Axial plunger pump batch production process optimization system based on digital twinning

Abstract

The invention discloses an axial plunger pump batch production process optimization system based on digital twinning, and belongs to the field of digital twinning, and the system comprises a data acquisition module which is used for obtaining a plurality of sensing data sets; the model construction module is used for constructing a statistical sub-model; the twin integration module is used for constructing a digital twin model of the target axial plunger pump; the simulation execution module is used for performing analog simulation on the target axial plunger pump to generate a plurality of batch simulation results; the scheme evaluation module is used for performing optimization evaluation on the plurality of batch production schemes to generate a production batch optimization scheme; and the process optimization module is used for intelligently optimizing the batch production process of the target axial plunger pump. The technical problems that in the existing axial plunger pump batch production process, the production efficiency is low, and the product quality is unstable are solved, and the technical effects of improving the axial plunger pump batch production efficiency and ensuring the product quality stability by building the digital twin model and conducting simulation optimization are achieved.

Description

Technical Field

[0001] The present invention relates to the field of digital twins, and in particular to a digital twin-based batch production process optimization system for axial piston pumps. Background Art

[0002] The performance and quality of axial piston pumps depend largely on the machining precision and assembly quality of their internal components. However, during the mass production of axial piston pumps, factors such as changes in the production environment, equipment wear, and operational differences make it difficult to ensure that the machining and assembly of each pump component are ideal, resulting in unstable product performance and quality. Traditional production models rely on pre-set process parameters that cannot be adjusted promptly based on actual production conditions. Therefore, during mass production, the lack of real-time monitoring and dynamic optimization of the pump's physical properties and production process affects the production efficiency of axial piston pumps and makes it impossible to guarantee consistent product quality. Summary of the Invention

[0003] This application provides a digital twin-based axial piston pump batch production process optimization system, aiming to solve the technical problems of low production efficiency and unstable product quality in the existing axial piston pump batch production process due to the lack of real-time monitoring and dynamic optimization of the pump's physical properties and production process.

[0004] The digital twin-based axial piston pump batch production process optimization system disclosed in the present application includes: a data acquisition module for obtaining multiple sensor data sets, the multiple sensor data sets including a first sensor data set and a second sensor data set; a model construction module for constructing a physical sub-model based on the first sensor data set and a statistical sub-model based on the second sensor data set; a twin integration module for integrating the physical sub-model with the statistical sub-model to construct a digital twin model of the target axial piston pump; a simulation execution module for simulating the target axial piston pump by executing multiple batch production plans through the digital twin model to generate multiple batch simulation results; a plan evaluation module for optimizing and evaluating multiple batch production plans based on multiple batch simulation results to generate a production batch optimization plan; a process optimization module for executing the production batch optimization plan according to the target production batch to intelligently optimize the batch production process of the target axial piston pump.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages: By adopting the data acquisition module to obtain multiple sensor data sets, including the first sensor data set and the second sensor data set, a data foundation is laid for the subsequent construction of the digital twin model; the model construction module is used to construct a physical sub-model based on the first sensor data set, and a statistical sub-model is constructed based on the second sensor data set, so as to realize comprehensive digital modeling of the axial piston pump and its production process; the physical sub-model and the statistical sub-model are integrated through the twin integration module to construct a digital twin model of the axial piston pump, which comprehensively reflects the physical properties and production process characteristics of the axial piston pump and provides an integrated digital platform for subsequent production process optimization; the simulation execution module is used to simulate multiple batch production plans through the digital twin model, generate corresponding batch simulation results, and quickly and cost-effectively evaluate the feasibility and effectiveness of different production plans. The results provide decision support for production process optimization; using the solution evaluation module, multiple batch production solutions are optimized and evaluated according to the batch simulation results, the optimal production batch optimization solution is generated, and the optimal combination of production batches and process parameters is found; through the process optimization module, the production batch optimization solution is executed according to the target production batch, and the batch production process of axial piston pumps is intelligently optimized to achieve dynamic optimization control of the production process and ensure the quality and efficiency of batch production. This technical solution solves the technical problems of low production efficiency and unstable product quality due to the lack of real-time monitoring and dynamic optimization of the pump physical characteristics and production process in the existing batch production process of axial piston pumps. It achieves the technical effect of improving the batch production efficiency of axial piston pumps and ensuring product quality stability by building a digital twin model and performing simulation optimization.

[0006] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 A structural schematic diagram of a digital twin-based axial piston pump batch production process optimization system is provided for the embodiment of the present application; Figure 2 A flow chart of generating multiple batch simulation results in a batch production process optimization system of axial piston pumps based on digital twins is provided for an embodiment of the present application.

