Performance optimization method and system for electromechanical equipment

Abstract

The invention provides a performance optimization method and system for electromechanical equipment, and relates to the technical field of intelligent control, and the method comprises the steps: obtaining a process index set and a quality index set through the interaction of a target electromechanical equipment scene; establishing a simulation analysis model of a production target of the equipment, and determining an adjustable range of a process index; obtaining a parameter optimization space and carrying out central composite design; calling the simulation model to perform analogue simulation to obtain a sample scheme matrix, and constructing and training a target agent model based on the matrix; the target agent model is used as an optimization target, and a multi-target optimization algorithm is used for optimizing the process index set of the electromechanical equipment, so that the technical problems of poor adaptability to complex scenes, low optimization accuracy and efficiency and difficulty in guaranteeing the intelligence and stability of equipment operation in the performance optimization process of the electromechanical equipment are solved; the accuracy and efficiency of performance optimization of the electromechanical equipment are improved, and then the production efficiency and the product quality are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control, in particular to a performance optimization method and system for electromechanical equipment. BACKGROUND

[0002] With the rapid development of industrialization and automation, electromechanical equipment is increasingly widely used in various fields, from manufacturing production lines to construction equipment in the construction industry, to key facilities in the fields of transportation and environmental protection. The performance of electromechanical equipment directly affects the overall system's operating efficiency, energy consumption, and environmental impact. Therefore, performance optimization of electromechanical equipment has become an important direction of current technological development.

[0003] Traditional optimization methods mainly rely on experience and trial and error, lack systematic analysis and scientific design guidance, and rely on technological development to provide a series of optimization techniques and algorithms. However, with the increasing complexity of electromechanical equipment and the increasing performance requirements, existing performance optimization methods have been difficult to meet actual needs.

[0004] In summary, the existing technology in the performance optimization process of electromechanical equipment often has poor adaptability to complex scenarios, low accuracy and efficiency of optimization, and difficulty in ensuring the intelligentization and stability of equipment operation. SUMMARY

[0005] The present application provides a performance optimization method and system for electromechanical equipment to solve the technical problems of poor adaptability to complex scenarios, low accuracy and efficiency of optimization, and difficulty in ensuring the intelligentization and stability of equipment operation in the performance optimization process of electromechanical equipment.

[0006] In view of the above problems, the present application provides a performance optimization method and system for electromechanical equipment.

[0007] In a first aspect, the present application provides a performance optimization method for electromechanical equipment, which is executed by a performance optimization system for electromechanical equipment, and the method comprises: interacting with a target electromechanical equipment scenario to obtain a target process index set and a target quality index set.

[0008] Based on a simulation method, a simulation analysis model of a target electromechanical equipment production target is established, and the adjustable range of multiple target process indexes in the target process index set is determined to obtain a parameter optimization space.

[0009] According to the parameter optimization space, a central composite design of multiple target process indexes is performed, and the simulation analysis model is called to perform simulation to obtain a sample scheme matrix, wherein the sample scheme matrix includes a process index parameter set and a corresponding quality index parameter set.

[0010] Based on the sample scheme matrix, a target agent model is constructed and trained, wherein the target agent model includes multiple sub-target agent models, and the sub-target agent models correspond one-to-one with multiple target quality indicators in the target quality indicator set.

[0011] Using the target proxy model as the optimization objective, a multi-objective optimization algorithm is applied to optimize the target process index set of the target electromechanical equipment and obtain an optimized process scheme.

[0012] Secondly, this application provides a performance optimization system for electromechanical equipment, the system comprising: The indicator acquisition module is used to interact with the target electromechanical equipment scenario and acquire the target process indicator set and the target quality indicator set.

[0013] The simulation model construction module is used to establish a simulation analysis model of the target electromechanical equipment production target based on the simulation method, and to determine the adjustable range of multiple target process indicators in the target process indicator set, and obtain the parameter optimization space.

[0014] The simulation module is used to perform a central composite design of multiple target process indicators based on the parameter optimization space, call the simulation analysis model to perform simulation, and obtain a sample scheme matrix, wherein the sample scheme matrix includes a set of process indicator parameters and a corresponding set of quality indicator parameters.

[0015] The surrogate model training module is used to construct and train a target surrogate model based on the sample scheme matrix. The target surrogate model includes multiple sub-target surrogate models, and the sub-target surrogate models correspond one-to-one with multiple target quality indicators in the target quality indicator set.

[0016] The process optimization module is used to optimize the target process index set of the target electromechanical equipment by applying a multi-objective optimization algorithm with the target proxy model as the optimization target, and to obtain an optimized process scheme.

[0017] One or more technical solutions provided in this application have at least the following technical effects or advantages: The performance optimization method for electromechanical equipment provided in this application obtains a set of target process indicators and a set of target quality indicators by interacting with a target electromechanical equipment scenario; based on simulation methods, a simulation analysis model of the production target of the target electromechanical equipment is established, and the adjustable range of multiple target process indicators in the target process indicator set is determined to obtain a parameter optimization space; according to the parameter optimization space, a central composite design of multiple target process indicators is performed, and the simulation analysis model is called to perform simulation to obtain a sample scheme matrix, wherein the sample scheme matrix includes a set of process indicator parameters and a corresponding set of quality indicator parameters; based on the sample scheme matrix, a target surrogate model is constructed and trained, wherein... The target proxy model includes multiple sub-target proxy models, and each sub-target proxy model corresponds one-to-one with multiple target quality indicators in the target quality indicator set. Using the target proxy model as the optimization objective, a multi-objective optimization algorithm is applied to optimize the target process indicator set of the target electromechanical equipment, thereby obtaining an optimized process solution. This solves the technical problems that often exist in the performance optimization process of electromechanical equipment, such as poor adaptability to complex scenarios, low accuracy and efficiency of optimization, and difficulty in ensuring the intelligence and stability of equipment operation. This makes the electromechanical equipment more adaptable to complex scenarios, improves the accuracy and efficiency of electromechanical equipment performance optimization, and thus improves production efficiency and product quality. Attached Figure Description

[0018] Figure 1 This application provides a schematic flowchart of a performance optimization method for electromechanical equipment.

