Energy management control method for extended-range electric logistics vehicle

Abstract

The present invention discloses an energy management control method for an extended-range electric logistics vehicle, including establishing a physical model of the extended-range electric logistics vehicle; performing cluster analysis of driving conditions, identifying driving modes by combining variational mode decomposition and extreme learning machines based on the assumption that the driving style will not mutate in a short period of time, and obtaining a driving mode recognition result; according to the driving mode recognition result, a dynamic programming algorithm is used to calculate the reference SOC trajectory under the global optimum under the entire driving condition with fuel economy as the goal, and obtain the SOC reference trajectory; the driving mode recognition result is combined with the reference SOC trajectory, and the optimal control action is obtained by minimizing the objective function, and finally the instantaneous power of the engine is limited to maintain the stable charging and discharging of the SOC. The present invention combines the predicted driving condition with the power allocation strategy, simplifies the energy management strategy framework and reduces the computational load of the energy management strategy.

Description

Technical Field

[0001] The present invention belongs to the technical field of hybrid vehicle energy management, and in particular relates to an energy management control method for an extended-range electric logistics vehicle. Background Art

[0002] Nowadays, in order to alleviate the crisis of environmental pollution and energy depletion, extended-range electric vehicles have received widespread attention in the automotive industry, such as urban commuter vehicles and logistics vehicles. Generally, extended-range electric vehicles are directly driven by a drive motor, and the energy comes from power batteries and other energy sources, which can be called an electric-electric coupling mode. Due to the particularity of the power source, the car is also named a series hybrid electric vehicle. Compared with traditional cars and pure electric vehicles, extended-range electric vehicles have two unique advantages. On the one hand, by refueling without a large-capacity battery, the weight and cost of the vehicle are greatly reduced, solving the problem of "charging anxiety". On the other hand, the mechanical connection between the wheel and the engine is decoupled, the chassis distribution is more flexible, and the engine can reasonably work in a high-efficiency area. In summary, extended-range electric vehicles have the following advantages: small engine, efficient auxiliary power unit, long service life and mature matching technology. However, due to the large size of the powertrain and auxiliary power unit, the cost of the whole vehicle is high. Therefore, extended-range electric vehicles are mainly used in special vehicles such as military vehicles, logistics vehicles and heavy commercial vehicles. In order to achieve low operating costs for extended-range electric vehicles, it is necessary to further explore and develop an adaptive power allocation strategy to obtain higher fuel economy by controlling the quasi-static characteristics of multiple power sources.

[0003] Improving the energy efficiency and emission performance of extended-range electric vehicles is its primary goal, which largely relies on the use of reasonable energy management strategies for power distribution among multiple power sources. With the development of intelligent control, driving mode recognition, model predictive control and reinforcement learning have been used to study real-time and optimal energy management strategies. Compared with other intelligent controls, driving mode recognition is more applicable, convenient and reliable. Fuzzy logic classifiers, neural networks, support vector machines, etc. are currently more popular solutions, among which Markov chain methods and dynamic programming are integrated into the optimal energy management system to identify the driving mode of battery and supercapacitor hybrid vehicles. In addition, driving mode recognition technology based on support vector machines has been introduced to achieve more efficient energy distribution. However, due to the huge computational load and changing traffic environment, intelligent driving mode recognition methods are difficult to execute in real time. Summary of the invention

[0004] In view of the deficiencies in the prior art, the technical problem that the present invention intends to solve is to provide an energy management control method for an extended-range electric logistics vehicle, which combines driving conditions with power allocation strategies.

[0005] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0006] A first aspect of the present invention provides an energy management control method for an extended-range electric logistics vehicle, comprising the following steps:

[0007] Step 1: Establish a physical model of the extended-range electric logistics vehicle;

[0008] Step 2: Perform a cluster analysis of driving conditions, and identify driving modes by combining variational mode decomposition and extreme learning machine based on the assumption that driving style will not change suddenly in a short period of time, and obtain the driving mode recognition result;

[0009] Step 3: Based on the driving mode recognition result obtained in step 2, a dynamic programming algorithm is used to calculate the reference SOC trajectory under the global optimal state under the entire driving condition with fuel economy as the goal, and the SOC reference trajectory is obtained;

[0010] Step 4: Combine the driving mode recognition results with the reference SOC trajectory, obtain the optimal control action by minimizing the objective function, and ultimately limit the instantaneous power of the engine to maintain a smooth charge and discharge of the SOC.