[0008] Explanation of the accompanying drawings: data acquisition module 11, model building module 12, twin integration module 13, simulation execution module 14, solution evaluation module 15, process optimization module 16. DETAILED DESCRIPTION

[0009] The overall idea of ​​the technical solution provided by this application is as follows: The embodiment of the present application provides a digital twin-based axial piston pump batch production process optimization system, which realizes digital and intelligent optimization of the entire production process by constructing a digital twin model of the axial piston pump.

[0010] Specifically, the data acquisition module first acquires the axial piston pump's physical characteristic data and production process data. The model building module then constructs physical and statistical sub-models, respectively. The twin integration module then integrates these sub-models into a complete digital twin model. Based on this, the simulation execution module uses the digital twin model to simulate different batch production scenarios, generating multiple batch simulation results. The scenario evaluation module optimizes and evaluates the production scenarios based on these multiple batch simulation results, generating the optimal batch optimization scenario. Subsequently, the process optimization module intelligently optimizes and dynamically controls the batch production process based on the batch optimization scenario.

[0011] In summary, this application realizes the full process intelligence of mass production of axial piston pumps from data acquisition, modeling and simulation to process optimization, effectively improves production efficiency and product quality stability, and realizes efficient and high-quality mass production of axial piston pumps.

[0012] After introducing the basic principles of the present application, the following will specifically introduce the non-limiting implementation methods of the present application in conjunction with the drawings in the specification.

[0013] like Figure 1 As shown, the embodiment of the present application provides an axial piston pump batch production process optimization system based on digital twin, which includes: The data acquisition module 11 is configured to obtain a plurality of sensor data sets, wherein the plurality of sensor data sets include a first sensor data set and a second sensor data set.

[0014] Specifically, various types of sensors, such as pressure sensors, temperature sensors, and flow sensors, are deployed at the mass production site of axial piston pumps to collect various data from the production process in real time. During data collection, the data acquisition module 11 divides the collected data into two parts: a first sensor data set and a second sensor data set, based on the data type and purpose. The first sensor data set includes data related to the physical characteristics of the axial piston pump, such as the pump's geometric dimensions and material properties; the second sensor data set primarily includes data related to the production process, such as the operating parameters of the production equipment and environmental conditions.

[0015] By classifying and managing the collected sensor data, the data acquisition module 11 provides high-quality, multi-dimensional data support for the subsequent digital twin model construction and production process optimization, laying a data foundation for the intelligent optimization of the mass production process of axial piston pumps.

[0016] The model building module 12 is configured to build a physical sub-model based on the first sensor data set and a statistical sub-model based on the second sensor data set.

[0017] Specifically, the model building module 12 first analyzes the first sensor data set acquired by the data acquisition module 11 to extract information related to the physical characteristics of the axial piston pump, such as the pump's geometric structure data and material property data. Then, based on the extracted geometric structure data and material property data, combined with pre-set boundary conditions and constraints, a physical sub-model is constructed that reflects the actual physical characteristics of the axial piston pump through methods such as finite element analysis and multi-body dynamics simulation. This model simulates the axial piston pump's behavior under actual operating conditions, such as stress, deformation, and fatigue, providing a basis for subsequent production process optimization. Simultaneously, the model building module 12 analyzes the second sensor data set acquired by the data acquisition module 11 to extract information related to the production process, such as process parameters of the production equipment. Based on this data, a statistical sub-model is constructed using data mining, machine learning, and other techniques to reflect the inherent connections between various factors in the mass production process of the axial piston pump. This model quantifies the impact of various factors on the performance of the axial piston pump during the production process and provides data-driven decision support for optimizing production process parameters.

[0018] Through the synergistic effect of the physical sub-model and the statistical sub-model, the model building module 12 can comprehensively and accurately describe the physical characteristics and production process of the axial piston pump, providing support for subsequent digital twin model integration and production process optimization.

[0019] The twin integration module 13 is used to integrate the physical sub-model with the statistical sub-model to construct a digital twin model of the target axial piston pump.

[0020] Specifically, first, the twin integration module 13 standardizes the physical sub-model and statistical sub-model constructed by the model construction module 12, unifying the data format and interface specifications to facilitate seamless integration and data interaction between models. Then, through multi-level, multi-granular model fusion technology, the physical sub-model and statistical sub-model are deeply integrated to construct a digital twin model that can fully reflect the actual characteristics of the target axial piston pump. During the integration process, attention is paid to the matching and synchronization of data and information flows between models to ensure that the simulation analysis at the physical level and the statistical modeling at the data level are closely integrated, mutually verified, and dynamically updated. Through model integration, the digital twin model reflects all characteristics of the target axial piston pump, including its geometric structure, material properties, and operating status, in real time in virtual space. It can also dynamically correct and optimize the model based on real-time production data, ensuring that it always maintains a high degree of consistency with the physical entity. Through the high-precision, dynamically updated digital twin model, a reliable and efficient digital test platform is provided for subsequent production process simulation optimization, providing tool support for the full process optimization of mass production of axial piston pumps.