[0019] Figure 2 This application provides a schematic diagram of a performance optimization system for electromechanical equipment.

[0020] Figure labeling: 11. Index acquisition module; 12. Simulation model construction module; 13. Simulation module; 14. Proxy model training module; 15. Process optimization module. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0022] Example 1 like Figure 1 As shown, this application provides a performance optimization method for electromechanical equipment, the method being executed by a performance optimization system for electromechanical equipment, the method comprising: Step S100: Interact with the target electromechanical equipment scene to obtain the target process index set and the target quality index set.

[0023] Electromechanical equipment is widely used in many fields, including automated production and smart healthcare. It is an indispensable part of the intelligent development of science and technology. In order to promote high-quality industrial production, the methods for optimizing the performance of electromechanical equipment need to be constantly updated and improved. Only by constantly adapting to the needs of the times for equipment can industrial efficiency be improved.

[0024] Currently, intelligent manufacturing of electromechanical equipment typically employs centralized control. In a specific working environment, the equipment interacts with the user (or operator) through a human-machine interface (HMI) or system to achieve specific goals or tasks. The user sends instructions or requests to the equipment via the HMI (such as a touchscreen or control panel). Upon receiving the instructions, the equipment executes corresponding operations according to preset programs or algorithms, such as adjusting equipment parameters, performing production tasks, and returning data. The equipment then feeds back the execution results or relevant information to the user or system through the HMI, allowing them to understand the equipment's status and task completion. Therefore, by interacting with the target electromechanical equipment scenario, a set of target process indicators and a set of target quality indicators can be obtained.

[0025] The target process indicator set focuses on the production process and process parameters of electromechanical equipment to ensure that products are manufactured according to predetermined process requirements. For example, process indicators may include equipment utilization rate and capacity efficiency. The target quality indicator set focuses on the quality characteristics and performance parameters of the product to ensure that the product meets target needs and expectations, such as control accuracy and compliance rate. By exploring interactive usage scenarios of the electromechanical equipment, sets containing both target process indicators and target quality indicators are obtained, forming the target process indicator set and target quality indicator set, laying the foundation for subsequent equipment optimization.

[0026] Step S200: Based on the simulation method, establish a simulation analysis model of the target electromechanical equipment production target, determine the adjustable range of multiple target process indicators in the target process indicator set, and obtain the parameter optimization space.

[0027] Furthermore, by utilizing simulation analysis technology, virtual models of electromechanical equipment can be established, and various parameters can be adjusted and tested in a virtual environment, thereby quickly evaluating the impact of different parameter schemes on equipment performance. Specifically, the type, function, production process, and key technological steps of the target electromechanical equipment should be clearly defined, the production process should be analyzed, and key technological indicators affecting production targets should be identified. Simulation software (such as Simulink, Arena, AnyLogic, etc.) or custom programming (such as Python, MATLAB, etc.) should be used to build a simulation model. This simulation model should be able to simulate the production process of the electromechanical equipment, including material flow, equipment operation, and changes in process parameters, and should be able to accept process parameters as input and output corresponding process indicator values.

[0028] For each target process indicator in the target process indicator set, analyze its adjustable range under the current production conditions. This adjustable range may be affected by various factors such as equipment performance, raw material properties, and operational limitations. Specifically, the specific value of the adjustable range can be determined by combining experimental data, expert experience, or historical simulations for evaluation.

[0029] Then, different combinations of process parameters are set in the simulation model to simulate the production process. The results of each simulation run are recorded, including the values ​​of each target process index. The simulation is repeated to cover different parameter combinations within the adjustable range. Furthermore, the collected data is analyzed to evaluate the impact of different parameter combinations on the target process index. Statistical analysis, optimization algorithms (such as genetic algorithms, particle swarm optimization, etc.), or machine learning techniques are used to identify the optimal parameter combination and determine the parameter optimization space, that is, the optimal or suboptimal parameter range that can achieve the production target.

[0030] By establishing a simulation analysis model of the target electromechanical equipment production target and obtaining the parameter optimization space, an adjustment constraint can be provided for parameter control to ensure that the equipment parameters can be optimized within the parameter range, thereby reducing computing power and improving optimization efficiency and accuracy.

[0031] Step S300: Perform a central composite design of multiple target process indicators based on the parameter optimization space, call the simulation analysis model to perform simulation, and obtain a sample scheme matrix, wherein the sample scheme matrix includes a set of process indicator parameters and a corresponding set of quality indicator parameters.

[0032] Specifically, after determining the parameter optimization space, a central composite design method is used for simulation design. Central composite design is a commonly used response surface methodology that can effectively find the optimal parameter combination in multi-factor experiments. Based on the number and range of process parameters, the central composite design method is used to generate experimental points. These experimental points include center points (representing the average value of the parameters), axial points (representing the extreme values ​​of the parameter range), and any additional points near the center points (used to estimate curvature). These points cover different regions of the parameter space to comprehensively evaluate the impact of parameters on process indicators. Furthermore, for each experimental point (i.e., a set of process parameter combinations) in the central composite design, a simulation analysis model is invoked to perform simulation, and the output process indicator values ​​are recorded. These simulation results constitute a sample scheme matrix. Each row of the sample scheme matrix represents an experimental point, including the process parameter combination and the corresponding process indicator value.

