New energy aluminum alloy part casting process multi-parameter collaborative optimization system and method thereof

Abstract

The invention relates to the technical field of aluminum alloy part casting processes, in particular to a multi-parameter collaborative optimization system and method for a new energy aluminum alloy part casting process. According to the technical scheme, the multi-parameter collaborative optimization system for the new energy aluminum alloy part casting process comprises a multivariate heterogeneous data acquisition module, a multi-physics field coupling optimization module, an intelligent execution control module, a virtual-real fusion verification module and a man-machine collaborative decision module; by deploying the melt state monitoring unit, the mold dynamic thermal field sensing unit and the equipment health monitoring unit, all-around and high-precision monitoring of melt, mold and equipment states is realized, and meanwhile, space-time alignment and noise elimination are performed on multi-source data through a Kalman filtering algorithm; and generating a characteristic matrix containing a melt fluidity index, a mold temperature gradient field and an equipment health state, and providing a high-quality data basis for subsequent optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of aluminum alloy part casting process, and particularly relates to a new energy aluminum alloy part casting process multi-parameter collaborative optimization system and a method thereof. BACKGROUND

[0002] The new energy aluminum alloy part casting process multi-parameter collaborative optimization system for automobile part production is a comprehensive intelligent manufacturing system integrating advanced data acquisition, multi-physical field coupling simulation, intelligent execution control, virtual-real fusion verification and man-machine collaborative decision-making functions. The system mainly aims at the lightweight, high-performance and efficient production requirements of new energy vehicles, constructs a die casting process digital twin model by real-time acquisition of melt state, mold temperature field, equipment running state and other multi-source heterogeneous data, and realizes full-cycle simulation and optimization of filling, solidification and stress.

[0003] The existing new energy aluminum alloy part casting process multi-parameter collaborative optimization system often only relies on a single or a small number of sensors, and the data acquisition dimension is limited, which cannot fully reflect the complex state of the die casting process.

[0004] In view of the problem that the existing new energy aluminum alloy part casting process multi-parameter collaborative optimization system often only relies on a single or a small number of sensors, and the data acquisition dimension is limited, which cannot fully reflect the complex state of the die casting process, the present application realizes all-around and high-precision monitoring of the melt, mold and equipment state by deploying a melt state monitoring unit, a mold dynamic thermal field sensing unit and an equipment health monitoring unit, and simultaneously generates a feature matrix containing melt fluidity index, mold temperature gradient field and equipment health state by using Kalman filtering algorithm for spatio-temporal alignment and noise elimination of multi-source data, thereby providing a high-quality data basis for subsequent optimization. SUMMARY

[0005] In order to overcome the problems proposed in the background art, the present application proposes a new energy aluminum alloy part casting process multi-parameter collaborative optimization system and a method thereof.

[0006] The technical scheme of the present application is as follows: a new energy aluminum alloy part casting process multi-parameter collaborative optimization system, which comprises the following modules:

[0007] A multi-element heterogeneous data acquisition module is used to acquire a plurality of multi-source heterogeneous data in real time, and the acquired basic data is transmitted to the system.

[0008] A multi-physical field coupling optimization module is used to construct a die casting process digital twin model, simulate a plurality of parameters for collaborative solution, and generate an optimal parameter combination.

[0009] An intelligent execution control module is used to optimize the parameter combination output by the optimization module for precise closed-loop control.

[0010] Virtual-real fusion verification module: combine historical defect database and real-time simulation results, overlay predicted defect positions on physical mold through AR glasses;

[0011] Man-machine collaborative decision-making module: used to provide a three-dimensional visual monitoring interface to display the dynamic changes of melt flow field and temperature field in real time.

[0012] As a preferred, the multi-source heterogeneous data acquisition module comprises:

[0013] A101: melt state monitoring unit, including a dual-color temperature meter, an ultrasonic liquid level sensor and an online viscometer, for real-time monitoring of aluminum alloy melt temperature gradient, fluidity and oxidation inclusion content, to provide basic data for parameter optimization;

[0014] A102: mold dynamic thermal field sensing unit, including an infrared thermal imaging array and a fiber Bragg grating sensor, for building a three-dimensional temperature field model of the mold to predict local hot spots and solidification sequence;

[0015] A103: equipment health monitoring unit, including a vibration sensor, a hydraulic oil pressure sensor and a mold wear monitoring module, for real-time assessment of equipment fatigue state to avoid process fluctuations.