[0021] The simulation execution module 14 is used to simulate the target axial piston pump by executing multiple batch production plans through the digital twin model to generate multiple batch simulation results.

[0022] Specifically, first, based on the production needs and equipment capabilities of the axial piston pump, multiple sets of alternative batch production plans are formulated, and each set of plans corresponds to a different combination of production batches, production rhythms, equipment configurations and other factors. Then, the simulation execution module 14 inputs multiple batch production plans into the digital twin model, and through parameter mapping and logical control, constructs a simulation scene in the virtual space that is highly corresponding to the physical production environment. Based on these simulation scenes, the digital twin model realistically simulates each batch production plan, and virtually reproduces the entire production and manufacturing process of the axial piston pump under different production plans. During the simulation process, the digital twin model simulates the processing, assembly, testing and other aspects of the axial piston pump product in real time, and dynamically presents key information such as equipment operating status, material flow, and product quality. At the same time, the various parameters and performance indicators in the simulation process are monitored in real time and recorded to form a complete simulation result data set. After high-fidelity simulation, the simulation execution module 14 generates batch simulation results corresponding to each batch production plan, and obtains multiple batch simulation results, which not only include predicted data on the quality and performance of axial piston pump products, but also include production efficiency, equipment utilization, material consumption and other production management-related statistical information, which comprehensively reflects the execution effect and optimization space of each batch production plan.

[0023] Through the batch simulation function of the simulation execution module 14, the feasibility and advantages and disadvantages of different batch production plans can be quickly and cost-effectively evaluated and compared in the virtual space, greatly reducing the risks and iteration costs of physical trial production, and providing a decision-making basis for the optimization of the batch production process of axial piston pumps.

[0024] The solution evaluation module 15 is used to optimize and evaluate the multiple batch production solutions according to the multiple batch simulation results to generate a production batch optimization solution.

[0025] Specifically, the solution evaluation module 15 first summarizes and analyzes the multiple batch simulation results generated by the simulation execution module 14, extracting key data and indicators relevant to production solution optimization, such as production efficiency, product qualification rate, and energy consumption. Then, based on these key data and indicators, the solution evaluation module 15 quantitatively evaluates and ranks the advantages and disadvantages of each batch production solution. Using predefined evaluation rules and weighting factors, the module comprehensively considers the impact of various indicators on production solution optimization and calculates a comprehensive score and ranking for each solution. Based on this evaluation and ranking, the module further analyzes the top-ranked solutions, summarizing their common characteristics and optimization patterns in terms of production batch size and process parameter settings. Based on this, it generates optimization strategies and improvement directions to guide the dynamic adjustment and optimization of production batch size and process parameters. The solution evaluation module 15 then applies the optimization strategy to the existing batch production solutions, revising and optimizing their key parameters. Through repeated iterations and simulation verification, the module continuously improves the performance and reliability of the optimized solutions, ultimately generating an optimized production batch solution that serves as the basis for improving the batch production process of axial piston pumps.

[0026] Through the optimization evaluation of the solution evaluation module 15, the advantages and disadvantages of different batch production solutions are objectively and efficiently evaluated, and a practical optimization solution is automatically generated based on the evaluation results, providing intelligent decision-making support for the batch production of axial piston pumps, effectively improving production efficiency and product quality, and reducing production costs and resource consumption.

[0027] The process optimization module 16 is used to execute the production batch optimization plan according to the target production batch to intelligently optimize the target axial piston pump batch production process.

[0028] Specifically, the production batch optimization plan generated by the plan evaluation module 15 is first matched with the target production batch to ensure that the optimization plan is applicable to the current production tasks and requirements. Then, the process optimization module 16 extracts key parameters and setting values ​​from the production batch optimization plan, such as production cycle time, equipment parameters, and process parameters, and distributes them to the control system and execution equipment at the production site, thereby optimizing and dynamically adjusting the axial piston pump batch production process in real time. During the optimization process, the execution effect of the production batch optimization plan is monitored and predicted in real time, and the optimization parameters are dynamically adjusted based on feedback data to ensure that the production process is always in an optimal state.

[0029] Through the intelligent optimization of process optimization module 16, digital twin technology is deeply integrated with on-site production control to achieve full-process optimization and dynamic adjustment of the mass production process of axial piston pumps, thereby improving production efficiency and product quality.

[0030] Furthermore, the embodiment of the present application also includes: Deploy multiple sensing devices based on the production process of the target axial piston pump and determine the sensing device group; Dividing the sensing device group according to sensing targets to generate a first sensing device group and a second sensing device group; Retrieving production batch data to determine multiple production batches, matching target axial piston pumps based on the multiple production batches, and determining a target production batch; According to the target production batch and the physical characteristics of the target axial piston pump, the target axial piston pump is batch-traversed using the first sensing device group to generate the first sensing data set; According to the target production batch and the production process, the target axial piston pumps are batch-traversed using the second sensing device group to generate the second sensing data set.

