A machining parameter automatic optimization method and system for a three-axis numerical control machine tool

Abstract

The present application relates to the technical field of numerical control machine tool control, in particular to a machining parameter automatic optimization method and system for three-axis numerical control machine tools, the specific steps are: collecting machine tool parameter data and workpiece parameter data, the machine tool parameter data includes the current voltage and tool wear state of the three-axis numerical control machine tool, and the workpiece parameter data includes workpiece size, vibration, deflection and surface roughness; based on the machine tool parameter data and the workpiece parameter data, a three-axis numerical control machine tool digital twin model is constructed by using a physical information neural network, real-time monitoring and feedback are performed, and machining efficiency and energy consumption are calculated; based on the machining efficiency and energy consumption, a numerical control machine tool machining process model and an energy consumption model are constructed, an optimization algorithm is used to maximize the machining efficiency and minimize the energy consumption, the machining process model reflects the quality efficiency index, and the energy consumption model reflects the energy consumption characteristics; an optimization function obtained by the optimization algorithm is used to construct a control instruction execution module, and comprehensive factory control is covered.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of numerical control machine tool control, in particular to a machining parameter automatic optimization method and system for a three-axis numerical control machine tool. BACKGROUND

[0002] In the field of numerical control technology control, with the continuous progress of technology, the machining parameter automatic optimization method of the numerical control machine tool is also constantly updated, but there are still some problems.

[0003] For a three-axis numerical control machine tool, the basic form of machining parameter optimization is to use mathematical methods, the core of which is to calculate the optimal machining parameters by establishing the mathematical relationship between the machine tool, the tool and the workpiece system. These methods are usually designed to quantify the physical phenomena in the cutting process to guide parameter selection, and adaptive control technology is one of the core directions of numerical control machining parameter automatic optimization, which can automatically adjust the machine tool feed rate to the optimal value and record the machining parameters in real time. The current technology mainly uses optimization algorithms, which overcome the shortcomings of traditional methods in dealing with complex, nonlinear and multi-objective problems through learning and reasoning. Starting from genetic algorithm and particle swarm optimization, the complexity is continuously improved.

[0004] The main challenges are high cost, data integrity and representativeness, integration with existing equipment and limitations of the technology itself, and the influence of parameter setting on the results also limits its robustness in practical applications. The application makes improvements on the automation system architecture, making the parameters of the machine tool more intuitive and easier for operators to control, achieving comprehensive factory control, improving its robustness and interpretability in actual industrial environments, and solving the problem of data sparsity, especially in the specific working conditions of three-axis machine tools. For this purpose, a machining parameter automatic optimization method and system for a three-axis numerical control machine tool are proposed. SUMMARY

[0005] The application is a machining parameter automatic optimization method and system for a three-axis numerical control machine tool, which is used for automatic optimization and debugging of three-axis numerical control machine tool machining parameters. The specific implementation steps are as follows: first, collect the parameter data of the three-axis numerical control machine tool and the various parameter data of the machined workpiece, then based on the parameter data, use a physical information neural network to construct a three-axis numerical control machine tool digital twin model, real-time monitoring and feedback, and calculate the machining efficiency and energy consumption, construct a numerical control machine tool machining process model and an energy consumption model, use an optimization algorithm to maximize the machining efficiency and minimize the energy consumption, and then use the optimization function obtained by the optimization algorithm to construct a control instruction execution module, covering comprehensive factory control.

[0006] The application fuses the characteristics of multi-source data, through the processing process model and energy consumption model constructed according to the processing efficiency and energy consumption, the ideal parameter combination that can maximize the processing efficiency is accurately calculated by using the optimization algorithm, compared with the traditional setting mode depending on experience or fixed parameter table, the invalid operation and waiting time in processing is effectively reduced, the material removal rate or workpiece processing quantity in unit time is greatly improved, the machining efficiency of the machine tool is improved, scientific and accurate processing parameter decision support is provided for the operators and production management personnel, and the processing problems caused by insufficient human experience or judgment errors are avoided.

