A data processing method for intelligent control processing system
Through distributed multi-sensor networks and nonlinear mapping, time evolution networks and feedback networks, multi-dimensional data processing problems in high-end manufacturing environments are solved, real-time optimization and prediction of intelligent control systems are realized, and processing accuracy and efficiency are improved.
Patent Information
- Application Number
- CN202411360565.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-27
- Publication Date
- 2024-12-20
- Estimated Expiration
- 2044-09-27
AI Technical Summary
Existing manufacturing systems are difficult to effectively monitor and process multi-dimensional data in complex high-end manufacturing environments, resulting in low control accuracy and efficiency, and real-time adjustment and prediction cannot be achieved, affecting processing quality and efficiency.
A distributed multi-sensor network is used to collect multi-dimensional data, perform feature conversion through nonlinear mapping functions, and use the time evolution network to predict future states, and dynamically adjust processing parameters and optimize control strategies based on the nonlinear feedback network and global optimization mechanism.
It realizes comprehensive monitoring and real-time optimization of the processing system, improves the system's adaptability and intelligence level, reduces processing errors and delays, and ensures the stability and accuracy of the processing process.
Smart Images
Figure CN118884841B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data processing of intelligent control processing systems, and in particular to a data processing method of intelligent control processing systems. Background Art
[0002] In the modern industrial manufacturing process, with the continuous improvement of product quality, production efficiency and equipment automation requirements, traditional processing systems have gradually become unable to meet the increasingly complex manufacturing needs. Especially in high-end manufacturing fields such as aerospace, automobile manufacturing, and precision instruments, the processing process involves a large number of equipment operating parameters, workpiece status, and environmental conditions. These factors interact and cross-influence each other, and put forward higher requirements for the control accuracy and processing effect of the system. At the same time, with the refinement of manufacturing technology, enterprises hope to improve the automation of production lines through more intelligent control systems, reduce dependence on manual operations, and reduce processing errors caused by human factors.
[0003] The current manufacturing environment is often full of uncertainty. The status of the equipment, the characteristics of the workpiece, and changes in the environment will directly affect the accuracy and efficiency of the processing process. Therefore, how to use intelligent control systems to monitor these complex factors in a comprehensive and real-time manner, and how to achieve intelligent adjustment of processing parameters through efficient data processing methods, has become a key challenge to improve manufacturing quality and production efficiency.
[0004] Most existing technologies use linear processing models or simple control logic, which cannot effectively deal with complex nonlinear relationships and the interactive effects of multi-dimensional data, and the control accuracy is difficult to guarantee. They cannot optimize processing parameters through deep feature mining, nor can they dynamically adjust and optimize according to real-time data, which affects the processing quality and efficiency. They fail to effectively use time series data to predict future states, making it difficult to achieve early adjustment of equipment and workpiece states, resulting in a delayed response of the system to emergencies. Summary of the invention
[0005] In view of the deficiencies in the above-mentioned prior art, the present invention provides a data processing method for an intelligent control processing system.
[0006] The present invention provides a data processing method for an intelligent control processing system, which specifically includes the following technical solutions:
[0007] A data processing method for an intelligent control processing system comprises the following steps:
[0008] S1. Collect multi-dimensional data from processing equipment and environment, use nonlinear mapping function to transform the data features, and use the mapped feature vector to predict the future processing state through the time evolution network;
[0009] S2. Based on the prediction of future processing status, a nonlinear feedback network is used to dynamically adjust processing parameters and optimize control strategies. Through the global optimization mechanism, global evaluation and adjustment are performed on multiple time nodes to ensure that the intelligent control processing system can maintain the optimal state at different stages.
[0010] Furthermore, the multi-dimensional data acquisition in S1 collects processing data from multiple dimensions through a distributed multi-sensor network. The multi-dimensional data comes from multiple sensors arranged at various key positions of the processing equipment to capture various status information of the environment and equipment. Each sensor can feedback different state variables in real time. The multi-dimensional data includes equipment operating parameters, workpiece status data, environmental variables, and historical processing data.
