Search

CN-121979112-A - Numerical control machine tool control method and system based on real-time processing data feedback

CN121979112ACN 121979112 ACN121979112 ACN 121979112ACN-121979112-A

Abstract

The application provides a control method and a control system of a numerical control machine based on real-time processing data feedback, and relates to the technical field of machine control, wherein the method comprises the steps of collecting multi-source data of the numerical control machine in the processing process; the method comprises the steps of predicting control quality based on multi-source data to obtain the prediction quality, generating on-line control constraint according to the prediction quality and a machine tool safety boundary, solving and optimizing the target of a machining period, a cutter abrasion increment and instantaneous power under the on-line control constraint to obtain target control parameters, issuing the target control parameters to execute, executing parameter rollback when abnormal limit exceeding is detected, and performing incremental update on the prediction control quality process by taking a rollback result as feedback. The application can solve the technical problem of poor control quality of the numerical control machine tool in the prior art, and achieves the technical effect of improving the control quality of the numerical control machine tool.

Inventors

  • SHEN WEIPING

Assignees

  • 南通百盛精密机械有限责任公司

Dates

Publication Date
20260505
Application Date
20251224

Claims (10)

  1. 1. The numerical control machine tool control method based on real-time processing data feedback is characterized by comprising the following steps of: Collecting multi-source data of the numerical control machine tool in the processing process; predicting control quality based on the multi-source data to obtain predicted quality; Generating an online control constraint according to the predicted quality and a machine tool safety boundary; under the on-line control constraint, carrying out solution optimization by taking a machining period, a cutter abrasion increment and instantaneous power as targets to obtain target control parameters; and issuing and executing the target control parameters, and executing parameter rollback when the abnormality is detected to exceed the limit, and incrementally updating the predicted control quality process by taking the rollback result as feedback.
  2. 2. The method for controlling a numerical control machine based on real-time processing data feedback according to claim 1, wherein collecting multi-source data of the numerical control machine during processing comprises: acquiring multi-source real-time data of a numerical control machine tool for timing and alignment, and outputting a multi-source data matrix with timing and alignment, wherein the multi-source real-time data comprises spindle current, feeding current, vibration, acoustic emission, temperature, encoder displacement and on-line measurement data; denoising and preprocessing the multi-source data matrix, and outputting a multi-source signal sequence; dividing the multi-source signal sequence into real-time sliding windows according to the multi-source signal sequence, and extracting multi-source data according to the real-time sliding windows.
  3. 3. The method for controlling a numerical control machine tool based on real-time processing data feedback according to claim 1, wherein predicting control quality based on the multi-source data, to obtain predicted quality, previously comprises: extracting a real-time window of the multi-source data, and acquiring a plurality of continuous history scale windows of the real-time window; Extracting multi-source sample data and control quality samples according to the continuous multiple history scale windows; Taking the multi-source sample data as a sample input set, taking the control quality sample as a sample output set, and training a control quality prediction channel to obtain a prediction control quality coefficient; and when the predicted control quality coefficient meets a control quality coefficient threshold value, obtaining a control quality prediction channel.
  4. 4. A method of controlling a numerically controlled machine tool based on real-time process data feedback as in claim 3, wherein extracting multi-source sample data and control quality samples based on the continuous plurality of historic scale windows comprises: Window adaptation is carried out on the multi-scale input samples and the multi-scale output samples of the continuous multiple history scale windows, and multi-source sample data and control quality samples are obtained; And configuring a dynamic window scale based on the real-time window, and filling insufficient-scale samples and cutting redundant-scale samples according to the dynamic window scale.
  5. 5. The method of claim 1, wherein generating an on-line control constraint based on the predicted quality and a machine tool safety boundary comprises: Introducing the machine tool safety boundary, and adjusting the machine tool safety boundary to generate a feasible interval by combining the predicted quality with the comparison of the machine tool safety boundary; mapping the feasible interval to the adjustable parameter of the machine tool to generate an on-line control constraint.
  6. 6. The method for controlling a numerical control machine based on real-time processing data feedback according to claim 1, wherein the on-line control constraint is solved and optimized with the processing cycle, the tool wear increment and the instantaneous power as targets, to obtain target control parameters, comprising: inputting the online control constraint into an optimization objective function and outputting a first optimization control parameter, wherein the optimization objective function comprises an optimization objective of the machining period, the cutter abrasion increment and the instantaneous power; and calculating the machining performance index according to the first optimized control parameter weight, solving the optimized control parameter with the minimum machining performance index value under the on-line control constraint, and obtaining the target control parameter.
  7. 7. The method for controlling a numerical control machine based on real-time process data feedback according to claim 1, wherein when an abnormality is detected beyond a limit, performing parameter rollback, incrementally updating a predictive control quality process with a rollback result as feedback, comprising: Acquiring multi-source real-time data to judge whether abnormal out-of-limit occurs or not, and obtaining an abnormal detection result; Executing a rollback strategy according to the abnormal detection result to obtain a rollback safety control parameter; and calculating errors of the safety control parameters and the target control parameters, and executing increment updating.
  8. 8. The method for controlling a numerical control machine tool based on real-time process data feedback according to claim 1, wherein the predicted quality includes tool wear, surface roughness, thermal geometry errors, and vibration stability margin.
  9. 9. The method for controlling a numerical control machine tool based on real-time processing data feedback according to claim 1, wherein the target control parameters include spindle rotation speed, feed speed, cutting depth, cutting width and trajectory acceleration and deceleration.
  10. 10. A numerically controlled machine tool control system based on real-time processing data feedback, characterized by comprising the steps of: The multi-source data set number module is used for collecting multi-source data of the numerical control machine tool in the machining process; The prediction quality obtaining module is used for predicting control quality based on the multi-source data to obtain prediction quality; The constraint generation module is used for generating an online control constraint according to the prediction quality and the machine tool safety boundary; the parameter obtaining module is used for solving and optimizing the processing period, the cutter abrasion increment and the instantaneous power serving as targets under the online control constraint to obtain target control parameters; And the execution module is used for issuing and executing the target control parameters, executing parameter rollback when the abnormality is detected to exceed the limit, and incrementally updating the predicted control quality process by taking the rollback result as feedback.

