Full-scene-oriented extended-range electric vehicle fixed-point strategy balance optimization system and method

Abstract

The invention discloses a full-scene-oriented extended-range electric vehicle fixed-point strategy balance optimization system and a full-scene-oriented extended-range electric vehicle fixed-point strategy balance optimization method. The system comprises a multi-scene working condition construction module, a fixed-point strategy model construction module, a design scheme sampling module, an approximate model construction module, a global multi-target optimization module, a weight coefficient calculation module and an optimal scheme determination module. According to the invention, the comprehensive performance of the extended-range electric vehicle in an actual travel scene of a user can be improved; according to the method, historical driving data of the user is combined, the method is closer to real travel of the user, index weight confirmation and multi-target design scheme sorting are considered in multiple aspects and multiple angles, and the method is more comprehensive and scientific.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy-saving technologies for passenger vehicles, and particularly relates to a system and method for optimizing the balance of a range extender electric vehicle fixed-point strategy for all scenarios. Background Art

[0002] With the continuous progress of automotive technologies and the increasing requirements of consumers for the comprehensive performance of vehicles, users pay more and more attention to the perception of vehicle performance during actual trips. The development of traditional hybrid vehicle energy management strategies mainly focuses on performance optimization under typical driving cycles (such as WLTC, CLTC, etc.). However, these strategies often lack a comprehensive consideration of the actual driving scenarios of users.

[0003] In the field of range extender electric vehicles, the design of fixed-point strategies also faces challenges. Most of the existing fixed-point strategies for range extender electric vehicles are developed based on the WLTC driving cycle and a single economic performance index. This mode ignores the complex and variable scenario requirements in actual user trips and also lacks a quantitative consideration of the comprehensive performance of multiple evaluation indicators. Therefore, the existing technologies have certain limitations in meeting the actual travel needs of users and cannot provide the best travel experience for users. Summary of the Invention

[0004] In order to solve the problems in the technical background, improve the comprehensive performance of range extender electric vehicles in the actual travel scenarios of users, and provide a better travel experience for users, the present invention proposes a system and method for optimizing the balance of a range extender electric vehicle fixed-point strategy for all scenarios.

[0005] A system for optimizing the balance of a range extender electric vehicle fixed-point strategy for all scenarios, which realizes one of the purposes of the present invention, includes:

[0006] A design scheme sampling module: used to take a certain operating parameter of the vehicle as a design factor, and obtain multiple design schemes through a sampling method according to the value range of the certain operating parameter. Each design scheme includes the values of the design factor at each operating point; the operating point is used to represent the operating state of the range extender under specific working conditions; each operating point corresponds to a different combination of rotational speed and torque, and the operating point is selected in the optimal economic region to give full play to the advantages of the range extender and take into account the performance requirements under different working conditions;

[0007] An approximate model construction module: used to construct an approximate model for evaluating the multiple design schemes; the input of the approximate model is the design scheme, and the output is the evaluation index value for evaluating the input design scheme;

[0008] Global multi-objective optimization module: It is used to construct an optimization model for finding the optimal design solution. The optimization model globally optimizes the evaluation index values output by the approximation model using a multi-objective optimization algorithm to obtain multiple frontier optimization solutions that can optimize multiple evaluation indexes simultaneously;

[0009] Weight coefficient calculation module: It is used to calculate the combined weight coefficient of each evaluation index by combining subjective and objective methods;

[0010] Optimal solution determination module: It is used to calculate the comprehensive contribution degree coefficient of each frontier optimization solution according to the combined weight coefficient of each evaluation index, and select the optimal design solution with the best balance from the multiple frontier optimization solutions according to the comprehensive contribution degree coefficient.

[0011] Furthermore, in the weight coefficient calculation module, the combined weight coefficient includes a subjective weight coefficient, and its determination method includes:

[0012] Score each pair of evaluation indexes according to a preset scoring standard; construct a judgment matrix according to the scores;

[0013] Normalize the judgment matrix to obtain the subjective weights of each evaluation index.

[0014] In the weight coefficient calculation module, by scoring each pair of evaluation indexes according to a preset scoring standard, constructing a judgment matrix and normalizing it to obtain the subjective weights, it can make full use of the experience and knowledge of experts or relevant personnel to determine the importance of different evaluation indexes according to actual needs and concerns, making the determination of weights more in line with the subjective expectations of vehicle performance and providing a more reasonable subjective basis for subsequent comprehensive evaluation.

[0015] Furthermore, in the weight coefficient calculation module, the combined weight coefficient includes an objective weight coefficient, and its determination method includes:

[0016] Score the influence of the design factors on the evaluation indexes of each design solution; construct a first decision matrix according to the scores;

[0017] Perform non-negative processing on the elements in the first decision matrix;

[0018] Calculate the information entropy of each evaluation index according to the first decision matrix after non-negative processing;

[0019] Calculate the objective weight coefficient of each evaluation index according to the information entropy.

[0020] In the weight coefficient calculation module, the technical effects of determining the objective weight coefficient by scoring according to the influence of design factors on evaluation indicators, constructing the first decision matrix, non-negative processing, and calculating information entropy include: determining the weight based on the objective influence of design factors on evaluation indicators, avoiding the limitations of subjective judgment, and being able to more accurately reflect the importance of each indicator in actual operation. The calculation of information entropy can quantify the uncertainty of indicators, making the determination of objective weights more scientific and objective. Combining subjective weights and objective weights can obtain a more comprehensive and accurate combined weight coefficient.

[0021] Further, the calculation method of the comprehensive contribution degree coefficient includes:

[0022] Weight the first decision matrix according to the combined weight coefficient to obtain a second decision matrix;

[0023] Determine the positive and negative ideal solution sets of each evaluation indicator according to the second decision matrix;

[0024] Calculate the Euclidean distances of the positive and negative ideal solution sets of each evaluation indicator;

[0025] Determine the development trend correlation matrix corresponding to the positive and negative ideal solution sets according to the second decision matrix; the development trend correlation matrix is used to represent the similarity degree of the development trends of each frontier optimization scheme and the positive and negative ideal solution sets on each evaluation indicator;

[0026] Obtain the correlation coefficients relative to the positive and negative ideal solution sets according to the development trend correlation matrix;

[0027] Calculate the relative fitting degrees of each frontier optimization scheme and the positive and negative ideal solution sets V+ and V- according to the Euclidean distances and correlation coefficients;

[0028] Determine the comprehensive contribution degree coefficient of each frontier optimization scheme according to the relative fitting degrees.

[0029] Further, the calculation method of the development trend correlation matrix includes:

[0030]

[0031] v j + =max(v 1j ,v 2j ,…,v mj );v j - =min(v 1j ,v 2j ,…,v mj )

[0032] In the formula:

[0033] G + represents the development trend correlation matrix corresponding to the positive ideal solution set;

[0034] G - represents the development trend correlation matrix corresponding to the negative ideal solution set;

[0035] v ij represents an element in the second decision matrix, and is used to represent the value of the j-th evaluation index of the i-th advanced optimization scheme.