[0033] By centrally designing multi-objective process indicators and corresponding parameters, and conducting simulation and result analysis based on a simulation analysis model, the optimal combination of process parameters is identified, thus achieving the optimization of multi-objective process parameters. Moreover, this optimization process takes into account the trade-offs between various objectives, ensuring higher production efficiency and lower costs.

[0034] Step S400: Based on the sample scheme matrix, construct and train the target agent model, wherein the target agent model includes multiple sub-target agent models, and the sub-target agent models correspond one-to-one with multiple target quality indicators in the target quality indicator set.

[0035] For example, process parameters and corresponding target quality indicators are extracted from the sample scheme matrix. The extracted data undergoes necessary preprocessing, such as data cleaning and normalization, to ensure data quality and consistency. Then, based on the sample scheme matrix, a target surrogate model is constructed and trained, which includes multiple sub-target surrogate models. During training, a suitable sub-target surrogate model is selected for each target quality indicator based on its characteristics (e.g., continuity, discreteness) and data scale. For example, for continuous target quality indicators, Support Vector Machine (SVM), Anonymous Neural Network (ANN), or Multinomial Regression models can be selected; for discrete target quality indicators, decision trees, Naive Bayes, etc., can be selected. The preprocessed data is then used to train and validate the sub-models. Finally, a suitable ensemble strategy is used to integrate the sub-models, such as weighted ensemble. Next, the integrated overall target surrogate model is trained again, and the trained model is used as the final target surrogate model.

[0036] The target surrogate model is a model used to simulate and predict the behavior of multiple target variables (i.e., target quality indicators) in complex systems or processes. In optimization and decision-making processes, directly manipulating or measuring target variables may be time-consuming or infeasible; therefore, surrogate models are used to approximate the behavior of target variables for optimization. The target surrogate model can simultaneously consider and handle multiple conflicting or interdependent target quality indicators and predict the value of the output target quality indicator based on input process parameters, significantly reducing computation time and cost. Furthermore, the target surrogate model also includes multiple sub-target surrogate models; each sub-target surrogate model is a component of the target surrogate model and is specifically designed to simulate and predict the behavior of a single target quality indicator. When constructing the target surrogate model, a corresponding sub-target surrogate model is typically built for each target quality indicator. Each sub-target surrogate model focuses on only one specific target quality indicator and predicts its value based on input parameters. Different sub-target surrogate models can employ different algorithms and model structures to adapt to the characteristics of different target quality indicators. Multiple sub-target surrogate models can be combined into a complete target surrogate model to handle multiple target quality indicators simultaneously. This approach offers greater adaptability and scalability during optimization, improving efficiency and making the optimization process more focused and flexible.

[0037] Step S500: Using the target proxy model as the optimization objective, apply a multi-objective optimization algorithm to optimize the target process index set of the target electromechanical equipment and obtain an optimized process scheme.

[0038] Optionally, the process indicators in the target process indicator set of the target electromechanical equipment are clearly defined, and a multi-objective optimization objective function is set based on each indicator. A suitable multi-objective optimization algorithm is selected, such as the Non-Dominated Sorting Genetic Algorithm (NSGA-II) or the Multi-Objective Particle Swarm Optimization Algorithm (MOPSO). These algorithms can handle multiple conflicting or interdependent optimization objectives and find a set of solutions that balance the performance of each objective. Based on the actual situation of the target electromechanical equipment, constraints on the optimization problem are defined. These constraints may include the range of process parameter values, equipment performance limitations, product quality standards, etc. That is, within the defined parameter optimization space, a set of process parameter combinations that satisfy the constraints is randomly generated as an initial population using a multi-objective optimization algorithm. Then, a target surrogate model is used to evaluate each individual in the initial population, calculating its corresponding target process indicator value. Following the strategy of the multi-objective optimization algorithm, a new population is continuously generated iteratively, and individuals in the new population are continuously evaluated. Iteration stops when there is no significant change after several consecutive iterations. One or more satisfactory solutions are selected from the final iterative results as optimized process schemes. These schemes will, while satisfying all constraints, balance the performance of each target process indicator as much as possible. Finally, the selected optimized process scheme is applied to actual electromechanical equipment for verification testing. Based on the test results, the process parameters are fine-tuned to obtain better performance. If the verification results meet expectations, the optimized process scheme is officially implemented.

[0039] Furthermore, in the interactive target electromechanical equipment scenario, to obtain the target process index set and the target quality index set, step S100 of this application also includes: Step S110: Interact with the target electromechanical equipment scene and obtain historical processing records.

[0040] Step S120: Based on the target electromechanical equipment optimization target, perform a compliance analysis of the historical processing records and calculate the compliance index.

[0041] Step S130: Extract the target quality index set based on the compliance index.

[0042] Step S140: Using the target quality index set as the target variable and multiple process indicators in the historical processing records as the original variables, perform principal component regression analysis to obtain the target process index set.