[0016] As a preferred, the multi-physical field coupling optimization module comprises:

[0017] A201: digital twin modeling unit, including a high-performance computing server cluster, a GPU accelerator card and a process simulation software, for building a digital twin of the die casting process to realize full-cycle simulation of filling, solidification and stress;

[0018] A202: parameter collaborative optimization unit, including a multi-objective genetic algorithm library, a neural network prediction model and a process parameter database, for optimizing injection speed, mold temperature and holding time based on Pareto frontier;

[0019] A203: dynamic boundary compensation unit, including an environmental parameter acquisition module and a real-time compensation algorithm, for dynamically adjusting the process window according to environmental fluctuations to ensure batch-to-batch stability.

[0020] As a preferred, the intelligent execution control module comprises:

[0021] A301: multi-axis collaborative control unit, including an EtherCAT bus controller, a servo motor drive system and a proportional overflow valve group, for realizing μs-level synchronous control of the injection mechanism, mold opening and closing, and cooling system;

[0022] A302: Process parameter closed-loop adjustment unit, including PID adaptive controller, fuzzy logic decision module and solid-state relay group, for real-time adjustment of mold temperature gradient distribution according to optimization results;

[0023] A303: Abnormal state protection unit, including emergency stop safety module, redundant sensor system and process parameter out-of-limit alarm, for preventing casting defects caused by parameter over-limit during die casting process.

[0024] As a preferred embodiment, the virtual-real fusion verification module comprises:

[0025] A401: Augmented reality defect prediction unit, including AR glasses, defect feature database and real-time rendering engine, for predicting casting shrinkage or crack position through historical data and superimposed display on physical mold;

[0026] A402: Process window exploration unit, including Monte Carlo simulation module and parameter sensitivity analysis software, for testing extreme process conditions in virtual environment and expanding feasible region boundary;

[0027] A403: Digital certificate generation unit, including blockchain storage module and process parameter hash algorithm, for generating tamper-proof digital certificate for optimized process scheme.

[0028] As a preferred embodiment, the man-machine collaborative decision-making module comprises:

[0029] A501: Three-dimensional visualization monitoring unit, including 4K industrial display screen and process parameter holographic projection, for real-time display of melt flow field, temperature field and stress field dynamic changes;

[0030] A502: Remote expert collaboration unit, including 5G communication module, remote desktop protocol and digital signature system, for supporting real-time consultation of multi-site experts and online modification of optimization target weight;

[0031] A503: Self-learning unit, including reinforcement learning framework and knowledge graph database, for automatically updating optimization model according to new batch data and realizing closed-loop evolution.

[0032] A new energy aluminum alloy part casting process multi-parameter collaborative optimization method, comprising a new energy aluminum alloy part casting process multi-parameter collaborative optimization system as described above, comprising the following steps:

[0033] S101: First, install multiple data acquisition devices at key nodes of the die casting equipment, and dynamically collect and preprocess multi-source heterogeneous data;

[0034] S102: Perform multi-physical field coupling modeling, dynamically adjust optimization model parameters according to environmental temperature and humidity data through transfer learning framework, and perform parameter collaborative optimization;

[0035] S103: The EtherCAT bus controller receives the optimization parameters, drives the servo motor and proportional valve to perform μs-level synchronous control.

[0036] As preferred, when performing dynamic acquisition and preprocessing of multi-source heterogeneous data, the following steps are included:

[0037] S201: Dual-color temperature measuring instruments, ultrasonic liquid level sensors, and fiber Bragg grating sensors are installed at key nodes of the die casting equipment to synchronously collect melt state, mold thermal distribution, and equipment vibration data;

[0038] S202: Kalman filtering algorithm is used to perform time and space alignment and noise elimination on multi-source data, generating a feature matrix containing melt fluidity index, mold temperature gradient field, and equipment health status;

[0039] S203: Wavelet packet transform is used to extract high-frequency mutation characteristics of the temperature field, and modal analysis is used to identify equipment vibration modes, forming a 256-dimensional feature vector for subsequent optimization.