[0031] In one feasible implementation, various types of sensors, such as pressure sensors, temperature sensors, flow sensors, and displacement sensors, are first installed at corresponding locations along the production line, based on key processes and links in the axial piston pump production process. This forms a complete sensor device group, enabling comprehensive monitoring and data collection across the entire production process. Next, the sensors are divided into a first and a second sensor device group, based on the sensing targets. The first sensor device group monitors the physical characteristics of the axial piston pump, such as pump body geometry and material properties, acquiring relevant data through methods such as optical dimensional measurement and ultrasonic testing. The second sensor device group monitors production process parameters, such as equipment operating status and process parameter execution, acquiring relevant data through methods such as vibration monitoring and process parameter collection. Subsequently, production batch data is retrieved to identify multiple production batches. A target axial piston pump is matched based on the multiple production batches to determine the target production batch. The production batch data records the total production quantity and batch production schedule for the axial piston pump, and the target axial piston pump is located within a specific production batch. By analyzing production batch data, the target production batch of the target axial piston pump is determined, providing a batch reference for subsequent data collection and analysis.

[0032] Subsequently, the first sensing device group is controlled to collect and record the physical characteristic parameters of each target axial piston pump during the production process of the target production batch, obtaining data on the pump body geometry, material properties, and other aspects, and aggregating these data to form a first sensing data set. Simultaneously, the second sensing device group is controlled to collect and record the production process parameters of each target axial piston pump during the production process of the target production batch, obtaining data on the equipment operating status, process parameter execution, and other aspects, and aggregating these data to form a second sensing data set.

[0033] By effectively collecting and classifying multi-source, heterogeneous data from the mass production of the target axial piston pump, the data foundation for the subsequent construction of the digital twin model was established. The first and second sensor datasets characterized the key attributes and behaviors of the axial piston pump from the perspectives of physical characteristics and production process, respectively, providing the necessary data support for a comprehensive understanding of its manufacturing process.

[0034] Furthermore, the embodiment of the present application also includes: parsing the first sensor data set to obtain geometric structure data and material property data of the target axial piston pump; Setting boundary conditions, performing dynamic analysis based on the geometric structure data and the material property data, and constructing the physical sub-model; parsing the second sensor data set to obtain operating performance data and production sensor data of the target axial piston pump; Regression analysis is performed based on the operating performance data and the production sensor data to construct the statistical sub-model.

[0035] In a feasible implementation, first, the first sensor data set is pre-processed by data cleaning, feature extraction, and other operations to extract key data related to the physical characteristics of the target axial piston pump, such as the three-dimensional dimensions of the pump body, surface roughness, material density, elastic modulus, etc., to provide accurate parameter input for subsequent physical modeling. Then, based on the acquisition of the pump body geometry and material property data, boundary conditions and constraints are set according to the actual working conditions of the target axial piston pump, such as pump inlet pressure, outlet pressure, speed, load, etc. Subsequently, numerical simulation methods such as multi-body dynamics simulation and finite element analysis are used to perform force analysis, stress and strain calculation, fatigue life prediction, etc. on the pump body to construct a mathematical model that can accurately reflect its physical behavior, namely the physical sub-model.

[0036] Simultaneously, data preprocessing and feature extraction are performed on the second sensor data set to obtain key data related to the actual operating performance and production process of the target axial piston pump, such as pump outlet flow rate, efficiency, vibration spectrum, production cycle time, and equipment operating conditions. This provides reliable sample data for subsequent statistical modeling. Next, based on the acquired axial piston pump operating performance and production process data, data mining methods such as multivariate regression and time series analysis are used to establish quantitative relationships between various performance indicators and production parameters, characterizing the key factors and patterns that influence pump performance during production. Through regression analysis, a statistical prediction model, or statistical sub-model, is generated to predict and optimize the axial piston pump manufacturing process.

[0037] By constructing a physical sub-model and a statistical sub-model, the inherent properties and behavioral characteristics of the axial piston pump are characterized at the mechanistic and data levels, respectively. The physical sub-model, based on the first sensor data set, utilizes physical laws and numerical simulation techniques to accurately simulate the structural characteristics and mechanical behavior of the pump body, providing a reliable theoretical tool for product design optimization and performance prediction. The statistical sub-model, based on the second sensor data set, utilizes data mining and machine learning techniques to reveal how various factors in the production process affect pump performance, providing data-driven decision support for process optimization and quality control. The collaborative integration of the physical and statistical sub-models improves the accuracy and practicality of the digital twin model, laying the foundation for optimizing the mass production process of axial piston pumps.