[0036] The above calculation method for determining the comprehensive contribution degree coefficient comprehensively considers each evaluation index and the relationship between different advanced optimization schemes and the ideal solution set, comprehensively evaluates the advantages and disadvantages of each advanced optimization scheme. Through the calculation of the relative fitting degree, the scheme closest to the ideal state under multiple evaluation indexes can be found more accurately, realizing the scientific screening and optimization of the design scheme, and improving the comprehensive performance and balance of the whole vehicle.

[0037] Furthermore, the system further includes:

[0038] Multi-scenario driving condition construction module: used to construct short driving segments according to the driving data of multiple vehicle models in the background; fit the full-scenario driving conditions of multiple users' trips through cluster analysis based on the short driving segments;

[0039] Fixed-point strategy model construction module: used to construct a fixed-point strategy model, and the fixed-point strategy model is used to determine the working point corresponding to each full-scenario driving condition according to multiple vehicle operation parameters in the full-scenario driving condition, so that the selection of the working point is closer to the actual trips of users and covers the full scenarios of users.

[0040] Furthermore, the method for determining the working point corresponding to each full-scenario driving condition includes:

[0041] When the value of the first vehicle operation parameter is less than or equal to the first set value: if the value of the second vehicle operation parameter is greater than or equal to the second set value, it is considered that the working point corresponding to the current driving condition is the fifth working point; otherwise, it is considered that the working point corresponding to the current driving condition is the fourth working point;

[0042] When the value of the first vehicle operation parameter is greater than the first set value and less than the third set value: if the value of the second vehicle operation parameter is greater than or equal to the second set value, it is considered that the working point corresponding to the current driving condition is the third working point; otherwise, it is considered that the working point corresponding to the current driving condition is the second working point;

[0043] When the first vehicle operating parameter value is greater than the third set value: If the second vehicle operating parameter value is less than the second set value, the operating point corresponding to the current working condition is considered as the first operating point; If the second vehicle operating parameter is greater than or equal to the second set value and less than the fourth set value, the operating point corresponding to the current working condition is considered as the third operating point; If the vehicle speed is greater than or equal to the fourth set value, the operating point corresponding to the current working condition is considered as the fourth operating point.

[0044] Further, before determining the operating point corresponding to each full-scenario working condition, it also includes the judgment of the following preconditions: The range extender is turned on and the first vehicle operating parameter value is less than or equal to the set maximum value.

[0045] The technical effects brought by the method for determining the operating point corresponding to each full-scenario working condition include: Determining the operating point according to the ranges of different vehicle operating parameter values (the first vehicle operating parameter and the second vehicle operating parameter). This clear method for determining the operating point can accurately determine the operating point of the range extender according to the actual operating state of the vehicle, enabling the range extender to be in a suitable operating mode under different working conditions, improving the working efficiency of the range extender and the performance of the whole vehicle, and avoiding problems such as increased energy consumption or performance degradation caused by the range extender operating in an inappropriate state.

[0046] A full-scenario-oriented fixed-point strategy balance optimization method for extended-range electric vehicles to achieve the second object of the present invention includes:

[0047] Construct short-trip segments based on multi-vehicle background driving data; Fit the full-scenario working conditions of multiple users' trips through cluster analysis according to the short-trip segments;

[0048] Construct a fixed-point strategy model, which is used to determine the operating point corresponding to each full-scenario working condition according to multiple vehicle operating parameters in the full-scenario working condition; The operating point is used to represent the operating state of the range extender under specific working conditions; Each operating point corresponds to a different combination of rotational speed and torque, and the operating point is selected in the optimal economic region to give play to the advantages of the range extender and take into account the performance requirements under different working conditions;

[0049] Taking a certain operating parameter of the vehicle as a design factor, obtaining multiple design schemes through a sampling method according to the value range of the certain operating parameter, and each design scheme includes the value of the design factor at each operating point;

[0050] Construct an approximation model for evaluating the multiple design schemes; The input of the approximation model is the design scheme, and the output is the evaluation index value for evaluating the input design scheme;

[0051] Construct an optimization model for finding the optimal design solution. The optimization model uses a multi-objective optimization algorithm to globally optimize the evaluation index values output by the approximate model, and obtains multiple frontier optimization solutions that can simultaneously optimize multiple evaluation indexes;

[0052] Calculate the combined weight coefficient of each evaluation index by combining subjective and objective methods;

[0053] Calculate the comprehensive contribution degree coefficient of each frontier optimization solution according to the combined weight coefficient of each evaluation index, and select the optimal design solution with the best balance from the multiple frontier optimization solutions according to the comprehensive contribution degree coefficient.

[0054] A non-transitory computer-readable storage medium for achieving the third object of the present invention, on which a computer program is stored. When the computer program is executed by a processor, the steps of the fixed-point strategy balance optimization method for range-extended electric vehicles for the whole scenario are implemented.

[0055] A computer program product for achieving the fourth object of the present invention, including a computer program / instructions. When the computer program / instructions are executed by a processor, the steps of the fixed-point strategy balance optimization method for range-extended electric vehicles for the whole scenario are implemented.

[0056] The beneficial effects of the present invention include:

[0057] The present invention can be directly applied to the field of energy-saving strategies for hybrid vehicles. After applying this method, the comprehensive performance of range-extended electric vehicles can be improved in the actual travel scenarios of users, providing a better travel experience for users; this method combines the historical driving data of users, is closer to the real travel of users, and the confirmation of index weights and the ranking of multi-objective design solutions are considered from multiple aspects and angles, and the method is more comprehensive and scientific; all work can be completed within one week by this design method, and it will not affect the original strategy development cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 is a schematic flowchart of an embodiment of the method of the present invention;

[0059] Figure 2 is a framework diagram of a multi-workpoint control strategy. DETAILED DESCRIPTION

[0060] The following detailed description is used to explain the technical solution of the claims of the present invention so that those skilled in the art can understand the claims. The protection scope of the present invention is not limited to the following specific implementation structures. What those skilled in the art make that includes the technical solution of the claims of the present invention and is different from the following specific implementation manners is also within the protection scope of the present invention.