[0040] As preferred, when performing multi-physical field coupling modeling and parameter collaborative optimization, the following steps are included:

[0041] S301: Based on ProCAST software, a multi-physical field coupling model of the die casting process is established, mold CAD geometric data is imported, boundary conditions of hot runner, mold temperature control, and alloy composition are set, and the filling-solidification-stress full-cycle process is simulated;

[0042] S302: NSGA-II algorithm is used to perform multi-objective optimization on injection speed, mold partition temperature gradient, and holding time parameters, target functions include casting density, residual stress, and production cycle, and a Pareto optimal solution set is generated;

[0043] S303: Transfer learning framework, dynamically adjust optimization model parameters according to environmental temperature and humidity data, compensate for the influence of external disturbances on the process window.

[0044] As preferred, when performing intelligent execution and virtual-real fusion verification, the following steps are included:

[0045] S401: The EtherCAT bus controller receives the optimization parameters, drives the servo motor and proportional valve to perform μs-level synchronous control of the injection curve and mold opening and closing timing, and the preset execution precision is 98% of the set value;

[0046] S402: Based on the historical defect database, LSTM network is used to predict the location of casting shrinkage or cracks, and the prediction result is superimposed on the physical mold through AR glasses to assist process personnel in advance intervention;

[0047] S403: Robustness testing of the optimized parameter combination is performed using Monte Carlo simulation, generating a dataset containing 10... 4 This virtual experiment, conducted at the Pareto frontier, verifies the feasibility of parameter combinations under extreme conditions.

[0048] The beneficial effects of this invention are:

[0049] 1. Compared with existing multi-parameter collaborative optimization systems for casting processes of new energy aluminum alloy parts, which often rely on only a single or a few sensors and have limited data acquisition dimensions, this solution deploys a melt state monitoring unit, a mold dynamic thermal field sensing unit, and an equipment health monitoring unit to achieve comprehensive and high-precision monitoring of the melt, mold, and equipment status. At the same time, the Kalman filtering algorithm is used to perform spatiotemporal alignment and noise elimination on multi-source data to generate a feature matrix containing melt flowability index, mold temperature gradient field, and equipment health status, providing a high-quality data foundation for subsequent optimization.

[0050] 2. Compared with existing multi-parameter collaborative optimization systems for casting processes of new energy aluminum alloy parts, which mostly use single-physics field modeling and ignore the multi-physics field coupling effect, resulting in low model accuracy and optimization algorithms that are mostly single-objective or simple multi-objective optimizations, making it difficult to find the global optimum, this solution establishes a multi-physics field coupling model of the die casting process based on ProCAST software, imports mold CAD geometric data, sets hot runner, mold temperature control, and alloy composition boundary conditions to achieve full-cycle simulation of filling-solidification-stress. The NSGA-II algorithm is used to perform multi-objective optimization of parameters such as injection speed, mold zone temperature gradient, and holding time. The objective functions include casting density, residual stress, and production cycle, generating a Pareto optimal solution set.

[0051] 3. Compared with existing multi-parameter collaborative optimization systems for casting processes of new energy aluminum alloy parts, which mostly rely on experience-based judgment or simple statistical analysis, resulting in low prediction accuracy and a lack of virtual-real fusion verification methods, the process scheme verification cycle is long and costly. This solution is based on a historical defect database and uses an LSTM network to predict the location of shrinkage cavities or cracks in the casting. The prediction results are superimposed on the physical mold through AR glasses to assist process personnel in early intervention. At the same time, the digital certificate generation unit generates an unalterable digital certificate for the optimized process scheme to ensure the traceability and credibility of the process scheme. Attached Figure Description

[0052] Figure 1 The diagram shown is a schematic of the process framework of a multi-parameter collaborative optimization system for casting new energy aluminum alloy parts according to the present invention.

[0053] Figure 2The diagram shown is a schematic of the multi-source heterogeneous data dynamic acquisition and preprocessing flow of a multi-parameter collaborative optimization method for casting process of new energy aluminum alloy parts according to the present invention.

