Optimization design method based on dual-motor mode switching rule

Abstract

The invention provides an optimization design method based on a dual-motor mode switching rule, and aims to solve the problem of frequent mode switching of a single / dual motor in a medium-low speed and medium-low torque region. Determining a single / double-motor mode switching rule with optimal energy consumption based on self-driving characteristics of the double motors and transmission parameters of the whole vehicle; by designing a filtering optimization algorithm, driving mode discrete points are traversed based on a mode switching rule, and outliers are judged and replaced according to the magnitude of domain values, mode numbers and median numbers of the discrete points. The optimized mode switching rule boundary is clearer, the mode switching frequency of the dual-motor system in the working process is greatly reduced on the premise that the dynamic property requirement is guaranteed and the energy consumption loss is very small, and the driving quality of the vehicle and the smoothness of the system are improved.

Description

Technical Field

[0001] The invention relates to an optimization design method based on dual-motor mode switching rules, and belongs to the technical field of electric vehicle power systems. Background Art

[0002] With the rapid development of electric vehicle technology, electric vehicles have shown great potential in solving the global energy crisis and reducing emissions. Electric vehicles have attracted widespread attention from researchers around the world due to their advantages such as low noise, low cost per kilometer and strong power. However, some inherent defects of electric vehicles, such as limited driving range, long charging requirements and inconvenient charging facilities, have limited the large-scale application of electric vehicles. The core of the electric vehicle power system technology field is to improve power, economy, comfort, reliability and safety while reducing costs. Recently, the dual-motor drive system of electric vehicles has attracted attention due to its high energy efficiency and strong power. In terms of mode switching of the dual-motor drive system, achieving a balance between economy and power under normal driving conditions is the key to the dual-motor drive system. By optimizing the mode switching rules, the boundaries of the mode switching rules are made clearer. Under the premise of ensuring power requirements and small energy consumption losses, the number of dual-motor system mode switching times during operation is significantly reduced, and the driving quality of the vehicle and the smoothness of the system are improved.

[0003] Based on this, in order to optimize the dual-motor mode switching rules and improve the vehicle driving quality and system smoothness, an optimization design method based on the dual-motor mode switching rules is proposed. By designing a filtering optimization algorithm, the outliers in the mode switching rules are removed and replaced. At the same time, the optimized mode switching rules are output to achieve smooth mode switching under the dual-motor drive system. Summary of the invention

[0004] The purpose of the present invention is to provide an optimization design method based on dual-motor mode switching rules to solve the problem of frequent mode switching of single / dual motors in medium and low speed and medium and low torque areas, and greatly reduce the number of mode switching of dual-motor systems during operation while ensuring power requirements and very little energy loss. This method proposes an optimization design method, which optimizes the mode switching rules by removing and replacing outliers while obtaining single / dual motor mode switching rules. The optimized mode switching rules can greatly reduce the number of mode switching times, and have the advantages of low computational cost and good optimization effect. Its technical content is as follows:

[0005] In the first step, the dual-motor two-speed electric drive axle is equipped with a power motor and an economic motor, which includes a total of eight working modes;

[0006] The second step is to evenly discretize the peak speed of the motor into 100 points based on the parameters of the dual motors to obtain the speed range of the two motors. and According to the transmission ratio of the vehicle power system, the driving torque-driving speed characteristic curve of the single / dual motor in each working mode is calculated respectively, as shown below:

[0007] Single motor drive:

[0008]

[0009] Dual motor drive:

[0010]

[0011] Where k1=1, 2, 3, 4; k2=5, 6, 7, 8 are the corresponding single / dual motor working mode states, r is the wheel radius, i0 is the main reducer transmission ratio, is the AMT transmission ratio in k1 mode, The AMT transmission ratio of the two motors in K2 mode is: To drive the vehicle speed, is the economic motor speed, T' MG_2 The torque interpolated from the economic motor drive characteristics, Output torque in k1 and k2 modes;

[0012] The third step is to find the maximum torque envelope of single / dual motor drive based on the characteristic curves of each working mode in the second step, and evenly discretize the torque on the envelope into 100 points downward, and use this as the working point. and

[0013] The fourth step is to traverse each operating point based on the operating point calculated in the third step and calculate the energy consumption under the single motor mode and the dual motor mode. For the mode that fails to meet the torque requirement of the operating point, its energy consumption is defined as INF. The energy consumption is expressed as follows:

[0014] Single motor mode:

[0015]

