An aero-engine part machining data collection and optimization application method

Abstract

The application provides an aero-engine part machining data collection and optimization application method, and relates to the technical field of aero-engines, in particular to: performing space-time mapping processing on time domain data or combing data and geometric space positions of part machining, storing machining process data into a process database, constructing a machining quality and efficiency prediction mapping relationship model, establishing a mapping relationship between process parameters and machining quality and efficiency prediction, simultaneously inputting process parameters and machining quality and efficiency data of the process database as boundary input, and obtaining a machining quality and efficiency prediction model after training; constructing a process optimization model and performing optimization; taking working condition data contained in the process database as original data, embedding the machining quality and efficiency prediction model and the process optimization model into the process database, continuously accumulating and iterating process data, so as to improve the accuracy of machining quality and efficiency prediction, optimize process parameters, and improve machining quality and efficiency.

Description

Technical Field

[0001] This invention relates to the field of aero-engine technology, and in particular to a method for collecting and optimizing the application of aero-engine part processing data. Background Technology

[0002] Aero-engines are the heart of fighter jets, representing a major national strategic need and a cutting-edge area of ​​international engineering science. Due to their broad technological scope, long R&D cycle, and large funding requirements, they are considered the "crown jewel" of modern industry. Currently, my country's processing level for difficult-to-machine materials such as titanium alloys, high-temperature alloys, and intermetallic compounds is relatively backward. In particular, there is a lack of systematic sorting, summarization, and evaluation of processing data and knowledge. Enabling tools such as data knowledge reuse reasoning and optimization applications are incomplete. Ultimately, the process parameters applied to parts processing vary from person to person, resulting in poor stability and consistency in the processing of typical aero-engine parts such as blades, integral bladed disks, casings, and turbine disks. This poses a significant challenge to the stability of mass-produced models and the progress of research and development models. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a method for collecting and optimizing processing data of aero-engine parts. This method enables the collection and spatiotemporal mapping of processing data, ensuring a one-to-one correspondence between processing data and the geometric spatial location of the processed parts. This allows for correlation analysis between processing results and part structure. Through extensive data accumulation and optimization algorithms for process parameters, the method optimizes process parameters for similar parts or features. Simultaneously, a processing quality and efficiency prediction model is constructed to predict parameters such as processing efficiency, processing accuracy, surface quality, tool life, processing deformation, processing chatter, and surface integrity. This guides actual processing, effectively improving production efficiency and saving processing costs.

[0004] A method for collecting and optimizing the application of machining data for aero-engine parts includes the following steps:

[0005] Step 1: Summarize the methods for acquiring processed data, define the types of data to be collected, and apply the corresponding acquisition methods to acquire processed data according to the data types;

[0006] The data types are divided into sorted data and collected data; sorted data is data that can be directly obtained, while collected data is data that needs to be detected and collected but cannot be directly obtained.

[0007] The data acquisition methods are divided into two categories. The acquisition schemes for data acquisition-type data are as follows: A spindle power or torque acquisition system is embedded based on the CNC machine tool's standard communication interface or a third-party interface to acquire spindle power or torque data; spindle motor current detection technology is used to detect cutting force and tool wear; vibration and deformation data during machining are detected through embedded non-disruptive sensors attached to the spindle and worktable, on-machine probes, and the CNC system's own motion axis vibration monitoring; and three-dimensional morphology measuring instruments and residual stress testing and analysis systems are used to measure and acquire data on the three-dimensional morphology and residual stress of the machined surface. The acquisition method for data sorting-type data is as follows: According to the acquisition strategy and specifications, the type of data to be sorted is defined, and data is acquired according to the rules.

[0008] Step 2: Perform spatiotemporal mapping processing on the time domain data or sorting data in the collected data and the geometric spatial position of the part processing, and store the processing data in the process database;

[0009] The spatiotemporal mapping processing method is based on the tool trajectory of typical part CNC machining. It introduces a volume element model to discretize the machining process in the time and space domains. A short time domain processing method is used to characterize the machining process signal into the corresponding short time domain signal features. A knowledge association between the unit volume element working condition and the short time machining process signal is established to form a spatiotemporal mapping model corresponding to the machining process data and the part machining position, so as to map the machining process data to the part space and store the machining process data in the process database.