[0007] A machining parameter optimization method for a three-axis numerical control machine tool, comprising the following steps: collecting machine tool parameter data and workpiece parameter data, the machine tool parameter data including current voltage and tool wear state of the three-axis numerical control machine tool, and the workpiece parameter data including workpiece size, vibration frequency, deflection and surface roughness; based on the machine tool parameter data and the workpiece parameter data, a digital twin model of the three-axis numerical control machine tool is constructed, real-time monitoring and feedback are performed, and processing efficiency and energy consumption are calculated; based on the processing efficiency and energy consumption, a machining process model and an energy consumption model of the numerical control machine tool are constructed, an optimization algorithm is used to maximize the processing efficiency and minimize the energy consumption, the machining process model monitors the quality efficiency index of the machining process, and the energy consumption model monitors the energy consumption of the machining process; an optimized function obtained by the optimization algorithm is used to construct a control instruction execution system, the control instruction execution system is used to run the optimized parameters, adaptively covers the overall factory control, and realizes multi-machine tool collaborative scheduling through a distributed network interface.

[0008] Preferably, the process of collecting machine tool parameter data and workpiece parameter data comprises:

[0009] The current, voltage and tool wear state of the cutting die tool are collected as the machine tool parameter data, and the workpiece size, vibration frequency, deflection and surface roughness of the workpiece are collected as the workpiece parameter data, and the collected data is preprocessed.

[0010] Preferably, the process of constructing a digital twin model of the three-axis numerical control machine tool based on the machine tool parameter data and the workpiece parameter data, real-time monitoring and feedback, and calculating the processing efficiency and energy consumption comprises:

[0011] The collected multi-source data is fused and processed by using a deep learning algorithm, noise and abnormal values are automatically identified and removed, feature parameters having a key influence on the processing efficiency and energy consumption are extracted, a fused data set is constructed, a large amount of historical data is used to train the model, the weights and biases of the network are adjusted, the performance of the model is monitored in real time during the training process, new data is periodically collected as the machine tool runs and the processing conditions change, and the model is updated and retrained.

[0012] Preferably, the process of constructing the machining process model and the energy consumption model based on machining efficiency and energy consumption comprises:

[0013] The machining data collected and accumulated in the digital twin model in real time is used to train the machining process model, the preliminary model based on physical laws is corrected and optimized, so that the model can reflect the complex nonlinear machining efficiency relationship in the actual machining process; the input variables and the corresponding energy consumption data collected in the actual machining process in the digital twin model are used to train and optimize the energy consumption model based on the basic relationship, and the parameters of the model are estimated and adjusted to fit the relationship between the actual energy consumption and each input variable.

[0014] Preferably, the process of maximizing machining efficiency and minimizing energy consumption using an optimization algorithm comprises:

[0015] The simulated annealing algorithm is selected, the machining efficiency after the machining model is corrected and optimized is called, and the energy consumption data after the energy consumption model is corrected and optimized is called, and the weighted sum is obtained, the determination of the weight coefficient is made according to the priority in the actual production scene, the optimization function is obtained, the machining process model and the energy consumption model constructed are integrated into the optimization algorithm, the machining parameters are taken as the decision variables, and the objective function is taken as the optimization direction, the optimization algorithm is run, the combination of the maximum machining efficiency and the minimum energy consumption is obtained, and it is analyzed and verified, and the machine tool control instruction set is given to the computer.

[0016] Preferably, the quality efficiency index and the energy consumption characteristics are called and weighted to obtain the optimization function, the comprehensive factory control is adaptively covered through the control instruction execution system, and the multi-machine tool collaborative scheduling is realized through the distributed network interface, and the process comprises:

[0017] According to the optimal machining parameters obtained from the optimization function, the corresponding control instructions are generated, the generated control instructions are converted into a format executable by the numerical control machine tool, the instructions are safety checked, and the control instruction execution system is sent to the control instruction execution system. The control instruction execution system transmits the instructions to multiple numerical control machine tools through a distributed network interface to realize factory-level collaborative control.