[0043] Specifically, the system interacts with the target electromechanical equipment in the corresponding scenario through API interfaces, database queries, or other data exchange methods to extract the historical processing records of the target electromechanical equipment. These records include various process parameters, equipment status, raw material information, and processing results. Furthermore, the optimization objectives of the target electromechanical equipment are defined. These objectives typically refer to indicators related to product quality, production efficiency, and energy consumption. A compliance analysis is performed on each processing task in the historical processing records, comparing the actual processing results with the optimization objectives. Based on the analysis results, one or more compliance indices are calculated, which can be simple pass rates or completion rates, or complex comprehensive evaluation indicators. Next, the relationship between the compliance indices and different quality indicators is analyzed to identify quality indicators that have a significant impact on the compliance indices. These quality indicators are extracted to form a target quality indicator set, which contains indicators that reflect the key quality characteristics of the product.

[0044] Furthermore, the target quality indicator set is used as the target variable (dependent variable), and multiple process indicators from historical processing records are used as original variables (independent variables) for principal component regression analysis. Principal component regression analysis (PCR) is a statistical method that combines principal component analysis (PCA) and multiple linear regression (MLR) to process data with multiple correlated variables. Therefore, in this process, PCA is first used to reduce the dimensionality of the original variables and extract the main principal components. These principal components represent most of the information in the original variables and are uncorrelated with each other. Then, these principal components are used as new independent variables to perform multiple linear regression analysis to predict the target quality indicator set. Finally, based on the results of the regression analysis, the process indicators that have a significant impact on the target quality indicators are identified, forming the target process indicator set. These process indicators are key parameters for optimizing the processing of the target electromechanical equipment.

[0045] By performing principal component regression analysis on historical data, the processing of target electromechanical equipment can be systematically analyzed and optimized, thereby improving product quality and production efficiency.

[0046] Furthermore, based on simulation methods, a simulation analysis model of the target electromechanical equipment production target is established. Step S200 of this application also includes: Step S210: Obtain the performance index set of the target electromechanical equipment, construct the finite element analysis model of the target electromechanical equipment, and obtain the equipment simulation analysis model.

[0047] Step S220: Based on the production target information of the target electromechanical equipment, construct a finite element analysis model of the production target of the target electromechanical equipment, and obtain the production target simulation analysis model.

[0048] Step S230: Fit the equipment simulation analysis model with the production target simulation analysis model to obtain the simulation analysis model.

[0049] Optionally, a simulation analysis model of the target electromechanical equipment can be constructed for virtual simulation to quickly assess the impact of different parameter schemes on equipment performance. Specifically, firstly, performance index data such as stress distribution, deformation, vibration frequency, and temperature distribution of the target electromechanical equipment under various test conditions, as well as data on the equipment's geometry, material properties, and boundary conditions, are obtained from the target electromechanical equipment performance index set. Further, the objectives of the simulation analysis model are determined, such as evaluating equipment performance and optimizing parameter design. Based on the type of equipment and simulation requirements, appropriate finite element analysis (FEA) simulation software is selected.

[0050] Finite element analysis (FEA) is a numerical method for solving complex physical problems. When constructing an FEA model, the target electromechanical equipment needs to be divided into multiple small, simple, interconnected elements (i.e., finite elements). Specialized FEA software (such as ANSYS, Abaqus, SolidWorks Simulation, etc.) is used to build the FEA model based on the equipment's geometry, material properties, boundary conditions, etc.

[0051] Specifically, a three-dimensional geometric model of the equipment is created in the simulation software based on its geometry and dimensions. Corresponding material properties, such as elastic modulus, Poisson's ratio, and density, are defined for different parts of the equipment. Furthermore, loads (such as force, torque, and temperature) and constraints (such as fixed supports and sliding supports) are defined in the model according to the actual working environment of the equipment to simulate its behavior under actual working conditions. Next, the geometric model of the equipment is meshed to generate a finite element mesh for simulation analysis. The mesh density and shape should be adjusted according to the complexity of the equipment and the simulation requirements. Then, solution parameters (such as solver type, solution accuracy, and number of iterations) are set in the FEA model, and the solver is run to perform simulation analysis. The simulation analysis results are obtained, and the performance indicators such as stress distribution, deformation, vibration frequency, and temperature distribution of the equipment are viewed. Based on the simulation analysis results, the simulation model is optimized, such as adjusting the equipment structure, material properties, or boundary conditions. The optimized simulation model is then rerun, and the iterative optimization process continues until the performance requirements are met, resulting in the final equipment simulation analysis model.

[0052] Similarly, a finite element analysis (FEA) model is constructed to represent the production objectives of the target electromechanical equipment. These objectives may involve aspects such as production efficiency, product quality, and production cost. Performance indicators related to the production objectives are defined; for example, production efficiency may be related to equipment uptime, downtime, and failure rate; product quality may be related to equipment processing accuracy, stability, and reliability. Based on these performance indicators, a FEA model related to the production objectives is constructed. The FEA model must include correct material properties (such as Young's modulus, Poisson's ratio, hardening index, and yield strength), boundary conditions (such as fixed constraints and loads), and any special settings directly related to the production objectives (such as stamping / bending speed and pressure). Appropriate loads and constraints are set, and the FEA model related to the production objectives is run for simulation analysis.

[0053] After the simulation is completed, key result data, such as maximum stress, maximum deformation, and springback, are exported or extracted from the FEA software. The extracted simulation results are then correlated with preset production target performance indicators (such as product qualification rate, production efficiency, and failure rate). Based on the simulation analysis results, the original FEA model is iterated and optimized to obtain a production target simulation analysis model.