[0010] The data stored in the process database are typical part machining site condition data, including process route, machining stage, machining method, machine tool equipment, clamping method, selected tool, programming strategy, cutting parameters, cooling and lubrication, spindle power, cutting force, cutting heat, surface morphology, tool wear, deformation, vibration, and surface integrity.

[0011] Step 3: Construct a mapping relationship model for predicting processing quality and efficiency, establish the mapping relationship between process parameters and the prediction of processing quality and efficiency, and continuously train and iterate the model using process parameters and processing quality and efficiency data from the process database as boundary inputs to obtain the processing quality and efficiency prediction model.

[0012] The specific construction of the machining quality and efficiency prediction mapping model is as follows: process parameters are used as input to the network structure, and surface roughness, tool life, and workpiece fatigue life evaluation parameters are used as output. Specifically, it includes a surface roughness prediction model, a tool life application prediction model, and a workpiece fatigue life prediction model to realize the mapping between process parameters and machining quality and efficiency.

[0013] The surface roughness prediction model is based on an artificial neural network. It takes cutting feed, feed rate, and cutting depth as inputs to the network structure and surface roughness as output, and establishes an N-Net multi-hidden-layer neural network model for predicting the surface roughness of machined parts by integrating local weights.

[0014] The tool life application prediction model is developed by collecting data during tool use, including cutting speed, depth of cut, feed rate, tool material information, cutting environment temperature and humidity, training a neural network model, optimizing model parameters, and evaluating model performance through cross-validation.

[0015] The workpiece fatigue life prediction model utilizes the morphology of the metal surface detected by a white light interferometer. After filtering the detected point cloud data, sparse outliers are removed and the point cloud data is smoothed. Then, defects, grooves, and cracks are identified based on the morphological feature image, and finally fatigue life prediction is achieved.

[0016] Step 4: Construct a process optimization model, conduct sample tests, and verify the optimization results of the process parameters. If the processing effect meets the requirements, the process optimization model is completed. If the processing effect does not meet the requirements, iterate the process optimization model repeatedly until the processing effect meets the requirements.

[0017] The process optimization model is based on neural networks and genetic algorithms, with processing quality and efficiency as the optimization objective and process data that matches the requirements for processing quality and efficiency evaluation as the basic constraint.

[0018] Step 5: Use the operating condition data contained in the process database as the raw data, and embed the processing quality and efficiency prediction model and process optimization model into the process database.

[0019] Step 6: Apply the processing quality and efficiency prediction model and process optimization model of the process database to continuously accumulate and iterate process data, thereby improving the accuracy of processing quality and efficiency prediction, optimizing process parameters, and improving processing quality and efficiency.

[0020] The machining quality and efficiency prediction model is as follows: enter the machining quality and efficiency prediction model in the process database, select the index to be predicted, input the initial conditions according to the requirements of the machining quality and efficiency prediction model, and output the prediction results; wherein the initial conditions include feed rate, cutting speed, cutting depth, part material, part structure, tool material, tool type, cutting method, and machining equipment, and the prediction result is the surface roughness Ra value;

[0021] The process optimization model is accessed through the process database. It selects the initial conditions for CNC machining, outputs optimized process parameters or a process plan, and, based on prompts from the process database, allows users to choose whether to save the optimized process parameters or plan to the database. The initial conditions include part material, part features, machining equipment, machining method, tool material, tool type, and machining operation information.

[0022] The beneficial effects of adopting the above technical solution are as follows:

[0023] This invention provides a method for collecting and optimizing processing data of aero-engine parts. In traditional CNC machining, the selection of process parameters generally relies on the experience of technicians and accumulated technical data. This approach suffers from limitations in the rationality of process parameter selection, which is constrained by the technicians' capabilities and lacks flexibility. This invention enables the effective collection of machining condition data. Through a large accumulation of data and a process parameter optimization algorithm, it optimizes process parameters for similar parts or features. Simultaneously, the constructed machining quality and efficiency prediction module can predict parameters such as machining efficiency, machining accuracy, surface quality, tool life, machining deformation, machining chatter, and surface integrity, thereby guiding actual machining and effectively improving production efficiency and saving machining costs. Attached Figure Description

[0024] Figure 1 This is an overall flowchart of the method for collecting and optimizing the application of aero-engine parts processing data according to the present invention;

[0025] Figure 2 This invention provides a data acquisition scheme for processing.