[0011] Furthermore, in step 1, the physical model of the extended-range electric logistics vehicle includes a three-cylinder engine fuel consumption model, a drive motor required power, a battery SOC and an actual vehicle speed obtained by analyzing the speed-time curve of the extended-range electric vehicle.

[0012] Furthermore, in step 2, a six-dimensional characteristic data matrix of five driving conditions is selected as the input of the control method, and a cluster analysis is performed to divide the five driving conditions into three driving modes, namely, a low-speed cycle, a medium-speed cycle and a high-speed cycle.

[0013] Furthermore, the driving mode recognition result obtained by combining variational mode decomposition and extreme learning machine includes: variational mode decomposition, feature synthesis and pattern recognition of vehicle speed time series, specifically including:

[0014] The vehicle historical speed is decomposed into eigenmode functions of different frequencies using variational modal analysis;

[0015] Select x intrinsic mode functions and sum them to form the identification signal;

[0016] An extreme learning machine is constructed, including an input layer, a hidden layer, and an output layer. The input weights and thresholds are randomly assigned, and then the output weights are calculated through the regularization principle. The average value of 10 repeated training values ​​is taken as the driving mode recognition result.

[0017] Furthermore, in step 4, the comprehensive energy consumption is defined as the sum of the engine fuel consumption and the motor equivalent fuel consumption, and is defined as the objective function. The state variable and the control variable are SOC and engine power, respectively. Constraints are set, and the approximate optimal solution of the equivalent fuel consumption minimum strategy is obtained by adjusting the equivalent factor under the current driving mode.

[0018] A second aspect of the present invention provides an energy management control device for an extended-range electric logistics vehicle, comprising:

[0019] The first module is used to establish the physical model of the extended-range electric logistics vehicle;

[0020] The second module is used to perform cluster analysis of driving conditions. Based on the assumption that driving style will not change suddenly in a short period of time, the driving mode is identified by combining variational mode decomposition and extreme learning machine to obtain the driving mode recognition result.

[0021] The third module is used to calculate the reference SOC trajectory under the global optimum under the entire driving condition by using a dynamic programming algorithm with fuel economy as the goal according to the driving mode recognition result obtained in the second module, and obtain the SOC reference trajectory;

[0022] The fourth module is used to combine the driving mode recognition results with the reference SOC trajectory, obtain the optimal control action by minimizing the objective function, and ultimately limit the instantaneous power of the engine to maintain smooth charging and discharging of the SOC.

[0023] A third aspect of the present invention provides a computer device, comprising: a processor and a memory communicatively connected to the processor;

[0024] The memory stores a computer program;

[0025] The processor executes the computer program stored in the memory to implement the above-mentioned energy management control method for an extended-range electric logistics vehicle.

[0026] A fourth aspect of the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the above-mentioned energy management control method for an extended-range electric logistics vehicle.

[0027] A fifth aspect of the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned energy management control method for an extended-range electric logistics vehicle.

[0028] Compared with the prior art, the beneficial effects of the present invention are: the present invention proposes an energy management strategy that combines the predicted driving conditions with the power allocation strategy, simplifies the energy management strategy framework and reduces the computational load of the energy management strategy. The control method of the present invention enables the extended-range electric logistics vehicle to adjust relevant parameters in real time during the life cycle of the vehicle, and limits the instantaneous power of the engine to maintain smooth charging and discharging of the SOC. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flow chart of an energy management control method for an extended-range electric logistics vehicle provided by the present invention;

[0030] Figure 2 A schematic diagram of the system configuration of the extended-range electric logistics vehicle provided by the present invention;

[0031] Figure 3 The cycle conditions provided by the present invention are: (a) five driving conditions (b) low speed cycle (c) medium speed cycle (d) high speed cycle;

[0032] Figure 4 A schematic flow chart of a driving mode recognition method provided by the present invention;

[0033] Figure 5 A schematic diagram of the extreme learning machine framework provided by the present invention;

[0034] Figure 6 This is a schematic diagram of the SOC change trend curve provided by the present invention. DETAILED DESCRIPTION

[0035] The specific embodiments of the present invention are given below. The specific embodiments are only used to further illustrate the present invention in detail and do not limit the protection scope of the present invention.