[0054] Finally, the performance indicators in the equipment simulation analysis model and the production target simulation analysis model are compared, differences are analyzed, equipment design or production parameters are adjusted, and fitting methods such as optimization algorithms and statistical analysis are used to fit the two models. The purpose of fitting is to combine the equipment simulation analysis model with the production target simulation analysis model to evaluate whether the equipment can meet the requirements of the production target. Ultimately, a simulation analysis model that can evaluate whether the target electromechanical equipment meets the production target requirements is obtained, which can be used to guide equipment design, production, optimization, and other work.

[0055] Furthermore, to determine the adjustable range of multiple target process indicators in the target process indicator set and obtain the parameter optimization space, step S200 of this application also includes: Step S240: Extract the first target process index based on the target process index set.

[0056] Step S250: Analyze the historical processing records to obtain the first historical process index range of the first target process index.

[0057] Step S260: Define an index expansion step size based on the adjustment step size of the first target process index, and configure a first expanded process index sequence based on the first historical process index range.

[0058] Step S270: Perform simulation of the first expanded process index sequence to obtain the first expanded quality index sequence.

[0059] Step S280: Fit and analyze the first extended process index sequence and the first extended quality index sequence to generate a first response curve, and combine the target electromechanical equipment scenario quality constraints to determine the range of the first target process index.

[0060] Step S290: Traverse the target process index set to obtain multiple target process index ranges and construct the parameter optimization space.

[0061] For example, when acquiring the parameter optimization space, a specific process indicator is first selected or extracted from the set of target process indicators as the first target process indicator. Next, historical process parameter ranges related to the first target process indicator are identified by analyzing historical processing records. This range may include process parameter data from different batches, different times, or different equipment conditions. An indicator expansion step size is defined based on the adjustment step size of the first target process indicator; this adjustment step size is typically determined based on the experience of process engineers or experimental data. Then, based on the first historical process indicator range, an expansion step size is used to configure a first expanded process indicator sequence containing multiple different process parameter values. A simulation model is then used to simulate each process parameter value in the first expanded process indicator sequence. During the simulation, it should be ensured that the simulation conditions are as consistent as possible with the actual production conditions to obtain accurate results. Through simulation, a first expanded quality indicator sequence corresponding to the first expanded process indicator sequence is obtained. Then, data fitting techniques are used to analyze the relationship between the first expanded process indicator sequence and the first expanded quality indicator sequence, generating a first response curve. Finally, based on the quality constraints of the target electromechanical equipment scenario (such as product quality requirements, production efficiency requirements, etc.), a reasonable range of first target process indicators is determined on the first response curve. This range should ensure that key indicators such as product quality and production efficiency can meet the requirements during actual production. Repeating the above steps, each process indicator in the target process indicator set is traversed to obtain its respective target process indicator range. These target process indicator ranges are then combined to form a multi-dimensional parameter optimization space. This space contains all possible combinations of process parameters and their corresponding key indicators such as product quality and production efficiency.

[0062] By constructing a parameter optimization space based on historical processing records and simulations, an optimization foundation is built for finding the best combination of processing parameters, thereby improving the efficiency of parameter optimization and the possibility of parameter combinations, and ultimately improving production efficiency and product quality.

[0063] Furthermore, based on the parameter optimization space, a central composite design of multiple target process indicators is performed, and the simulation analysis model is invoked to perform simulation to obtain a sample scheme matrix. Step S300 of this application also includes: Step S310: Based on the target quality index set and combined with the target electromechanical equipment scenario quality constraints, construct a quality evaluation function.

[0064] Step S320: Using multiple target process indicators as design variables and the quality evaluation function as the target response, establish a central composite design and obtain sample factor combinations.

[0065] Step S330: Input the sample factor combination into the simulation analysis model to perform simulation, obtain the simulation result sequence, and perform numerical calculation of the simulation result sequence according to the target response to obtain the sample scheme matrix.

[0066] Furthermore, based on the target quality index set, and combined with the quality constraints of the target electromechanical equipment scenario (such as working environment, safety standards, etc.), weights are assigned to each quality index. Then, a comprehensive quality evaluation function is constructed based on the assigned weights and the target quality indexes. This function reflects the balance between different indicators and assigns weighted scores to each indicator according to actual needs. Next, key target process indicators are determined, which will be adjusted as design variables in the experiment. A central composite design is used to generate a series of sample factor combinations, and these combinations are input into the established simulation analysis model for execution, obtaining a simulation result sequence corresponding to each sample factor combination. Then, the previously constructed quality evaluation function is used to numerically calculate these simulation result sequences, yielding a comprehensive quality score for each sample. The comprehensive quality scores of all samples are organized into a matrix, i.e., the sample scheme matrix, which will be used for subsequent data analysis and process optimization.

[0067] The sample scheme matrix is ​​a table containing multiple sample points and their corresponding quality scores. Each row represents a sample point, which includes the process indicator settings and corresponding quality score for that point. By analyzing this matrix, it is possible to identify which combinations of process indicators produce the optimal product quality, thereby guiding the actual production process.

[0068] Central composite design can effectively evaluate the main effects and interaction effects between variables. By combining statistical experimental design, simulation and numerical analysis methods, the optimal combination of process parameters can be found efficiently to improve product quality and performance.

[0069] Furthermore, based on the sample scheme matrix, the target agent model is constructed and trained. Step S400 of this application further includes: Step S410: Based on the target electromechanical equipment scenario, determine the target kernel function of the proxy model.

[0070] Step S420: Based on the target kernel function and the sample scheme matrix, calculate and obtain the hyperparameters of the target kernel function to obtain the surrogate model.

[0071] Step S430: Using the historical processing records, configure evaluation indicators to evaluate the performance of the proxy model. If the performance evaluation result meets the preset performance constraints, then output the proxy model as the target proxy model.