[0018] A machining parameter optimization system for a three-axis numerical control machine tool, comprising:

[0019] A data collection module for collecting machine tool parameter data and workpiece parameter data;

[0020] A digital twin module for monitoring and feeding back machine tool parameter data and workpiece parameter data, and calculating machining efficiency and energy consumption in real time;

[0021] A machining process model and energy consumption model module for reflecting the complex nonlinear machining efficiency relationship in the actual machining process and fitting the relationship between the actual energy consumption and each input variable;

[0022] Optimization algorithm module: use optimization algorithm to maximize processing efficiency and minimize energy consumption, retrieve and perform weighted summation to obtain optimization function;

[0023] Control instruction execution system module: build adaptive control instruction execution, execute optimal optimization parameter scheme, and transmit instructions to multiple numerical control machine tools to realize factory-level collaborative control.

[0024] Compared with the prior art, the beneficial effects of the present application are:

[0025] 1. The processing process model and energy consumption model proposed in the present application realize fine control of the energy consumption of each component of the machine tool. The average energy consumption of the three-axis numerical control machine tool is reduced during the machining process, which saves a large amount of energy cost for the enterprise in long-term production. The optimal parameter combination that maximizes the processing efficiency is accurately calculated by using the optimization algorithm, which effectively reduces the invalid operation and waiting time in processing, greatly improves the material removal rate or the number of workpieces processed per unit time, and improves the processing efficiency of the machine tool.

[0026] 2. The optimization algorithm proposed in the present application fully considers the minimization of energy consumption while pursuing the maximization of processing efficiency by analyzing the processing process model and the energy consumption model. Through real-time monitoring and feedback mechanism, the system can timely adjust the processing parameters to adapt to the changes of the workpiece and machine tool state, prevent energy waste caused by serious tool wear and motor overload due to unreasonable parameters, further improve the energy utilization efficiency, and reduce economic losses and environmental pressure caused by unnecessary energy consumption.

[0027] 3. The optimization function proposed in the present application retrieves the optimized processing efficiency and energy consumption data after correction of the processing model, and performs weighted summation to build control instructions. The control instruction execution system not only optimizes and controls the parameters of a single machine tool, but also focuses on the coverage of overall factory control. Through deep integration with the factory production management system, it realizes the whole process closed-loop control from production planning, processing task allocation to machine tool operation execution, improves the collaboration and efficiency of the whole production operation of the factory, and enables the enterprise to respond to complex production mode requirements such as multi-variety and small-batch in a more flexible and efficient way. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 A processing parameter automatic optimization method flowchart for a three-axis numerical control machine tool is proposed for the embodiments of the present application;

[0029] Figure 2 A process diagram for obtaining the optimization function of the embodiments of the present application;

[0030] Figure 3A processing parameter automatic optimization system module diagram for a three-axis numerical control machine tool is provided for the embodiments of the application. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work are within the protection scope of the application.

[0032] In the prior art, the processing parameter optimization technology for a three-axis numerical control machine tool mainly adopts adaptive control technology. They overcome the deficiencies of traditional methods in dealing with complex, nonlinear and multi-objective problems through learning and reasoning. Starting from genetic algorithm and particle swarm optimization, the complexity is constantly improved. However, the main challenges are high cost, data integrity and representativeness, integration with existing equipment and limitations of the technology itself. In addition, the parameter setting has a great influence on the results, which also limits the robustness in practical application. The application improves the automation system architecture, makes the parameters of the machine tool more intuitive and easier for operators to control, and realizes overall factory control.

[0033] The application relates to the technical field of numerical control machine tool control, in particular to a processing parameter automatic optimization method and system for a three-axis numerical control machine tool.