[0054] Figure 3 The diagram shown is a schematic of the multi-physics field coupling modeling and parameter collaborative optimization process of a multi-parameter collaborative optimization method for casting new energy aluminum alloy parts according to the present invention.

[0055] Figure 4 The diagram shows the intelligent execution and virtual-real fusion verification process of a multi-parameter collaborative optimization method for casting processes of new energy aluminum alloy parts according to the present invention. Detailed Implementation

[0056] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0057] Please see Figure 1 This invention provides an embodiment: a multi-parameter collaborative optimization system for the casting process of new energy aluminum alloy components, comprising the following modules:

[0058] Multi-source heterogeneous data acquisition module: used to acquire multiple heterogeneous data sources in real time and transmit the acquired basic data to the system;

[0059] Multiphysics coupling optimization module: used to build a digital twin model of the die casting process, simulate multiple parameters for collaborative solution, and generate the optimal parameter combination;

[0060] Intelligent execution control module: used to optimize the parameter combination of the module output for precise closed-loop control;

[0061] Virtual-Real Fusion Verification Module: Combining historical defect databases with real-time simulation results, the predicted defect locations are overlaid and displayed on the physical mold using AR glasses;

[0062] Human-machine collaborative decision-making module: used to provide a three-dimensional visualization monitoring interface to display the dynamic changes of melt flow field and temperature field in real time.

[0063] As a preferred option, the multi-source heterogeneous data acquisition module includes:

[0064] A101: Melt condition monitoring unit, including a dual-color thermometer, an ultrasonic level sensor and an online viscometer, is used to monitor the temperature gradient, fluidity and oxide inclusion content of aluminum alloy melt in real time, providing basic data for parameter optimization;

[0065] A102: Mold dynamic thermal field sensing unit, including infrared thermal imaging array and fiber Bragg grating sensor, used to construct a three-dimensional temperature field model of the mold and predict local hot spots and solidification sequence;

[0066] A103: Equipment health monitoring unit, including vibration sensor, hydraulic oil pressure sensor and mold wear monitoring module, is used to assess equipment fatigue status in real time and avoid process fluctuations.

[0067] As a preferred option, the multiphysics coupling optimization module includes:

[0068] A201: Digital Twin Modeling Unit, including a high-performance computing server cluster, GPU accelerator card and process simulation software, is used to construct a digital twin of the die casting process and realize full-cycle simulation of filling-solidification-stress.

[0069] A202: Parameter Co-optimization Unit, including a multi-objective genetic algorithm library, neural network prediction model and process parameter database, is used to optimize injection speed, mold temperature and holding time based on Pareto front.

[0070] A203: Dynamic boundary compensation unit, including an environmental parameter acquisition module and a real-time compensation algorithm, is used to dynamically adjust the process window according to environmental fluctuations to ensure the stability of different batches.

[0071] Preferably, the intelligent execution control module includes:

[0072] A301: Multi-axis collaborative control unit, including EtherCAT bus controller, servo motor drive system and proportional overflow valve group, used to realize μs-level synchronous control of injection mechanism, mold opening and closing, and cooling system;

[0073] A302: Closed-loop adjustment unit for process parameters, including a PID adaptive controller, a fuzzy logic decision module and a solid-state relay group, used to adjust the mold temperature gradient distribution in real time according to the optimization results;

[0074] A303: Abnormal state protection unit, including emergency stop safety module, redundant sensor system and process parameter over-limit alarm, used to prevent casting defects caused by parameter over-limit during the die casting process.

[0075] As a preferred option, the virtual-real fusion verification module includes:

[0076] A401: Augmented Reality Defect Prediction Unit, including AR glasses, a defect feature database and a real-time rendering engine, is used to predict the location of shrinkage cavities or cracks in castings using historical data and overlay them onto the solid mold.

[0077] A402: Process window exploration unit, including Monte Carlo simulation module and parameter sensitivity analysis software, is used to test extreme process conditions in a virtual environment and expand the boundaries of the feasible domain.

[0078] A403: Digital certificate generation unit, including blockchain evidence storage module and process parameter hash algorithm, used to generate tamper-proof digital certificates for optimized process solutions.