[0038] Furthermore, the embodiment of the present application also includes: Construct the LASSO regression expression: ; in, To minimize the loss function, is the number of samples of operating performance data or production sensor data, is the number of features of the operational performance data or production sensor data, For the The true target value of the sample, For the The first sample eigenvalues, is the regression intercept, for The weight coefficient of is the regularization parameter, and are two positive integers that can be changed incrementally; Using the LASSO regression expression, regression analysis is performed on the operating performance data and the production sensor data to obtain a minimized loss function; The regularization parameter is cross-validated based on the minimization loss function to construct the statistical sub-model.

[0039] In a preferred embodiment, the statistical sub-model is constructed using the LASSO regression algorithm. First, the LASSO regression expression is constructed as follows: ; in, To minimize the loss function, is the number of samples of operating performance data or production sensor data, is the number of features of the operational performance data or production sensor data, For the The true target value of the sample, For the The first sample eigenvalues, is the regression intercept, for The weight coefficient of is the regularization parameter, and are two positive integers that can be changed incrementally. The first term of the LASSO regression expression describes the sum of squared errors between the model prediction value and the true value, and the second term is the weighted sum of the absolute values ​​of the weight coefficients. The goal of LASSO regression is to find an optimal set of weight coefficients. and intercept Make the loss function minimize.

[0040] Then, the obtained operating performance data and production sensor data are respectively substituted into the LASSO regression expression, and the loss function is minimized. , estimate the weight coefficient of each feature and the regression intercept Among them, the operating performance data and production sensor data are substituted into the expression for fitting in turn to obtain different loss functions and regression coefficients. In order to select the optimal regularization parameter To avoid overfitting or underfitting, the cross-validation method is used to divide the data set into a training set and a validation set. Repeat the regression analysis under the given value, calculate the prediction error on the validation set, and select the one with the smallest prediction error As the optimal parameter. Then, using the optimal Re-perform LASSO regression on all data to obtain the final statistical sub-model, including the weight coefficient and regression intercept of each feature.

[0041] By constructing a statistical sub-model based on LASSO regression, the operating performance data and production sensor data of the axial piston pump are effectively utilized to reveal the intrinsic relationship between various performance indicators and production parameters, providing a basis for the optimization of production process parameters.

[0042] Further, such as Figure 2 As shown, the embodiment of the present application also includes: Retrieving historical production data record archives, and obtaining batch production trends based on the historical production data archives; Performing a production demand analysis according to the batch production trend, and determining the plurality of batch production plans based on the demand analysis results and the batch production capacity of the production equipment; Performing a production environment analysis on the multiple batch production plans and constructing multiple simulation scenarios, wherein the multiple simulation scenarios correspond to the multiple batch production plans; The multiple simulation scenarios are synchronized to the digital twin model, the multiple batch production plans are executed for simulation, and multiple batch simulation results are generated. The multiple batch simulation results correspond to the multiple simulation scenarios.

[0043] In a feasible implementation method, first, the historical production data record archives of axial piston pumps, such as production plans, order delivery, and capacity utilization, are extracted from production management systems such as MES and ERP, and statistical analysis is performed on these data to explore potential trends and patterns, such as seasonal changes in output and dynamic adjustments to production rhythms, to form a trend curve or trend model reflecting the characteristics of batch production of axial piston pumps, and to obtain batch production trends. Then, based on the obtained batch production trends, combined with production demand information such as market forecasts and customer orders, the production demand for axial piston pumps in the future is analyzed and predicted to obtain a quantitative demand target as the demand analysis result. Next, considering factors such as the design capacity, equipment performance, and process level of the existing production line, its batch production capacity is evaluated, and based on this, several feasible batch production plans are generated to obtain multiple batch production plans, each of which corresponds to a different production batch, production plan, equipment configuration, etc. Afterwards, for each batch production plan, a comprehensive analysis of the production environment involved is conducted, including production line layout, logistics routes, staffing, fixtures, environmental parameters, etc. Based on this, a virtual production scenario is constructed, enabling dynamic simulation of the axial piston pump batch production process under this plan in virtual space, resulting in multiple simulation scenarios. After the multiple simulation scenarios are completed, they are integrated with the digital twin model to form a complete virtual production system. Then, according to each batch production plan, a complete production process simulation is executed in the corresponding simulation scenario, including virtual assembly, virtual machining, virtual testing, and other links. Various production indicators and quality data such as output, pass rate, cycle time, energy consumption, etc. are counted in real time to obtain multiple batch simulation results.

[0044] By leveraging digital twins, rapid, low-cost, and high-fidelity simulation verification of mass production plans for axial piston pumps is achieved, shortening the plan evaluation and optimization cycle, improving the scientific nature and accuracy of decision-making, and providing a reliable decision-making basis for subsequent process optimization.

[0045] Furthermore, the embodiment of the present application also includes: Determining multiple pieces of production equipment information and multiple pieces of production process information based on the production process; Evaluate the mass production capability of the target axial piston pump based on the multiple pieces of production equipment information combined with the multiple pieces of production process information, and generate a mass production evaluation result; Performing a cost-benefit analysis based on the batch production evaluation results to determine production cost-benefit data; The feasibility of the batch production evaluation results is determined according to the production cost-effectiveness data, and the multiple batch production plans are formulated based on the feasible evaluation results.