[0034] Embodiment one

[0035] The embodiments of the application disclose a processing parameter automatic optimization method and system for a three-axis numerical control machine tool, which is suitable for further optimization of the processing parameters of the three-axis numerical control machine tool, improves processing efficiency and reduces energy consumption. Figure 1A factory processes aerospace parts, and in the face of high-precision and complex-shaped titanium alloy aircraft engine blade processing tasks, it needs to use three-axis CNC machine tools. First, collect machine tool parameter data and workpiece parameter data, the machine tool parameter data includes the current voltage and tool wear state of the three-axis CNC machine tool, and the workpiece parameter data includes the workpiece size, vibration frequency, deflection and surface roughness; based on the machine tool parameter data and the workpiece parameter data, a three-axis CNC machine tool digital twin model is constructed using a physical information neural network, real-time monitoring and feedback are performed, and the processing efficiency and energy consumption are calculated; based on the processing efficiency and energy consumption, a CNC machine tool processing process model and an energy consumption model are constructed, and an optimization algorithm is used to maximize the processing efficiency and minimize the energy consumption, the processing process model monitors the quality efficiency indicators of the processing process, and the energy consumption model monitors the energy consumption of the processing process; through the optimization function obtained by the optimization algorithm, a control instruction execution module is constructed to cover the overall factory control.

[0036] Further, collect machine tool parameter data and workpiece parameter data, corresponding to the above steps, the specific implementation process includes:

[0037] Collect the current and voltage data of the tool working as an important part of the machine tool parameter data, and pay attention to the tool wear state and quantify it as a monitorable parameter data. Current and voltage data can reflect the running load and state of machine tool motor and other electrical equipment, while tool wear state is directly related to machining precision and surface quality; Collect the workpiece parameters comprehensively, including the size, vibration frequency, deflection and surface roughness of the workpiece. These parameters reflect the state and processing quality of the workpiece in the processing process from different angles; use the generative adversarial network to pre-process the machine tool parameter data and workpiece parameter data, automatically and adaptively learn and remove complex, nonlinear noise and outliers.

[0038] This embodiment collects data through detailed specification of the specific content and method of data collection, ensuring that the collected multi-dimensional data comprehensively and accurately reflects the real-time state of the three-axis CNC machine tool and its processed workpiece. This accurate data collection provides high-quality raw data for subsequent construction of digital twin model, processing process model and energy consumption model, and the comprehensive collection of workpiece parameters enables the optimization system to more accurately grasp the influence of the processing process on the workpiece quality, which helps to improve the accuracy and reliability of the model, enabling the model to better simulate and predict the actual situation in the processing process, and provides a solid basis for the optimization algorithm.

[0039] Further, based on the machine tool parameter data and the workpiece parameter data, a three-axis CNC machine tool digital twin model is constructed, real-time monitoring and feedback are performed, and the processing efficiency and energy consumption are calculated, corresponding to the above steps, the specific implementation process includes:

[0040] The collected data is pre-processed by removing noise, normalization and feature extraction operations. The interference information in the data is eliminated by removing noise to make the data accurately reflect the real processing state. The normalization process converts data of different dimensions and numerical ranges to a unified scale, completing the preconditions for subsequent data analysis and modeling. The convolutional neural network is used to identify the feature parameters that have a key impact on processing efficiency and energy consumption, and the multi-dimensional data is fused to provide information for subsequent construction of accurate digital twin models and processing process models, energy consumption models. Then a large amount of historical data is used to train the model, adjust the weights and biases of the network, and build the digital twin model. During the training process, the performance of the model is monitored in real time to ensure that the model can accurately simulate and predict the actual behavior in the processing process. As the machine tool runs and the processing conditions change, new data is collected regularly to update and retrain the model to maintain the accuracy and adaptability of the model.

[0041] By constructing a digital twin model, the embodiment realizes the mechanism of real-time monitoring of model performance and periodic collection of new data for model updating and retraining, so that the digital twin model can timely capture the changes in machine tool running state and processing conditions. This real-time and adaptability ensures that the model can maintain good prediction performance under different processing tasks and working conditions, providing accurate real-time data support for optimization algorithms, thereby realizing dynamic optimization and precise control of the processing process.