[0072] Specifically, the kernel function is a key component in the surrogate model, determining the similarity measure between data points. A suitable kernel function is selected or designed based on the characteristics of the target electromechanical equipment scenario (such as data distribution, noise level, complexity, etc.). Common kernel functions include linear kernels, polynomial kernels, and radial basis function (RBF) kernels. Hyperparameters are parameters that need to be set before training the surrogate model, and they significantly impact model performance. A sample scheme matrix (i.e., a table containing different combinations of process indicators and their corresponding quality scores) is used as training data. Optimization algorithms (such as grid search, random search, Bayesian optimization, etc.) are used to search the hyperparameter space to find the optimal hyperparameter combination for the surrogate model's performance. Furthermore, historical processing records are used as test data to evaluate the surrogate model's performance. Evaluation metrics that comprehensively reflect the surrogate model's prediction accuracy, robustness, and generalization ability are configured, such as mean squared error, coefficient of determination, and cross-validation score. The test data is then input into the surrogate model to obtain prediction results, which are then evaluated using the evaluation metrics. Finally, based on preset performance constraints (such as prediction error less than a certain threshold, determination coefficient greater than a certain value, etc.), it is determined whether the performance of the surrogate model meets the requirements. If the performance constraints are met, the surrogate model is output as the target surrogate model; otherwise, the kernel function, hyperparameters, or data collection needs to be adjusted, and the above steps are repeated until a surrogate model that meets the performance constraints is obtained. The preset performance constraints can be set by the user based on specific equipment conditions and requirements. Once a surrogate model that meets the performance constraints is obtained, it can be output as the target surrogate model and used in subsequent practical applications. The target surrogate model can quickly predict product quality scores based on new combinations of process indicators, thereby guiding the actual production process and improving production efficiency and product quality.

[0073] Furthermore, taking the target proxy model as the optimization objective, a multi-objective optimization algorithm is applied to optimize the target process index set of the target electromechanical equipment to obtain an optimized process scheme. Then, step S500 of this application further includes: Step S510: Based on the simulation analysis model, perform verification analysis on the optimized process scheme and obtain verification analysis results, wherein the verification analysis results include multiple verification quality index values.

[0074] Step S520: Based on the target electromechanical equipment scenario quality constraints, perform binary verification on multiple verification quality index values. If all multiple verification quality index values ​​meet the target electromechanical equipment scenario quality constraints, then apply the optimized process scheme to optimize the performance of the target electromechanical equipment.

[0075] Optionally, using the previously established simulation analysis model, the parameter settings of the optimized process scheme are input into the model and the simulation analysis model is run to simulate the operation of the optimized process scheme in the actual production environment. Multiple verification quality index values ​​are obtained through simulation; these index values ​​reflect the impact of the optimized process scheme on the performance of the target electromechanical equipment. During the simulation process, all relevant verification quality index values ​​are recorded and collected. These index values ​​comprehensively reflect the performance of the target electromechanical equipment, ensuring that the collected verification quality index values ​​are accurate, reliable, and comparable for subsequent binary verification. Next, binary verification is performed on the multiple verification quality index values ​​according to the quality constraints of the target electromechanical equipment scenario. Binary verification is a simple verification method that compares each verification quality index value with its corresponding quality constraint to determine whether it meets the requirements. Therefore, for each verification quality index value, it is checked whether it meets the quality constraints of the target electromechanical equipment scenario. If all verification quality index values ​​meet the quality constraints, the verification passes; otherwise, the verification fails. If the verification is successful, meaning that multiple verification quality indicators meet the quality constraints of the target electromechanical equipment scenario, then the optimized process scheme can be considered effective. In this case, the optimized process scheme can be applied to the actual target electromechanical equipment for performance optimization, including adjusting equipment process parameters, improving equipment structure, or optimizing production processes. After applying the optimized process scheme, it is necessary to continue monitoring the performance of the target electromechanical equipment to ensure that the optimization effect meets expectations, and to make necessary adjustments and improvements based on the actual situation. Through the above steps, the effectiveness of the optimized process scheme is ensured, actual production risks are reduced, the performance of the target electromechanical equipment is improved, and the controllability of the production process is enhanced.

[0076] Through the technical solutions of the above embodiments, the performance optimization method for electromechanical equipment provided in this application solves the technical problems that often exist in the performance optimization process of electromechanical equipment, such as poor adaptability to complex scenarios, low accuracy and efficiency of optimization, and difficulty in ensuring the intelligence and stability of equipment operation. By interacting with the target electromechanical equipment scenario, the target process index set and the target quality index set are clarified, which gives the performance optimization work a clear goal and direction. Through the simulation analysis model established based on the simulation method, the production target of the target electromechanical equipment can be accurately simulated, and the adjustable range of the target process index can be determined, improving the optimization efficiency and accuracy. By performing a central composite design of multiple target process indices in the parameter optimization space, the parameter space can be explored comprehensively and efficiently. The sample scheme matrix obtained through simulation covers a variety of possible process index parameter sets and their corresponding quality index parameter sets, ensuring comprehensive and efficient optimization of the target optimization parameter combination, and improving the adaptability, optimizability, and stability of parameter optimization. By constructing and training a target surrogate model based on a sample scheme matrix, including multiple sub-target surrogate models, and assigning each sub-model a target quality index, the model can accurately predict the quality index parameter sets corresponding to different process parameter sets. This provides a reliable basis for optimization and improves the accuracy of the optimization results. Using the target surrogate model as the optimization objective and applying a multi-objective optimization algorithm, multiple target quality indices can be considered simultaneously to find the optimal set of process index parameters, i.e., the optimized process scheme. This optimization method can significantly improve the performance of electromechanical equipment, meet multiple requirements, and make the equipment more adaptable to complex scenarios. In summary, the optimized process scheme obtained through the above method helps improve production efficiency and product quality, making the electromechanical equipment more adaptable to complex scenarios.