[0061] An embodiment of the present invention provides an optimization method for the balance of the fixed-point strategy of a range-extended electric vehicle for the full scenario, as follows: Figure 1 shown, including the following steps:

[0062] Step 1: Construction of multi-scenario working conditions based on big data

[0063] In order to better reflect the real travel experience of users, it is necessary to construct a full-scenario usage working condition that is closer to the actual travel of users. The construction method includes:

[0064] First, collect and clean the in-vehicle data. Specifically, it includes: using the cloud big data system to obtain the background driving data of multiple vehicle models, with a transmission frequency of 1 Hz, including longitude, latitude, vehicle speed, battery power, driving mileage, pedal opening, etc.; through system intelligent classification, short travel segments are constructed based on the vehicle speed change characteristics, and each short travel segment represents a working condition, such as idle condition, acceleration condition, deceleration condition, constant speed condition. And for the data missing within 5 s, linear interpolation is used to complete it, and the missing data exceeding 5 s is directly deleted; at the same time, the data with a vehicle speed of 0 km / h and a duration exceeding 180 s is excluded; the data is filtered, and the mean value of the neighborhood data of the original data is used to replace the original data to form a mean value sequence.

[0065] Secondly, fit n multi-scenario working conditions of users' actual travel according to the short travel segments. Specifically, it includes: extracting the segment feature parameters, and the segment feature parameters include 24 feature parameters such as the proportion of idle, acceleration, deceleration, and constant speed conditions, maximum vehicle speed, average vehicle speed, and the proportion of each speed segment, mileage, etc. to form an initial feature parameter matrix; performing standardization processing on the initial feature parameter matrix to eliminate the influence of the dimension and magnitude of each feature parameter, and calculating the contribution degrees b1, b2, b3,..., b 24 , and arranging them in descending order, selecting the front feature parameters with a cumulative contribution rate exceeding 80% as the main components, reducing the dimension of the feature parameter matrix, and obtaining a feature parameter matrix composed of the main components. Further, clustering processing is carried out on the feature parameter matrix composed of the main components, and continuously iteratively adjusted to make the Euclidean distances of the speed and acceleration of the same type of working conditions as small as possible, and the distances between different types of working conditions as large as possible; fitting out n multi-scenario working conditions [Scene1, Scene2..., Scene n ;

[0066] After the fitting is completed, calculate the total duration of each working condition. The calculation method of the total duration of each working condition is as follows:

[0067]

[0068] In the formula, t m is the total duration of the m-th working condition after fitting; t cycleis the total duration of all working conditions after fitting; t all is the total duration of the collected data; t m,i is the duration of the i-th short stroke segment in the m-th working condition.

[0069] Each multi-scenario working condition contains a series of information related to vehicle operation. In addition to basic information such as speed and acceleration involved in the construction process, it also includes segment feature parameters, such as the proportion of idle speed, acceleration, deceleration, constant speed working conditions, maximum speed, average speed, the proportion of each speed segment, mileage, etc. These information describe the vehicle's operating state under this working condition from multiple dimensions and are important bases for subsequent fixed-point strategy optimization.

[0070] The technical effects of calculating the total duration of each working condition are as follows: In the actual travel scenario, the occurrence times of various working conditions (such as idle speed, acceleration, deceleration, constant speed, etc.) are different. By calculating the total duration, these differences can be quantified, providing data support for more accurate simulation of actual travel in the future. For example, if the total duration of the idle speed working condition accounts for a large proportion in actual travel, then when optimizing the fixed-point strategy, more attention needs to be paid to the working state of the range extender under the idle speed working condition.

[0071] The relationship between the above total duration and the working condition combination is as follows: The working condition combination is based on the principle of the closest Euclidean distance of speed and acceleration, and the calculation of the total duration is to perform statistical analysis in the time dimension for each combined working condition after the working condition combination is completed. Different working condition combinations represent different travel scenario segments, and the total duration is a measure of the proportion of these scenario segments in the overall travel time. The two cooperate with each other to jointly construct a more realistic full-scenario working condition.

[0072] Finally, assign the weight ratio of each working condition; the assignment method includes:

[0073] Divide the total duration of each working condition by the total duration of the collected data to obtain the actual proportion of each working condition;

[0074] Analyze the weight settings of the corresponding scenarios in the WLTC working condition. As a standard working condition, the weights of its various scenarios have certain reference value. Consider the above actual proportion and the weights of the corresponding scenarios in the WLTC working condition comprehensively. The weighted average method can be used to assign different weight coefficients to the actual scenario proportion and the WLTC working condition weights respectively, and calculate the weight ratio of each working condition; these weight ratios will be used in subsequent links such as the construction of the fixed-point strategy model and the sampling of the design scheme, playing a key role in optimizing the balance of the fixed-point strategy of the range-extended electric vehicle.

[0075] Step 2: Build a fixed-point strategy model

[0076] The fixed operating point (i.e., fixed point) strategy can fully utilize the advantages of the non-mechanical connection between the range extender and the powertrain, contribute to improving NVH in the low-speed range, enhancing fuel economy in the medium and high-speed ranges, and can also take into account performance requirements such as thermal management. In the embodiment of the present invention, a five-operating-point control strategy is adopted. The operating points are used to represent the operating states of the range extender under specific working conditions, and each operating point is set according to the weight ratio of each multi-scenario working condition obtained in the above steps; it can be understood as fitting multiple typical working conditions (such as high speed, urban areas, etc.), allocating weights to these working conditions based on the collected data, and then weighting indicators such as fuel consumption of these working conditions into one; then as Figure 2 shown, determine the operating points corresponding to each working condition according to the first vehicle operating parameter SOC and the second vehicle operating parameter vehicle speed of each multi-scenario working condition; the method for determining the operating points includes:

[0077] When the SOC is less than the set minimum value of SOC, turn on the range extender, otherwise turn off the range extender;

[0078] After the range extender is turned on, if the SOC is greater than the set maximum value of SOC, turn off the range extender, and then make the following judgment:

[0079] When the value of SOC is less than or equal to the first set value SOC1: If the vehicle speed is greater than or equal to the second set value V low (i.e., the set low-speed value), it is considered that the operating point corresponding to the current working condition is the fifth operating point E; otherwise, it is considered that the operating point corresponding to the current working condition is the fourth operating point D;

[0080] When the value of SOC is greater than the first set value SOC1 and less than the third set value SOC2: If the vehicle speed is greater than or equal to the second set value V low , it is considered that the operating point corresponding to the current working condition is the third operating point C; otherwise, it is considered that the operating point corresponding to the current working condition is the second operating point B;

[0081] When the value of SOC is greater than the third set value SOC2: If the vehicle speed is less than the second set value V low , it is considered that the operating point corresponding to the current working condition is the first operating point A; if the vehicle speed is greater than or equal to the second set value V low and less than the fourth set value V high , it is considered that the operating point corresponding to the current working condition is the third operating point C; if the vehicle speed is greater than or equal to the set high-speed value V high , it is considered that the operating point corresponding to the current working condition is the fourth operating point D.