[0042] Further, based on the processing efficiency and energy consumption, the processing process model and the energy consumption model are constructed, and the process includes:

[0043] The various data in the processing process are collected in real time in the digital twin model, including machine tool parameters and workpiece parameters, as input variables. At the same time, the processing efficiency is taken as the output variable, a preliminary processing process model is constructed based on the cutting force model, and the support vector machine algorithm is used to correct and optimize the preliminary processing process model. Through the analysis of a large amount of historical processing data, the deviations and errors in the preliminary model are identified and corrected, so that the model can reflect the complex nonlinear relationship in the actual processing process. The input variables in the actual processing process, including cutting speed, feed speed and cutting depth, and the corresponding energy consumption data, including spindle motor energy consumption and feed motor energy consumption, are collected from the digital twin model. Based on these data, the basic relationship of the energy consumption model is established. A linear regression model is usually used to preliminarily fit the relationship between energy consumption and each input variable. A data-driven modeling method is used to train the energy consumption model by adjusting the parameters of the model to better fit the actual energy consumption data and improve the prediction accuracy of the model. In the training process, the real-time collected and accumulated processing data in the digital twin model are used to train the processing process model. The model can identify the real causal variables that affect the processing efficiency and energy consumption by integrating the structural causal model, and provide interpretable basis for optimization decision-making.

[0044] The built processing process model and energy consumption model are used as a simulator of the reinforcement learning environment, a weighted function of processing efficiency and energy consumption is used as a reward function, policy learning is performed in the simulation environment of the digital twin model, the policy learning can dynamically adjust the processing parameters according to the real-time monitored machine tool parameter data and workpiece parameter data, achieve adaptive update, improve the processing data efficiency, the digital twin model, the processing process model and the energy consumption model form a dynamic two-way self-correction closed loop mechanism, the digital twin model is the core of real-time data aggregation and state mapping, the monitoring data not only trains and optimizes the processing process model and the energy consumption model, on the contrary, the prediction results and optimization suggestions of the two models are also fed back to the digital twin model in real time, driving the update and virtual simulation of the digital twin state, so as to form more accurate digital mapping and more effective guidance. When the digital twin model and the processing model and the energy consumption model are bidirectionally self-corrected, the prediction uncertainty is introduced for evaluation, the uncertainty of the model prediction result is quantified, when the prediction uncertainty is high, a more cautious parameter adjustment strategy is automatically triggered; when the prediction uncertainty is low, more bold adjustment of the processing parameters is allowed to further optimize the processing efficiency and energy consumption, so that the dynamic adjustment of the processing parameters is more intelligent and flexible, and the complex and changeable actual processing environment can be better coped with.

[0045] The processing process model of the embodiment can more accurately capture the complex nonlinear relationship between various parameters in the processing process by modifying and optimizing the preliminary model based on physical laws; the energy consumption model can more accurately fit the relationship between actual energy consumption and each input variable by training and optimizing the energy consumption model using a data-driven modeling method; the processing process model and the energy consumption model can reduce processing time and scrap rate, reduce production cost, reduce energy consumption of machine tools, and reduce production cost and environmental impact by improving processing efficiency, accurately predicting and optimizing energy consumption.

[0046] Further, the implementation process of maximizing the processing efficiency and minimizing the energy consumption using the optimization algorithm includes:

[0047] The built machining process model and energy consumption model are integrated into the optimization algorithm with machining parameters as decision variables. The two models provide predicted values of machining efficiency and energy consumption as part of the objective function of the optimization algorithm. Starting from an initial solution, a multi-objective optimization algorithm based on dynamic game strategy is used, which regards the machining process as a dynamic game process between multiple decision-making subjects, and adjusts the strategies of each decision-making subject in real time. In each iteration, the algorithm decides whether to accept a new solution based on the combination of machining efficiency and energy consumption of the current solution. If the new solution has better objective function values, it is accepted; if the new solution is slightly worse, it is also accepted with a certain probability to avoid local optimization, so that the two objectives of machining efficiency and energy consumption can achieve dynamic balance under different working conditions and constraints.

[0048] The optimization algorithm of the embodiment can effectively avoid local optimization and improve the globality of the optimization result. In three-axis CNC machine tool machining, the simultaneous optimization of machining efficiency and energy consumption can significantly improve the machining efficiency. During the optimization process, the optimization results are analyzed and verified in real time, and compared and evaluated with actual machining data to ensure that the optimized machining parameters can effectively improve the machining efficiency and reduce the energy consumption in actual production.