[0077] Example 2 Based on the same inventive concept as the performance optimization method for electromechanical equipment in the foregoing embodiments, such as Figure 2 As shown, this application provides a performance optimization system for electromechanical equipment, the system comprising: The indicator acquisition module 11 is used to interact with the target electromechanical equipment scenario and acquire the target process indicator set and the target quality indicator set.

[0078] The simulation model construction module 12 is used to establish a simulation analysis model of the target electromechanical equipment production target based on the simulation method, and to determine the adjustable range of multiple target process indicators in the target process indicator set, and obtain the parameter optimization space.

[0079] The simulation module 13 is used to perform a central composite design of multiple target process indicators based on the parameter optimization space, call the simulation analysis model to perform simulation, and obtain a sample scheme matrix, wherein the sample scheme matrix includes a set of process indicator parameters and a corresponding set of quality indicator parameters.

[0080] The surrogate model training module 14 is used to construct and train a target surrogate model based on the sample scheme matrix. The target surrogate model includes multiple sub-target surrogate models, and the sub-target surrogate models correspond one-to-one with multiple target quality indicators in the target quality indicator set.

[0081] The process optimization module 15 is used to optimize the target process index set of the target electromechanical equipment by applying a multi-objective optimization algorithm with the target proxy model as the optimization target, and to obtain an optimized process scheme.

[0082] Furthermore, the indicator acquisition module 11 is also used to perform the following steps: Interact with the target electromechanical equipment scene and obtain historical processing records.

[0083] Based on the target electromechanical equipment optimization objective, a compliance analysis of the historical processing records is performed to calculate and obtain the compliance index.

[0084] Based on the compliance index, extract the target quality index set.

[0085] Using the target quality index set as the target variable and multiple process indicators in the historical processing records as the original variables, principal component regression analysis is performed to obtain the target process index set.

[0086] Furthermore, the simulation model construction module 12 is also used to perform the following steps: Obtain the performance index set of the target electromechanical equipment, construct the finite element analysis model of the target electromechanical equipment, and obtain the equipment simulation analysis model.

[0087] Based on the production target information of the target electromechanical equipment, a finite element analysis model of the production target of the target electromechanical equipment is constructed, and a simulation analysis model of the production target is obtained.

[0088] The simulation analysis model of the equipment is fitted to the simulation analysis model of the production target to obtain the simulation analysis model.

[0089] Furthermore, the simulation model construction module 12 is also used to perform the following steps: Based on the target process index set, the first target process index is extracted.

[0090] The historical processing records are analyzed to obtain the first historical process index range of the first target process index.

[0091] Based on the adjustment step size of the first target process indicator, define the indicator expansion step size, and configure the first expanded process indicator sequence based on the first historical process indicator range.

[0092] Simulation of the first extended process index sequence is performed to obtain the first extended quality index sequence.

[0093] The first extended process index sequence and the first extended quality index sequence are fitted and analyzed to generate a first response curve. Combined with the quality constraints of the target electromechanical equipment scenario, the range of the first target process index is determined.

[0094] The target process index set is traversed to obtain multiple target process index ranges, and the parameter optimization space is constructed.

[0095] Furthermore, the simulation module 13 is also used to perform the following steps: Based on the target quality index set and combined with the target electromechanical equipment scenario quality constraints, a quality evaluation function is constructed.

[0096] Using multiple target process indicators as design variables and the quality evaluation function as the target response, a central composite design is established to obtain sample factor combinations.

[0097] The sample factors are input into the simulation analysis model for simulation, the simulation result sequence is obtained, and the numerical calculation of the simulation result sequence is performed based on the target response to obtain the sample scheme matrix.

[0098] Furthermore, the surrogate model training module 14 is also used to perform the following steps: Based on the target electromechanical equipment scenario, the target kernel function of the proxy model is determined.

[0099] Based on the target kernel function and the sample scheme matrix, the hyperparameters of the target kernel function are calculated and obtained to obtain the surrogate model.

[0100] The proxy model is evaluated using the historical processing records and configured evaluation indicators. If the performance evaluation results meet the preset performance constraints, the proxy model is output as the target proxy model.

[0101] Furthermore, the process optimization module 15 is also used to perform the following steps: Based on the simulation analysis model, the optimized process scheme is verified and analyzed to obtain the verification analysis results, which include multiple verification quality index values.

[0102] Based on the target electromechanical equipment scenario quality constraints, a binary verification is performed on multiple verification quality index values. If all of the multiple verification quality index values ​​meet the target electromechanical equipment scenario quality constraints, then the optimized process scheme is applied to optimize the performance of the target electromechanical equipment.

[0103] Through the foregoing detailed description of the performance optimization method for electromechanical equipment, those skilled in the art can clearly understand the performance optimization system for electromechanical equipment in this embodiment. As for the apparatus disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and relevant parts can be referred to the method section description.