[0016] Dual motor mode: For a certain operating point, in the dual motor drive mode, the torque of the power motor and the economic motor must meet Decompose it into 100 torque combinations, and solve the torque combination of the power motor and the economic motor that minimizes energy consumption among the 100 torque combinations under this operating point:

[0017]

[0018] In the formula, is the transmission ratio of AMT1 and AMT2 in k2 mode, T MG1 ,T MG2is the output torque of the power motor and the economic motor, η MG is the driving efficiency of the motor, E min (x,y) is the minimum energy consumption corresponding to the torque combination, is the energy consumption in k1 and k2 modes;

[0019] Step 5: Based on the motor energy consumption of each mode under each operating point calculated in step 4, the lowest energy consumption is used as the control target to determine the working mode distribution with the best energy consumption for single / dual motors;

[0020] Step 6: Based on the single-motor working mode distribution and the dual-motor working mode distribution with the lowest energy consumption determined in step 5, establish single / dual-motor mode switching rules X1 and X2, and use the mode switching rules X1 and X2 as inputs of the filtering optimization algorithm;

[0021] Step 7: traverse the mode switching rules X1 and X2, select the area window with point X(x, y) as the center point in the two-dimensional rule table, select the size of the area window as a 3×3 block, use the block data as the sliding input signal (data sequence), and extract the neighborhood window value sequence r(k), where k ranges from 1 to 8 and is arranged from top to bottom and from left to right according to the matrix number. r(k) is defined as follows:

[0022] r(k)=X(x+Δ i (k),y+Δ j (k)),k=1,2,...,8

[0023] In the formula, Δ i (k), Δ j (k) are the offsets of the kth element relative to the center point X(x, y) in the row and column directions, respectively. The value of the offset is determined by the value of k;

[0024] Step 8: Boundary value processing: determine whether the center point X(x,y) is a boundary value. The specific determination method includes:

[0025] Step 8.1: If the domain window value sequence of the center point X(x,y) exceeds the index range, the center point X(x,y) is the boundary value of the two-dimensional table, and symmetric filling is used to expand the domain window data of the center point X(x,y);

[0026] Step 8.2: If the domain window value sequence of the center point X(x,y) does not exceed the index range, the center point X(x,y) is not a boundary value, and proceed to the next step;

[0027] The ninth step is to determine whether the target point X(x,y) is an outlier. The specific determination method includes:

[0028] Step 9.1: If the target point X(x, y) and the domain value r(k) are all equal, it means that the target point is not an outlier, and the point X(x, y) is directly used as the output of the algorithm;

[0029] Step 9.2: If there is a deviation between the target point X(x,y) and the domain value r(k), the mode t of the domain value r(k) is calculated and used as the basis for judging the outlier point. If the target point X(x,y) is equal to the mode t, it indicates that the target point is not an outlier point, and the point X(x,y) is directly used as the output of the algorithm;

[0030] Step 9.3: If there is a deviation between the target point and the mode t, the MAD value of the target point X(x, y) is calculated as the characterization of the outlier. The MAD value is defined as follows:

[0031] MAD=median(|X i -median(X)|)

[0032] Where, X i is an element in the data sequence, where i = 1, 2, 3, ... 8; median(X) is the median of the data sequence; |X i -median(X)| represents the absolute difference between each data point and the median;

[0033] Step 9.4: Calculate the difference between the target point X(x,y) and the median(X) of the data sequence. If the difference is less than 3 times the MAD value, the mode t is used as the output of the algorithm.

[0034] Step 10: If the difference in step 9.4 is greater than 3 times the MAD value, the target point X(x,y) is an outlier, and the suboptimal solution X(x,y) of the target point is replaced by c Assign the value to X(x,y), and return to step 9 to recalculate;

[0035] Step 11: Determine whether the traversal of the single / dual motor mode switching rules has been completed. If not, return to step 8. If it has been reached, proceed to the next step.

[0036] Step 12: Output the optimized single / dual motor drive mode switching rules and end the algorithm.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] (1) Compared with the original mode switching rules, the optimization design method of the dual-motor mode switching rule of the present invention can reduce the number of mode switching times while taking into account the power and economy of the drive system, greatly reduce the impact during the mode switching process, and improve the driving quality of the vehicle and the smoothness of the system;

[0039] (2) The optimization design method of the dual-motor mode switching rules described in the present invention optimizes the torque distribution of the dual-motor system by performing torque distribution based on the minimum energy consumption of the dual-motor system, thereby ensuring both the economy and power of the dual-motor system. During the mode optimization process, real-time monitoring can be performed, the calculation cost is low, and the iterative optimization is fast, thereby improving the system efficiency.