[0049] Further, the optimization function obtained by the optimization algorithm is used to optimize and run the parameters, and the specific implementation includes: Figure 2

[0050] The corrected and optimized machining efficiency data is retrieved from the optimized machining process model, and the corrected and optimized energy consumption data is retrieved from the energy consumption model. The retrieved machining efficiency data and energy consumption data are weighted and summed to obtain an optimization function, which considers the balance between machining efficiency and energy consumption. The weight coefficients are determined according to the priority in the actual production scene; according to the scene, a higher weight is given to the machining efficiency and a lower weight is given to the energy consumption. The optimal machining parameters are calculated according to the optimization function, and a multi-objective optimization function is constructed again through the optimization function, which aims to find a set of Pareto optimal solutions. Each point in this solution set represents a combination of efficiency and energy consumption. According to the Pareto frontier, a series of optimal choices are provided for the operator or higher-level control system, rather than just one optimal combination. Among these Pareto optimal solutions, manual decision-making is made according to the most urgent needs, and intelligent selection is made by the upper system.

[0051] ​Afterwards, the corresponding control instructions are generated, the generated control instructions are converted into G code format that can be recognized and executed by the numerical control machine tool, and then the instructions are safety checked to ensure that the instructions will not cause the machine tool to exceed the range, collide or other safety problems; the instruction execution module is internally provided with a multi-modal semantic understanding ware and an interactive instruction generation system, which can not only understand the more complex intentions of the operator, but also combine the intentions with a physical information neural network model to generate more accurate processing parameters and instructions, the system can reversely convert the generated G code instructions into natural language descriptions that are easy for the operator to understand, and combine the digital twin model to perform visual simulation, provide real-time interpretation of instruction execution and visual feedback of potential impact, greatly enhancing the depth and efficiency of human-machine collaboration.

[0052] The safety checked control instructions are sent to the control instruction execution system, the control instruction execution system transmits the instructions to multiple numerical control machine tools through a distributed network interface, and factory-level collaborative control is achieved.

[0053] The embodiment can accurately convert the optimized processing parameters into machine tool executable instructions by constructing a control instruction execution module, ensuring that the optimization strategy is effectively executed, and the optimized processing parameters can significantly improve the processing efficiency, reduce the processing time and the scrap rate. At the same time, stable processing parameters help to improve the processing quality of workpieces, ensure that the size accuracy and surface roughness meet the requirements, and through the distributed network interface, the collaborative scheduling of multiple numerical control machine tools can be realized, and the overall production efficiency and equipment utilization of the factory can be improved.

[0054] Although embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made thereto without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

[0055] Embodiment two

[0056] The embodiment of the present application discloses a federated learning collaborative optimization method and system for energy storage battery safety prevention and control, B factory manufactures automobile parts, and needs to efficiently and energy-savingly process a large number of aluminum alloy engine cylinder bodies, and the traditional processing mode is difficult to simultaneously meet the production requirements of high efficiency, low energy consumption and stable quality, and a processing parameter automatic optimization method and system for three-axis numerical control machine tools need to be used. Figure 3A kind of processing parameter automatic optimization system module diagram for three-axis numerical control machine tool, the processing parameter automatic optimization system for three-axis numerical control machine tool includes data collection module, digital twin module, processing model and energy consumption model module, optimization algorithm module and control instruction execution system module.

[0057] Further, machine tool parameter data and workpiece parameter data are collected by the data collection module, including collecting current, voltage data of three-axis numerical control machine tool and tool wear state in real time through sensor during processing, while collecting the size, vibration frequency, deflection and surface roughness of workpiece.And the collected machine tool parameter data and workpiece parameter data are preprocessed to remove redundant information and noise.

[0058] Further, a physical information neural network three-axis numerical control machine tool digital twin model is constructed by the digital twin module, real-time monitoring and calculating processing efficiency and energy consumption.Convolutional neural network algorithm is used to identify and eliminate outliers, and then feature parameters related to processing efficiency and energy consumption such as current, voltage and tool wear state data are extracted, and then a large amount of historical data is used to train the model, adjust the weight and bias of the network, and construct the digital twin model.In subsequent training, data is updated regularly, and the model is retrained.