[0082] Each operating point corresponds to a different combination of rotational speed and torque to achieve optimal performance. The meanings represented by the above five operating points are as follows:

[0083] The first operating point A: low rotational speed and low torque, suitable for low-speed driving or idling, optimizing NVH performance. The second operating point B: medium rotational speed and medium torque, suitable for medium and low-speed driving, balancing fuel economy and power demand. The third operating point C: high rotational speed and high torque, suitable for high-speed driving, providing sufficient power and optimizing fuel economy. The fourth operating point D: a specific combination of rotational speed and torque, suitable for a specific vehicle speed range, further optimizing performance. The fifth operating point E: the operating point under extreme conditions, used to handle special situations (such as when the battery power is extremely low).

[0084] Among them, the operating points are selected within the optimal economic region, and subsequent balance optimization is carried out based on the basic strategy.

[0085] Step 3: Sampling of the design scheme and construction of the approximation model

[0086] Taking the rotational speed in the operating points as the design factor, determine the value range of the design factor in each operating point, that is, clarify the possible rotational speed change range of the range extender at each operating point. Design the test scheme according to the rotational speed change range (such as optimal Latin hypercube sampling, etc.), and the number of samples ≥ m1×(m1 + 1) / 2, where m1 represents the number of design variables. For example, when optimizing for 5 operating conditions, m1 is 5; a total of m design schemes are obtained. For example, design scheme A: the first operating point A = 1200 RPM, the second operating point B = 1800 RPM, the third operating point C = 2200 RPM, the fourth operating point D = 2800 RPM, the fifth E = 3200 RPM; design scheme B: the first operating point A = 1500 RPM, the second operating point B = 2000 RPM, the third operating point C = 2500 RPM, the fourth operating point D = 3000 RPM, the fifth E = 3500 RPM. Construct an approximation model according to the evaluation indexes of each design scheme. The approximation model is used to construct the functional relationship between each evaluation index and the rotational speed corresponding to the 5 operating points of the range extender. Its input data is the rotational speed value of each operating point, and the output data is each evaluation index. The evaluation indexes include vehicle performance indexes such as acceleration time, fuel consumption, and power consumption. In order to reduce the number of calls to the simulation model, improve the optimization efficiency of the algorithm, and evaluate the approximation model with the coefficient of determination (R 2 ), when the constructed approximation model R 2 > 0.9, it meets the accuracy requirements, and the closer the value is to 1, the higher the fitting and prediction accuracy. The specific calculation method is as follows:

[0087]

[0088] Among them, k is the number of sampling points, which is 18 in this article; y i are the predicted value and the actual value of the i-th sample point respectively, represents the mean value of all sample points.

[0089] Step 4: Confirmation of Evaluation Index Weights

[0090] In order to evaluate the comprehensive performance of the whole vehicle during the strategy optimization process, the weights of each evaluation index output in Step 3 are confirmed by combining subjective and objective methods.

[0091] The subjective weight is confirmed by comparing the importance degrees of each evaluation index. First, the importance of the evaluation index is scored based on the scoring criteria in Table 1 below, and the following judgment matrix A is constructed according to the scores:

[0092]

[0093] where a ij represents the score of the i-th evaluation index relative to the j-th evaluation index, and the scoring criteria are shown in Table 1:

[0094] Table 1 Scoring Criteria

[0095]

[0096] Then, the judgment matrix A is normalized to obtain the subjective weights w s of each evaluation index, where n is the total number of evaluation indexes, and w si represents the subjective weight of the i-th evaluation index:

[0097] w s =(w s1 , w s2 , w s3 , …, w sn ) T

[0098] The objective weight is confirmed by comparing the information amounts contained in the data of each evaluation index. The greater the influence of the change of the design factor (the rotational speed in this embodiment) on the evaluation index, the greater the weight, and vice versa. The specific steps are as follows:

[0099] First, construct the first decision matrix B:

[0100]

[0101] where b ij is the value of the j-th evaluation index of the i-th set of design schemes; m represents the number of design schemes obtained by the optimal Latin hypercube sampling method.

[0102] Secondly, due to the differences in attributes and magnitudes of different evaluation indexes, some data characteristics are likely to be lost. To reduce this influence and improve the speed and reliability of data processing, each evaluation index value in the decision matrix B is subjected to standard non-negative processing, and the processing method is as follows:

[0103]

[0104] Then, obtain the information entropy of each evaluation index, where e i represents the information entropy of the i-th evaluation index:

[0105]

[0106] Among them,

[0107] Finally, calculate the objective weight coefficient of each evaluation index according to the following formula:

[0108]

[0109] Among them, is the objective weight coefficient of the j-th evaluation index; n is the number of evaluation indexes.

[0110] Finally, obtain the objective weight coefficient w0:

[0111] w o =(w o1 , w o2 , w o3 ,…, w on ) T

[0112] Combine the calculated subjective weight coefficient w s and the objective weight coefficient w0 for comprehensive calculation to obtain the combined weight coefficient w for representing all evaluation indexes, where the calculation formula of the combined weight coefficient w i for each evaluation index includes:

[0113]

[0114] Step 5: Global multi-objective optimization

[0115] First, combine each design factor and evaluation index to construct the following optimization model, where Value1, Value1, …, Value n represent the values of each evaluation index, and the smaller the value of the evaluation index, the better. The following fraction represents the constraint on the value of each evaluation index; n1, n2, …, n 2k represent the set value range of each evaluation index.

[0116]

[0117] The approximate model constructs the functional relationship between each evaluation index and the variable (the rotational speed in this embodiment). By assigning weights, it can be simply understood that multiple evaluation indexes are combined into one for optimization. Then, for the approximate models of each evaluation index determined in step 3, global optimization is carried out in combination with the combined weight coefficient w determined in step 4, and the NSGA-II (Non-dominated Sorting Genetic) algorithm is adopted. The specific parameter settings are as follows:

[0118] Table 2 Important Parameter Settings of NSGA-II Algorithm

[0119]

[0120]

[0121] Finally, after the algorithm optimization, the Pareto front is obtained, and a total of m frontier optimization solutions [Solution 1, Solution 2,..., Solution m] are obtained, which is the same as the number of design solutions obtained by the optimal Latin hypercube sampling method mentioned above.

[0122] Step 6: Calculate the comprehensive contribution degree coefficient

[0123] For the m frontier optimization solutions obtained in step 5, the Euclidean distance and development trend similarity between each design solution and the ideal solution set are comprehensively considered for solution selection.

[0124] First, in combination with the weight coefficient w of each evaluation index determined in step 4, the second decision matrix V is obtained by weighting. Its function is to integrate the importance of different evaluation indexes into the decision matrix, highlight the influence of important indexes on the solution evaluation, make the subsequent calculation and evaluation better reflect the relative importance of each evaluation index, avoid ignoring important indexes in the evaluation process, and improve the accuracy of solution evaluation. w ij is the combined weight coefficient of the evaluation index.