[0011] Furthermore, in the distributed multi-sensor network, each sensor node collects data in real time and performs preliminary noise reduction and filtering operations to remove random errors and interference signals in the data to ensure the accuracy of the data; each sensor collects data according to a specific time period and integrates the data into a time series. ,in, , Indicated by The data value collected by the sensor, that is, Collect data in multiple dimensions; is the total number of dimensions;
[0012] The nonlinear mapping function is used to transform the collected data after preliminary processing. The mapping process not only retains the characteristic information of the input data, but also enhances the adaptability to different processing scenarios by adding nonlinear dynamic weights. The mapping formula of the input data is:
[0013] ,in, Represents the feature vector after mapping; is the nonlinear mapping weight, controlling the The degree of influence of the input data of each dimension on the final feature map; is the frequency parameter in the sine function, which is used to adjust the input data cyclical changes; It is the phase shift in the sine function, which controls how the input data is relatively shifted in different dimensions and adjusts the synchronization of each data dimension; is the exponential decay coefficient, which controls the decay rate of input data in different dimensions with time or input amplitude changes; is the frequency parameter in the cosine function, which represents the oscillation frequency of each data dimension and adjusts the periodic characteristics of different data dimensions; in particular, It represents the rate of change of each dimension data in the time domain, and can reflect the dynamic characteristics of real-time data changing over time.
[0014] Furthermore, the data processed by nonlinear mapping is not directly used for control decision-making, but for adaptive dynamic regulation. The mapped feature vector is processed by the time evolution network to predict the future processing state and perform multiple adaptive iterative adjustments. The state changes of the intelligent control processing system in the future are determined by analyzing the time series of the mapped feature vector. The formula for time evolution is:
[0015] ,in, Indicates that intelligent control processing system Status prediction at each moment; represents the number of input features considered during the time evolution; For the The input mapping feature value comes from the mapping formula Dimensional data features; is the oscillation amplitude parameter, which controls the degree of fluctuation of the input feature in time; is the angular frequency, which represents the oscillation frequency of the feature in the time dimension; is a noise term, simulating random disturbances from the external environment. Through this time evolution formula, the state evolution trajectory of the processing equipment in the future can be predicted in the time domain.
[0016] In the formula represents the acceleration term, and the changing trend of the system state during the entire processing process is calculated by integration. The periodic oscillation is simulated, which can reflect the influence of periodic disturbance in the machining process.
[0017] Furthermore, based on the prediction of future states, a nonlinear feedback network is used to dynamically adjust the processing parameters, and the processing parameters are optimized through recursive feedback control by combining the current state prediction and historical data.
[0018] Furthermore, the recursive feedback formula for control decision is:
[0019]
[0020] ,in, is the initial control parameter, based on the initial state and the initial mapping features , calculate the initial control parameters; for The moment control parameters, such as spindle speed or tool feed rate, is the learning rate, controlling the step size of the adjustment, is the output after time evolution state features, represented by System status value at the moment; Indicates The sensitivity parameter of each state characteristic controls the response degree of the intelligent control processing system to the state feedback; For the The historical control weight of each state feature, which controls the influence of the history on the current control strategy; is the number of feedback dimensions, indicating the number of feature dimensions involved in feedback control calculation; and is the weight coefficient of the initial optimization, which is used to balance the influence of mapping characteristics and state estimation on control parameters;
[0021] In the recursive feedback formula is the activation function, ensuring that the control parameters respond nonlinearly to the state evolution results. is the adjustment amplitude function of the control parameter, which is used to ensure the stability of the control parameter in extreme cases; the historical feedback term in the formula Combining past control parameter change rates with current status ensures learning from historical data and continuous optimization of future control strategies.
[0022] Furthermore, in order to ensure the global optimality of the intelligent control machining system, a global optimization and multiple feedback mechanism is introduced. By globally evaluating and adjusting the operating status at multiple time nodes, the weights and feedback paths at each level are dynamically updated to ensure that each stage of the machining process can obtain an optimized control strategy.
[0023] Furthermore, the specific recursive adjustment formula of the global optimization and multiple feedback mechanism is: ,in, represents the result of global optimization, i.e., the feedback adjustment after global evaluation of the control parameters at multiple time nodes, are the control parameters at different time nodes, Indicates Time nodes; Indicates the number of time nodes for global optimization; is the time decay coefficient, which controls the influence of historical data on the current feedback path during the global optimization process. is the global optimization coefficient, which controls the intensity of global feedback adjustment. for The feature vector at the moment; is the entire time range over which the state is accumulated;
[0024] Perform a global evaluation at each time node, and optimize and adjust the feedback path based on current and historical data to ensure that the entire intelligent control processing system can respond to changes in the external environment and processing status in real time; Represents the acceleration of the control parameter, which is mapped to the input feature The combination of time decay coefficient Control the impact of historical data on current feedback; integral term The global state information of the time evolution layer is combined with the control decision to ensure that the system can maintain the global optimum during long-term operation.
