Intelligent scheduling and collaborative operation system and method based on aluminum alloy part manufacturing and machining

Abstract

The invention discloses an intelligent scheduling and collaborative operation system and method based on aluminum alloy part manufacturing and machining, and relates to the technical field of aluminum alloy part manufacturing and machining. The system comprises the following components: a data acquisition and transmission module, which is used for arranging a plurality of sensors in each link of aluminum alloy part casting to acquire material performance and processing parameter data, encrypting and transmitting the data to a processing center, and carrying out data cleaning, format conversion and normalization preprocessing; through cooperative operation of machine learning and a fluid dynamics simulation technology, real-time, accurate and intelligent regulation and control are carried out on the whole aluminum alloy casting process, the machine learning analysis module can monitor material performance in real time and pre-judge the future change trend of the material performance, the fluid dynamics simulation module can deeply analyze casting defect generation roots, and the casting defect generation quality is improved. And the defects of air holes and shrinkage porosity can be prevented in advance, stable product performance is ensured, and the strict requirements of automobile parts and the like for high-quality aluminum alloy products are met.

Description

Technical Field

[0001] The present invention relates to the technical field of aluminum alloy component manufacturing and processing, and in particular to an intelligent scheduling and collaborative operation system and method based on the manufacturing and processing of aluminum alloy components. Background Art

[0002] In today's automobile manufacturing industry, the use of aluminum alloy parts not only helps to reduce the overall weight of the car and improve fuel economy, but also enhances the vehicle's handling and safety performance. It has become an important supporting material for promoting the development of the automobile industry towards lightweight and energy-saving. With the continuous advancement of science and technology and the growing market demand, the automobile industry has put forward more stringent requirements on the quality, performance and production efficiency of aluminum alloy parts. High-precision, high-performance aluminum alloy parts can significantly improve the overall quality and competitiveness of the car, and efficient production process is the key factor in meeting market demand and reducing production costs.

[0003] In the traditional casting production of aluminum alloy parts, existing production methods and technical means have many difficult-to-overcome defects. On the one hand, the traditional production process lacks real-time and accurate monitoring methods for the performance changes of aluminum alloy materials at different processing stages and under complex environmental conditions. During the casting process, the performance of aluminum alloy materials will be affected by various factors such as temperature, humidity, and pressure. The changes in these factors are often difficult to detect and accurately grasp in a timely manner. Due to the inability to reasonably adjust the production process in advance, it is difficult to effectively ensure product quality. On the other hand, for common defects such as porosity, shrinkage, and cold shuts that are prone to occur during the casting process, traditional detection methods mainly rely on post-casting inspection methods such as non-destructive testing and destructive testing. These inspection methods not only cannot detect and effectively prevent defects in a timely manner during the casting process, but once problems are detected, they often result in material waste and production delays. In addition, the adjustment of process parameters in traditional production processes mainly relies on the operator's experience and subjective judgment, lacking a scientific and systematic decision-making basis. This leads to low production efficiency and high production costs, which seriously restricts the development of the aluminum alloy parts manufacturing industry. Summary of the Invention

[0004] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide an intelligent scheduling and collaborative operation system and method for the manufacturing and processing of aluminum alloy parts. It can use a variety of high-precision sensors through the data acquisition and transmission module to collect the performance data and processing parameters of aluminum alloy materials under different environmental conditions in real time and comprehensively at each key link of casting production. The machine learning analysis module uses a machine learning algorithm based on the large amount of collected data to build a high-precision dynamic prediction model of material properties, which can monitor material properties in real time and dynamically evaluate their future evolution trends, and predict performance changes in advance. The fluid dynamics simulation module uses professional software to perform high-precision simulation of the flow, filling and solidification processes of aluminum alloy liquid in the mold, and deeply analyzes the causes and locations of casting defects. The intelligent decision-making and scheduling module organically integrates the results of the first two modules, uses intelligent algorithms for comprehensive analysis and decision-making, and provides a scientific and reasonable guidance plan for the production process. The parameter adjustment and optimization feedback module automatically adjusts the process parameters according to the decision results, performs performance compensation and optimization adjustment, and forms a closed-loop optimization system to continuously optimize the material performance prediction model and the fluid dynamics simulation model, so that the system can better adapt to the complex and changeable aluminum alloy casting production process.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: On the one hand, based on the intelligent scheduling and collaborative operation system for the manufacturing and processing of aluminum alloy parts, the system includes the following components:

[0006] Data acquisition and transmission module: Various sensors are deployed in each link of aluminum alloy parts casting to collect material properties and processing parameter data, which are encrypted and transmitted to the processing center for data cleaning, format conversion and normalization pre-processing;

[0007] Machine learning analysis module: Integrates collected data and extracts key features, builds and trains machine learning models to learn the relationship between performance and features, monitors in real time, predicts material performance changes based on current and historical data, and issues warnings in the event of anomalies;

[0008] Fluid dynamics simulation module: This module uses professional software to combine the physical properties of aluminum alloy liquid and the mold geometry to simulate its flow, filling, and solidification processes within the mold. The module considers the influence of microstructure, sets parameters based on actual process conditions, and compares simulation results with actual casting inspection results to optimize the model and accurately predict casting defects.