[0036] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and the like are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", and the like may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0037] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood by specific circumstances.

[0038] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0039] Embodiment 1:

[0040] like Figure 1 As shown, the present invention provides an energy management control method for an extended-range electric logistics vehicle, comprising the following steps:

[0041] Step S1: Establishing a physical model of the extended-range electric logistics vehicle;

[0042] Step S2: performing cluster analysis of the working conditions, and based on the assumption that the driving style will not change suddenly in a short period of time, combining variational mode decomposition and extreme learning machine to identify the driving mode, and obtaining a driving mode recognition result;

[0043] Step S3: using a dynamic programming algorithm to guide battery energy distribution, taking fuel economy as a goal, and calculating a reference SOC trajectory under the global optimum under the entire driving condition;

[0044] Step S4: Combine the driving mode recognition result with the reference SOC trajectory, obtain the optimal control action by minimizing the objective function, and ultimately limit the instantaneous power of the engine to maintain a smooth charge and discharge of the SOC.

[0045] Specifically, in step S1, a physical model is established according to the basic parameters of the vehicle provided in Table 1, including a three-cylinder engine fuel consumption model, a motor power consumption model, and a battery model.

[0046] The basic parameters of the vehicle are shown in Table 1. The overall configuration of the vehicle is shown in Table 1. Figure 2 shown.

[0047] Table 1

[0048]

[0049] The fuel consumption model of a three-cylinder engine is as follows:

[0050]

[0051] Among them, Q e is the fuel consumption, T Eis the engine torque, ω E is the engine speed, b E is the engine fuel consumption rate, ρ f is the fuel density.

[0052] The motor power consumption model is as follows:

[0053]

[0054] Among them, Q f is the motor power consumption, T M is the motor torque, ω M is the motor speed, η M is the motor efficiency.

[0055] The battery model is as follows:

[0056]

[0057] Among them, SOC init is the initial SOC, U VOC is the open circuit voltage, R B is the internal resistance of the battery, P B is the battery power.

[0058] Specifically, in step S2, 5 operating conditions, including CLTC-P, WLTC, NEDC, WLTP, and UDDS, and the corresponding six-bit feature data matrix, including maximum vehicle speed, average vehicle speed, maximum acceleration, average acceleration, maximum deceleration, and average deceleration, are selected. The five driving conditions are divided into three driving modes by using the cluster analysis method. In the reconstructed driving cycle, a large number of driving modules have similar basic profiles. The three driving modes are defined as low-speed cycle, medium-speed cycle, and high-speed cycle, respectively, as shown in FIG. Figure 3 shown.

[0059] Driving mode recognition flow chart Figure 4 As shown in the figure, there are three main processes: variational mode decomposition of vehicle speed time series, feature synthesis and pattern recognition. Variational mode decomposition is a signal decomposition and estimation method. Based on the assumption that driving style will not change suddenly in a short period of time, variational mode analysis is used to decompose the vehicle's historical speed into intrinsic mode functions of different frequencies. The random and complex characteristics of the generated vehicle speed can be clearly displayed in the signal. Feature synthesis is to select x intrinsic mode functions and sum them to form a recognition signal. The structure of the extreme learning machine is as follows: Figure 5 As shown, it includes input layer, hidden layer and output layer. This method randomly assigns input weights and thresholds, and then calculates output weights through regularization principle. Extreme learning machine is used as the classification key of driving mode recognition method to extract the intricate input-output mapping between historical mode and predicted mode.

[0060] Pattern recognition improves the real-time and robustness of training by introducing extreme learning machines, which can eliminate redundant data and is the key to driving mode recognition. The classification accuracy is often affected by the feature dimension, so the node parameters of the hidden layer of the single hidden layer neural network are randomly obtained to ensure that the convergence of the network is not affected. According to the random characteristics of the extreme learning machine, the average value of 10 repeated training values ​​is finally defined as the output, which is the driving mode recognition result.