[0079] As a preferred option, the human-machine collaborative decision-making module includes:

[0080] A501: A 3D visualization monitoring unit, including a 4K industrial display screen and a holographic projection of process parameters, used to display the dynamic changes of melt flow field, temperature field and stress field in real time;

[0081] A502: Remote Expert Collaboration Unit, including a 5G communication module, remote desktop protocol and digital signature system, is used to support real-time consultations among experts in multiple locations and to modify and optimize target weights online;

[0082] A503: Autonomous Learning Unit, comprising a reinforcement learning framework and a knowledge graph database, used to automatically update and optimize the model based on new batches of data, achieving closed-loop evolution.

[0083] Please see Figures 2-4 In this embodiment, a multi-parameter collaborative optimization method for casting processes of new energy aluminum alloy components includes the following steps:

[0084] S101: First, multiple data acquisition devices are installed at key nodes of the die-casting equipment to dynamically acquire and preprocess multi-source heterogeneous data.

[0085] S102: Perform multiphysics coupling modeling, and dynamically adjust and optimize model parameters based on environmental temperature and humidity data through a transfer learning framework to perform parameter co-optimization;

[0086] S103: The EtherCAT bus controller receives optimized parameters and drives the servo motor and proportional valve for μs-level synchronous control.

[0087] Preferably, the dynamic acquisition and preprocessing of multi-source heterogeneous data includes the following steps:

[0088] S201: Install a dual-color thermometer, an ultrasonic level sensor, and a fiber Bragg grating sensor at key nodes of the die-casting equipment to simultaneously collect data on melt state, mold heat distribution, and equipment vibration.

[0089] S202: The Kalman filtering algorithm is used to perform spatiotemporal alignment and noise removal on multi-source data to generate a feature matrix containing melt flowability index, mold temperature gradient field and equipment health status.

[0090] S203: High-frequency abrupt change features of the temperature field are extracted by wavelet packet transform, and the vibration modes of the equipment are identified by modal analysis to form a 256-dimensional feature vector for subsequent optimization.

[0091] Preferably, the following steps are included when performing multiphysics coupling modeling and parameter co-optimization:

[0092] S301: Based on ProCAST software, a multi-physics coupling model of the die casting process is established, mold CAD geometric data is imported, hot runner, mold temperature control and alloy composition boundary conditions are set, and the full cycle process of filling-solidification-stress is simulated.

[0093] S302: The NSGA-II algorithm is used to perform multi-objective optimization of parameters such as injection speed, mold zone temperature gradient and holding time. The objective functions include casting density, residual stress and production cycle, and a Pareto optimal solution set is generated.

[0094] S303: Transfer learning framework, which dynamically adjusts and optimizes model parameters based on environmental temperature and humidity data to compensate for the impact of external disturbances on the process window.

[0095] As a preferred method, the following steps are included when performing intelligent execution and virtual-real fusion verification:

[0096] S401: The EtherCAT bus controller receives optimized parameters and drives the servo motor and proportional valve to perform μs-level synchronous control of the injection curve and mold opening and closing sequence, with a preset execution accuracy of 98% of the set value.

[0097] S402: Based on a historical defect database, an LSTM network is used to predict the location of shrinkage cavities or cracks in castings. The prediction results are then superimposed on the physical mold using AR glasses to assist process engineers in early intervention.

[0098] S403: Robustness testing of the optimized parameter combination is performed using Monte Carlo simulation, generating a dataset containing 10... 4 This virtual experiment, conducted at the Pareto frontier, verifies the feasibility of parameter combinations under extreme conditions.

[0099] Example

[0100] A certain model of new energy vehicle requires the production of battery trays. The traditional process uses gravity casting, which has problems such as shrinkage cavities and large dimensional fluctuations.

[0101] S501: The data acquisition module monitors the melt temperature gradient and mold thermal field in real time. The data is processed by Kalman filtering to generate a three-dimensional temperature field model.

[0102] S502: The multiphysics coupling module uses ProCAST to simulate the filling process, combines the NSGA-II algorithm to optimize the injection speed, and sets the mold temperature gradient from a fixed 280℃ to a gradient of 260℃ / 275℃ / 290℃ in the left / right / middle zones.