[0009] Intelligent Decision-Making and Scheduling Module: This module integrates material performance evaluation and casting defect risk data, uses intelligent algorithms to comprehensively assess production status risk levels and potential problems, formulates decision-making and scheduling strategies for adjusting process parameters and arranging production tasks, and collects feedback data during execution to evaluate and adjust strategies accordingly.

[0010] Parameter adjustment and optimization feedback module: After receiving decision instructions, it automatically adjusts the processing and casting process parameters, optimizes the equipment operation status according to changes in material properties, and feeds back simulation and production result data to the analysis and simulation module to optimize the corresponding model, forming a closed-loop optimization system.

[0011] Furthermore, the data acquisition and transmission module arranges multiple sensors in various links of aluminum alloy component casting to collect material performance and processing parameter data. Specifically, during the pre-casting preparation, casting process and post-casting processing stages, the hardness, strength, and toughness data of the aluminum alloy material under different temperatures, humidity and air pressures, as well as the temperature, stress and strain parameters during the processing process are collected.

[0012] Furthermore, the machine learning analysis module integrates the aluminum alloy material performance data and processing process parameter data transmitted by the data acquisition and transmission module, and uses feature extraction technology to screen out key features closely related to the aluminum alloy material performance from the data. Based on the extracted key features, a machine learning algorithm is used to construct a material performance dynamic prediction model. The trained material performance dynamic prediction model receives the latest data from the data acquisition and transmission module in real time, monitors the aluminum alloy material performance in real time, and predicts the performance changes of the aluminum alloy material in future processing stages or under different environmental conditions based on the changing trends of the current material performance data and historical data. When abnormal changes or deviations from the expected range are detected in the material performance, an early warning signal is issued in time.

[0013] Furthermore, the machine learning analysis module uses a machine learning algorithm to construct a dynamic prediction model for material properties, and its algorithm formula is: Among them, Y is the predicted aluminum alloy material performance index, including hardness, strength and toughness, w j is the jth basic prediction model f j The weight of (X), and The weights will be adaptively adjusted according to the performance of the model in different data segments, f j (X) is the jth basic prediction model, X is the input eigenvector, j = 1, 2, …, m, m is the number of basic prediction models, X is the eigenvector composed of key characteristic parameters related to the performance of aluminum alloy materials extracted from the data, ∈ is the error term, reflecting the influence of other factors not considered in the model on the material properties.

[0014] Furthermore, the w j is the jth basic prediction model f j The weight of (X) is adjusted as follows: in and They are the weight after adjustment and the weight before adjustment, Δw is the adjustment step size, Errormin is the minimum error among all base models, Error j is the error of the jth base model.

[0015] Furthermore, the fluid dynamics simulation module uses professional fluid dynamics simulation software to establish a numerical simulation model of the aluminum alloy liquid in the mold based on the actual physical and chemical properties of the aluminum alloy liquid and the precise three-dimensional geometric model of the mold. When setting the simulation parameters, the process conditions in actual production are fully considered, the simulation model is run, and the flow, filling and solidification processes of the aluminum alloy liquid in the mold are numerically calculated. Through simulation calculations, the flow velocity distribution, temperature field changes and pressure distribution information of the aluminum alloy liquid at different times are obtained, the simulation results are analyzed, and visualization technology is used to display the flow trajectory and solidification process of the aluminum alloy liquid. The simulation results are compared with the casting quality inspection data in actual production to verify the accuracy of the simulation model. If there is a large deviation between the simulation results and the actual situation, the cause is analyzed and the simulation model is optimized.

[0016] Furthermore, the intelligent decision-making and scheduling module integrates material performance evaluation and casting defect risk data, and uses an intelligent algorithm to comprehensively evaluate the production status risk level and potential problems. The algorithm formula is: Among them, S is the comprehensive decision score, reflecting the final production decision result, μ i is the fuzzy weight, which is determined by the fuzzy reasoning system based on the material properties and casting defect risk factors. i is the optimized value of the i-th decision objective, i=1,2,…,p, p is the number of decision objectives, μ i is the fuzzy weight of the i-th decision target, 0≤μ i ≤1, and

[0017] Furthermore, when the parameter adjustment and optimization feedback module receives the decision instruction from the intelligent decision and scheduling module, the process parameter automatic adjustment unit quickly adjusts the processing parameters and casting process parameters. The process parameter adjustment formula is: Among them, v new is the adjusted process parameter, v old is the process parameter before adjustment, α is the integral adjustment coefficient, and by testing different α values, the influence of process parameter adjustment on the performance of aluminum alloy materials is observed, and the α value with the best performance adjustment effect is selected. t0 is the start time, t1 is the end time, and E target is the target performance index value, determined according to product quality requirements and production standards, E currentis the current performance index value. The performance data of the aluminum alloy material is collected in real time by the data acquisition and transmission module and evaluated by the machine learning analysis module. β is the performance change adjustment coefficient, which is determined according to material properties and production experience. Different aluminum alloy materials have different sensitivities to performance changes. The appropriate β value is determined through experiments and experience. ΔP is the performance change of the aluminum alloy material obtained by the machine learning analysis module.