[0059] Further, a processing model and an energy consumption model are constructed by the processing model and the energy consumption model module.The real-time collected processing is input variable in the digital twin model, and the processing efficiency is output variable, and a preliminary processing model is constructed based on the cutting force model;The real-time collected energy consumption data is used as input variable, and a preliminary energy consumption model is established, and the energy consumption model is trained by using data-driven modeling method, and the accuracy of the model is improved by adjusting the parameters of the model.

[0060] Further, the optimization algorithm module maximizes the processing efficiency and minimizes the energy consumption.Select simulated annealing algorithm as optimization algorithm, integrate the constructed processing model and energy consumption model into the algorithm, use cutting speed, feed speed, cutting depth and other processing parameters as decision variables, run the optimization algorithm, and get the optimal processing parameter combination that can maximize the processing efficiency and minimize the energy consumption under the current working condition.

[0061] Further, an adaptive control instruction execution mode is constructed by controlling the control instruction execution system module to execute the optimal optimization parameter scheme. The control instruction execution module is constructed according to the optimization function, control instructions are generated and converted into an instruction format recognizable by the machine tool, after safety verification, the instructions are sent to multiple machine tools in the workshop through a distributed network interface, collaborative scheduling is realized, the optimized machining parameters improve the machining efficiency of the engine cylinder, reduce energy consumption, effectively improve the production benefit and energy utilization efficiency of the workshop, while ensuring the machining quality of the cylinder, the production cost is significantly reduced.

[0062] The embodiment of the application makes a more effective and reasonable decision on the automatic optimization of the three-axis numerical control machine tool machining parameters through a machining parameter automatic optimization system for a three-axis numerical control machine tool, which integrates the data collection module, the digital twin module, the machining process model and the energy consumption model module, the optimization algorithm module and the control instruction execution system module for module integrated data management, comprehensive factory control.

[0063] Although the embodiments of the application have been shown and described, it can be understood by those of ordinary skill in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the application, and the scope of the application is defined by the appended claims and their equivalents.

Claims

1. A method for automatic optimization of machining parameters for a three-axis CNC machine tool, characterized in that, The method comprises the following steps: Collecting machine tool parameter data and workpiece parameter data, the machine tool parameter data including current voltage and tool wear state of a three-axis numerical control machine tool, and the workpiece parameter data including workpiece size, vibration frequency, deflection and surface roughness; Based on the machine tool parameter data and the workpiece parameter data, a three-axis numerical control machine tool digital twin model is constructed by using a physical information neural network, real-time monitoring and feedback are performed, and processing efficiency and energy consumption are calculated; Based on the processing efficiency and the energy consumption, a numerical control machine tool processing process model and an energy consumption model are constructed, a dynamic two-way self-correction closed loop mechanism is formed by the digital twin model, the processing process model and the energy consumption model, the digital twin model is the core of real-time data aggregation and state mapping, the monitoring data of the digital twin model not only trains and optimizes the processing process model and the energy consumption model, but also feeds back the prediction results and optimization suggestions of the two models to the digital twin model in real time, so as to drive the update and virtual simulation of the digital twin state; the constructed processing process model and energy consumption model are used as a simulator of a reinforcement learning environment, a weighted function of the processing efficiency and the energy consumption is used as a reward function, and strategy learning is performed in a simulation environment of the digital twin model, the strategy learning can dynamically adjust processing parameters according to real-time monitored machine tool parameter data and workpiece parameter data; an optimization algorithm is used to maximize the processing efficiency and minimize the energy consumption, the processing process model monitors quality efficiency indexes of a processing process, and the energy consumption model monitors energy consumption of the processing process, prediction uncertainty is introduced in the dynamic two-way self-correction closed loop mechanism, the uncertainty of a model prediction result is quantified, when the prediction uncertainty is high, a cautious parameter adjustment strategy is automatically triggered; and when the prediction uncertainty is low, a bold adjustment of the processing parameters is allowed to further optimize the processing efficiency and the energy consumption; A quality efficiency index and energy consumption characteristic weighted sum are called to obtain an optimization function, a blockchain is used as an underlying architecture of a control instruction execution system, an intelligent contract is deployed on the blockchain, and is used for defining and executing rules, the intelligent contract verifies uploaded parameters according to preset rules, the parameters passing the verification are added to a ledger of the blockchain, if the verification fails, the intelligent contract refuses to store the data, and feeds back error information to an uploading user; the control instruction execution system adaptively covers comprehensive factory control, and multi-machine tool collaborative scheduling is realized through a distributed network interface.

