Brushless direct current motor loss optimization method and system
By acquiring and configuring the loss database of brushless DC motors, and performing consistency evaluation and expansion, and combining variation optimization technology to generate loss optimization solutions, the problem of poor loss optimization effect of brushless DC motors in different environments is solved, and the problem of poor loss optimization effect of brushless DC motors in different environments is realized. More efficient loss optimization.
Patent Information
- Application Number
- CN202411374090.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-29
- Publication Date
- 2025-02-14
AI Technical Summary
In the prior art, brushless DC motors are difficult to achieve the corresponding expected loss optimization effect under different application environments.
By obtaining the target application information of the target brushless DC motor, configuring the original loss database, and performing consistency evaluation and neighborhood expansion, an expansion loss database is generated. Then, based on the current temperature field distribution, the variation optimization is carried out in the original and expanded loss database, a target loss optimization scheme is generated, and the motor is lost optimization.
It realizes precise optimization of the loss of brushless DC motors in different application environments, and improves the loss optimization effect.
Smart Images
Figure CN119442585A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of brushless DC motors, and in particular to a method and system for optimizing the loss of a brushless DC motor. Background Art
[0002] As an efficient and stable power source, brushless DC motors have been widely used in many fields such as electric vehicles, aerospace, and industrial automation. However, with the diversification of application environments and the complexity of load requirements, the loss problem of motors has gradually become prominent and has become a key factor restricting the improvement of motor performance. Traditional loss optimization methods are often based on fixed parameters or empirical formulas, which cannot accurately reflect the actual loss of motors under different environments and load conditions, resulting in limited optimization effects. Therefore, a method that can adapt to different application environments and accurately optimize motor losses has become an urgent problem to be solved in the industry. Summary of the invention
[0003] The embodiments of the present application provide a brushless DC motor loss optimization method and system, which solves the technical problem in the prior art that the brushless DC motor is difficult to achieve the corresponding expected loss optimization effect in different application environments.
[0004] In view of the above problems, an embodiment of the present application provides a method and system for optimizing the losses of a brushless DC motor.
[0005] A first aspect of an embodiment of the present application provides a method for optimizing losses of a brushless DC motor, the method comprising:
[0006] Acquiring target application information of a target brushless DC motor, wherein the application information includes target environment information and target load information;
[0007] Using the target environment information and the target load information as constraints, configuring an original loss database for the target brushless DC motor;
[0008] Performing consistency evaluation on the original loss database, and performing neighborhood expansion on the original loss database according to a first expansion scale according to a consistency evaluation result to generate an expanded loss database;
[0009] Collecting the current target temperature field distribution of the target brushless DC motor, performing mutation optimization in the original loss database and the expanded loss database based on the target temperature field distribution, and generating a target loss optimization solution;
[0010] The target brushless DC motor is subjected to loss optimization based on the target loss optimization scheme.
[0011] A second aspect of the embodiments of the present application provides a brushless DC motor loss optimization system, the system comprising:
[0012] An information acquisition module, the information acquisition module is used to acquire target application information of the target brushless DC motor, wherein the application information includes target environment information and target load information;
[0013] A configuration module, the configuration module is used to configure an original loss database for the target brushless DC motor based on the target environment information and the target load information;
[0014] A domain expansion module, the domain expansion module is used to perform consistency evaluation on the original loss database, and perform neighborhood expansion on the original loss database according to a first expansion scale according to a consistency evaluation result to generate an expanded loss database;
[0015] An optimization module, the optimization module is used to collect the current target temperature field distribution of the target brushless DC motor, perform variation optimization in the original loss database and the expanded loss database based on the target temperature field distribution, and generate a target loss optimization solution;
[0016] A loss optimization module, wherein the loss optimization module is used to optimize the loss of the target brushless DC motor based on the target loss optimization scheme.