[0018] A method for intelligent scheduling and collaborative operation based on the manufacturing and processing of aluminum alloy parts, which is applicable to the intelligent scheduling and collaborative operation system based on the manufacturing and processing of aluminum alloy parts according to any one of claims 1 to 8, is characterized in that the method comprises the following specific steps:

[0019] Comprehensive data collection: Utilizing various sensors in the data collection and transmission module, we collect aluminum alloy material performance data and various parameter data during the processing process at every stage of aluminum alloy parts casting production, and transmit the data to the data processing center;

[0020] Model building and performance evaluation: The machine learning analysis module uses machine learning algorithms to build a dynamic material performance prediction model based on the collected data, and conducts real-time and dynamic monitoring and evaluation of aluminum alloy material performance;

[0021] Simulation analysis and defect prediction: The fluid dynamics simulation module uses simulation software to simulate and analyze the flow, filling, and solidification processes of aluminum alloy liquid in the mold, predict possible casting defects, and determine the location and type of defects;

[0022] Comprehensive analysis and intelligent decision-making: The intelligent decision-making scheduling module integrates the material performance evaluation results and casting defect prediction results, and uses intelligent algorithms to conduct comprehensive analysis and make decisions;

[0023] Parameter adjustment and optimization: If the material performance evaluation results show that the aluminum alloy material performance deviates from expectations, the processing parameters are automatically adjusted to compensate for the performance, and the casting process parameters and equipment operating status are optimized according to the changes in material properties;

[0024] Data feedback and model optimization: The data generated by the fluid dynamics simulation is fed back to the machine learning operation module to optimize the material performance prediction model. At the same time, the casting result data in actual production is fed back to the fluid dynamics simulation module to correct the fluid dynamics simulation model and achieve closed-loop optimization of the system.

[0025] Compared with the existing technology, the intelligent scheduling and collaborative operation system and method based on aluminum alloy parts manufacturing and processing have the following beneficial effects:

[0026] 1. The present invention uses the collaborative operation of machine learning and fluid dynamics simulation technology to perform real-time, precise, and intelligent control of the entire aluminum alloy casting process. The machine learning analysis module can monitor material properties in real time and predict their future changing trends. The fluid dynamics simulation module can deeply analyze the root causes of casting defects and prevent porosity and shrinkage defects in advance, ensuring stable product performance and meeting the stringent requirements of automotive parts and other industries for high-quality aluminum alloy products.

[0027] 2. The present invention uses the intelligent decision-making and scheduling module to comprehensively consider the material performance evaluation results and casting defect risks, quickly make decisions and feed back to the parameter adjustment and optimization feedback module, so that the production process can be flexibly adjusted according to real-time conditions, significantly improving production efficiency and shortening the production cycle. At the same time, the closed-loop optimization system continuously optimizes two key models, allowing the system to better adapt to complex production conditions, reduce production costs, and enhance corporate competitiveness.

[0028] Other advantages, objects and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0030] Figure 1 This is a schematic diagram of the structure of the intelligent scheduling and collaborative operation system for the manufacturing of aluminum alloy parts;

[0031] Figure 2 This is a flow chart of the intelligent scheduling and collaborative operation method based on the manufacturing of aluminum alloy parts. DETAILED DESCRIPTION

[0032] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0033] Example 1

[0034] As the core component of the engine, the quality of the automotive aluminum alloy engine cylinder directly determines the key performance indicators of the engine, such as power output, fuel economy, reliability and durability.

[0035] During the aluminum alloy smelting process, high-precision, high-stability thermocouple temperature sensors are used to monitor the melting temperature of the aluminum alloy liquid. These sensors can quickly respond to temperature changes, ensuring real-time and accurate temperature data of the aluminum alloy liquid during the smelting process, providing a reliable basis for subsequent smelting process control. Stress sensors are densely arranged in key areas of the casting mold, such as the gate, cavity wall, and cooling channels. These stress sensors use the resistance strain gauge principle to accurately measure stress changes in all directions of the mold during the pouring and cooling processes. At the same time, hardness sensors are installed in the working area of ​​the processing machine tool to quickly and accurately detect the hardness of various parts of the engine cylinder after processing. These sensors collect performance data and processing parameters of the aluminum alloy material at different stages in real time at a set high frequency, and transmit them encrypted via a high-speed and stable optical fiber network. After arriving at the system's data processing center, the data is first cleaned and then converted to a standard format that the system can recognize. Finally, normalization is performed. Linear normalization is used to keep the data in the numerical range [0, 1] to facilitate subsequent data analysis and model construction.