[0103] S503: The intelligent execution module controls the servo motor via EtherCAT bus to achieve μs-level synchronous injection.

[0104] The mold cooling system dynamically adjusts the water flow path based on optimized parameters;

[0105] S504: The virtual-real fusion module predicts the location of shrinkage cavities in castings, and AR glasses guide workers to place chills in advance; S505: The human-machine collaboration module allows remote experts to adjust target weights in real time via 5G network, and after the system's closed-loop evolution, the process yield rate increases from 51% to 69%.

[0106]

[0107]

[0108] Conclusion: The casting qualification rate increased from 85% to over 96%, and the annual production capacity of a single line increased by 30%.

[0109] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A multi-parameter collaborative optimization system for casting processes of new energy aluminum alloy components; characterized in that: It consists of the following modules: Multi-source heterogeneous data acquisition module: used to acquire multiple heterogeneous data sources in real time and transmit the acquired basic data to the system; Multiphysics coupling optimization module: used to build a digital twin model of the die casting process, simulate multiple parameters for collaborative solution, and generate the optimal parameter combination; Intelligent execution control module: used to optimize the parameter combination of the module output for precise closed-loop control; Virtual-Real Fusion Verification Module: Combining historical defect databases with real-time simulation results, the predicted defect locations are overlaid and displayed on the physical mold using AR glasses; Human-machine collaborative decision-making module: used to provide a three-dimensional visualization monitoring interface to display the dynamic changes of melt flow field and temperature field in real time.

2. The multi-parameter collaborative optimization system for casting process of new energy aluminum alloy parts according to claim 1, characterized in that: The multi-source heterogeneous data acquisition module includes: A101: Melt condition monitoring unit, including a dual-color thermometer, an ultrasonic level sensor and an online viscometer, is used to monitor the temperature gradient, fluidity and oxide inclusion content of aluminum alloy melt in real time, providing basic data for parameter optimization; A102: Mold dynamic thermal field sensing unit, including infrared thermal imaging array and fiber Bragg grating sensor, used to construct a three-dimensional temperature field model of the mold and predict local hot spots and solidification sequence; A103: Equipment health monitoring unit, including vibration sensor, hydraulic oil pressure sensor and mold wear monitoring module, is used to assess equipment fatigue status in real time and avoid process fluctuations.

3. The multi-parameter collaborative optimization system for casting process of new energy aluminum alloy parts according to claim 2, characterized in that: The multiphysics coupling optimization module includes: A201: Digital Twin Modeling Unit, including a high-performance computing server cluster, GPU accelerator card and process simulation software, is used to construct a digital twin of the die casting process and realize full-cycle simulation of filling-solidification-stress. A202: Parameter Co-optimization Unit, including a multi-objective genetic algorithm library, neural network prediction model and process parameter database, is used to optimize injection speed, mold temperature and holding time based on Pareto front. A203: Dynamic boundary compensation unit, including an environmental parameter acquisition module and a real-time compensation algorithm, is used to dynamically adjust the process window according to environmental fluctuations to ensure the stability of different batches.

4. The multi-parameter collaborative optimization system for casting process of new energy aluminum alloy parts according to claim 3, characterized in that: The intelligent execution control module includes: A301: Multi-axis collaborative control unit, including EtherCAT bus controller, servo motor drive system and proportional overflow valve group, used to realize μs-level synchronous control of injection mechanism, mold opening and closing, and cooling system; A302: Closed-loop adjustment unit for process parameters, including a PID adaptive controller, a fuzzy logic decision module and a solid-state relay group, used to adjust the mold temperature gradient distribution in real time according to the optimization results; A303: Abnormal state protection unit, including emergency stop safety module, redundant sensor system and process parameter over-limit alarm, used to prevent casting defects caused by parameter over-limit during the die casting process.