[0025] The beneficial effects of the technical solution of the present invention are:
[0026] 1. The present invention collects multi-dimensional data from equipment, workpieces and environment in real time through a distributed multi-sensor network, ensuring comprehensive monitoring of various operating parameters of the processing system, workpiece status and environmental variables; by introducing nonlinear mapping functions, feature conversion is performed on data of different dimensions, key information of the data is retained, and adaptability to complex processing scenarios is enhanced; the mapped data is processed through a complex mathematical model, which effectively improves the system's ability to understand the processing status and ensures the accuracy and real-time performance of data collection and processing;
[0027] 2. By designing a time evolution network, the present invention can perform time series analysis of the system state based on the collected data and predict the operating state of the processing equipment in the future; the time evolution formula introduces the processing of acceleration terms and periodic oscillations, simulating the dynamic changes of the equipment state during the processing process, so that the system can adjust the control parameters in advance according to the upcoming state changes, thereby reducing delays and errors in the processing process, and significantly improving the system's adaptability and intelligence level;
[0028] 3. The present invention designs a new nonlinear feedback network, which dynamically optimizes the processing parameters through recursive feedback control. It not only combines the current system state prediction, but also comprehensively considers the influence of historical data, so as to ensure that the control parameters can respond to the state changes of the system in real time; the activation function and historical feedback term in the recursive formula effectively ensure the stability and continuous optimization of the control parameters, avoid control errors caused by extreme situations during the processing, and improve the stability and accuracy of the entire processing process;
[0029] 4. Global optimization and multiple feedback mechanisms evaluate multiple time nodes of the processing system, and control the impact of historical data on the current control strategy through the time decay coefficient; combine the input mapping characteristics and the cumulative feedback of the system status to ensure that the system can continue to perform global optimization; ensure that in the long-term operation process, the intelligent control processing system can maintain the global optimal state, achieve efficient control and quality assurance in the processing process, and reduce the system performance degradation caused by long-term factors such as equipment aging and environmental changes. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 It is a system flow chart of the present invention. DETAILED DESCRIPTION
[0031] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0032] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0033] The specific scheme of the data processing method of an intelligent control processing system provided by the present invention is described in detail below with reference to the accompanying drawings.
[0034] Refer to the attached Figure 1 , which shows a flow chart of a data processing method for an intelligent control processing system provided by an embodiment of the present invention, the method comprising the following steps:
[0035] S1. Collect multi-dimensional data from processing equipment and environment, use nonlinear mapping function to transform the data features, and use the mapped feature vector to predict the future processing state through the time evolution network;
[0036] Through the distributed multi-sensor network, processing data from multiple dimensions is collected. The input data of data collection contains multi-dimensional data, which comes from multiple sensors arranged in various key parts of the processing equipment to capture various state information of the environment and equipment. Each sensor can feedback different state variables in real time. The specific multi-dimensional data collected includes:
[0037] Equipment operating parameters: including the equipment's spindle speed, feed speed, tool speed, worktable displacement, bearing temperature, spindle vibration frequency, etc.;
[0038] Workpiece status data: Sensors installed in the workpiece processing area monitor the workpiece processing status in real time, such as surface roughness, processing accuracy, temperature gradient, etc.
[0039] Environmental variables: including temperature, humidity, air pressure and other environmental parameters of the processing workshop;
[0040] Historical processing data: Before each processing task starts, the previous operation records of the equipment, historical workpiece processing data and equipment maintenance data are automatically called;
[0041] Or other dimensional data of the prediction item according to the device requirements.
[0042] In order to ensure the real-time and integrity of the collected data, the sensor network adopts a distributed structure. Each sensor node collects data in real time and performs preliminary noise reduction and filtering operations to remove random errors and interference signals in the data to ensure the accuracy of the data. The noise reduction and filtering operations can be carried out using existing technologies. Each sensor collects data according to a specific time period and integrates the data into a time series. ,in, , Indicated by The data value collected by the sensor, that is, Collect data in multiple dimensions; is the total number of dimensions.