[0036] The pre-processed data is deeply integrated, and the feature extraction technology based on machine learning is used to screen out the key features closely related to the engine cylinder performance from the massive data, such as melting temperature, mold stress, processing time, cooling rate, alloy composition ratio, etc. Based on these key features, a dynamic prediction model of material properties is constructed using machine learning algorithms to predict the performance indicators of the engine cylinder, such as strength, wear resistance, sealing, thermal conductivity, etc. Figure 1 As shown, the model formula is: Among them, Y is the predicted performance index of aluminum alloy material (such as hardness, strength, toughness, etc.), w j is the jth basic prediction model f j The weight of (X), and f j (X) is the jth basic prediction model, X is the input eigenvector, j = 1, 2, ..., m, m is the number of basic prediction models, X is the eigenvector composed of key characteristic parameters (such as temperature, stress, strain, processing time, etc.) related to the performance of aluminum alloy materials extracted from the data, ∈ is the error term, reflecting the influence of other factors not considered in the model on the material properties. When the model weight is initially determined, according to the complexity of each basic prediction model and its performance on historical data, the complex model that can theoretically capture the data characteristics more accurately is selected. During the model operation process, new production data is continuously introduced to test each basic prediction model, and the weight is dynamically adjusted according to the size of the prediction error. The specific adjustment formula is: in and They are the weights after adjustment and before adjustment, Δw is a smaller adjustment step, and the Error is determined according to the actual situation. min is the minimum error among all base models, Error j is the error of the jth basic model. Through continuous learning and adjustment, real-time monitoring of engine cylinder performance is achieved. When the performance indicators are predicted to deviate from the expected range, early warning signals are issued in a timely manner, and corresponding performance change trend analysis and possible cause inference are provided.

[0037] Using professional fluid dynamics simulation software, based on precise physical and chemical properties of the aluminum alloy liquid, such as density, viscosity, and surface tension, as well as a precise 3D geometric model of the engine cylinder mold obtained through 3D scanning and precise modeling, the flow, filling, and solidification processes of the aluminum alloy liquid within the mold are simulated with high precision. During the simulation process, the process parameters used in actual production are fully considered. Through simulation calculations, detailed information such as the flow velocity distribution, temperature field changes, and pressure distribution of the aluminum alloy liquid at different times are obtained. Visualization techniques such as streamline plots, isotherm plots, and pressure contour maps are used to intuitively present the flow trajectory and solidification process of the aluminum alloy liquid. By analyzing the simulation results, the location, type, and cause of possible casting defects such as shrinkage cavities, air holes, cold shuts, and slag inclusions are accurately predicted. The results are then compared and verified with the quality inspection data of the engine cylinder block in actual production. During this comparison and verification process, statistical methods are used to quantitatively analyze the simulation results and actual inspection results. Based on the analysis results, the parameters and structure of the simulation model are continuously optimized to improve the accuracy of the predictions.

[0038] The dynamic evaluation results of engine cylinder performance obtained by the machine learning analysis module are deeply integrated with the casting defect risk data obtained by the fluid dynamics simulation module. Based on fuzzy logic and multi-objective optimization methods, a comprehensive assessment of the production status is conducted to determine the risk level and potential problems in product quality, production efficiency, cost control, etc. in the current production process. A comprehensive decision-making formula based on fuzzy logic and multi-objective optimization is adopted: Among them, S is the comprehensive decision score, reflecting the final production decision result, μ i is the fuzzy weight, which is determined by the fuzzy reasoning system based on factors such as material properties and casting defect risks. iis the optimization value of the i-th decision goal, i = 1, 2, ..., p, p is the number of decision goals, when the weight is initially determined, according to the actual needs and experience of automobile engine cylinder production, the decision goals such as material performance optimization, defect risk reduction, production efficiency improvement, and cost control are given corresponding weights. For example, in the stage where product quality requirements are extremely high, the weight of the material performance optimization goal may be set to a higher value, such as 0.4. When the production task is tight, the weight of the production efficiency improvement goal is appropriately increased. As production progresses, new production data is continuously collected, including changes in raw material quality (such as fluctuations in impurity content in aluminum alloys), equipment operating status (such as equipment wear and tear, failure frequency), market demand fluctuations ( Such as changes in order quantity and delivery time), etc., the weights are adjusted in real time through the fuzzy inference system, and the input data is fuzzified, inferred and defuzzified according to the pre-set fuzzy rules and membership functions to obtain the adjusted weights. According to the comprehensive evaluation results, scientific and reasonable production decisions and scheduling strategies are formulated, such as adjusting the pouring speed, optimizing the processing parameters (such as cutting speed, feed rate), reasonably arranging the sequence of production tasks, adjusting the raw material procurement plan, etc., to achieve optimal control of the production process. When formulating decisions and scheduling strategies, various constraints in the production process are considered, such as the production capacity of the equipment, the supply of raw materials, the compatibility of the process, etc., to ensure the feasibility and effectiveness of the strategy.