[0104] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for performance optimization of electromechanical equipment, characterized in that, The method includes: Interact with the target electromechanical equipment scenario to obtain the target process index set and the target quality index set; Based on simulation methods, a simulation analysis model of the target electromechanical equipment production target is established, and the adjustable range of multiple target process indicators in the target process indicator set is determined to obtain the parameter optimization space. Based on the parameter optimization space, a central composite design of multiple target process indicators is performed, and the simulation analysis model is called to perform simulation to obtain a sample scheme matrix, wherein the sample scheme matrix includes a set of process indicator parameters and a corresponding set of quality indicator parameters. Based on the sample scheme matrix, a target agent model is constructed and trained, wherein the target agent model includes multiple sub-target agent models, and the sub-target agent models correspond one-to-one with multiple target quality indicators in the target quality indicator set; Using the target proxy model as the optimization objective, a multi-objective optimization algorithm is applied to optimize the target process index set of the target electromechanical equipment and obtain an optimized process scheme.

2. The method as described in claim 1, characterized in that, Interact with the target electromechanical equipment scenario to obtain the target process indicator set and the target quality indicator set, including: Interact with the target electromechanical equipment scene and obtain historical processing records; Based on the target electromechanical equipment optimization objectives, a compliance analysis of the historical processing records is performed to calculate and obtain the compliance index; Based on the compliance index, extract the target quality indicator set; Using the target quality index set as the target variable and multiple process indicators in the historical processing records as the original variables, principal component regression analysis is performed to obtain the target process index set.

3. The method as described in claim 2, characterized in that, Based on simulation methods, a simulation analysis model for the production target of the target electromechanical equipment is established, including: Obtain the performance index set of the target electromechanical equipment, construct the finite element analysis model of the target electromechanical equipment, and obtain the equipment simulation analysis model; Based on the production target information of the target electromechanical equipment, a finite element analysis model of the production target of the target electromechanical equipment is constructed, and a simulation analysis model of the production target is obtained. The simulation analysis model of the equipment is fitted to the simulation analysis model of the production target to obtain the simulation analysis model.

4. The method as described in claim 3, characterized in that, Determine the adjustable range of multiple target process indicators in the target process indicator set to obtain the parameter optimization space, including: Based on the target process index set, extract the first target process index; Analyze the historical processing records to obtain the first historical process index range of the first target process index; Based on the adjustment step size of the first target process indicator, define the indicator expansion step size, and configure the first expanded process indicator sequence based on the first historical process indicator range. Simulation of the first extended process index sequence is performed to obtain the first extended quality index sequence; Fit analysis is performed on the first extended process index sequence and the first extended quality index sequence to generate a first response curve, and the range of the first target process index is determined by combining the target electromechanical equipment scenario quality constraints. The target process index set is traversed to obtain multiple target process index ranges, and the parameter optimization space is constructed.

5. The method as described in claim 4, characterized in that, Based on the parameter optimization space, a central composite design of multiple target process indicators is performed. The simulation analysis model is then invoked to conduct simulation, and a sample scheme matrix is ​​obtained, including: Based on the target quality index set and combined with the target electromechanical equipment scenario quality constraints, a quality evaluation function is constructed. Using multiple target process indicators as design variables and the quality evaluation function as the target response, a central composite design is established to obtain sample factor combinations. The sample factors are input into the simulation analysis model for simulation, the simulation result sequence is obtained, and the numerical calculation of the simulation result sequence is performed based on the target response to obtain the sample scheme matrix.

6. The method as described in claim 5, characterized in that, Based on the sample scheme matrix, a target agent model is constructed and trained, including: Based on the target electromechanical equipment scenario, determine the target kernel function of the proxy model; Based on the target kernel function and the sample scheme matrix, the hyperparameters of the target kernel function are calculated and obtained to obtain the surrogate model; The proxy model is evaluated using the historical processing records and configured evaluation indicators. If the performance evaluation results meet the preset performance constraints, the proxy model is output as the target proxy model.

7. The method as described in claim 1, characterized in that, Using the target proxy model as the optimization objective, a multi-objective optimization algorithm is applied to optimize the target process index set of the target electromechanical equipment to obtain an optimized process scheme. This process further includes: Based on the simulation analysis model, the optimized process scheme is verified and analyzed to obtain the verification analysis results, wherein the verification analysis results include multiple verification quality index values. Based on the target electromechanical equipment scenario quality constraints, a binary verification is performed on multiple verification quality index values. If all of the multiple verification quality index values ​​meet the target electromechanical equipment scenario quality constraints, then the optimized process scheme is applied to optimize the performance of the target electromechanical equipment.

8. A performance optimization system for electromechanical equipment, characterized in that, The system is used to implement the performance optimization method for electromechanical equipment according to any one of claims 1-7, the system comprising: The indicator acquisition module is used to interact with the target electromechanical equipment scenario and acquire the target process indicator set and the target quality indicator set. The simulation model construction module is used to establish a simulation analysis model of the target electromechanical equipment production target based on the simulation method, and to determine the adjustable range of multiple target process indicators in the target process indicator set, and obtain the parameter optimization space. The simulation module is used to perform a central composite design of multiple target process indicators based on the parameter optimization space, call the simulation analysis model to perform simulation, and obtain a sample scheme matrix, wherein the sample scheme matrix includes a set of process indicator parameters and a corresponding set of quality indicator parameters. The proxy model training module is used to construct and train a target proxy model based on the sample scheme matrix. The target proxy model includes multiple sub-target proxy models, and the sub-target proxy models correspond one-to-one with multiple target quality indicators in the target quality indicator set. The process optimization module is used to optimize the target process index set of the target electromechanical equipment by applying a multi-objective optimization algorithm with the target proxy model as the optimization target, and to obtain an optimized process scheme.

Images