[0043] The nonlinear mapping function is used to transform the collected data after preliminary processing. The mapping process not only retains the characteristic information of the input data, but also enhances the adaptability to different processing scenarios by adding nonlinear dynamic weights. The mapping formula of the input data is: ,in, Represents the feature vector after mapping; is the nonlinear mapping weight, controlling the The degree of influence of the input data of each dimension on the final feature map; is the frequency parameter in the sine function, which is used to adjust the input data cyclical changes; It is the phase shift in the sine function, which controls how the input data is relatively shifted in different dimensions and adjusts the synchronization of each data dimension; is the exponential decay coefficient, which controls the decay rate of input data in different dimensions with time or input amplitude changes; is the frequency parameter in the cosine function, which represents the oscillation frequency of each data dimension and adjusts the periodic characteristics of different data dimensions; in particular, It represents the rate of change of each dimension data in the time domain, and can reflect the dynamic characteristics of real-time data changing over time.
[0044] The data processed by nonlinear mapping is not directly used for control decision-making, but for adaptive dynamic regulation. The mapped feature vector is processed by the time evolution network to predict the future processing state and make multiple adaptive iterative adjustments. By analyzing the time series of the mapped feature vector, the state changes of the intelligent control processing system in the future are determined. The formula for time evolution is: ,in, Indicates that intelligent control processing system Status prediction at each moment; represents the number of input features considered during the time evolution; For the The input mapping feature value comes from the mapping formula Dimensional data features; is the oscillation amplitude parameter, which controls the degree of fluctuation of the input feature in time; is the angular frequency, which represents the oscillation frequency of the feature in the time dimension; is a noise term, simulating random disturbances from the external environment. Through this time evolution formula, the state evolution trajectory of the processing equipment in the future can be predicted in the time domain.
[0045] In the formula represents the acceleration term, and the changing trend of the system state during the entire processing process is calculated by integration. The periodic oscillation is simulated, which can reflect the influence of periodic disturbance in the machining process.
[0046] S2. Based on the prediction of future processing status, a nonlinear feedback network is used to dynamically adjust processing parameters and optimize control strategies. Through the global optimization mechanism, global evaluation and adjustment are performed on multiple time nodes to ensure that the intelligent control processing system can maintain the optimal state at different stages.
[0047] Based on the prediction of future states, a nonlinear feedback network is used to dynamically adjust the processing parameters. Combining the current state prediction and historical data, the processing parameters are optimized through recursive feedback control. The recursive formula for control decision is:
[0048] ,in, is the initial control parameter, based on the initial state and the initial mapping features , calculate the initial control parameters; for The moment control parameters, such as spindle speed or tool feed rate, is the learning rate, controlling the step size of the adjustment, is the output after time evolution state features, represented by System status value at the moment; Indicates The sensitivity parameter of each state characteristic controls the response degree of the intelligent control processing system to the state feedback; For the The historical control weight of each state feature, which controls the influence of the history on the current control strategy; is the number of feedback dimensions, indicating the number of feature dimensions involved in feedback control calculation; and is the weight coefficient of the initial optimization, which is used to balance the influence of mapping characteristics and state estimation on the control parameters.
[0049] In the recursive feedback formula is the activation function, ensuring that the control parameters respond nonlinearly to the state evolution results. is the adjustment amplitude function of the control parameter, which is used to ensure the stability of the control parameter in extreme situations. Combining past control parameter change rates with current status ensures learning from historical data and continuous optimization of future control strategies.
[0050] In order to ensure the global optimality of the intelligent control processing system, a global optimization and multiple feedback mechanism is introduced. By globally evaluating and adjusting the operating status at multiple time nodes, the weights and feedback paths at each level are dynamically updated to ensure that each stage of the processing process can obtain an optimized control strategy. The recursive adjustment formula for global optimization is: ,in, represents the result of global optimization, i.e., the feedback adjustment after global evaluation of the control parameters at multiple time nodes, are the control parameters at different time nodes, Indicates Time nodes; Indicates the number of time nodes for global optimization; is the time decay coefficient, which controls the influence of historical data on the current feedback path during the global optimization process. is the global optimization coefficient, which controls the intensity of global feedback adjustment. for The feature vector at the moment; is the entire time range over which the state is accumulated.