[0039] After receiving the decision instructions from the intelligent decision-making and scheduling module, the process parameter automatic adjustment unit quickly and accurately adjusts the casting and processing parameters. Specifically, based on the real-time changes and predicted results of the engine cylinder performance, it uses an adjustment strategy based on feedback control and dynamic programming to optimize the process parameters. The adjustment formula is: Among them, v new is the adjusted process parameter, v old is the process parameter before adjustment, α is the integral adjustment coefficient, and by testing different α values, the influence of process parameter adjustment on the performance of aluminum alloy materials is observed, and the α value with the best performance adjustment effect is selected. t0 is the start time, t1 is the end time, and E target is the target performance index value, determined according to product quality requirements and production standards, E currentis the current performance index value. The performance data of the aluminum alloy material is collected in real time by the data acquisition and transmission module and evaluated by the machine learning analysis module. β is the performance change adjustment coefficient, which is determined according to the material properties. Different aluminum alloy materials have different sensitivities to performance changes. The appropriate β value is determined through experiments. ΔP is the performance change of the aluminum alloy material obtained by the machine learning analysis module (such as hardness change, strength change, etc., which can be quantified as a numerical value). At the same time, according to the changes in the performance of the aluminum alloy material, the operating status of the equipment is optimized in advance, such as adjusting the cutting parameters of the processing equipment (such as tool angle, cutting depth), the flow rate and temperature of the mold cooling system, etc. When adjusting the operating status of the equipment, the performance characteristics and process requirements of the equipment are fully considered to ensure that the equipment operates in the best state. The detailed data generated by the fluid dynamics simulation module, such as the flow velocity distribution of the aluminum alloy liquid, temperature field changes, pressure change curves, etc., as well as the quality inspection data of the engine cylinder in actual production, defect types and locations, and other casting result data, are fed back to the machine learning analysis module and the fluid dynamics simulation module in real time. The machine learning analysis module optimizes and updates the material performance prediction model based on the feedback data, and uses the online learning algorithm to continuously adjust the parameters and structure of the model to improve the prediction accuracy of the model. The fluid dynamics simulation module corrects the simulation model based on the feedback data, adjusts the simulation parameters (such as the physical properties of the aluminum alloy liquid, the thermal conductivity coefficient of the mold, etc.) and the model structure (such as the accuracy of the grid division), and further improves the accuracy and reliability of the model, forming a continuously optimized closed-loop system.

[0040] Example 2

[0041] As an important part of the vehicle's driving system, aluminum alloy wheels not only bear key functions such as supporting the vehicle's weight, transmitting driving and braking forces, but their appearance and quality also directly affect the vehicle's overall aesthetics, driving safety, handling stability, and fuel economy.

[0042] High-precision infrared temperature sensors are installed at key locations inside the aluminum alloy melting furnace. This sensor can quickly and accurately measure the melting temperature of the aluminum alloy liquid, with a wide measurement range and an accuracy of up to ±1°C. It can monitor subtle temperature changes during the melting process in real time, ensuring that the aluminum alloy liquid reaches the appropriate melting temperature range and providing stable raw material conditions for the subsequent casting process. Various types of sensors are carefully arranged at different depths on the surface and inside the wheel hub casting mold. Among them, the temperature sensor can accurately measure the temperature distribution of the mold during the pouring and cooling process, helping to understand the heat transfer characteristics of the mold. The stress sensor uses a high-precision piezoresistive stress sensor, which can monitor the stress changes of the mold under the pressure and thermal stress of the aluminum alloy liquid in real time. It has high sensitivity and can detect tiny stress fluctuations. In the processing link, high-precision hardness sensors and non-contact optical size measurement sensors are used. The hardness sensor can perform hardness testing on different parts of the wheel hub after processing to ensure that the wheel hub hardness meets the design requirements. These sensors collect the performance data and processing parameters of the aluminum alloy material at different processing stages in real time and comprehensively according to the set frequency, such as Figure 2 As shown in the figure, the data is encrypted and transmitted via high-speed Ethernet. After the data is transmitted to the data processing center of the system, the data is first cleaned and then format converted. The various formats of data output by different types of sensors (such as analog signals, digital signals, image data, etc.) are uniformly converted into a standard format that the system can process. Finally, normalization is performed and the normalization function is used to map the data to the interval [0, 1] to make the data comparable and consistent, which is convenient for subsequent data analysis and model training.

[0043] The collected and pre-processed data is deeply integrated, and the machine learning feature extraction algorithm is used to accurately screen out key features closely related to the performance of aluminum alloy wheels from massive data. These key features include the melting temperature change curve, the distribution characteristics of the mold temperature field, the time series of stress changes, processing parameters (such as cutting speed, feed rate, processing time) and the chemical composition of aluminum alloys. Based on these key features, a dynamic material performance prediction model is constructed using a machine learning algorithm to predict key performance indicators such as the strength, toughness, fatigue resistance, and dynamic balance performance of the wheel.