[0017] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0018] First, the target application information of the target brushless DC motor is obtained, wherein the application information includes target environment information and target load information. Then, the original loss database is configured for the target brushless DC motor with the target environment information and target load information as constraints. Next, the original loss database is evaluated for consistency, and the original loss database is expanded in the neighborhood according to the first expansion scale based on the consistency evaluation result to generate an expanded loss database. After that, the current target temperature field distribution of the target brushless DC motor is collected, and mutation optimization is performed in the original loss database and the expanded loss database based on the target temperature field distribution to generate a target loss optimization scheme. Finally, the loss of the target brushless DC motor is optimized based on the target loss optimization scheme. The technical problem that it is difficult for brushless DC motors to achieve the corresponding expected loss optimization effect under different application environments in the prior art is solved, and the technical effect of improving the loss optimization effect of brushless DC motors is achieved. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0020] Figure 1 A schematic flow chart of a method for optimizing the loss of a brushless DC motor provided in an embodiment of the present application;
[0021] Figure 2 A schematic diagram of the structure of a brushless DC motor loss optimization system provided in an embodiment of the present application.
[0022] Explanation of the reference numerals: information acquisition module 11, configuration module 12, domain expansion module 13, optimization module 14, loss optimization module 15. DETAILED DESCRIPTION
[0023] The embodiments of the present application provide a brushless DC motor loss optimization method and system, thereby solving the technical problem in the prior art that the brushless DC motor is difficult to achieve the corresponding expected loss optimization effect in different application environments.
[0024] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0025] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices.
[0026] Embodiment 1
[0027] like Figure 1 As shown, an embodiment of the present application provides a method for optimizing losses of a brushless DC motor, wherein the method comprises:
[0028] Acquiring target application information of a target brushless DC motor, wherein the application information includes target environment information and target load information;
[0029] The target application information of the target brushless DC motor includes target environment information and target load information. The target environment information mainly includes environmental factors such as temperature, humidity, and air pressure when the motor is running. For example, a high temperature environment may cause the internal temperature of the motor to rise, increasing heat loss; a high humidity environment may reduce the insulation performance of the motor and increase electrical loss. The target load information refers to the load characteristics that the motor needs to withstand during operation, including the size, type, and change pattern of the load. The load information directly determines the operating state and power demand of the motor, and has a direct impact on the loss of the motor. For example, heavy loads or frequently changing loads may cause the motor to overheat, increasing mechanical and electrical losses. By obtaining the target application information of the target brushless DC motor, you can have a more comprehensive understanding of the motor's operating environment and load conditions, providing important data support for loss optimization.
[0030] Using the target environment information and the target load information as constraints, configuring an original loss database for the target brushless DC motor;
[0031] Target environmental information, such as temperature, humidity, and air pressure, directly affects the heat dissipation, insulation, and electromagnetic properties of the motor, thereby determining the loss level of the motor. Similarly, target load information, including load size, change frequency, and type, also profoundly affects the operating status and loss of the motor. Therefore, when configuring the original loss database, it is necessary to use these environmental information and load information as key constraints to obtain the loss data of the motor under different environments and loads to form the original loss database. The original loss database is the loss condition of the brushless DC motor in operation that meets the target environmental information and target load information, including different motor loss conditions under different temperature fields, and the loss optimization scheme adopted.
[0032] Further, the method comprises:
[0033] Performing data mining with the target environment information and the target load information as constraints to generate an initial loss data set;
[0034] Performing format standardization processing on the initial loss data set to generate a standardized loss data set;
[0035] Using a semantic recognition network layer to identify the data missing degree of the standardized loss data set, to obtain a data missing degree set;
[0036] Determine whether the data missing degree set is less than or equal to a preset data missing degree threshold. If so, add the corresponding standardized loss data into the original loss database.