[0026] Figure 3 Spatiotemporal mapping processing scheme for processing data

[0027] Figure 4 Data Structure for Predictive Models of Processing Quality Effects Detailed Implementation

[0028] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0029] A method for collecting and optimizing the application of machining data for aero-engine parts, such as Figure 1 As shown, taking the machining data processing of an annular casing as an example, the steps include:

[0030] Step 1: Summarize the methods for acquiring processed data, define the types of data to be collected, and apply the corresponding acquisition methods to acquire processed data according to the data types;

[0031] The data types are divided into organized data and collected data. Organized data is directly obtainable data, including: material (grade, heat treatment state), blank form, typical structural features, process route, machining stage, datum selection, allowance allocation, and processes (machine tools, positioning and clamping methods, fixtures, cutting tools, machined surfaces, tool paths, cutting parameters), etc. Collected data is data that needs to be detected and collected but cannot be directly obtained, including: cutting force / heat, tool wear, machining efficiency, machining accuracy, surface quality, machining deformation, chatter, surface integrity (surface roughness, surface microhardness, surface residual stress, burns), and time-domain data, etc.

[0032] In this embodiment, a data acquisition module is designed to collect information on the material (grade, heat treatment status) of the annular casing, blank form, typical structural features, process route, machining stage, datum selection, allowance allocation, and processes (machine tool equipment, positioning and clamping method, fixture, cutting tool, machined surface, tool path, cutting parameters), etc.

[0033] The data acquisition methods are divided into two categories. The acquisition schemes for data acquisition-type data are as follows: A spindle power or torque acquisition system is embedded based on the CNC machine tool's standard communication interface or a third-party interface to acquire spindle power or torque data; spindle motor current detection technology is used to detect cutting force and tool wear; vibration and deformation data during machining are detected through embedded non-disruptive sensors attached to the spindle and worktable, on-machine probes, and the CNC system's own motion axis vibration monitoring; and three-dimensional morphology measuring instruments and residual stress testing and analysis systems are used to measure and acquire data on the three-dimensional morphology and residual stress of the machined surface. The acquisition method for data sorting-type data is as follows: According to the acquisition strategy and specifications, the type of data to be sorted is defined, and data is acquired according to the rules.

[0034] Step 2 involves performing spatiotemporal mapping between the time-domain data from the collected data set or the data from the organized data set and the geometric spatial location of the part during machining, and then storing the machining process data in the process database, such as... Figure 2 As shown;

[0035] The spatiotemporal mapping processing method is as follows: Figure 3 As shown, the tool path for CNC machining of typical parts is introduced. The machining process is discretized in time and space using a volume element model. The machining process signal is characterized as a corresponding short-time domain signal feature using a short-time domain processing method. A knowledge association between the unit volume element working condition and the short-time machining process signal is established to form a spatiotemporal mapping model between the machining process data and the machining position of the part, so as to map the machining process data to the part space and store the machining process data in the process database.

[0036] The data stored in the process database are typical part machining site condition data, including process route, machining stage, machining method, machine tool equipment, clamping method, selected tool, programming strategy, cutting parameters, cooling and lubrication, spindle power, cutting force, cutting heat, surface morphology, tool wear, deformation, vibration, and surface integrity.

[0037] Step 3: Construct a mapping model for predicting processing quality and efficiency, establishing the mapping relationship between process parameters and the prediction of processing quality and efficiency, such as... Figure 4 As shown, the process parameters and processing quality and efficiency data from the process database are used as boundary inputs to continuously train and iterate the model to obtain a processing quality and efficiency prediction model; ultimately ensuring the accuracy of the prediction model.

[0038] The specific steps for constructing the machining quality and efficiency prediction mapping model are as follows: process parameters are used as input to the network structure, and surface roughness, tool life, and workpiece fatigue life evaluation parameters are used as output. Specifically, it includes a surface roughness prediction model, a tool life application prediction model, and a workpiece fatigue life prediction model to realize the mapping between process parameters and machining quality and efficiency. At the same time, the process parameters and machining quality and efficiency data in the process process database are used as boundary inputs to continuously train the model, ultimately ensuring the accuracy of the prediction model.