[0044] When the model weights are initially determined, the structural complexity of each basic prediction model, the fitting effect on historical data, and the sensitivity to different features are comprehensively considered. During the operation of the model, new production data are continuously introduced to test each basic prediction model. A variety of performance evaluation indicators are used to comprehensively evaluate the prediction accuracy of the model. The weights are dynamically adjusted according to the evaluation results, and the weights are iteratively updated using an optimization algorithm to enable the weights to continuously adapt to changes in data and improve the overall prediction performance of the model. By monitoring the wheel hub performance in real time, when abnormal fluctuations or deviations from the expected range are predicted, detailed early warning information is issued in a timely manner, including the specific changes in performance indicators, possible cause analysis, and impact assessment on subsequent production.

[0045] By using professional fluid dynamics simulation software, combined with the physical and chemical properties of aluminum alloy liquid measured through precise experiments (such as density, viscosity, surface tension, thermal conductivity, etc.) and the precise geometric model of the wheel hub mold obtained through high-precision 3D scanning and reverse engineering technology, a highly realistic simulation of the flow, filling and solidification process of aluminum alloy liquid in the mold is carried out. In the simulation process, various process conditions in actual production are fully considered. Through simulation calculation, detailed information such as flow velocity distribution, temperature field change, pressure distribution, solid phase ratio change of aluminum alloy liquid at different times is obtained, and visualization technology such as 3D dynamic simulation demonstration, contour map, Vector diagrams, etc., can intuitively display the flow trajectory, filling process and solidification process of the aluminum alloy liquid. Through in-depth analysis of the simulation results, the possible casting defects such as shrinkage, pores, cracks, cold shut, segregation, etc. can be accurately predicted, and the location, type and cause of the resulting defects can be carefully predicted. The simulation results are carefully compared and verified with the quality inspection data of the wheel hub in actual production. The accuracy and reliability of the simulation results are evaluated by statistical hypothesis testing and other methods. According to the comparison results, the parameters (such as material properties and boundary conditions) and structure (such as the density and method of grid division) of the simulation model are optimized and adjusted to continuously improve the accuracy and predictive ability of the simulation model.

[0046] The wheel hub performance evaluation results obtained by the machine learning analysis module are deeply integrated with the casting defect risk data obtained by the fluid dynamics simulation module analysis. Based on the fuzzy logic and multi-objective optimization methods, a comprehensive and integrated evaluation of the production status is conducted to determine the risk level and potential problems in product quality, production efficiency, cost control, equipment maintenance, etc. in the current production process. When the weights are initially determined, the decision-making goals such as material performance optimization, defect risk reduction, production efficiency improvement, cost control, and equipment maintenance are given corresponding weights based on the industry standards for automobile aluminum alloy wheel production, the company's actual production experience, and the focus of market demand. For example, when producing high-end wheel hub products, the weights of material performance optimization and defect risk reduction goals may be set to The weights of the production efficiency improvement targets are set relatively high, such as 0.3 and 0.3 respectively. When the production task is urgent, the weight of the production efficiency improvement target is appropriately increased. As the production continues, new production data is continuously collected, including batch differences of raw materials (such as purity fluctuations of aluminum alloys, changes in impurity content), equipment operating status data (such as equipment vibration parameters, motor current, temperature sensor readings), dynamic changes in market demand (such as changes in order quantity of different styles of wheels, delivery requirements), etc. The weights are adjusted in real time through the fuzzy inference system. According to the pre-defined fuzzy rule base and membership function, the input data is fuzzified and the adjusted weight value is obtained through logical reasoning, so that the weight can better reflect the actual situation and needs of current production.

[0047] Based on the comprehensive evaluation results, scientific, reasonable and feasible production decision-making and scheduling strategies are formulated. These strategies include adjusting casting process parameters (such as casting temperature, speed, and casting method), optimizing processing parameters (such as tool selection, cutting parameter adjustment, and processing sequence optimization), reasonably arranging the priority and sequence of production tasks, adjusting the procurement plan and inventory management of raw materials, formulating equipment maintenance plans and preventive maintenance strategies, etc., in order to achieve optimal control of the production process and rational allocation of resources. When formulating decisions and scheduling strategies, various constraints in the production process are fully considered, such as equipment production capacity limitations, raw material supply stability, compatibility between processes, and limitations of human and time resources, to ensure the feasibility and effectiveness of the strategy.

[0048] After receiving the decision instructions issued by the intelligent decision-making and scheduling module, the process parameter automatic adjustment unit quickly and accurately adjusts the casting and processing parameters. Specifically, based on the real-time monitoring data and prediction results of the wheel hub performance, the process parameters are optimized and adjusted using an algorithm based on feedback control and dynamic programming. By real-time monitoring of parameters such as the temperature of the aluminum alloy liquid, the temperature of the mold, and the pressure changes during the pouring process, combined with the prediction of the wheel hub performance by the machine learning model, the pouring temperature is dynamically adjusted to ensure that the flow and filling process of the aluminum alloy liquid in the mold is more stable, thereby reducing the occurrence of casting defects.