[0037] Using data mining technology, data matching the target environment information and target load information are extracted from a large amount of historical data or experimental data. These data constitute the initial loss data set, which reflects the loss of the motor under different environmental and load conditions. In order to ensure the comparability and consistency of the data, it is necessary to standardize the format of the initial loss data set, including unifying the data unit, converting the data format, processing abnormal values, etc., and finally generate a standardized loss data set. The semantic recognition network layer is used to identify the data missing degree of the standardized loss data set to obtain the data missing degree set, which reflects the degree and distribution of missing information in the data. The semantic recognition network layer can intelligently analyze the integrity and consistency of the data and identify missing or abnormal data points. The quality of the data is judged according to the preset data missing degree threshold. If the value in the data missing degree set is less than or equal to the preset threshold, it means that the data integrity is good and can be added to the original loss database.
[0038] Performing consistency evaluation on the original loss database, and performing neighborhood expansion on the original loss database according to a first expansion scale according to a consistency evaluation result to generate an expanded loss database;
[0039] The original loss database is evaluated for consistency, and based on the consistency evaluation results, the original loss database is expanded in the neighborhood according to the first expansion scale to generate an expanded loss database. In the actual operation of the target brushless DC motor, the environment and load will fluctuate. Therefore, if the original loss database has a high consistency, it indicates that the sample quality is not high and needs to be expanded. The first expansion scale is reasonably set according to the characteristics of the motor, application requirements, and data quality.
[0040] Furthermore, the consistency evaluation of the original loss database is performed, and the method includes:
[0041] Searching the original loss database based on the temperature field distribution to generate an original temperature field distribution set;
[0042] Traversing the original temperature field distribution set to extract the local temperature maximum value, the local temperature minimum value and the number of local hot spots, and generating multiple local temperature maximum values, multiple local temperature minimum values and multiple local hot spot numbers;
[0043] Using a box plot, respectively, the box lengths of the plurality of local temperature maxima, the plurality of local temperature minima, and the plurality of local hot spots are calculated to generate a first box length, a second box length, and a third box length;
[0044] Performing weighted calculation on the first box length, the second box length and the third box length to generate an original box length;
[0045] The difference between the original box length and the predetermined box length is calculated, and the ratio of the difference to the predetermined box length is used as the consistency evaluation result.
[0046] According to the temperature field distribution information of the current motor, the original loss database is searched. By comparing the characteristics and patterns of the temperature field distribution, data similar to the current motor state is screened out from the original loss database to generate an original temperature field distribution set, which contains temperature field distribution data similar to the current motor state. The original temperature field distribution set is traversed to extract the local temperature maximum, local temperature minimum and local hot spot numbers, and multiple local temperature maximums, multiple local temperature minimums and multiple local hot spot numbers are obtained, which respectively describe the temperature extremes and hot spot distributions of different parts of the motor, and reflect the local characteristics and hot spot distribution of the motor temperature field.
[0047] The box plot is a graphical tool used to display data distribution. It usually consists of five statistics, including the minimum value, the lower quartile, the median, the upper quartile, and the maximum value, which can intuitively reflect the concentration, dispersion, and outliers of the data. By calculating the five-number summary (minimum value, lower quartile, median, upper quartile, and maximum value) of multiple local temperature maxima, multiple local temperature minima, and multiple local hot spots, the box length of the box plot, that is, the distance between the upper quartile and the lower quartile, is calculated based on the five-number summary, thereby obtaining the first box length, the second box length, and the third box length. These box lengths represent the distribution of the corresponding parameters, respectively. Among them, the box with a shorter box length indicates that the data is more concentrated and has a higher consistency; the longer box may mean that the data distribution is more scattered and has a poor consistency. The first box length, the second box length, and the third box length are weighted to generate the original box length. The original box length comprehensively considers the influence of multiple temperature field distribution parameters and can more comprehensively reflect the consistency of the motor temperature field distribution. Calculate the difference between the original box length and the predetermined box length, and use the ratio of the difference to the predetermined box length as the consistency evaluation result. The predetermined box length is a threshold set according to experience or standards, which represents the ideal state of consistency of the motor temperature field distribution. By comparing the difference between the original box length and the predetermined box length, the degree of consistency between the current motor temperature field distribution and the ideal state can be judged. The smaller the difference, the lower the consistency evaluation result obtained, indicating that the data is more concentrated, the sample types are fewer, and the sample quality is lower. When expanding the scale to optimize, a larger scale needs to be used. The larger the difference, the more dispersed the data is, and the number of sample types is large enough. At this time, when expanding the scale to optimize, a smaller scale can be used.