[0125] V=(v ij ) m×n =(w ij b ij ) m×n

[0126] Secondly, determine the positive and negative ideal solution sets V + 、V - of each evaluation index for each frontier optimization solution; the positive ideal solution set V + is the set of optimal values of each evaluation index in all solutions, and the negative ideal solution set V -It is the set of the worst values of all evaluation indicators in all solutions. The role of determining these two solution sets is to provide a reference standard for subsequent comparison of each frontier optimization solution. By comparing with the positive and negative ideal solution sets, it can be clearly seen the gap between each solution and the optimal and worst cases in each evaluation indicator, so as to measure the quality of the solution;

[0127] V + = [v1 + , v2 + , …, v m +

[0128] V - = [v1 - , v2 - ,..., v m -

[0129] where, v j + = max(v 1j , v 2j , …, v mj ); v j - = min(v 1j , v 2j , …, v mj ).

[0130] Then, determine the Euclidean distances d + , d - between the positive and negative ideal solution sets V + , d - of each frontier optimization solution; Calculating the Euclidean distance can quantify the degree of difference in space between each frontier optimization solution and the positive and negative ideal solution sets, and can intuitively reflect the degree of closeness of the solution to the ideal situation. The smaller the distance, the closer the solution is to the ideal state, providing a clear quantitative index for the preliminary screening of the solution and facilitating the sorting and comparison of the solutions;

[0131]

[0132] After that, construct the development trend correlation matrices G + , G - between each frontier optimization solution and the positive and negative ideal solution sets V + , G - ; The development trend correlation matrix is used to represent the degree of similarity in the development trend of each frontier optimization solution and the positive and negative ideal solution sets in each evaluation indicator. Its role is to evaluate the similarity between the solution and the ideal solution set from the perspective of the development trend, make up for the deficiency of only considering the distance factor, can more comprehensively reflect the comprehensive performance of the solution, and avoid ignoring the deficiency of the solution in the overall development trend due to the distance advantage at a certain moment;​​

[0133]

[0134] Among them, ρ is the discrimination coefficient, and generally ρ∈(0,1) is selected in the research.

[0135] Subsequently, calculate the correlation coefficients of each frontier optimization scheme with respect to the positive and negative ideal solution sets V + 、V - ; The correlation coefficient is used to measure the degree of association between the scheme and the ideal solution set. The dimensionless treatment eliminates the influence of different index dimensions on the calculation results, making the correlation coefficients under different evaluation indexes comparable and enabling a more accurate comprehensive evaluation of the scheme.

[0136]

[0137] Perform dimensionless treatment on the above d + 、d - as well as r + 、r - :

[0138]

[0139]

[0140] Immediately afterwards, combine the Euclidean distances d + 、d - and the correlation coefficients r + 、r - between each frontier optimization scheme and the ideal solution sets V + 、V - for comprehensive evaluation to obtain the relative fitting degrees F + 、F - i + 、F i - of each frontier optimization scheme with the positive and negative ideal solution sets V; The relative fitting degree comprehensively considers the distance between the scheme and the ideal solution set and the degree of association of the development trend. It evaluates the advantages and disadvantages of the scheme comprehensively from multiple dimensions, provides a more comprehensive and accurate evaluation index, can screen out the schemes that perform well in multiple aspects, and avoids the one-sidedness of single-index evaluation;

[0141]

[0142] Among them, α and β are the preference degrees of the decision maker for the ideal solution method and the grey relational analysis decision, and the two satisfy α + β = 1.

[0143] Finally, comprehensively consider the positive and negative ideal solution sets V + 、V -The relative fitting degree is used to determine the comprehensive contribution coefficient of each frontier optimization scheme according to the following formula:

[0144]

[0145] H i represents the comprehensive contribution coefficient of the i-th frontier optimization scheme;

[0146] The comprehensive contribution coefficient synthesizes the relative fitting degrees of the positive and negative ideal solution sets, comprehensively evaluates each frontier optimization scheme, and provides a unified quantitative standard for finally selecting the scheme with the best balance. It quickly and accurately selects the scheme with the best overall vehicle performance from multiple frontier optimization schemes, improving the scientificity and rationality of scheme selection.

[0147] Step 7. Select the optimal design scheme

[0148] Based on the comprehensive contribution coefficient for sorting and decision-making, the frontier optimization scheme corresponding to the maximum value of the comprehensive contribution coefficient is the optimal design scheme with the best balance. The optimal design scheme corresponds to the optimal overall vehicle performance at the rotational speeds of each operating point.

[0149] So far, the balance optimization work of the fixed-point strategy for range-extended electric vehicles for all scenarios has been completed. This work can improve the overall vehicle performance of range-extended electric vehicles in the actual travel scenarios of users, provide a better travel experience for users, and has good engineering application value and application prospects for the work in the field of vehicle energy conservation.

[0150] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0151] The embodiment of the present invention also provides a balance optimization system for the fixed-point strategy of range-extended electric vehicles for all scenarios, including:

[0152] Multi-scenario working condition construction module: used to construct short-trip segments based on the background driving data of multiple vehicle models; fit multiple full-scenario working conditions of users' actual travel through cluster analysis according to the short-trip segments;

[0153] Fixed-point strategy model construction module: used to construct a fixed-point strategy model, which is used to determine the operating point corresponding to each full-scenario working condition according to multiple vehicle operating parameters in the full-scenario working condition; the operating point is used to represent the operating state of the range extender under specific working conditions; each operating point corresponds to a different combination of rotational speed and torque, and the operating point is selected in the optimal economic region to give play to the advantages of the range extender and take into account the performance requirements under different working conditions;

[0154] Design scheme sampling module: used for taking a certain operating parameter of the vehicle as a design factor, and obtaining multiple design schemes through a sampling method according to the value range of the certain operating parameter, each design scheme including the value of the design factor at each working point;

[0155] An approximate model building module: used to build an approximate model for evaluating the multiple design solutions; the input of the approximate model is the design solution, and the output is an evaluation index value for evaluating the input design solution;

[0156] Global multi-objective optimization module: used to construct an optimization model for finding the optimal design solution. The optimization model uses a multi-objective optimization algorithm to perform global optimization on the evaluation index values ​​output by the approximate model to obtain multiple cutting-edge optimization solutions that can simultaneously optimize multiple evaluation indicators;

[0157] Weight coefficient calculation module: used to calculate the combined weight coefficient of each evaluation index by combining subjective and objective methods;

[0158] The optimal solution determination module is used to calculate the comprehensive contribution coefficient of each frontier optimization solution according to the combined weight coefficient of each evaluation index, and select the optimal design solution with the best balance from the multiple frontier optimization solutions according to the comprehensive contribution coefficient.