[0049] At the same time, the operating status of the equipment is optimized in advance according to the changes in the performance of the aluminum alloy material. For example, when the hardness of the aluminum alloy is predicted to change, the cutting parameters of the processing equipment, such as the cutting angle, cutting depth and feed speed of the tool, are adjusted in advance to ensure the processing quality. When the temperature distribution of the mold is monitored to be uneven, the flow and temperature of the mold cooling system are adjusted in time to make the mold temperature field more uniform and improve the casting quality. When adjusting the operating status of the equipment, the performance characteristics, service life and maintenance requirements of the equipment are fully considered to ensure that the equipment operates in the best condition and reduce the probability of equipment failure.

[0050] The detailed data generated by the fluid dynamics simulation module, such as the flow velocity vector diagram of the aluminum alloy liquid, the temperature field change curve, the pressure distribution cloud diagram, etc., as well as the quality inspection data of the wheel hub in actual production (such as hardness test report, size inspection report, dynamic balance inspection report), defect type and location and other casting result data, are fed back to the machine learning analysis module and the fluid dynamics simulation module in real time. The machine learning analysis module optimizes and updates the material property prediction model based on the feedback data, and uses the online learning algorithm to continuously adjust the parameters and structure of the model to improve the model's adaptability to new data and prediction accuracy. The fluid dynamics simulation module corrects the simulation model based on the feedback data, adjusts the simulation parameters (such as the physical properties of the aluminum alloy liquid, the heat transfer coefficient of the mold, boundary conditions, etc.) and the model structure (such as the accuracy of the grid division, the grid type, etc.), further improving the accuracy and reliability of the simulation model, forming a continuously optimized closed-loop system. Through continuous feedback and optimization, the production process can continuously adapt to various changing factors and continuously improve the production quality and production efficiency of the wheel hub.

[0051] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. Based on the intelligent scheduling and collaborative operation system for aluminum alloy parts manufacturing, it is characterized by: The system includes the following components: Data acquisition and transmission module: Various sensors are deployed in each link of aluminum alloy parts casting to collect material properties and processing parameter data, which are encrypted and transmitted to the processing center for data cleaning, format conversion and normalization pre-processing; Machine learning analysis module: Integrates collected data and extracts key features, builds and trains machine learning models to learn the relationship between performance and features, monitors in real time, predicts material performance changes based on current and historical data, and issues warnings in the event of anomalies; Fluid dynamics simulation module: This module uses professional software to combine the physical properties of aluminum alloy liquid and the mold geometry to simulate its flow, filling, and solidification processes within the mold. The module considers the influence of microstructure, sets parameters based on actual process conditions, and compares simulation results with actual casting inspection results to optimize the model and accurately predict casting defects. Intelligent Decision-Making and Scheduling Module: This module integrates material performance evaluation and casting defect risk data, uses intelligent algorithms to comprehensively assess production status risk levels and potential problems, formulates decision-making and scheduling strategies for adjusting process parameters and arranging production tasks, and collects feedback data during execution to evaluate and adjust strategies accordingly. Parameter adjustment and optimization feedback module: After receiving decision instructions, it automatically adjusts the processing and casting process parameters, optimizes the equipment operation status according to changes in material properties, and feeds back simulation and production result data to the analysis and simulation module to optimize the corresponding model, forming a closed-loop optimization system.

2. The intelligent scheduling and collaborative operation system for aluminum alloy parts manufacturing and processing according to claim 1 is characterized in that: The data acquisition and transmission module arranges multiple sensors in each link of aluminum alloy component casting to collect material performance and processing parameter data. Specifically, during the pre-casting preparation, casting process and post-casting processing stages, it collects the hardness, strength, and toughness data of the aluminum alloy material under different temperatures, humidity and air pressures, as well as the temperature, stress and strain parameters during the processing process.

3. The intelligent scheduling and collaborative operation system for aluminum alloy parts manufacturing and processing according to claim 1 is characterized in that: The machine learning analysis module integrates the aluminum alloy material performance data and processing process parameter data transmitted by the data acquisition and transmission module, uses feature extraction technology to screen out key features closely related to the aluminum alloy material performance from the data, and uses machine learning algorithms to construct a material performance dynamic prediction model based on the extracted key features. The trained material performance dynamic prediction model receives the latest data from the data acquisition and transmission module in real time, monitors the aluminum alloy material performance in real time, and predicts the performance changes of the aluminum alloy material in future processing stages or under different environmental conditions based on the changing trends of the current material performance data and historical data. When abnormal changes or deviations from the expected range of material performance are detected, an early warning signal is issued in a timely manner.

4. The intelligent scheduling and collaborative operation system for aluminum alloy parts manufacturing and processing according to claim 1 is characterized in that: The machine learning analysis module uses a machine learning algorithm to construct a dynamic prediction model for material properties. The algorithm formula is: Among them, Y is the predicted aluminum alloy material performance index, including hardness, strength and toughness, w j is the jth basic prediction model f j The weight of (X), and The weights will be adaptively adjusted according to the performance of the model in different data segments, f j (X) is the jth basic prediction model, X is the input eigenvector, j = 1, 2, …, m, m is the number of basic prediction models, X is the eigenvector composed of key characteristic parameters related to the performance of aluminum alloy materials extracted from the data, ∈ is the error term, reflecting the influence of other factors not considered in the model on the material properties.