Description

Numerical control machine tool control method and system based on real-time processing data feedback Technical Field The application relates to the technical field of machine tool control, in particular to a numerical control machine tool control method and system based on real-time processing data feedback. Background Along with the wide application of the numerical control machine tool in the fields of high-end manufacture, aerospace, automobile precision part machining and the like, the machine tool has higher and higher requirements on quality, efficiency and stability in the machining process. The traditional numerical control machine tool control method mostly depends on preset technological parameters and experience models to guide the machining process, and the method can obtain a good effect under a stable working condition, but often has insufficient performance in a complex and changeable actual machining environment. At present, the existing researches and technologies focus on single-aspect improvement, such as tool state monitoring based on spindle current or vibration signals or thermal error compensation based on temperature field modeling, but the methods often lack fusion and unified processing of multi-source data, so that prediction results are not comprehensive enough and cannot effectively reflect the real running state of a machine tool under complex working conditions. Even if part of researches introduce a real-time monitoring and prediction model, most of the researches are unidirectional monitoring, namely only abnormal detection is carried out and closed loop linkage with control parameters is lacked, so that a fracture exists between prediction and control. In summary, in the prior art, due to lack of linkage of fusion and control links of multi-source real-time data, the quality prediction result is single and is difficult to guide dynamic optimization of machine tool parameters, so that the stability of the processing process, the consistency of product quality and the improvement of production efficiency are further affected. Disclosure of Invention The application aims to provide a numerical control machine tool control method and a numerical control machine tool control system based on real-time processing data feedback, which are used for solving the technical problems that in the prior art, due to lack of linkage of fusion and control links of multi-source real-time data, a quality prediction result is single, dynamic optimization of machine tool parameters is difficult to guide, and stability of a processing process, consistency of product quality and improvement of production efficiency are further affected. In view of the above problems, the application provides a method and a system for controlling a numerical control machine tool based on real-time processing data feedback. In a first aspect, the application provides a control method of a numerical control machine based on real-time processing data feedback, which is realized by a control system of the numerical control machine based on real-time processing data feedback, and comprises the steps of collecting multi-source data of the numerical control machine in the processing process; the method comprises the steps of predicting control quality based on multi-source data to obtain the prediction quality, generating on-line control constraint according to the prediction quality and a machine tool safety boundary, solving and optimizing the target control parameter by taking machining period, cutter abrasion increment and instantaneous power as targets under the on-line control constraint, issuing and executing the target control parameter, executing parameter rollback when abnormal limit is detected, and carrying out incremental update on the prediction control quality process by taking a rollback result as feedback. The numerical control machine tool control method based on the real-time processing data feedback preferably further comprises the steps of collecting multi-source real-time data of the numerical control machine tool for timing and alignment, outputting a multi-source data matrix after timing and alignment, wherein the multi-source real-time data comprises main shaft current, feeding current, vibration, acoustic emission, temperature, encoder displacement and on-line measurement data, denoising and preprocessing the multi-source data matrix, outputting a multi-source signal sequence, dividing the multi-source signal sequence into real-time sliding windows according to the multi-source signal sequence, and extracting the multi-source data according to the real-time sliding windows. The numerical control machine tool control method based on real-time processing data feedback preferably further comprises the steps of extracting a real-time window of the multi-source data, obtaining a plurality of continuous historical scale windows of the real-time window, extracting multi-source sample data and control qua