[0051] Perform a global evaluation at each time node, and optimize and adjust the feedback path based on current and historical data to ensure that the entire intelligent control processing system can respond to changes in the external environment and processing status in real time; Represents the acceleration of the control parameter, which is mapped to the input feature The combination of time decay coefficient Control the impact of historical data on current feedback; integral term The global state information of the time evolution layer is combined with the control decision to ensure that the system can maintain the global optimum during long-term operation.
[0052] The present invention collects multi-dimensional data from equipment, workpieces and environment in real time through a distributed multi-sensor network, ensuring comprehensive monitoring of various operating parameters of the processing system, workpiece status and environmental variables; by introducing nonlinear mapping functions, feature conversion is performed on data of different dimensions, thereby retaining key information of the data and enhancing adaptability to complex processing scenarios; the mapped data is processed through a complex mathematical model, effectively improving the system's ability to understand the processing status and ensuring the accuracy and real-time performance of data collection and processing;
[0053] By designing a time evolution network, the present invention can perform time series analysis of the system state based on the collected data and predict the operating state of the processing equipment in the future; the time evolution formula introduces the processing of acceleration terms and periodic oscillations, simulating the dynamic changes of the equipment state during the processing process, so that the system can adjust the control parameters in advance according to the upcoming state changes, thereby reducing delays and errors in the processing process, and significantly improving the system's adaptability and intelligence level;
[0054] The present invention designs a new nonlinear feedback network, which dynamically optimizes the processing parameters through recursive feedback control. It not only combines the current system state prediction, but also comprehensively considers the influence of historical data, so as to ensure that the control parameters can respond to the state changes of the system in real time. The activation function and historical feedback term in the recursive formula effectively ensure the stability and continuous optimization of the control parameters, avoid control errors caused by extreme conditions during the processing, and improve the stability and accuracy of the entire processing process.
[0055] Global optimization and multiple feedback mechanisms evaluate multiple time nodes of the processing system and control the impact of historical data on the current control strategy through the time decay coefficient. Combining the input mapping characteristics and the cumulative feedback of the system status ensures that the system can continue to perform global optimization. This ensures that during long-term operation, the intelligent control processing system can maintain the global optimal state, achieve efficient control and quality assurance during the processing, and reduce the system performance degradation caused by long-term factors such as equipment aging and environmental changes.
[0056] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0057] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.
[0058] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.
Claims
1. A data processing method for an intelligent control processing system, It is characterized in that The following steps are involved: S1. Collect multi-dimensional data from processing equipment and environment, use nonlinear mapping function to transform the data features, and use the mapped feature vector to predict the future processing state through the time evolution network; Multi-dimensional data acquisition collects processing data from multiple dimensions through a distributed multi-sensor network. The multi-dimensional data comes from multiple sensors arranged at various key parts of the processing equipment to capture various status information of the environment and equipment. Each sensor can feedback different state variables in real time. The multi-dimensional data includes equipment operating parameters, workpiece status data, environmental variables and historical processing data. In the distributed multi-sensor network, each sensor node collects data in real time and performs preliminary noise reduction and filtering operations to remove random errors and interference signals in the data to ensure the accuracy of the data; each sensor collects data according to a time period and integrates the data into a time series. ,in, , Indicated by The data value collected by the sensor, that is, Collect data in multiple dimensions; is the total number of dimensions; The nonlinear mapping function is used to transform the collected data after preliminary processing. The mapping process not only retains the characteristic information of the input data, but also enhances the adaptability to different processing scenarios by adding nonlinear dynamic weights. The mapping formula of the input data is: ,in, Represents the feature vector after mapping; is the nonlinear mapping weight, controlling the The degree of influence of the input data of each dimension on the final feature map; is the frequency parameter in the sine function, which is used to adjust the input data cyclical changes; It is the phase shift in the sine function, which controls how the input data is relatively shifted in different dimensions and adjusts the synchronization of each data dimension; is the exponential decay coefficient, which controls the decay rate of input data in different dimensions with time or input amplitude changes; It is the frequency parameter in the cosine function, which indicates the oscillation frequency of each data dimension and adjusts the periodic characteristics of different data dimensions; Indicates the rate of change of each dimension data in the time domain, which can reflect the dynamic characteristics of real-time data changing over time; S2. Based on the prediction of future processing status, a nonlinear feedback network is used to dynamically adjust processing parameters and optimize control strategies. Through the global optimization mechanism, global evaluation and adjustment are performed on multiple time nodes to ensure that the intelligent control processing system can maintain the optimal state at different stages.