[0048] Furthermore, according to the consistency evaluation result, the original loss database is neighborhood expanded according to a first expansion scale to generate an expanded loss database, and the method includes:
[0049] Adjusting the predetermined expansion scale based on the consistency evaluation result to obtain the first expansion scale, wherein the predetermined expansion scale is the expansion scale corresponding to the predetermined box length;
[0050] Expanding the target environment information and the target load information based on the first expansion scale to obtain a target environment information expansion interval and a target load information expansion interval;
[0051] Performing data mining with the target environment information expansion interval and the target load information expansion interval as constraints to generate a neighborhood loss database;
[0052] The original loss database and the neighborhood loss database are combined to generate the expanded loss database.
[0053] According to the consistency evaluation result, the predetermined expansion scale is adjusted to obtain the first expansion scale. The predetermined expansion scale is preliminarily set, based on the expansion range corresponding to the predetermined box length, and represents the expected expansion degree before the actual consistency evaluation. The consistency evaluation result provides feedback on the matching degree between the current motor temperature field distribution and the existing data in the database. If the consistency evaluation result is good, it means that the current database is relatively complete and may not need an excessively large expansion scale; on the contrary, if the consistency evaluation result is poor, a larger expansion scale is required to cover more possible situations. Therefore, according to the consistency evaluation result, the predetermined expansion scale is appropriately adjusted to obtain a more appropriate first expansion scale. Based on the first expansion scale, the target environment information and the target load information are expanded to obtain the target environment information expansion interval and the target load information expansion interval, that is, according to the first expansion scale, the scope of the target environment information and the target load information is expanded to cover more possible operating scenarios and load conditions. With the target environment information expansion interval and the target load information expansion interval as constraints, data mining is performed to generate a neighborhood loss database. Data mining is a technology that extracts useful information from a large amount of data. Using data mining technology, we can find and extract data related to motor losses within the expanded range of environmental information and load information to build a neighborhood loss database, which contains the loss data of motors under a wider range of operating conditions. The original loss database and the neighborhood loss database are combined to generate an expanded loss database. By merging these two databases, an expanded loss database containing both original data and expanded data is obtained.
[0054] Collecting the current target temperature field distribution of the target brushless DC motor, performing mutation optimization in the original loss database and the expanded loss database based on the target temperature field distribution, and generating a target loss optimization solution;
[0055] The current target temperature field distribution of the target brushless DC motor is collected, and mutation optimization is performed in the original loss database and the expanded loss database to generate a target loss optimization plan. The target loss optimization plan provides motor operating parameter adjustment suggestions for the current target temperature field distribution to optimize loss performance.
[0056] Furthermore, the current target temperature field distribution of the target brushless DC motor is collected, and variation optimization is performed in the original loss database and the expanded loss database based on the target temperature field distribution. The method includes:
[0057] Extract data from the original loss database and the expanded loss database according to a preset crossover factor to generate a mutation optimization loss database;
[0058] Extracting first loss data from the mutation optimization loss database, and calculating the similarity between the temperature field distribution of the first loss data and the target temperature field distribution to generate a first fitness;
[0059] Taking the first loss data as a starting point, searching in the mutation optimization loss database according to a preset variation step length to obtain second loss data, and calculating the similarity between the temperature field distribution of the second loss data and the target temperature field distribution to generate a second fitness;
[0060] Determine whether the first fitness is less than or equal to the second fitness, and if so, use the second fitness as the stage fitness;
[0061] After multiple iterations of optimization, when the preset number of iterations is met, the loss optimization solution in the loss data corresponding to the maximum fitness value of the stage is used as the target loss optimization solution.