[0061] Specifically, in step S3, according to the driving mode recognition result obtained in step 2, a dynamic programming algorithm is adopted. The present invention takes fuel economy as a goal, engine power as a control variable, SOC value as a state variable, and the calculation step size is set to 1s. The SOC trajectory under the global optimal state under the entire driving condition is calculated, and finally the SOC reference trajectory is obtained.

[0062] The SOC reference trajectory is expressed as:

[0063] SOC ref (t+T w )=SOC(t)-K(d)·T ω ;

[0064] Wherein, K is a function of the aforementioned driving mode d. w is the predicted duration.

[0065] Specifically, in step S4, based on the driving mode recognition result of step 2 and the SOC reference trajectory obtained in step 3, the comprehensive energy consumption is defined as the sum of the engine fuel consumption and the motor equivalent fuel consumption, and it is defined as the objective function, and the state variable and the control variable are SOC and engine power, respectively.

[0066] Therefore, the optimal control problem can be expressed as:

[0067]

[0068] Among them, J is the comprehensive energy consumption, f(x(t), u(t)) is the state equation, F(x(u, t), u, t) is the cost function, and the calculation of the cost function is related to the three-cylinder engine fuel consumption and motor power consumption formula in step 1.

[0069] The equivalent factors under different cycles may cause the logistics vehicle to be unable to operate normally outside the allowed SOC boundary. By adjusting the equivalent factors under the current driving mode, an approximate optimal solution for the strategy of minimizing equivalent fuel consumption is obtained. The final effect of the present invention is as shown in the attached figure. Figure 6 As shown, the optimal control action is obtained by minimizing the objective function, and ultimately the instantaneous power of the engine is limited to maintain a smooth charge and discharge of the SOC.

[0070] Embodiment 2:

[0071] An energy management control device for an extended-range electric logistics vehicle, comprising:

[0072] The first module is used to establish the physical model of the extended-range electric logistics vehicle;

[0073] The second module is used to perform cluster analysis of driving conditions. Based on the assumption that driving style will not change suddenly in a short period of time, the driving mode is identified by combining variational mode decomposition and extreme learning machine to obtain the driving mode recognition result.

[0074] The third module is used to calculate the reference SOC trajectory under the global optimum under the entire driving condition by using a dynamic programming algorithm with fuel economy as the goal according to the driving mode recognition result obtained in the second module, and obtain the SOC reference trajectory;

[0075] The fourth module is used to combine the driving mode recognition results with the reference SOC trajectory, obtain the optimal control action by minimizing the objective function, and ultimately limit the instantaneous power of the engine to maintain smooth charging and discharging of the SOC.

[0076] Embodiment three:

[0077] A computer device comprises: a processor and a memory communicatively connected to the processor;

[0078] The memory stores a computer program;

[0079] The processor executes the computer program stored in the memory to implement the above-mentioned energy management control method for an extended-range electric logistics vehicle.

[0080] Embodiment 4:

[0081] A computer-readable storage medium stores a computer program, which, when executed by a processor, is used to implement the above-mentioned energy management control method for an extended-range electric logistics vehicle.

[0082] Embodiment five:

[0083] A computer program product includes a computer program, which, when executed by a processor, implements the above-mentioned energy management control method for an extended-range electric logistics vehicle.

[0084] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

[0085] Any matters not described in the present invention are applicable to the prior art.