[0040] (3) The optimization design method of the dual-motor mode switching rule described in the present invention has good scalability and adaptability, and can be applied to different types of dual-motor drive vehicles, including hybrid electric vehicles, pure electric vehicles, and plug-in hybrid electric vehicles, etc., and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flow chart of the optimization design method of an embodiment of the present invention.

[0042] Figure 2 This is the working mode of the electric drive bridge in the embodiment of the present invention.

[0043] Figure 3 It is a characteristic curve diagram of each working mode of the single / dual motor according to an embodiment of the present invention.

[0044] Figure 4 It is a single / dual motor mode switching rule table of an embodiment of the present invention.

[0045] Figure 4 (a) is the single motor mode switching rule; Figure 4 (b) is the dual motor mode switching rule.

[0046] Figure 5 This is the mode switching rule optimization result of the embodiment of the present invention.

[0047] Figure 5 (a) Optimization result of single motor mode switching rule; Figure 5 (b) Optimization result of dual-motor mode switching rules. DETAILED DESCRIPTION

[0048] The present invention will be further described below in conjunction with the accompanying drawings:

[0049] The present invention provides an optimization design method based on dual motor mode switching rules, such as Figure 1 As shown, it is characterized by:

[0050] The first step is that the dual-motor two-speed electric drive axle is equipped with a power motor and an economic motor, which includes eight working modes. The specific working modes are as follows: Figure 2 As shown:

[0051] Single motor mode:

[0052] Power motor MG1+AMT1 first gear (mode 1); Power motor MG1+AMT1 second gear (mode 2);

[0053] Economic motor MG2+AMT2 first gear (mode 3); Economic motor MG2+AMT2 second gear (mode 4);

[0054] Dual motor mode:

[0055] Power motor MG1+AMT1 first gear and economy motor MG2+AMT2 first gear (mode 5);

[0056] Power motor MG1+AMT1 first gear and economy motor MG2+AMT2 second gear (mode 6);

[0057] Power motor MG1+AMT1 second gear and economy motor MG2+AMT2 first gear (mode 7);

[0058] Power motor MG1+AMT1 second gear and economy motor MG2+AMT2 second gear (mode 8);

[0059] The second step is to evenly discretize the peak speed of the motor into 100 points based on the parameters of the dual motors to obtain the speed range of the two motors. and According to the transmission ratio of the vehicle power system, the driving torque-driving speed characteristic curve of the single / dual motor in each working mode is calculated respectively, as shown below:

[0060] Single motor drive:

[0061]

[0062] Dual motor drive:

[0063]

[0064] Where k1=1, 2, 3, 4; k2=5, 6, 7, 8 are the corresponding single / dual motor working mode states, r is the wheel radius, i0 is the main reducer transmission ratio, is the AMT transmission ratio in k1 mode, The AMT transmission ratio of the two motors in K2 mode is: To drive the vehicle speed, is the economic motor speed, T' MG_2 The torque interpolated from the economic motor drive characteristics, Output torque in k1 and k2 modes;

[0065] The driving torque-driving speed characteristic curve is as follows: Figure 3 As shown;

[0066] The third step is to find the maximum torque envelope of single / dual motor drive based on the characteristic curves of each working mode in the second step, and evenly discretize the torque on the envelope into 100 points downward, and use this as the working point. and

[0067] The fourth step is to traverse each operating point based on the operating point calculated in the third step and calculate the energy consumption under the single motor mode and the dual motor mode. For the mode that fails to meet the torque requirement of the operating point, its energy consumption is defined as INF. The energy consumption is expressed as follows:

[0068] Single motor mode:

[0069]

[0070] Dual motor mode: For a certain operating point, in the dual motor drive mode, the torque of the power motor and the economic motor must meet Decompose it into 100 torque combinations, and solve the torque combination of the power motor and the economic motor that minimizes energy consumption among the 100 torque combinations under this operating point:

[0071]

[0072] In the formula, is the transmission ratio of AMT1 and AMT2 in k2 mode, T MG1 ,T MG2 is the output torque of the power motor and the economic motor, η MG is the driving efficiency of the motor, E min (x,y) is the minimum energy consumption corresponding to the torque combination, is the energy consumption in k1 and k2 modes;

[0073] Step 5: Based on the motor energy consumption of each mode under each operating point calculated in step 4, the lowest energy consumption is used as the control target to determine the working mode distribution with the best energy consumption for single / dual motors;