2. The method for automatic optimization of machining parameters for a three-axis CNC machine tool according to claim 1, characterized in that, The process of collecting the machine tool parameter data and the workpiece parameter data comprises: Collecting current, voltage and tool wear state of a cutting die tool as machine tool parameter data, and collecting workpiece size, vibration frequency, deflection and surface roughness of a workpiece as workpiece parameter data, and performing data preprocessing operations.

3. The method for automatic optimization of machining parameters for a three-axis CNC machine tool according to claim 1, characterized in that, The process of constructing the three-axis numerical control machine tool digital twin model based on the machine tool parameter data and the workpiece parameter data, performing real-time monitoring and feedback, and calculating processing efficiency and energy consumption comprises: The collected multi-source data is further fused by using a deep learning algorithm, noise and outliers are automatically identified and removed, characteristic parameters that have a key influence on processing efficiency and energy consumption are extracted, a fused data set is constructed, a large amount of historical data is used to train the model, the weights and biases of the network are adjusted, the performance of the model is monitored in real time during the training process, and as the machine tool runs and the processing conditions change, new data is collected periodically to update and retrain the model.

4. The method for automatic optimization of machining parameters for a three-axis CNC machine tool according to claim 1, characterized in that, The process of constructing a numerical control machine tool processing process model and an energy consumption model based on the processing efficiency and energy consumption includes: The real-time collected and accumulated processing data is used to train the processing process model, so that the model can reflect the complex nonlinear processing efficiency relationship in the actual processing process; the input variables and corresponding energy consumption data collected in the actual processing process are used to fit the relationship between actual energy consumption and each input variable.

5. The method for automatic optimization of machining parameters for a three-axis CNC machine tool according to claim 1, characterized in that, The process of maximizing processing efficiency and minimizing energy consumption using an optimization algorithm includes: The constructed processing process model and energy consumption model are integrated into the optimization algorithm, the processing parameters are used as decision variables, the optimization algorithm is run to obtain the combination of maximum processing efficiency and minimum energy consumption, and the combination is analyzed and verified.

6. The method for automatic optimization of machining parameters for a three-axis CNC machine tool according to claim 1, characterized in that, The quality efficiency index and energy consumption characteristics are retrieved and weighted summed to obtain an optimization function, a control instruction execution system is adaptively covered to comprehensively control the factory, and a distributed network interface is used to realize multi-machine tool collaborative scheduling, and the process includes: The processing efficiency of the modified and optimized processing model is retrieved, the energy consumption data of the modified and optimized energy consumption model is retrieved, and the data is weighted summed, the weight coefficients are determined according to the priority in the actual production scene, an optimization function is obtained, the optimal processing parameters are calculated according to the optimization function, and corresponding control instructions are generated.

7. A system for automatic optimization of machining parameters for a three-axis CNC machine tool, characterized by, The method of claim 1 includes the following modules: A data collection module for collecting machine tool parameter data and workpiece parameter data; A digital twin module for monitoring and feeding back machine tool parameter data and workpiece parameter data, and calculating processing efficiency and energy consumption in real time; A processing process model and energy consumption model module for reflecting the complex nonlinear processing efficiency relationship in the actual processing process and fitting the relationship between actual energy consumption and each input variable; An optimization algorithm module for maximizing processing efficiency and minimizing energy consumption using an optimization algorithm, retrieving and weighting to obtain an optimization function; A control instruction execution system module for constructing adaptive control instruction execution, executing the optimal optimization parameter scheme, and transmitting the instructions to multiple numerical control machine tools to realize factory-level collaborative control.

Images