[0046] In one feasible implementation, a detailed analysis of the axial piston pump production process is first conducted to identify key information such as the equipment type, machining accuracy, and process parameters required for each process step. The performance parameters, machining capabilities, and process range of each piece of equipment are then compiled to generate structured information on multiple production equipment and production processes, providing foundational data for subsequent production capacity assessment. Based on this information, the production capacity of the production line at different batch levels is then assessed, taking into account factors such as the number and performance of the equipment, as well as the rationality of the process route and the balance of the production cycle. This includes factors such as maximum production capacity, bottleneck processes, and average cycle time. Quantified batch production assessment results are then generated to identify the achievable output and constraints for each batch plan.

[0047] Next, based on the batch production capacity assessment results, further cost-benefit analysis is conducted, including fixed cost accounting, variable cost accounting, and economic benefit analysis. Fixed cost accounting accounts for fixed costs associated with batch production, such as equipment depreciation, factory rental, and management salaries. Variable cost accounting accounts for variable costs directly related to batch size, such as raw material costs, direct labor costs, and energy consumption. Economic benefit analysis, based on fixed and variable costs, combines product sales prices and market demand forecasts to calculate financial indicators such as total cost, unit cost, gross profit, and return on investment for different batch production scenarios. Through cost-benefit analysis, quantitative production cost-benefit data is obtained, reflecting the economic value and risk level of different batch production scenarios. Subsequently, the batch production assessment results and the obtained production cost-benefit data are comprehensively evaluated to screen batch production scenarios with balanced capacity and benefits and manageable risks, and prioritize them. From these scenarios, several of the most feasible scenarios are selected as multiple batch production scenarios for subsequent simulation verification and optimization decisions.

[0048] Through the generation and optimization of batch production plans, the influencing factors of multiple dimensions such as production capacity, cost-effectiveness, and risk control are fully considered, which improves the scientific nature and accuracy of plan formulation and provides a basis for subsequent simulation verification.

[0049] Furthermore, the embodiment of the present application also includes: Synchronize the multiple simulation scenarios to the digital twin model to set the time step, and determine the simulation step and simulation duration; executing the plurality of batch production plans in sequence according to the simulation duration and the simulation step size to perform cluster computing and generate a plurality of parallel simulation results; The multiple parallel simulation results are dynamically updated in combination with the simulation step size, and simulation compensation is performed on the multiple parallel simulation results according to the dynamic update result to generate the multiple batch simulation results.

[0050] In a preferred embodiment, first, the constructed multiple simulation scenarios are integrated with the digital twin model so that they can be jointly simulated in a unified virtual environment. At the same time, in order to balance simulation accuracy and computational efficiency, the concept of time step is introduced, that is, the simulation process is discretized into several time segments, each of which is called a simulation step. The length of the simulation step is set according to the actual needs of the mass production of axial piston pumps. In addition, the simulation duration of the entire simulation experiment is determined to cover a complete production cycle or several batches to ensure the representativeness and reliability of the simulation results.

[0051] After determining the simulation step size and duration, a high-performance computing cluster was used to perform parallel simulations of multiple batch production scenarios. Each batch production scenario was split into multiple discrete events corresponding to the simulation step size. These events were then assigned to different computing nodes. Parallel computing was then used to simultaneously simulate the execution of multiple production scenarios at various time intervals. Each computing node independently ran a digital twin model instance, processing the assigned production events and generating the corresponding simulation result data. The use of cluster parallel computing improved simulation speed and efficiency, enabling the system to complete large-scale production scenario simulation experiments within an acceptable timeframe.

[0052] After the parallel simulation is complete, the simulation result data distributed across different nodes is aggregated and synchronized. Specifically, the simulation data generated by each node in the current time segment is collected, using the simulation step size as a unit, and stored centrally in a database to form a complete, dynamically updated set of simulation results. At the same time, due to the potential for communication delays and data inconsistencies between nodes in parallel simulation, the aggregated simulation results are compensated and corrected as necessary, such as timestamp alignment and data interpolation, to ensure the continuity and accuracy of the simulation results. The compensated simulation result data is then classified and organized according to the batch production plan, generating complete batch simulation results corresponding to each plan, resulting in multiple batch simulation results.

[0053] Through efficient parallel simulation of axial piston pump batch production plans, digital twin models and high-performance computing technologies are fully utilized to shorten simulation time, improve simulation scale and accuracy, and provide high-quality data support for subsequent production decisions.