[0062] Optionally, the preset crossover factor is the ratio of data extraction from the two databases, which is a fixed value. According to the preset crossover factor, a certain number of data records are randomly extracted from the original loss database and the expanded loss database, and these extracted data records will constitute the initial content of the mutation optimization loss database. A loss data is extracted from the mutation optimization loss database as the first loss data, and the first loss data will be used as the starting point of the optimization process. The similarity between the temperature field distribution of the first loss data and the target temperature field distribution is calculated by Euclidean distance, cosine similarity, etc., to generate the first fitness. Taking the first loss data as the starting point, search in the mutation optimization loss database according to the preset variable step length to find the second loss data adjacent to the first loss data, wherein the variable step length is set as the amplitude of a single movement in the mutation optimization database, which is a fixed value. Calculate the similarity between the temperature field distribution of the second loss data and the target temperature field distribution, and generate the second fitness. Determine whether the first fitness is less than or equal to the second fitness. If so, it means that the second loss data is closer to the target temperature field distribution, so the second fitness is used as the stage fitness, and the current optimal solution is updated to the second loss data. Otherwise, keep the first fitness as the stage fitness, and continue to search for the next round with the current optimal solution as the starting point. Repeat this process for multiple iterations. Each iteration will update the current optimal solution based on the fitness comparison result, and gradually approach the true optimal solution. When the preset number of iterations is met, stop searching, and output the loss optimization solution in the loss data corresponding to the maximum value of the stage fitness as the target loss optimization solution.
[0063] Further, the method comprises:
[0064] Determine whether the maximum value of the stage fitness of the mutation optimization loss database meets the preset fitness, and if not, adjust the preset crossover factor using a crossover update formula to generate a first crossover factor;
[0065] Extract data from the original loss database and the expanded loss database based on the first crossover factor to generate a first updated mutation optimization loss database;
[0066] Based on the first update mutation optimization loss database, obtaining a first update loss optimization solution;
[0067] Determine whether the stage fitness of the first updated loss optimization scheme is greater than the stage fitness of the target loss optimization scheme. If so, update the first updated loss optimization scheme to the target loss optimization scheme.
[0068] Preferably, the preset crossover factor is adjustable to meet the randomness and diversity and avoid falling into local optimization. Determine whether the maximum stage fitness of the mutation optimization loss database meets the preset fitness. The preset fitness is a threshold used to determine whether the currently found optimal solution is good enough. If the maximum stage fitness does not reach the preset fitness, it means that the current optimal solution is not ideal enough and the optimization strategy needs to be further adjusted. If the maximum stage fitness does not meet the preset fitness, the preset crossover factor is adjusted using the crossover update formula to generate a first crossover factor. The crossover factor plays a role in controlling the data extraction range in the mutation optimization. Adjusting the crossover factor can change the number and distribution of data extracted from the original loss database and the expanded loss database, thereby affecting the effect of subsequent optimization. Data is extracted from the original loss database and the expanded loss database based on the first crossover factor to generate a first updated mutation optimization loss database. Based on the first updated mutation optimization loss database, a new round of optimization process is carried out to obtain a first updated loss optimization scheme. Determine whether the stage fitness of the first updated loss optimization scheme is greater than the stage fitness of the target loss optimization scheme. If yes, it means that the first updated loss optimization scheme is better than the current target loss optimization scheme, so the first updated loss optimization scheme is updated to the target loss optimization scheme. By continuously iterating and adjusting the cross factor, the optimal solution that satisfies the preset fitness can be gradually approached, thereby generating a target loss optimization scheme for the target temperature field distribution.
[0069] Furthermore, the cross update formula is:
[0070]
[0071] Among them, l m is the first cross factor, l m1 is the preset crossover factor, f max is the maximum stage fitness of the mutation optimization loss database constructed based on the preset crossover factor, f′ x The default fitness.