[0159] In some embodiments, in the multi-scenario working condition construction module, the method of constructing the full-scenario working condition includes:

[0160] Data collection: Use the cloud big data system to obtain the background driving data of multiple models, including longitude, latitude, vehicle speed, battery power, mileage, pedal opening, etc.; construct short-trip segments based on the vehicle speed change characteristics, and each short-trip segment represents a working condition, such as idling condition, acceleration condition, deceleration condition, and constant speed condition.

[0161] Data preprocessing: traverse each short trip segment, use linear interpolation to fill in the missing data within a short duration (such as within 5 seconds), and directly delete the missing data exceeding 5 seconds; at the same time, eliminate the data with a speed of 0 km / h and a duration exceeding the set duration (such as 180 seconds); filter the data, and replace the original data with the mean of the original data neighborhood data to form a mean sequence.

[0162] Condition fitting: Fitting the multi-scenario conditions of multiple users' travel based on all pre-processed short trip segments. The fitting methods include:

[0163] Extracting the characteristic parameters of the trip segments, wherein the characteristic parameters include idle speed, acceleration, deceleration, uniform speed ratio, maximum speed, average speed, ratio of each speed segment, mileage and other multiple (24) characteristic parameters to form an initial characteristic parameter matrix;

[0164] Standardize the initial characteristic parameter matrix to eliminate the influence of the dimension and magnitude of each characteristic parameter;

[0165] Solve the covariance matrix of the standardized characteristic parameter matrix; the covariance matrix can reflect the correlation between each characteristic parameter;

[0166] Calculate the contribution degrees b1, b2, b3, …, b of each characteristic parameter according to the covariance matrix 24 ; the contribution degree of the characteristic parameter is determined by calculating the eigenvalue. The larger the eigenvalue, the stronger the explanatory ability of the corresponding characteristic parameter to the data variation, and the greater the contribution degree;

[0167] Select the front characteristic parameters with the cumulative contribution rate exceeding the set ratio (such as 80%) as the principal components to obtain the characteristic parameter matrix composed of the principal components; the purpose of this step is to reduce the dimension, reduce the complexity of the data, and at the same time retain the main characteristic information of the data. For example, if the cumulative contribution rate of the first 10 characteristic parameters exceeds 80%, then these 10 characteristic parameters are selected to construct a new characteristic parameter matrix.

[0168] Perform clustering processing on the characteristic parameter matrix composed of the principal components, and select a suitable clustering algorithm, such as the K-Means algorithm. The K-Means algorithm is a commonly used clustering algorithm, and its goal is to divide the data points into K clusters (here K = n), so that the data points within the same cluster have a high similarity, and the data points in different clusters have a low similarity.

[0169] Determine the clustering goal as combining the working conditions with the closest Euclidean distances of speed and acceleration. In the K-Means algorithm, a distance metric method needs to be defined. Here, the Euclidean distance is selected to measure the similarity between working conditions. The calculation formula of the Euclidean distance is: where (x1, x2, …, x M ) and (y1, y2, …, y M ) are the feature vectors of two working conditions, and M is the number of features.

[0170] Set the number of clusters as n (that is, the number of multi-scenario working conditions to be fitted), and initialize n cluster centers. The cluster centers can be randomly selected or selected by some heuristic methods, such as selecting data points with a far distance as the initial cluster centers to improve the clustering effect.

[0171] Start the iterative clustering process:

[0172] Calculate the Euclidean distance from each data point (that is, each working condition) to the n cluster centers, and assign each data point to the cluster where the closest cluster center is located;

[0173] Recalculate the cluster center of each cluster, that is, calculate the mean value of all data points within the cluster as the new cluster center;

[0174] Repeat the above two steps until the cluster center no longer changes or changes very little, or the assignment of data points no longer changes. At this time, the clustering process is completed, and n multi-scenario working conditions

Scene1, Scene2 ……, Scene n

[0175] In some embodiments, after the fitting is completed, it is necessary to calculate the total duration of each working condition. The formula is:

[0176]

[0177] where t m is the total duration of the m-th working condition after fitting; t cycle is the total duration of all working conditions after fitting; t all is the total duration of the collected data; t m,i is the duration of the i-th short stroke segment of the m-th working condition.

[0178] Through the above formula, the time proportion of each working condition in the entire set of fitted working conditions can be obtained, further improving the construction of multi-scenario working conditions.

[0179] In some embodiments, it also includes allocating the weight proportion of each working condition according to the total duration. The allocation method includes:

[0180] Divide the total duration of each working condition by the total duration of the collected data to obtain the actual proportion of each working condition;

[0181] Analyze the weight settings corresponding to each scenario in the WLTC working condition. As a standard working condition, the weights of each scenario in the WLTC working condition have certain reference value. The actual scenario proportion and the weights of the corresponding scenarios in the WLTC working condition are considered comprehensively. The weighted average method can be used to assign different weight coefficients to the actual scenario proportion and the WLTC working condition weights respectively, and calculate the comprehensive weight of each working condition.

[0182] In some embodiments, in the weight coefficient calculation module, the combined weight coefficient includes a subjective weight coefficient, and its calculation method includes:

[0183] Score the evaluation indicators pairwise according to the preset scoring criteria; construct a judgment matrix A according to the scores;

[0184]

[0185] Normalize the judgment matrix A to obtain the subjective weights of each evaluation indicator

[0186] w s=(w s1 , w s2 , w s3 , …, w sn ) T 。

[0187] In some embodiments, in the weight coefficient calculation module, the combined weight coefficient includes an objective weight coefficient, and the determination method thereof includes:

[0188] Score according to the influence of the design factors on the evaluation indexes of each design scheme; the greater the influence of the change of the design factor (the rotational speed in this embodiment) on the evaluation indexes of each design scheme, the greater the weight, and vice versa. Construct the first decision matrix B according to the score;

[0189]

[0190] wherein, b ij is the value of the j-th evaluation index of the i-th set of design schemes; m represents the number of design schemes obtained by the optimal Latin hypercube sampling method.

[0191] Perform non-negative processing on the elements in the first decision matrix according to the following formula:

[0192]

[0193] Calculate the information entropy of each evaluation index according to the first decision matrix after non-negative processing; the calculation method includes:

[0194]

[0195] wherein,

[0196] Calculate the objective weight coefficient of each evaluation index according to the information entropy:

[0197]

[0198] wherein, is the objective weight coefficient of the j-th evaluation index; n is the number of evaluation indexes.