[0054] Furthermore, the embodiment of the present application also includes: Performing optimization evaluation on the plurality of batch simulation results to generate a simulation optimization evaluation result; generating optimization suggestions based on the simulation optimization evaluation results; Based on the optimization suggestions, a multi-index optimization channel is constructed to optimize and analyze the multiple batch production plans, and generate optimization feedback data; The optimization feedback data is used as an index to traverse the multiple batch production plans, the traversal results are verified, and the production batch optimization plan is output.

[0055] In a preferred embodiment, the generated multiple batch simulation results are first analyzed and evaluated, focusing on the performance differences between the various solutions in key indicators such as production efficiency, product quality, and cost-effectiveness. A comprehensive diagnosis of the strengths and weaknesses of each solution is conducted through a combination of quantitative scoring and qualitative description, forming a structured simulation optimization evaluation result. Then, based on the simulation optimization evaluation results, the optimization clues and improvement directions implicit in the simulation optimization evaluation results are further explored to generate a series of practical optimization suggestions to guide the adjustment and optimization of batch production solutions. For example, production batches can be adjusted to match market demand and production capacity constraints; resource allocation can be optimized to improve equipment utilization and personnel efficiency; process parameters can be improved to shorten production cycles and improve product quality; and logistics routes can be optimized to reduce work-in-process inventory and shorten delivery cycles.

[0056] After obtaining optimization recommendations, a multi-metric optimization pipeline is first constructed to evaluate the combined effectiveness of different optimization recommendation combinations. Specifically, the optimization recommendations are divided into several optimization dimensions, such as production batch size, process parameters, and equipment configuration, with each dimension corresponding to a set of optional optimization actions. Intelligent optimization algorithms, such as heuristic search and evolutionary computing, are then used to search for the optimal combination strategy across multiple optimization dimensions. The effectiveness of each strategy is evaluated through simulation experiments, generating corresponding optimization feedback data, including predicted values ​​of optimized key indicators, a list of optimization actions, and estimated optimization benefits. After completing the multi-metric optimization analysis, the optimization feedback data is used to modify and update the original batch production plan to produce an optimized production plan. Simulation verification is then re-performed based on the optimized production plan to evaluate and confirm the optimization results. If the verification results meet expectations, the plan is output as the final production batch optimization plan. If the verification results still show room for improvement, the optimization strategy is further adjusted until a satisfactory production batch optimization plan is achieved.

[0057] Through closed-loop optimization of the axial piston pump batch production plan, we fully utilize simulation results and artificial intelligence algorithms to automatically generate optimization suggestions, and search for the best strategy combination through multi-indicator optimization channels, thereby improving optimization efficiency and effectiveness.

[0058] In summary, the digital twin-based axial piston pump mass production process optimization system provided by the embodiments of the present application has the following technical effects: The data acquisition module is used to obtain multiple sensor data sets, including a first sensor data set and a second sensor data set, providing the data foundation for building the digital twin model. The model construction module is used to construct a physical sub-model based on the first sensor data set and a statistical sub-model based on the second sensor data set, thereby achieving digital modeling of the axial piston pump and its production process. The twin integration module is used to integrate the physical sub-model with the statistical sub-model to construct a digital twin model of the target axial piston pump, providing an integrated digital platform for production process optimization. The simulation execution module is used to simulate the target axial piston pump using the digital twin model by executing multiple batch production plans, generating multiple batch simulation results to provide decision support for production process optimization. The solution evaluation module is used to optimize and evaluate multiple batch production plans based on the multiple batch simulation results, generate a production batch optimization plan, and determine the optimal process parameter combination. The process optimization module is used to execute the production batch optimization plan according to the target production batch, and intelligently optimize the batch production process of the target axial piston pump to ensure the quality and efficiency of batch production.

[0059] Any steps of the system described above can be stored as computer instructions or programs in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any system in the embodiments of the present application. No unnecessary restrictions are made here.

[0060] Furthermore, the terms "first" or "second" as described above may not only represent an order relationship but may also represent a specific concept and / or refer to the selectability of multiple elements, either individually or in combination. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, if such modifications and variations fall within the scope of this application and its equivalents, this application is intended to include such modifications and variations.

Claims

1. Axial piston pump mass production process optimization system based on digital twin, characterized by: The system comprises: a data acquisition module, configured to obtain a plurality of sensor data sets, wherein the plurality of sensor data sets include a first sensor data set and a second sensor data set; a model building module, configured to build a physical sub-model based on the first sensor data set and a statistical sub-model based on the second sensor data set; A twin integration module, configured to integrate the physical sub-model with the statistical sub-model to construct a digital twin model of the target axial piston pump; a simulation execution module, configured to simulate a target axial piston pump by executing multiple batch production plans through the digital twin model, and generate multiple batch simulation results; A scheme evaluation module, configured to optimize and evaluate the plurality of batch production schemes according to the plurality of batch simulation results, and generate a production batch optimization scheme; The process optimization module is used to execute the production batch optimization plan according to the target production batch to intelligently optimize the target axial piston pump batch production process.