[0039] The surface roughness prediction model is based on an artificial neural network. It takes cutting feed, feed rate, and cutting depth as inputs to the network structure and surface roughness as output, and establishes an N-Net multi-hidden-layer neural network model for predicting the surface roughness of machined parts by integrating local weights.

[0040] The tool life application prediction model is developed by collecting data during tool use, including cutting speed, depth of cut, feed rate, tool material information, cutting environment temperature and humidity, training a neural network model, optimizing model parameters, and evaluating model performance through cross-validation to continuously improve the model's prediction accuracy.

[0041] The workpiece fatigue life prediction model utilizes the morphology of the metal surface detected by a white light interferometer. After filtering the detected point cloud data, sparse outliers are removed and the point cloud data is smoothed. Then, defects, grooves, and cracks are identified based on the morphological feature image, and finally fatigue life prediction is achieved.

[0042] Step 4: Develop process optimization algorithms, construct process optimization models, conduct sample tests, and verify the optimization results of process parameters. If the processing effect meets the requirements, the process optimization model is completed. If the processing effect does not meet the requirements, iterate the process optimization model repeatedly until the processing effect meets the requirements.

[0043] The process optimization model is based on neural networks and genetic algorithms, with processing quality and efficiency as the optimization objective and process data that matches the requirements for processing quality and efficiency evaluation as the basic constraint.

[0044] This embodiment takes surface quality and processing efficiency as optimization objectives, and takes parameters such as processing equipment, processing materials, service environment, and processing requirements as basic constraints.

[0045] Step 5: Use the operating condition data contained in the process database as the raw data, and embed the processing quality and efficiency prediction model and process optimization model into the process database.

[0046] Step 6: Apply the processing quality and efficiency prediction model and process optimization model of the process database to continuously accumulate and iterate process data, thereby improving the accuracy of processing quality and efficiency prediction, optimizing process parameters, and improving processing quality and efficiency.

[0047] The machining quality and efficiency prediction model is as follows: enter the machining quality and efficiency prediction model in the process database, select the index to be predicted, input the initial conditions according to the requirements of the machining quality and efficiency prediction model, and output the prediction results; wherein the initial conditions include feed rate, cutting speed, cutting depth, part material, part structure, tool material, tool type, cutting method, and machining equipment, and the prediction result is the surface roughness Ra value;

[0048] The process optimization model is accessed through the process database. It selects the initial conditions for CNC machining, outputs optimized process parameters or a process plan, and, based on prompts from the process database, allows users to choose whether to save the optimized process parameters or plan to the database. The initial conditions include part material, part features, machining equipment, machining method, tool material, tool type, and machining operation information.

[0049] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for collecting and optimizing the application of machining data for aero-engine parts, characterized in that, Includes the following steps: Step 1: Summarize the methods for acquiring processed data, define the types of data to be collected, and apply the corresponding acquisition methods to acquire processed data according to the data types; Step 2: Perform spatiotemporal mapping processing on the time domain data or sorting data in the collected data and the geometric spatial position of the part processing, and store the processing data in the process database; The spatiotemporal mapping processing method is based on the tool path of typical CNC machining of parts. It introduces a volume element model to discretize the machining process in the time and space domains. A short time domain processing method is used to characterize the machining process signal into the corresponding short time domain signal features. Establish a knowledge association between unit volume element working conditions and short-term machining process signals to form a spatiotemporal mapping model corresponding to machining process data and part machining position, so as to map the machining process data to the part space and store the machining process data in the process database; Step 3: Construct a mapping relationship model for predicting processing quality and efficiency, establish the mapping relationship between process parameters and the prediction of processing quality and efficiency, and continuously train and iterate the model using process parameters and processing quality and efficiency data from the process database as boundary inputs to obtain the processing quality and efficiency prediction model. Step 4: Construct a process optimization model, conduct sample tests, and verify the optimization results of the process parameters. If the processing effect meets the requirements, the process optimization model is completed. If the processing effect does not meet the requirements, iterate the process optimization model repeatedly until the processing effect meets the requirements. Step 5: Use the operating condition data contained in the process database as the raw data, and embed the processing quality and efficiency prediction model and process optimization model into the process database. Step 6: Apply the processing quality and efficiency prediction model and process optimization model of the process database to continuously accumulate and iterate process data, thereby improving the accuracy of processing quality and efficiency prediction, optimizing process parameters, and improving processing quality and efficiency.

