirement is greater than the maximum torque that the engine can provide, the engine and the motor jointly provide torque ( Figure 3 The area above the maximum torque curve of the engine is shown in Figure 2).

[0077] The third standard case is a real-time energy management strategy based on minimizing equivalent fuel consumption:

[0078] In order to minimize the battery energy output, when the vehicle driving force F t When it is a negative value, the energy consumption is constant; when the vehicle driving force F t When it is positive, the battery outputs energy, and the energy consumption function is P(t). In a predicted time T, the energy consumption E t The values ​​can be described as:

[0079]

[0080] The basic principle is to save a certain amount of fuel through the use of electric energy; the implementation method is to convert the electric energy consumed or recovered by the motor into fuel consumption through a conversion factor (equivalence factor), and the objective function is converted into overall fuel consumption. By solving the minimum objective function, the output torque of the motor can be determined.

[0081] The essence of the adaptive equivalent fuel consumption algorithm is to use some algorithms to automatically adjust the equivalent factor to minimize the equivalent fuel consumption. Currently, most of them use PI controllers to adaptively adjust the equivalent factor.

[0082] PID control is a linear control, which is based on the constant r (t) The actual output value y (t) The control deviation e(t) is:

[0083] e(t)=r(t)-y(t);

[0084] Continuous situation

[0085]

[0086] Discrete Case

[0087]

[0088] Δu(k)=k p (e(k)-e(k-1))+k i e(k)+k d e(k)-2e(k-1)+e(k-2);

[0089] According to the research, the equivalence factor can be described as:

[0090]

[0091] Where S (u+1) is the equivalent factor of the next iteration cycle, S (u) is the equivalent factor of the current iteration cycle, k p is the proportional factor of the deviation between the target SOC and the SOC at time t, k i is the integral factor of the deviation between the target SOC and the SOC at time t. The equivalent factor S of the previous iteration interval [u, u+1] (u) The theoretical optimal equivalent factor S that maintains SOC balance within this range is optimal(u) If the SOC is much lower, the cost of the hybrid system consuming electric energy is relatively small, and the power battery still has a tendency to reduce SOC; if the SOC is not increased, (u) , then in the next iteration interval [u+1, u+2], the battery will still reduce the SOC, and continue to make the SOC tend to deviate from the SOC. At this time, it is necessary to increase S (u) Make it close to S optimal(u) , which will increase the battery SOC. The overall goal is to monitor SOC data parameters and related power output in real time, introduce equivalent factors, and use an adaptive equivalent fuel consumption minimum algorithm to maintain SOC stability as the optimization goal for comprehensive control.

[0092] The above is only a preferred embodiment of the present invention. It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the structure of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent.

Claims

1. An optimal intelligent energy management method under multiple alternating working conditions, characterized in that: The method includes, Step 1: Obtaining a standard power domain of an automobile, current traffic condition information, and a current power domain of an automobile, and establishing an automobile power domain database according to the standard power domain of the automobile, the current traffic condition information, and the current power domain of the automobile; Step 2: constructing a classification set model based on the automobile power domain database; Step 3: Perform real-time detection and monitoring of the subsystems in the classification set model and obtain data in the subsystems; Step 4: Optimize the data in the subsystem to output an energy control decision signal.

2. The optimal intelligent energy management method under multiple alternating working conditions according to claim 1 is characterized in that: The method for establishing the automobile power domain database is to obtain the standard state value and standard parameters in the automobile standard power domain, obtain the current state value and current parameters of the current automobile power domain, and obtain a state difference value by subtracting the standard state value from the current state value. When the state difference value is greater than a preset state difference threshold, the standard state value is taken as the optimal state value; otherwise, the current state value is taken as the optimal state value. Subtracting the standard parameter from the current standard parameter to obtain a parameter difference value, and when the parameter difference value is greater than a preset parameter difference threshold, the standard parameter is taken as the optimal parameter; otherwise, the current parameter is taken as the optimal parameter; An automobile power domain database is established according to the optimal parameters, optimal state values ​​and current traffic condition information.

3. The optimal intelligent energy management method under multiple alternating working conditions according to claim 1 is characterized in that: The method of constructing the classification ensemble model is: The first step: extracting keywords from the data in the automobile power domain database and indexing the keywords; Step 2: Classify the keywords and generate multiple sets; Step 3: Set up a classifier, input the set into the classifier for training to obtain a subsystem corresponding to the set, and construct the classification set model based on the subsystem.

4. The optimal intelligent energy management method under multiple alternating working conditions according to claim 3 is characterized in that: The keywords include engine operating conditions, road conditions and vehicle mileage.

5. The optimal intelligent energy management method under multiple alternating working conditions according to claim 1 is characterized in that: The method for optimizing the data in the subsystem is: The data in the preset automobile standard database are integrated into standard cases, and the data in the subsystem are integrated into existing cases. The case similarity between the existing cases and the standard cases is calculated. When the case similarity is less than the preset standard similarity, the data in the subsystem corresponding to the existing case is used as an energy control decision signal; when the case similarity is greater than or equal to the standard similarity, the data in the automobile standard database corresponding to the standard case is used as an energy control decision signal.

6. The optimal intelligent energy management method under multiple alternating working conditions according to claim 5, characterized in that: The data in the automobile standard database and the data in the subsystem are integrated into the form of tables or images.

7. The optimal intelligent energy management method under multiple alternating working conditions according to claim 5, characterized in that: The expression for calculating the case similarity between the existing case and the standard case is: Where n is the number of case attributes, w j is the weight of the i-th attribute, p is a direct match, q is an indirect match, C i For the existing case, C r For the standard case, S(C i , C r ) is the case similarity.

8. The optimal intelligent energy management method under multiple alternating working conditions according to claim 1, characterized in that: The method for real-time detection and monitoring of the subsystems in the classification set model is: Obtain the standard parameters in the preset automobile standard database, and subtract the standard parameters from the parameters in the subsystem to obtain a data difference. When the data difference is greater than or equal to the preset data alarm difference, a fault alarm is issued; when the data difference is less than the preset data alarm difference, no fault alarm is issued.

9. The optimal intelligent energy management method under multiple alternating working conditions according to claim 8, characterized in that: After the energy control decision signal is output, the energy control decision signal is recorded in the automobile standard database.

10. The optimal intelligent energy management method under multiple alternating working conditions according to claim 1, characterized in that: A decision priority is set, and data in the subsystem is optimized according to the decision priority and an energy control decision signal is output.

Images