2. The digital twin-based axial piston pump mass production process optimization system according to claim 1 is characterized in that: The plurality of sensor data sets, the system comprises: Deploy multiple sensing devices based on the production process of the target axial piston pump and determine the sensing device group; Dividing the sensing device group according to sensing targets to generate a first sensing device group and a second sensing device group; Retrieving production batch data to determine multiple production batches, matching target axial piston pumps based on the multiple production batches, and determining a target production batch; According to the target production batch and the physical characteristics of the target axial piston pump, the target axial piston pump is batch-traversed using the first sensing device group to generate the first sensing data set; According to the target production batch and the production process, the target axial piston pumps are batch-traversed using the second sensing device group to generate the second sensing data set.

3. The digital twin-based axial piston pump mass production process optimization system according to claim 1, characterized in that: A physical sub-model is constructed based on the first sensor data set, and a statistical sub-model is constructed based on the second sensor data set, the system comprising: parsing the first sensor data set to obtain geometric structure data and material property data of the target axial piston pump; Setting boundary conditions, performing dynamic analysis based on the geometric structure data and the material property data, and constructing the physical sub-model; parsing the second sensor data set to obtain operating performance data and production sensor data of the target axial piston pump; Regression analysis is performed based on the operating performance data and the production sensor data to construct the statistical sub-model.

4. The digital twin-based axial piston pump mass production process optimization system according to claim 3 is characterized in that: Based on the operational performance data and the production sensor data, regression analysis is performed to construct the statistical sub-model, and the system includes: Construct the LASSO regression expression: ； in, To minimize the loss function, is the number of samples of operating performance data or production sensor data, is the number of features of the operational performance data or production sensor data, For the The true target value of the samples, For the The first sample eigenvalues, is the regression intercept, for The weight coefficient of is the regularization parameter, and are two positive integers that can be changed incrementally; Using the LASSO regression expression, regression analysis is performed on the operating performance data and the production sensor data to obtain a minimized loss function; The regularization parameter is cross-validated based on the minimization loss function to construct the statistical sub-model.

5. The digital twin-based axial piston pump mass production process optimization system according to claim 2, characterized in that: The digital twin model is used to execute multiple batch production plans to simulate the target axial piston pump and generate multiple batch simulation results. The system includes: Retrieving historical production data record archives, and obtaining batch production trends based on the historical production data archives; Performing a production demand analysis according to the batch production trend, and determining the plurality of batch production plans based on the demand analysis results and the batch production capacity of the production equipment; Performing a production environment analysis on the multiple batch production plans and constructing multiple simulation scenarios, wherein the multiple simulation scenarios correspond to the multiple batch production plans; The multiple simulation scenarios are synchronized to the digital twin model, the multiple batch production plans are executed for simulation, and multiple batch simulation results are generated. The multiple batch simulation results correspond to the multiple simulation scenarios.

6. The digital twin-based axial piston pump mass production process optimization system according to claim 5, characterized in that: A production demand analysis is performed based on the batch production trend, and multiple batch production plans are determined based on the demand analysis results and the batch production capacity of the production equipment. The system includes: Determining multiple pieces of production equipment information and multiple pieces of production process information based on the production process; Evaluate the mass production capability of the target axial piston pump based on the multiple pieces of production equipment information combined with the multiple pieces of production process information, and generate a mass production evaluation result; Performing a cost-benefit analysis based on the batch production evaluation results to determine production cost-benefit data; The feasibility of the batch production evaluation results is determined according to the production cost-effectiveness data, and the multiple batch production plans are formulated based on the feasible evaluation results.

7. The digital twin-based axial piston pump mass production process optimization system according to claim 5, characterized in that: Synchronizing the multiple simulation scenarios to the digital twin model, executing the multiple batch production plans for simulation, and generating multiple batch simulation results, the system includes: Synchronize the multiple simulation scenarios to the digital twin model to set the time step, and determine the simulation step and simulation duration; executing the plurality of batch production plans in sequence according to the simulation duration and the simulation step size to perform cluster computing and generate a plurality of parallel simulation results; The multiple parallel simulation results are dynamically updated in combination with the simulation step size, and simulation compensation is performed on the multiple parallel simulation results according to the dynamic update result to generate the multiple batch simulation results.

8. The digital twin-based axial piston pump mass production process optimization system according to claim 1, characterized in that: Optimizing and evaluating the plurality of batch production plans based on the plurality of batch simulation results to generate a production batch optimization plan, the system includes: Performing optimization evaluation on the plurality of batch simulation results to generate a simulation optimization evaluation result; generating optimization suggestions based on the simulation optimization evaluation results; Based on the optimization suggestions, a multi-index optimization channel is constructed to optimize and analyze the multiple batch production plans, and generate optimization feedback data; The optimization feedback data is used as an index to traverse the multiple batch production plans, the traversal results are verified, and the production batch optimization plan is output.

Images