5. The multi-parameter collaborative optimization system for casting process of new energy aluminum alloy parts according to claim 4, characterized in that: The virtual-real fusion verification module includes: A401: Augmented Reality Defect Prediction Unit, including AR glasses, a defect feature database and a real-time rendering engine, is used to predict the location of shrinkage cavities or cracks in castings using historical data and overlay them onto the solid mold. A402: Process window exploration unit, including Monte Carlo simulation module and parameter sensitivity analysis software, is used to test extreme process conditions in a virtual environment and expand the boundaries of the feasible domain. A403: Digital certificate generation unit, including blockchain evidence storage module and process parameter hash algorithm, used to generate tamper-proof digital certificates for optimized process solutions.

6. The multi-parameter collaborative optimization system for casting process of new energy aluminum alloy parts according to claim 5, characterized in that: The human-machine collaborative decision-making module includes: A501: A 3D visualization monitoring unit, including a 4K industrial display screen and a holographic projection of process parameters, used to display the dynamic changes of melt flow field, temperature field and stress field in real time; A502: Remote Expert Collaboration Unit, including a 5G communication module, remote desktop protocol and digital signature system, is used to support real-time consultations among experts in multiple locations and to modify and optimize target weights online; A503: Autonomous learning unit, including a reinforcement learning framework and a knowledge graph database, used to automatically update and optimize the model based on new batches of data, achieving closed-loop evolution.

7. A multi-parameter collaborative optimization method for casting processes of new energy aluminum alloy components, characterized in that... The multi-parameter collaborative optimization system for casting process of new energy aluminum alloy parts according to claim 6 includes the following steps: S101: First, multiple data acquisition devices are installed at key nodes of the die-casting equipment to dynamically acquire and preprocess multi-source heterogeneous data. S102: Perform multiphysics coupling modeling, and dynamically adjust and optimize model parameters based on environmental temperature and humidity data through a transfer learning framework to perform parameter co-optimization; S103: The EtherCAT bus controller receives optimized parameters and drives the servo motor and proportional valve for μs-level synchronous control.

8. The multi-parameter collaborative optimization method for casting process of new energy aluminum alloy parts according to claim 7, characterized in that: The following steps are included in the dynamic acquisition and preprocessing of multi-source heterogeneous data: S201: Install a dual-color thermometer, an ultrasonic level sensor, and a fiber Bragg grating sensor at key nodes of the die-casting equipment to simultaneously collect data on melt state, mold heat distribution, and equipment vibration. S202: The Kalman filtering algorithm is used to perform spatiotemporal alignment and noise removal on multi-source data to generate a feature matrix containing melt flowability index, mold temperature gradient field and equipment health status. S203: High-frequency abrupt change features of the temperature field are extracted by wavelet packet transform, and the vibration modes of the equipment are identified by modal analysis to form a 256-dimensional feature vector for subsequent optimization.

9. The multi-parameter collaborative optimization method for casting process of new energy aluminum alloy parts according to claim 7, characterized in that: The following steps are included in multiphysics coupling modeling and parameter co-optimization: S301: Based on ProCAST software, a multi-physics coupling model of the die casting process is established, mold CAD geometric data is imported, hot runner, mold temperature control and alloy composition boundary conditions are set, and the full cycle process of filling-solidification-stress is simulated. S302: The NSGA-II algorithm is used to perform multi-objective optimization of parameters such as injection speed, mold zone temperature gradient and holding time. The objective functions include casting density, residual stress and production cycle, and a Pareto optimal solution set is generated. S303: Transfer learning framework, which dynamically adjusts and optimizes model parameters based on environmental temperature and humidity data to compensate for the impact of external disturbances on the process window.

10. The multi-parameter collaborative optimization method for casting process of new energy aluminum alloy parts according to claim 7, characterized in that: The following steps are included in the intelligent execution and virtual-real fusion verification process: S401: The EtherCAT bus controller receives optimized parameters and drives the servo motor and proportional valve to perform μs-level synchronous control of the injection curve and mold opening and closing sequence, with a preset execution accuracy of 98% of the set value. S402: Based on a historical defect database, an LSTM network is used to predict the location of shrinkage cavities or cracks in castings. The prediction results are then superimposed on the physical mold using AR glasses to assist process engineers in early intervention. S403: Robustness testing of the optimized parameter combination is performed using Monte Carlo simulation, generating a dataset containing 10... 4 This virtual experiment, conducted at the Pareto frontier, verifies the feasibility of parameter combinations under extreme conditions.

Images