[0072] The crossover update formula is used to adjust the preset crossover factor so that the range of data extraction can be dynamically adjusted according to the gap between the current stage fitness and the preset fitness during the mutation optimization process. The crossover update formula calculates the difference between the maximum stage fitness and the preset fitness, and multiplies it by the preset crossover factor to obtain a new crossover factor. m is the first cross factor, l m1 is the preset crossover factor, f max is the maximum stage fitness of the mutation optimization loss database constructed based on the preset crossover factor, f′ x The default fitness.
[0073] The target brushless DC motor is subjected to loss optimization based on the target loss optimization scheme.
[0074] The operating parameters of the target brushless DC motor are adjusted according to the target loss optimization plan to achieve loss optimization of the target brushless DC motor, thereby effectively reducing the loss of the motor, improving the efficiency, and thus improving the overall performance and service life of the motor.
[0075] In summary, the embodiments of the present application have at least the following technical effects:
[0076] First, the target application information of the target brushless DC motor is obtained, wherein the application information includes target environment information and target load information. Then, the original loss database is configured for the target brushless DC motor with the target environment information and target load information as constraints. Next, the original loss database is evaluated for consistency, and the original loss database is expanded in the neighborhood according to the first expansion scale based on the consistency evaluation result to generate an expanded loss database. After that, the current target temperature field distribution of the target brushless DC motor is collected, and mutation optimization is performed in the original loss database and the expanded loss database based on the target temperature field distribution to generate a target loss optimization scheme. Finally, the loss of the target brushless DC motor is optimized based on the target loss optimization scheme. The technical problem that it is difficult for brushless DC motors to achieve the corresponding expected loss optimization effect under different application environments in the prior art is solved, and the technical effect of improving the loss optimization effect of brushless DC motors is achieved.
[0077] Embodiment 2
[0078] Based on the same inventive concept as the method for optimizing the loss of a brushless DC motor in the aforementioned embodiment, Figure 2 As shown, the present application provides a brushless DC motor loss optimization system, and the system and method embodiments in the present application are based on the same inventive concept. The system includes:
[0079] An information acquisition module 11, wherein the information acquisition module 11 is used to acquire target application information of a target brushless DC motor, wherein the application information includes target environment information and target load information;
[0080] A configuration module 12, the configuration module 12 is used to configure an original loss database for the target brushless DC motor based on the target environment information and the target load information;
[0081] A domain expansion module 13, the domain expansion module 13 is used to perform consistency evaluation on the original loss database, and perform neighborhood expansion on the original loss database according to a first expansion scale according to a consistency evaluation result to generate an expanded loss database;
[0082] An optimization module 14, the optimization module 14 is used to collect the current target temperature field distribution of the target brushless DC motor, perform variation optimization in the original loss database and the expanded loss database based on the target temperature field distribution, and generate a target loss optimization solution;
[0083] The loss optimization module 15 is used to optimize the loss of the target brushless DC motor based on the target loss optimization scheme.
[0084] Furthermore, the configuration module 12 is used to execute the following method:
[0085] Performing data mining with the target environment information and the target load information as constraints to generate an initial loss data set;
[0086] Performing format standardization processing on the initial loss data set to generate a standardized loss data set;
[0087] Using a semantic recognition network layer to identify the data missing degree of the standardized loss data set, to obtain a data missing degree set;
[0088] Determine whether the data missing degree set is less than or equal to a preset data missing degree threshold. If so, add the corresponding standardized loss data into the original loss database.
[0089] Furthermore, the domain expansion module 13 is used to execute the following method:
[0090] Searching the original loss database based on the temperature field distribution to generate an original temperature field distribution set;
[0091] Traversing the original temperature field distribution set to extract the local temperature maximum value, the local temperature minimum value and the number of local hot spots, and generating multiple local temperature maximum values, multiple local temperature minimum values and multiple local hot spot numbers;
[0092] Using a box plot, respectively, the box lengths of the plurality of local temperature maxima, the plurality of local temperature minima, and the plurality of local hot spots are calculated to generate a first box length, a second box length, and a third box length;
[0093] Performing weighted calculation on the first box length, the second box length and the third box length to generate an original box length;
[0094] The difference between the original box length and the predetermined box length is calculated, and the ratio of the difference to the predetermined box length is used as the consistency evaluation result.