[0074] Step 6: Based on the single motor working mode distribution and dual motor working mode distribution with the lowest energy consumption determined in step 5, establish single / dual motor mode switching rules X1 and X2, such as Figure 4 As shown, the mode switching rules X1 and X2 are used as inputs of the filtering optimization algorithm;

[0075] Step 7: traverse the mode switching rules X1 and X2, select the area window with point X(x, y) as the center point in the two-dimensional rule table, select the size of the area window as a 3×3 block, use the block data as the sliding input signal (data sequence), and extract the neighborhood window value sequence r(k), where k ranges from 1 to 8 and is arranged from top to bottom and from left to right according to the matrix number. r(k) is defined as follows:

[0076] r(k)=X(x+Δ i (k),y+Δ j (k)),k=1,2,...,8

[0077] In the formula, Δ i (k), Δ j (k) are the offsets of the kth element relative to the center point X(x, y) in the row and column directions, respectively. The value of the offset is determined by the value of k;

[0078] Step 8: Boundary value processing: determine whether the center point X(x,y) is a boundary value. The specific determination method includes:

[0079] Step 8.1: If the domain window value sequence of the center point X(x,y) exceeds the index range, the center point X(x,y) is the boundary value of the two-dimensional table, and symmetric filling is used to expand the domain window data of the center point X(x,y);

[0080] Step 8.2: If the domain window value sequence of the center point X(x,y) does not exceed the index range, the center point X(x,y) is not a boundary value, and proceed to the next step;

[0081] The ninth step is to determine whether the target point X(x,y) is an outlier. The specific determination method includes:

[0082] Step 9.1: If the target point X(x, y) and the domain value r(k) are all equal, it means that the target point is not an outlier, and the point X(x, y) is directly used as the output of the algorithm;

[0083] Step 9.2: If there is a deviation between the target point X(x,y) and the domain value r(k), the mode t of the domain value r(k) is calculated and used as the basis for judging the outlier point. If the target point X(x,y) is equal to the mode t, it indicates that the target point is not an outlier point, and the point X(x,y) is directly used as the output of the algorithm;

[0084] Step 9.3: If there is a deviation between the target point and the mode t, the MAD value of the target point X(x, y) is calculated as the characterization of the outlier. The MAD value is defined as follows:

[0085] MAD=median(|X i -median(X)|)

[0086] Where, X i is an element in the data sequence, where i = 1, 2, 3, ... 8; median(X) is the median of the data sequence; |X i -median(X)| represents the absolute difference between each data point and the median;

[0087] Step 9.4: Calculate the difference between the target point X(x,y) and the median(X) of the data sequence. If the difference is less than 3 times the MAD value, the mode t is used as the output of the algorithm.

[0088] Step 10: If the difference in step 9.4 is greater than 3 times the MAD value, the target point X(x,y) is an outlier, and the suboptimal solution X(x,y) of the target point is replaced by c Assign the value to X(x,y), and return to step 9 to recalculate;

[0089] Step 11: Determine whether the traversal of the single / dual motor mode switching rules has been completed. If not, return to step 8. If it has been reached, proceed to the next step.

[0090] Step 12: Output the optimized single / dual motor drive mode switching rules and end the algorithm.

[0091] The optimization result of the mode switching rule obtained by the optimization design method is as follows: Figure 5 As shown in the figure, compared with the mode switching rule before optimization, the boundary of the optimized mode switching rule is clearer and can be directly applied to the dual-motor mode switching control. The optimized rule can determine the optimal energy consumption working mode of the dual-motor according to the working conditions in different usage scenarios; at the same time, by judging and replacing outliers, the number of dual-motor system mode switching times during operation is greatly reduced while ensuring the power requirements and the energy loss is very small, thereby improving the vehicle's driving quality and the smoothness of the system.