[0199] Finally, obtain the objective weight coefficient w0:

[0200] w o =(w o1 , w o2 , w o3 , …, w on ) T 。

[0201] In some embodiments, the calculation method of the comprehensive contribution degree coefficient includes:

[0202] Weight the first decision matrix according to the combined weight coefficients to obtain the second decision matrix \(V=(v_{ij})\) ij ) m×n =(w_{j}b_{ij}) ij b ij ) m×n ; \(w_{j}\) is the combined weight coefficient of the evaluation index; ij

[0203] Determine the positive ideal solution set \(V^{+}\) and the negative ideal solution set \(V^{-}\) for each evaluation index according to the second decision matrix; + - ;

[0204] \(V^{+}=[v_{1}^{+},v_{2}^{+},\cdots,v_{m}^{+}]\) + =[v1 + ,v2 + ,...,v m + )

[0205] \(V^{-}=[v_{1}^{-},v_{2}^{-},\cdots,v_{m}^{-}]\) - =[v1 - ,v2 - ,...,v m - )

[0206] \(v_{j}^{+}=\max(v_{1j},v_{2j},\cdots,v_{nj})\); \(v_{j}^{-}=\min(v_{1j},v_{2j},\cdots,v_{nj})\); j + =max(v 1j ,v 2j ,…,v mj );v j - =min(v 1j ,v 2j ,…,v mj );

[0207] Calculate the Euclidean distances of the positive and negative ideal solution sets for each evaluation index;

[0208] Determine the development trend correlation matrix corresponding to the positive and negative ideal solution sets according to the second decision matrix; the development trend correlation matrix is used to represent the similarity degree of the development trends of each frontier optimization scheme and the positive and negative ideal solution sets on each evaluation index;

[0209] Obtain the correlation coefficients with respect to the positive and negative ideal solution sets according to the development trend correlation matrix;

[0210] Calculate the relative fitting degrees of each frontier optimization scheme with the positive and negative ideal solution sets \(V^{+}\), \(V^{-}\) according to the Euclidean distances and the correlation coefficients; + 、V - ;

[0211] Determine the comprehensive contribution degree coefficients of each frontier optimization scheme according to the relative fitting degrees.

[0212] Furthermore, the calculation method of the development trend correlation matrix includes:

[0213]

[0214] v j + = max(v 1j , v 2j , …, v mj ); v j - = min(v 1j , v 2j , …, v mj )

[0215] In the formula:

[0216] G + represents the development trend correlation matrix corresponding to the positive ideal solution set;

[0217] G - represents the development trend correlation matrix corresponding to the negative ideal solution set;

[0218] v ij represents an element in the second decision matrix and is used to represent the value of the j-th evaluation index of the i-th frontier optimization plan.

[0219] In some embodiments, the method for determining the operating point corresponding to each full-scenario operating condition includes:

[0220] When the first vehicle operating parameter value (SOC in this embodiment) is less than or equal to the first set value: If the second vehicle operating parameter value (vehicle speed in this embodiment) is greater than or equal to the second set value, it is considered that the operating point corresponding to the current operating condition is the fifth operating point; otherwise, it is considered that the operating point corresponding to the current operating condition is the fourth operating point;

[0221] When the first vehicle operating parameter value is greater than the first set value and less than the third set value: If the second vehicle operating parameter value is greater than or equal to the second set value, it is considered that the operating point corresponding to the current operating condition is the third operating point; otherwise, it is considered that the operating point corresponding to the current operating condition is the second operating point;

[0222] When the first vehicle operating parameter value is greater than the third set value: If the second vehicle operating parameter value is less than the second set value, it is considered that the operating point corresponding to the current operating condition is the first operating point; if the second vehicle operating parameter is greater than or equal to the second set value and less than the fourth set value, it is considered that the operating point corresponding to the current operating condition is the third operating point; if the vehicle speed is greater than or equal to the fourth set value, it is considered that the operating point corresponding to the current operating condition is the fourth operating point.

[0223] In some embodiments, before determining the operating point corresponding to each full-scenario operating condition, the following preconditions are also included for judgment: the range extender is turned on and the first vehicle operating parameter value is less than or equal to the set maximum value.

[0224] An embodiment of the present invention also provides an optimization method for the balance of the fixed-point strategy of a range-extended electric vehicle for the full scenario, including:

[0225] Taking a certain operating parameter of the vehicle as a design factor, multiple design schemes are obtained through a sampling method according to the value range of the certain operating parameter. Each design scheme includes the value of the design factor at each operating point; the certain operating parameter in this embodiment is the engine speed;

[0226] Construct an approximation model for evaluating the multiple design schemes; the input of the approximation model is the design scheme, and the output is the evaluation index value for evaluating the input design scheme;

[0227] Construct an optimization model for finding the optimal design scheme. The optimization model uses a multi-objective optimization algorithm to globally optimize the evaluation index values output by the approximation model, and obtains multiple frontier optimization schemes that can simultaneously optimize multiple evaluation indexes;

[0228] Calculate the combined weight coefficient of each evaluation index in a subjective and objective combined manner;

[0229] Calculate the comprehensive contribution coefficient of each frontier optimization scheme according to the combined weight coefficient of each evaluation index, and select the design scheme with the best balance from the multiple frontier optimization schemes according to the comprehensive contribution coefficient.

[0230] In some embodiments, it also includes determining the operating point of the vehicle. The determination method includes: constructing short-trip segments based on multi-vehicle background driving data; fitting multiple full-scenario operating conditions of user trips through clustering analysis according to the short-trip segments; determining the operating point corresponding to each full-scenario operating condition according to multiple vehicle operating parameters in the full-scenario operating conditions. The operating point is used to represent the operating state of the range extender under specific operating conditions; each operating point corresponds to a different combination of speed and torque, and the operating point is selected in the optimal economic region to give play to the advantages of the range extender and take into account the performance requirements under different operating conditions.

[0231] An embodiment of the present invention also provides a non-transitory computer-readable storage medium. The computer-readable storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a processor, each step of the method described in the present invention is implemented, which will not be elaborated here.

[0232] A computer-readable storage medium may be an internal storage unit of the data transmission device or computer device provided in any of the foregoing embodiments, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device.

[0233] Furthermore, the computer-readable storage medium may also include both the internal storage unit and the external storage device of the computer device. The computer-readable storage medium is used to store the computer program and other programs and data required by the computer device. The computer-readable storage medium may also be used to temporarily store data to be output or already output.

[0234] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0235] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0236] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0237] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable apparatus to generate a computer-implemented process, thereby providing instructions for implementing the steps of the process Figure 1 one process or multiple processes and / or blocks Figure 1 steps for the functions specified in one block or multiple blocks.

[0238] An embodiment of the present invention further provides a computer program product, including a computer program / instructions, and when the computer program / instructions are executed by a processor, the steps of the method for optimizing the balance of the fixed-point strategy of the range-extended electric vehicle for the full scenario are implemented.

[0239] The content not detailedly described in this specification belongs to the prior art well known to those skilled in the art.