[0095] Furthermore, the domain expansion module 13 is used to execute the following method:
[0096] Adjusting the predetermined expansion scale based on the consistency evaluation result to obtain the first expansion scale, wherein the predetermined expansion scale is the expansion scale corresponding to the predetermined box length;
[0097] Expanding the target environment information and the target load information based on the first expansion scale to obtain a target environment information expansion interval and a target load information expansion interval;
[0098] Performing data mining with the target environment information expansion interval and the target load information expansion interval as constraints to generate a neighborhood loss database;
[0099] The original loss database and the neighborhood loss database are combined to generate the expanded loss database.
[0100] Furthermore, the optimization module 14 is used to execute the following method:
[0101] Extract data from the original loss database and the expanded loss database according to a preset crossover factor to generate a mutation optimization loss database;
[0102] Extracting first loss data from the mutation optimization loss database, and calculating the similarity between the temperature field distribution of the first loss data and the target temperature field distribution to generate a first fitness;
[0103] Taking the first loss data as a starting point, searching in the mutation optimization loss database according to a preset variation step length to obtain second loss data, and calculating the similarity between the temperature field distribution of the second loss data and the target temperature field distribution to generate a second fitness;
[0104] Determine whether the first fitness is less than or equal to the second fitness, and if so, use the second fitness as the stage fitness;
[0105] After multiple iterations of optimization, when the preset number of iterations is met, the loss optimization solution in the loss data corresponding to the maximum fitness value of the stage is used as the target loss optimization solution.
[0106] Furthermore, the optimization module 14 is used to execute the following method:
[0107] Determine whether the maximum value of the stage fitness of the mutation optimization loss database meets the preset fitness, and if not, adjust the preset crossover factor using a crossover update formula to generate a first crossover factor;
[0108] Extract data from the original loss database and the expanded loss database based on the first crossover factor to generate a first updated mutation optimization loss database;
[0109] Based on the first update mutation optimization loss database, obtaining a first update loss optimization solution;
[0110] Determine whether the stage fitness of the first updated loss optimization scheme is greater than the stage fitness of the target loss optimization scheme. If so, update the first updated loss optimization scheme to the target loss optimization scheme.
[0111] Furthermore, the optimization module 14 is used to execute the following method:
[0112] The cross update formula is:
[0113]
[0114] Among them, l m is the first cross factor, l m1 is the preset crossover factor, f max is the maximum stage fitness of the mutation optimization loss database constructed based on the preset crossover factor, f′ x The default fitness.
[0115] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0116] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
[0117] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. A method for optimizing the loss of a brushless DC motor, characterized in that: The method comprises: Acquiring target application information of a target brushless DC motor, wherein the application information includes target environment information and target load information; Using the target environment information and the target load information as constraints, configuring an original loss database for the target brushless DC motor; Performing consistency evaluation on the original loss database, and performing neighborhood expansion on the original loss database according to a first expansion scale according to a consistency evaluation result to generate an expanded loss database; Collecting the current target temperature field distribution of the target brushless DC motor, performing mutation optimization in the original loss database and the expanded loss database based on the target temperature field distribution, and generating a target loss optimization solution; The target brushless DC motor is subjected to loss optimization based on the target loss optimization scheme.
2. The method according to claim 1, characterized in that The method comprises: Performing data mining with the target environment information and the target load information as constraints to generate an initial loss data set; Performing format standardization processing on the initial loss data set to generate a standardized loss data set; Using a semantic recognition network layer to identify the data missing degree of the standardized loss data set, to obtain a data missing degree set; Determine whether the data missing degree set is less than or equal to a preset data missing degree threshold. If so, add the corresponding standardized loss data into the original loss database.