5. The intelligent scheduling and collaborative operation system for aluminum alloy parts manufacturing and processing according to claim 4 is characterized in that: The w j is the jth basic prediction model f j The weight of (X) is adjusted as follows: in and They are the weight after adjustment and the weight before adjustment, Δw is the adjustment step size, Error min is the minimum error among all base models, Error j is the error of the jth base model.

6. The intelligent scheduling and collaborative operation system based on aluminum alloy parts manufacturing and processing according to claim 1 is characterized in that: The fluid dynamics simulation module uses professional fluid dynamics simulation software to establish a numerical simulation model of the aluminum alloy liquid in the mold based on the actual physical and chemical properties of the aluminum alloy liquid and the precise three-dimensional geometric model of the mold. When setting the simulation parameters, the process conditions in actual production are fully considered. The simulation model is run to perform numerical calculations on the flow, filling and solidification processes of the aluminum alloy liquid in the mold. Through the simulation calculations, the flow velocity distribution, temperature field changes and pressure distribution information of the aluminum alloy liquid at different times are obtained. The simulation results are analyzed, and the flow trajectory and solidification process of the aluminum alloy liquid are displayed using visualization technology. The simulation results are compared with the casting quality inspection data in actual production to verify the accuracy of the simulation model. If there is a large deviation between the simulation results and the actual situation, the cause is analyzed and the simulation model is optimized.

7. The intelligent scheduling and collaborative operation system for aluminum alloy parts manufacturing and processing according to claim 1 is characterized in that: The intelligent decision-making and scheduling module integrates material performance evaluation and casting defect risk data, and uses an intelligent algorithm to comprehensively evaluate the production status risk level and potential problems. The algorithm formula is: Among them, S is the comprehensive decision score, reflecting the final production decision result, μ i is the fuzzy weight, which is determined by the fuzzy reasoning system based on the material properties and casting defect risk factors. i is the optimized value of the i-th decision objective, i = 1, 2, ..., p, p is the number of decision objectives, μ i is the fuzzy weight of the i-th decision target, 0≤μ i ≤1, and 8. The intelligent scheduling and collaborative operation system for aluminum alloy parts manufacturing and processing according to claim 1 is characterized in that: When the parameter adjustment and optimization feedback module receives the decision instruction of the intelligent decision and scheduling module, the process parameter automatic adjustment unit quickly adjusts the processing parameters and casting process parameters. The process parameter adjustment formula is: Among them, v new is the adjusted process parameter, v old is the process parameter before adjustment, α is the integral adjustment coefficient, and by testing different α values, the influence of process parameter adjustment on the performance of aluminum alloy materials is observed, and the α value with the best performance adjustment effect is selected. t0 is the start time, t1 is the end time, and E target is the target performance index value, determined according to product quality requirements and production standards, E current is the current performance index value, β is the performance change adjustment coefficient, and ΔP is the performance change of the aluminum alloy material obtained by the machine learning analysis module.

9. A method for intelligent scheduling and collaborative operation based on the manufacturing and processing of aluminum alloy parts, which is applicable to the intelligent scheduling and collaborative operation system based on the manufacturing and processing of aluminum alloy parts according to any one of claims 1 to 8, characterized in that: The method comprises the following specific steps: Comprehensive data collection: Utilizing various sensors in the data collection and transmission module, we collect aluminum alloy material performance data and various parameter data during the processing process at every stage of aluminum alloy parts casting production, and transmit the data to the data processing center; Model building and performance evaluation: The machine learning analysis module uses machine learning algorithms to build a dynamic material performance prediction model based on the collected data, and conducts real-time and dynamic monitoring and evaluation of aluminum alloy material performance; Simulation analysis and defect prediction: The fluid dynamics simulation module uses simulation software to simulate and analyze the flow, filling, and solidification processes of aluminum alloy liquid in the mold, predict possible casting defects, and determine the location and type of defects; Comprehensive analysis and intelligent decision-making: The intelligent decision-making scheduling module integrates the material performance evaluation results and casting defect prediction results, and uses intelligent algorithms to conduct comprehensive analysis and make decisions; Parameter adjustment and optimization: If the material performance evaluation results show that the aluminum alloy material performance deviates from expectations, the processing parameters are automatically adjusted to compensate for the performance, and the casting process parameters and equipment operating status are optimized according to the changes in material properties; Data feedback and model optimization: The data generated by the fluid dynamics simulation is fed back to the machine learning operation module to optimize the material performance prediction model. At the same time, the casting result data in actual production is fed back to the fluid dynamics simulation module to correct the fluid dynamics simulation model and achieve closed-loop optimization of the system.

Images