2. According to claim 1, a data processing method for an intelligent control processing system, It is characterized in that The data processed by nonlinear mapping is not directly used for control decision-making, but for adaptive dynamic regulation. The mapped feature vector is processed by the time evolution network to predict the future processing state and make multiple adaptive iterative adjustments. The state change of the intelligent control processing system in the future is determined by analyzing the time series of the mapped feature vector. The formula of time evolution is: ,in, Indicates that intelligent control processing system Status prediction at each moment; represents the number of input features considered during the time evolution; For the The input mapping feature value comes from the mapping formula Dimensional data features; is the oscillation amplitude parameter, which controls the degree of fluctuation of the input feature in time; is the angular frequency, which represents the oscillation frequency of the feature in the time dimension; is a noise term, simulating random disturbances from the external environment. Through this time evolution formula, the state evolution trajectory of the processing equipment in the future can be predicted in the time domain. In the formula represents the acceleration term, and the changing trend of the system state during the entire processing process is calculated by integration. The periodic oscillation is simulated, which can reflect the influence of periodic disturbance in the machining process.
3. According to claim 2, a data processing method for an intelligent control processing system, It is characterized in that Based on the prediction of future states, a nonlinear feedback network is used to dynamically adjust the processing parameters, and the current state prediction and historical data are combined to optimize the processing parameters through recursive feedback control.
4. According to claim 3, a data processing method for an intelligent control processing system, It is characterized in that The recursive feedback formula for control decision is: ,in, is the initial control parameter, based on the initial state and the initial mapping features , calculate the initial control parameters; for The moment a control parameter, such as spindle speed or tool feed, is the learning rate, controlling the step size of the adjustment, is the output after time evolution state features, represented by System status value at the moment; Indicates The sensitivity parameter of each state characteristic controls the response degree of the intelligent control processing system to the state feedback; For the The historical control weight of each state feature, which controls the influence of the history on the current control strategy; is the number of feedback dimensions, indicating the number of feature dimensions involved in feedback control calculation; and is the weight coefficient of the initial optimization, which is used to balance the influence of mapping characteristics and state estimation on control parameters; In the recursive feedback formula is the activation function, ensuring that the control parameters respond nonlinearly to the state evolution results. is the adjustment amplitude function of the control parameter, which is used to ensure the stability of the control parameter in extreme cases; the historical feedback term in the formula Combining past control parameter change rates with current status ensures learning from historical data and continuous optimization of future control strategies.
5. According to claim 1, a data processing method for an intelligent control processing system, It is characterized in that In order to ensure the global optimality of the intelligent control machining system, a global optimization and multiple feedback mechanism is introduced. By globally evaluating and adjusting the operating status at multiple time nodes, the weights and feedback paths at each level are dynamically updated to ensure that each stage of the machining process can obtain an optimized control strategy.
6. The data processing method of the intelligent control processing system according to claim 5, It is characterized in that The specific recursive adjustment formula of the global optimization and multiple feedback mechanism is: ,in, represents the result of global optimization, i.e., the feedback adjustment after global evaluation of the control parameters at multiple time nodes, are the control parameters at different time nodes, Indicates Time nodes; Indicates the number of time nodes for global optimization; is the time decay coefficient, which controls the influence of historical data on the current feedback path during the global optimization process. is the global optimization coefficient, which controls the intensity of global feedback adjustment. for The feature vector at the moment; is the entire time range over which the state is accumulated; Perform a global evaluation at each time node, and optimize and adjust the feedback path based on current and historical data to ensure that the entire intelligent control processing system can respond to changes in the external environment and processing status in real time; Represents the acceleration of the control parameter, which is mapped to the input feature The combination of time decay coefficient Control the impact of historical data on current feedback; integral term The global state information of the time evolution layer is combined with the control decision to ensure that the system can maintain the global optimum during long-term operation.