Claims

1. A fixed-point strategy balance optimization system for range-extended electric vehicles facing all scenarios, characterized in that Including: Design scheme sampling module: Used to take a certain operating parameter of the vehicle as a design factor, and obtain multiple design schemes through sampling according to the value range of the certain operating parameter. Each design scheme includes the value of the design factor at each working point; the working point is used to represent the operating state of the range extender under specific working conditions; Approximation model construction module: Used to construct an approximation model for evaluating the multiple design schemes; the input of the approximation model is the design scheme, and the output is the evaluation index value for evaluating the input design scheme; Global multi-objective optimization module: Used to construct an optimization model for finding the optimal design scheme. The optimization model uses a multi-objective optimization algorithm to globally optimize the evaluation index values output by the approximation model, and obtains multiple frontier optimization schemes that can simultaneously optimize multiple evaluation indexes; Weight coefficient calculation module: Used to calculate the combined weight coefficient of each evaluation index in a subjective and objective combined manner; Optimal scheme determination module: Used to calculate the comprehensive contribution degree coefficient of each frontier optimization scheme according to the combined weight coefficient of each evaluation index, and select the optimal design scheme with the best balance from the multiple frontier optimization schemes according to the comprehensive contribution degree coefficient.

2. The optimized system for the balance of the fixed-point strategy of the range-extended electric vehicle for the full scenario according to claim 1, characterized in that, The combined weight coefficient includes a subjective weight coefficient, and its calculation method includes: Scoring each pair of evaluation indexes according to a preset scoring standard; constructing a judgment matrix according to the scoring; Performing normalization processing on the judgment matrix to obtain the subjective weight coefficients of each evaluation index.

3. The full-scenario-oriented range-extended electric vehicle fixed-point strategy balance optimization system according to claim 1 or 2, characterized in that, In the weight coefficient calculation module, the combined weight coefficient includes an objective weight coefficient, and its determination method includes: Scoring according to the influence of the design factor on the evaluation index of each design scheme; constructing a first decision matrix according to the scoring; Performing non-negative processing on the elements in the first decision matrix; Calculating the information entropy of each evaluation index according to the first decision matrix after non-negative processing; Calculating the objective weight coefficient of each evaluation index according to the information entropy.

4. The full-scenario oriented range-extended electric vehicle fixed-point strategy balance optimization system according to claim 3, characterized in that, The calculation method of the comprehensive contribution degree coefficient includes: Weighting the first decision matrix according to the combined weight coefficient to obtain a second decision matrix; Determining the positive and negative ideal solution sets of each evaluation index according to the second decision matrix; Calculating the Euclidean distance between the positive and negative ideal solution sets of each evaluation index; Determining a development trend correlation matrix corresponding to the positive and negative ideal solution sets according to the second decision matrix; the development trend correlation matrix is used to represent the similarity degree of the development trends of each frontier optimization scheme and the positive and negative ideal solution sets on each evaluation index; Obtaining the correlation coefficient with respect to the positive and negative ideal solution sets according to the development trend correlation matrix; Calculating the relative fitting degree of each frontier optimization scheme with the positive and negative ideal solution sets V+ and V- according to the Euclidean distance and the correlation coefficient; Determining the comprehensive contribution degree coefficient of each frontier optimization scheme according to the relative fitting degree.

5. The full-scenario-oriented range-extended electric vehicle fixed-point strategy balance optimization system according to claim 4, wherein, The calculation method of the development trend correlation matrix includes: v j + = max(v 1j , v 2j , …, v mj ); v j - = min(v 1j , v 2j , …, v mj ) Where: G + represents the development trend correlation matrix corresponding to the positive ideal solution set; G - represents the development trend correlation matrix corresponding to the negative ideal solution set; v ij It represents an element in the second decision matrix and is used to represent the value of the j-th evaluation index of the i-th frontier optimization solution.

6. The full-scenario-oriented range-extended electric vehicle fixed-point strategy balance optimization system according to claim 1, characterized in that Also including: Multi-scenario working condition construction module: Used to construct short travel segments according to multi-vehicle background driving data; Fitting multiple full-scenario working conditions for users' travel through clustering analysis according to the short travel segments; Fixed-point strategy model construction module: used to construct a fixed-point strategy model, which is used to determine the working point corresponding to each full-scenario working condition according to multiple vehicle operating parameters in the full-scenario working condition; the working point is used to represent the operating state of the range extender under specific working conditions.

7. The full-scenario-oriented extended-range electric vehicle fixed-point strategy balance optimization system according to claim 6, characterized in that The method for determining the working point corresponding to each full-scenario working condition includes: When the value of the first vehicle operating parameter is less than or equal to the first set value: if the value of the second vehicle operating parameter is greater than or equal to the second set value, it is considered that the working point corresponding to the current working condition is the fifth working point; otherwise, it is considered that the working point corresponding to the current working condition is the fourth working point; When the value of the first vehicle operating parameter is greater than the first set value and less than the third set value: if the value of the second vehicle operating parameter is greater than or equal to the second set value, it is considered that the working point corresponding to the current working condition is the third working point; otherwise, it is considered that the working point corresponding to the current working condition is the second working point; When the value of the first vehicle operating parameter is greater than the third set value: if the value of the second vehicle operating parameter is less than the second set value, it is considered that the working point corresponding to the current working condition is the first working point; if the second vehicle operating parameter is greater than or equal to the second set value and less than the fourth set value, it is considered that the working point corresponding to the current working condition is the third working point; if the vehicle speed is greater than or equal to the fourth set value, it is considered that the working point corresponding to the current working condition is the fourth working point.

8. An optimization method for the balance of the fixed-point strategy of a range-extended electric vehicle for all scenarios, characterized in that, It includes: Taking a certain operating parameter of the vehicle as a design factor, obtaining multiple design schemes through a sampling method according to the value range of the certain operating parameter, and each design scheme includes the value of the design factor at each working point; The working point is used to represent the operating state of the range extender under specific working conditions; Constructing an approximate model for evaluating the multiple design schemes; the input of the approximate model is the design scheme, and the output is the evaluation index value for evaluating the input design scheme; Constructing an optimization model for finding the optimal design scheme, and the optimization model uses a multi-objective optimization algorithm to globally optimize the evaluation index values output by the approximate model to obtain multiple frontier optimization schemes that can optimize multiple evaluation indexes simultaneously; Calculating the combined weight coefficient of each evaluation index in a subjective and objective combined manner; Calculating the comprehensive contribution degree coefficient of each frontier optimization scheme according to the combined weight coefficient of each evaluation index, and selecting the optimal design scheme with the best balance from the multiple frontier optimization schemes according to the comprehensive contribution degree coefficient.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it realizes the steps of the full-scenario-oriented fixed-point strategy balance optimization method for range-extended electric vehicles as described in claim 8.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by a processor, it realizes the steps of the full-scenario-oriented fixed-point strategy balance optimization method for range-extended electric vehicles as described in claim 8.

Images