2. The method for collecting and optimizing the application of machining data for aero-engine parts according to claim 1, characterized in that, The data types mentioned in step 1 are divided into sorted data and collected data; sorted data is data that can be directly obtained, while collected data is data that needs to be detected and collected but cannot be directly obtained. The data acquisition methods are divided into two categories. The acquisition schemes for data acquisition-type data are as follows: A spindle power or torque acquisition system is embedded based on the CNC machine tool's standard communication interface or a third-party interface to acquire spindle power or torque data; spindle motor current detection technology is used to detect cutting force and tool wear; vibration and deformation data during machining are detected through embedded non-disruptive sensors attached to the spindle and worktable, on-machine probes, and the CNC system's own motion axis vibration monitoring; and three-dimensional morphology measuring instruments and residual stress testing and analysis systems are used to measure and acquire data on the three-dimensional morphology and residual stress of the machined surface. The acquisition method for data sorting-type data is as follows: According to the acquisition strategy and specifications, the types of data to be sorted are defined, and data is acquired according to the rules.

3. The method for collecting and optimizing the application of machining data for aero-engine parts according to claim 1, characterized in that, The data stored in the process database are typical part machining site condition data, including process route, machining stage, machining method, machine tool equipment, clamping method, selected tool, programming strategy, cutting parameters, cooling and lubrication, spindle power, cutting force, cutting heat, surface morphology, tool wear, deformation, vibration, and surface integrity.

4. The method for collecting and optimizing the application of machining data for aero-engine parts according to claim 1, characterized in that, The construction of the machining quality and efficiency prediction mapping model in step 3 specifically involves taking process parameters as input to the network structure and taking surface roughness, tool life, and workpiece fatigue life evaluation parameters as output. Specifically, it includes a surface roughness prediction model, a tool life application prediction model, and a workpiece fatigue life prediction model to realize the mapping between process parameters and machining quality and efficiency.

5. The method for collecting and optimizing the application of aero-engine part machining data according to claim 4, characterized in that, The surface roughness prediction model is based on an artificial neural network. It takes cutting feed, feed rate, and cutting depth as inputs to the network structure and surface roughness as output, and establishes an N-Net multi-hidden-layer neural network model for predicting the surface roughness of machined parts by integrating local weights. The tool life application prediction model is developed by collecting data during tool use, including cutting speed, depth of cut, feed rate, tool material information, cutting environment temperature and humidity, training a neural network model, optimizing model parameters, and evaluating model performance through cross-validation. The workpiece fatigue life prediction model utilizes the morphology of the metal surface detected by a white light interferometer. After filtering the detected point cloud data, it removes sparse outliers and smooths the point cloud data. Then, it identifies defects, grooves, and cracks based on the morphological feature image, and finally realizes fatigue life prediction.

6. The method for collecting and optimizing the application of machining data for aero-engine parts according to claim 1, characterized in that, The process optimization model described in step 4 is based on neural networks and genetic algorithms, with processing quality and efficiency as the optimization objective and process data that matches the requirements for processing quality and efficiency evaluation as the basic constraints.

7. The method for collecting and optimizing the application of machining data for aero-engine parts according to claim 1, characterized in that, Step 6 describes the processing quality and efficiency prediction model: Enter the processing quality and efficiency prediction model in the process database, select the indicators to be predicted, input the initial conditions according to the requirements of the processing quality and efficiency prediction model, and output the prediction results; The initial conditions include feed rate, cutting speed, depth of cut, part material, part structure, tool material, tool type, cutting method, and machining equipment. The predicted result is the surface roughness Ra value. The process optimization model: The process optimization model enters the process database, selects the initial conditions for CNC machining, outputs optimized process parameters or process schemes, and selects whether to save the optimized process parameters or process schemes into the process database according to the prompts of the process database. The initial conditions include part material, part features, processing equipment, processing method, tool material, tool type, and processing procedure information.

Images