Claims

1. An optimization design method based on dual motor mode switching rules, characterized in that: In the first step, the dual-motor two-speed electric drive axle is equipped with a power motor and an economic motor, which includes a total of eight working modes; The second step is to evenly discretize the peak speed of the motor into 100 points based on the parameters of the dual motors to obtain the speed range of the two motors. and According to the transmission ratio of the vehicle power system, the driving torque-driving speed characteristic curve of the single / dual motor in each working mode is calculated respectively, as shown below: Single motor drive: Dual motor drive: Where k1=1, 2, 3, 4; k2=5, 6, 7, 8 are the corresponding single / dual motor working mode states, r is the wheel radius, i0 is the main reducer transmission ratio, is the AMT transmission ratio in k1 mode, The AMT transmission ratio of the two motors in K2 mode is: To drive the vehicle speed, is the economic motor speed, T' MG_2 The torque interpolated from the economic motor drive characteristics, Output torque in k1 and k2 modes; The third step is to find the maximum torque envelope of single / dual motor drive based on the characteristic curves of each working mode in the second step, and evenly discretize the torque on the envelope into 100 points downward, and use this as the working point. and The fourth step is to traverse each operating point based on the operating point calculated in the third step and calculate the energy consumption under the single motor mode and the dual motor mode. For the mode that fails to meet the torque requirement of the operating point, its energy consumption is defined as INF. The energy consumption is expressed as follows: Single motor mode: Dual motor mode: For a certain operating point, in the dual motor drive mode, the torque of the power motor and the economic motor must meet Decompose it into 100 torque combinations, and solve the torque combination of the power motor and the economic motor that minimizes energy consumption among the 100 torque combinations under this operating point: In the formula, is the transmission ratio of AMT1 and AMT2 in k2 mode, T MG1 ,T MG2 is the output torque of the power motor and the economic motor, η MG is the driving efficiency of the motor, E min (x,y) is the minimum energy consumption corresponding to the torque combination, is the energy consumption in k1 and k2 modes; Step 5: Based on the motor energy consumption of each mode under each operating point calculated in step 4, the lowest energy consumption is used as the control target to determine the working mode distribution with the best energy consumption for single / dual motors; Step 6: Based on the single-motor working mode distribution and the dual-motor working mode distribution with the lowest energy consumption determined in step 5, establish single / dual-motor mode switching rules X1 and X2, and use the mode switching rules X1 and X2 as inputs of the filtering optimization algorithm; Step 7: traverse the mode switching rules X1 and X2, select the area window with point X(x, y) as the center point in the two-dimensional rule table, select the size of the area window as a 3×3 block, use the block data as the sliding input signal (data sequence), and extract the neighborhood window value sequence r(k), where k ranges from 1 to 8 and is arranged from top to bottom and from left to right according to the matrix number. r(k) is defined as follows: r(k)=X(x+Δ i (k),y+Δ j (k)),k=1,2,...,8 In the formula, Δ i (k), Δ j (k) are the offsets of the kth element relative to the center point X(x, y) in the row and column directions, respectively. The value of the offset is determined by the value of k; Step 8: Boundary value processing: determine whether the center point X(x,y) is a boundary value. The specific determination method includes: Step 8.1: If the domain window value sequence of the center point X(x,y) exceeds the index range, the center point X(x,y) is the boundary value of the two-dimensional table, and symmetric filling is used to expand the domain window data of the center point X(x,y); Step 8.2: If the domain window value sequence of the center point X(x,y) does not exceed the index range, the center point X(x,y) is not a boundary value, and proceed to the next step; The ninth step is to determine whether the target point X(x,y) is an outlier. The specific determination method includes: Step 9.1: If the target point X(x, y) and the domain value r(k) are all equal, it means that the target point is not an outlier, and the point X(x, y) is directly used as the output of the algorithm; Step 9.2: If there is a deviation between the target point X(x,y) and the domain value r(k), the mode t of the domain value r(k) is calculated and used as the basis for judging the outlier point. If the target point X(x,y) is equal to the mode t, it indicates that the target point is not an outlier point, and the point X(x,y) is directly used as the output of the algorithm; Step 9.3: If there is a deviation between the target point and the mode t, the MAD value of the target point X(x, y) is calculated as the characterization of the outlier. The MAD value is defined as follows: MAD=median(|X i -median(X)|) Where, X i is an element in the data sequence, where i = 1, 2, 3, ... 8; median(X) is the median of the data sequence; |X i -median(X)| represents the absolute difference between each data point and the median; Step 9.4: Calculate the difference between the target point X(x,y) and the median(X) of the data sequence. If the difference is less than 3 times the MAD value, the mode t is used as the output of the algorithm. Step 10: If the difference in step 9.4 is greater than 3 times the MAD value, the target point X(x,y) is an outlier, and the suboptimal solution X(x,y) of the target point is replaced by c Assign the value to X(x,y), and return to step 9 to recalculate; Step 11: Determine whether the traversal of the single / dual motor mode switching rules has been completed. If not, return to step 8. If it has been reached, proceed to the next step. Step 12: Output the optimized single / dual motor drive mode switching rules and end the algorithm.

Images