Claims

1. An energy management control method for an extended-range electric logistics vehicle, characterized in that: The steps include: Step 1: Establish a physical model of the extended-range electric logistics vehicle, wherein the physical model includes a three-cylinder engine fuel consumption model, a drive motor required power, a battery SOC and an actual vehicle speed obtained by analyzing the speed-time curve of the extended-range electric vehicle; Step 2: Perform a cluster analysis of driving conditions, and identify driving modes by combining variational mode decomposition and extreme learning machine based on the assumption that driving style will not change suddenly in a short period of time, and obtain the driving mode recognition result; The driving mode recognition result obtained by combining variational mode decomposition and extreme learning machine includes: variational mode decomposition, feature synthesis and pattern recognition of vehicle speed time series, specifically including: The vehicle historical speed is decomposed into eigenmode functions of different frequencies using variational modal analysis; Select x intrinsic mode functions and sum them to form the identification signal; Construct an extreme learning machine, including input layer, hidden layer and output layer, randomly assign input weights and thresholds, and then calculate the output weights through the regularization principle, and take the average of 10 repeated training values ​​as the driving mode recognition result; Step 3: Based on the driving mode recognition result obtained in step 2, a dynamic programming algorithm is used to calculate the reference SOC trajectory under the global optimal state under the entire driving condition with fuel economy as the goal, and the SOC reference trajectory is obtained; Step 4: Combine the driving mode recognition results with the reference SOC trajectory, obtain the optimal control action by minimizing the objective function, and ultimately limit the instantaneous power of the engine to maintain a stable charge and discharge of the SOC; Specifically, the comprehensive energy consumption is defined as the sum of the engine fuel consumption and the motor equivalent fuel consumption, and it is defined as the objective function. The state variable and control variable are SOC and engine power respectively. Constraints are set, and the approximate optimal solution of the equivalent fuel consumption minimum strategy is obtained by adjusting the equivalent factor under the current driving mode.

2. The energy management control method of an extended-range electric logistics vehicle according to claim 1, characterized in that: In step 2, a six-dimensional characteristic data matrix of five driving conditions is selected as the input of the control method, and cluster analysis is performed to divide the five driving conditions into three driving modes, namely, a low-speed cycle, a medium-speed cycle and a high-speed cycle.

3. An energy management control device for an extended-range electric logistics vehicle, characterized in that: include: The first module is used to establish a physical model of the extended-range electric logistics vehicle, wherein the physical model of the extended-range electric logistics vehicle includes a three-cylinder engine fuel consumption model, a drive motor required power, a battery SOC and an actual vehicle speed obtained by analyzing the speed-time curve of the extended-range electric vehicle; The second module is used to perform cluster analysis of driving conditions. Based on the assumption that driving style will not change suddenly in a short period of time, the driving mode is identified by combining variational mode decomposition and extreme learning machine to obtain the driving mode recognition result. The driving mode recognition result obtained by combining variational mode decomposition and extreme learning machine includes: variational mode decomposition, feature synthesis and pattern recognition of vehicle speed time series, specifically including: The vehicle historical speed is decomposed into eigenmode functions of different frequencies using variational modal analysis; Select x intrinsic mode functions and sum them to form the identification signal; Construct an extreme learning machine, including input layer, hidden layer and output layer, randomly assign input weights and thresholds, and then calculate the output weights through the regularization principle, and take the average of 10 repeated training values ​​as the driving mode recognition result; The third module is used to calculate the reference SOC trajectory under the global optimum under the entire driving condition by using a dynamic programming algorithm with fuel economy as the goal according to the driving mode recognition result obtained in the second module, and obtain the SOC reference trajectory; The fourth module is used to combine the driving mode recognition results with the reference SOC trajectory, obtain the optimal control action by minimizing the objective function, and ultimately limit the instantaneous power of the engine to maintain smooth charging and discharging of the SOC; specifically, the comprehensive energy consumption is defined as the sum of the engine fuel consumption and the motor equivalent fuel consumption, and it is defined as the objective function. The state variable and control variable are SOC and engine power respectively. Constraints are set, and the approximate optimal solution of the equivalent fuel consumption minimum strategy is obtained by adjusting the equivalent factor under the current driving mode.

4. A computer device comprising: A processor and a memory communicatively connected to the processor; The memory stores a computer program; The processor executes the computer program stored in the memory to implement the energy management control method for an extended-range electric logistics vehicle as described in any one of claims 1-2.

5. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, it is used to implement the energy management control method of an extended-range electric logistics vehicle as described in any one of claims 1-2.

6. A computer program product, comprising a computer program, which, when executed by a processor, implements an energy management control method for an extended-range electric logistics vehicle as described in any one of claims 1-2.

Images