3. The method according to claim 1, characterized in that Performing consistency evaluation on the original loss database, the method comprising: Searching the original loss database based on the temperature field distribution to generate an original temperature field distribution set; Traversing the original temperature field distribution set to extract the local temperature maximum value, the local temperature minimum value and the number of local hot spots, and generating multiple local temperature maximum values, multiple local temperature minimum values and multiple local hot spot numbers; Using a box plot, respectively, the box lengths of the plurality of local temperature maxima, the plurality of local temperature minima, and the plurality of local hot spots are calculated to generate a first box length, a second box length, and a third box length; Performing weighted calculation on the first box length, the second box length and the third box length to generate an original box length; The difference between the original box length and the predetermined box length is calculated, and the ratio of the difference to the predetermined box length is used as the consistency evaluation result.
4. The method according to claim 3, characterized in that According to the consistency evaluation result, the original loss database is neighborhood expanded according to a first expansion scale to generate an expanded loss database, and the method includes: Adjusting the predetermined expansion scale based on the consistency evaluation result to obtain the first expansion scale, wherein the predetermined expansion scale is the expansion scale corresponding to the predetermined box length; Expanding the target environment information and the target load information based on the first expansion scale to obtain a target environment information expansion interval and a target load information expansion interval; Performing data mining with the target environment information expansion interval and the target load information expansion interval as constraints to generate a neighborhood loss database; The original loss database and the neighborhood loss database are combined to generate the expanded loss database.
5. The method according to claim 1, characterized in that The current target temperature field distribution of the target brushless DC motor is collected, and variation optimization is performed in the original loss database and the expanded loss database based on the target temperature field distribution. The method includes: Extract data from the original loss database and the expanded loss database according to a preset crossover factor to generate a mutation optimization loss database; Extracting first loss data from the mutation optimization loss database, and calculating the similarity between the temperature field distribution of the first loss data and the target temperature field distribution to generate a first fitness; Taking the first loss data as a starting point, searching in the mutation optimization loss database according to a preset variation step length to obtain second loss data, and calculating the similarity between the temperature field distribution of the second loss data and the target temperature field distribution to generate a second fitness; Determine whether the first fitness is less than or equal to the second fitness, and if so, use the second fitness as the stage fitness; After multiple iterations of optimization, when the preset number of iterations is met, the loss optimization solution in the loss data corresponding to the maximum fitness value of the stage is used as the target loss optimization solution.
6. The method according to claim 5, characterized in that The method comprises: Determine whether the maximum value of the stage fitness of the mutation optimization loss database meets the preset fitness, and if not, adjust the preset crossover factor using a crossover update formula to generate a first crossover factor; Extract data from the original loss database and the expanded loss database based on the first crossover factor to generate a first updated mutation optimization loss database; Based on the first update mutation optimization loss database, obtaining a first update loss optimization solution; Determine whether the stage fitness of the first updated loss optimization scheme is greater than the stage fitness of the target loss optimization scheme. If so, update the first updated loss optimization scheme to the target loss optimization scheme.
7. The method according to claim 6, characterized in that The cross update formula is: Among them, l m is the first cross factor, l m1 is the preset crossover factor, f max is the maximum stage fitness of the mutation optimization loss database constructed based on the preset crossover factor, f ′ x The default fitness.
8. A brushless DC motor loss optimization system, characterized in that: A method for optimizing the loss of a brushless DC motor according to any one of claims 1 to 7, the system comprising: An information acquisition module, the information acquisition module is used to acquire target application information of the target brushless DC motor, wherein the application information includes target environment information and target load information; A configuration module, the configuration module is used to configure an original loss database for the target brushless DC motor based on the target environment information and the target load information; A domain expansion module, the domain expansion module is used to perform consistency evaluation on the original loss database, and perform neighborhood expansion on the original loss database according to a first expansion scale according to a consistency evaluation result to generate an expanded loss database; An optimization module, the optimization module is used to collect the current target temperature field distribution of the target brushless DC motor, perform variation optimization in the original loss database and the expanded loss database based on the target temperature field distribution, and generate a target loss optimization solution; A loss optimization module is used to optimize the loss of the target brushless DC motor based on the target loss optimization scheme.