A smart manufacturing system for large, complex, thin-walled parts

Abstract

This invention discloses an intelligent manufacturing system for large, complex, thin-walled parts, including a process database, an intelligent control system, a machining system, a measurement feedback system, and a monitoring system. This invention provides an automated and intelligent intelligent manufacturing system for the entire machining process of large, complex, thin-walled parts.

Description

Technical Field

[0001] This invention belongs to the field of intelligent manufacturing technology and relates to an intelligent manufacturing system for large, complex, thin-walled parts. Background Technology

[0002] Large, complex, thin-walled components, due to their unique characteristics of low mass, high load-bearing capacity, and high strength, are widely used in engineering fields such as shipbuilding, automotive, and aerospace, as parts for skins, load-bearing structures, and fuel tanks. While these components are generally formed using milling, a series of difficulties and shortcomings still exist in their milling manufacturing process:

[0003] (1) Due to the large amount of material removed and the complete release of material stress, large machining deformation is easily generated. In addition, thin-walled parts have low rigidity and are prone to machining vibration during milling, which seriously affects the surface quality after forming.

[0004] (2) In order to avoid the impact of machining deformation on the forming process during the forming process, heat treatment process needs to be interspersed in the machining material removal process. However, the arrangement of the process flow is based solely on the experience of the process personnel, which may cause hidden dangers to the performance of the product.

[0005] (3) The workload of production and processing operators is high. During the processing, it is necessary to manually inspect the surface quality and machining deformation at any time and adjust the machine tool cutting parameters, resulting in low product processing efficiency.

[0006] (4) The forming process of thin-walled parts is complex and requires transfer between various workshops and equipment, which makes it extremely difficult to arrange the production tasks in the workshop and seriously affects the production efficiency of the products. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide an intelligent manufacturing system for large, complex, thin-walled parts.

[0008] To achieve the above objectives, the present invention employs the following technical solutions:

[0009] A smart manufacturing system for large, complex, thin-walled parts includes a process database, an intelligent control system, a machining system, a measurement feedback system, and a monitoring system.

[0010] The process database is used to collect and store process data and process experience data during the manufacturing process, perform knowledge modeling on the process data and process experience data to form an expert knowledge system, and output process routes and processing parameters by combining the input workpiece information.

[0011] The intelligent control system is used to receive process routes and processing parameters from the process database, generate NC programs, and perform intelligent scheduling of production plans.

[0012] After receiving instructions from the intelligent control system, the machining system completes the machining and heat treatment processes according to the NC program and process specifications.

[0013] The monitoring system uses vibration sensors, sound sensors, temperature sensors, and surveillance cameras to monitor the operation of the equipment, and the obtained equipment operation information is fed back to the intelligent control system in real time.

[0014] The measurement feedback system uses computer vision and point cloud technology to reconstruct the workpiece in three dimensions, detect its deformation in real time, and feed it back to the process database.

[0015] As a preferred approach, the process database adjusts the machining parameters and process route based on the machining deformation information fed back by the measurement feedback system.

[0016] As a preferred embodiment, the process data includes machine tool speed, feed rate, depth of cut, and the corresponding key dimensions of the workpiece to be processed.

[0017] As a preferred approach, the process experience data includes the process route, machining allowances for each process step, and heat treatment nodes.

[0018] As a preferred approach, the process database establishes a dynamic multi-objective optimization algorithm for product manufacturability. Based on the process route proposed by the expert knowledge system, the dynamic multi-objective optimization algorithm is used to obtain the most suitable processing parameters for each process step. The model of the dynamic multi-objective optimization algorithm is as follows:

[0019]

[0020] Where x represents the three machining elements (machine speed, feed rate, depth of cut), critical dimension, and machining time; X represents the actual three machining elements, critical dimension, and machining time; μ represents the three machining elements, critical dimension, and machining time in the database; n represents the number of data sets in the process database; and R represents the variation limit of the three machining elements, critical dimension, and machining time.

[0021] As a preferred embodiment, the monitoring system integrates a fault diagnosis neural network algorithm, into which machine tool temperature (T1, T2...Tm), sound signal (H1, H2...Hk), and vibration signal (f1, f2...fn) are input, and the probability of equipment failure is output.

[0022] The present invention has the following advantages:

[0023] This invention provides an automated and intelligent intelligent manufacturing system for the entire machining process of large, complex, thin-walled parts. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0025] The present invention will now be described in detail with reference to the accompanying drawings.

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0027] Example:

[0028] A smart manufacturing system for large, complex, thin-walled parts includes a process database, an intelligent control system, a machining system, a measurement feedback system, and a monitoring system.

[0029] (1) A database of part process specifications and machining data is established based on past process experience and machining data. This database is supported by the PDM system (process experience) and QMS system (machining data) commonly used in digital workshops. Process experience includes process routes, machining allowances for each process and heat treatment nodes; machining data includes the three machining elements (machine tool speed, feed rate, depth of cut) and the inspection status of the key dimensions of the product corresponding to the three machining elements.

[0030] (2) Based on the data in the processing database, after obtaining clearly calibrated and reliable processing data through data cleaning and data mining techniques, the processing data is modeled using a transformer natural language processing model based on a seq2seq architecture to construct an expert knowledge system. The structural features, raw material models, materials, key dimensions, etc. of the target processed parts are input into the expert knowledge system in the form of questions and answers to obtain the processing route given by the expert knowledge system.

[0031] (3) By utilizing the influence of the process experience and processing parameters stored in the database on the machining deformation and processing efficiency of parts, a dynamic multi-objective optimization algorithm for product manufacturability is established.

[0032] The model of the dynamic multi-objective optimization algorithm is as follows:

[0033]

[0034] Where x represents the three machining elements, critical dimensions, and machining time; X represents the actual three machining elements, critical dimensions, and machining time; μ represents the three machining elements, critical dimensions, and machining time in the database; n represents the number of data sets in the database; and R represents the variation limit of the three machining elements, critical dimensions, and machining time.

[0035] Based on the process route proposed by the expert knowledge system, the dynamic multi-objective optimization algorithm for product manufacturability is used to obtain the most suitable processing parameters for each process.

[0036] (4) After receiving the process route and processing parameters from the process database, the intelligent control system can generate nc programs and perform intelligent scheduling of production plans.

[0037] (5) After receiving instructions from the intelligent control system, the machining system can complete the corresponding machining and heat treatment processes according to the nc program and process specifications.

[0038] (6) The monitoring system uses vibration sensors, sound sensors, temperature sensors and monitoring cameras to monitor the operation of equipment such as machine tools and thermal aging furnaces. The monitoring system integrates a fault diagnosis neural network algorithm. The machine tool temperature (T1,T2...Tm), sound signal (H1,H2...Hk), and vibration signal (f1,f2...fn) are input into the fault diagnosis neural network algorithm, and the probability of equipment failure is output. The obtained equipment operation information is fed back to the intelligent control system in real time.

[0039] (7) During the machining process, the measurement feedback system uses computer vision and point cloud technology to reconstruct the workpiece in three dimensions. Its deformation is detected in real time and fed back to the process database. After receiving the deformation information, the process database updates the process experience data in the database and optimizes the machining parameters and process routes accordingly. This achieves closed-loop control of the entire system to ensure product quality.

[0040] This invention is not limited to the specific embodiments described above. The invention extends to any new feature or combination disclosed in this specification, as well as any new method or process step or combination disclosed herein.

Claims

1. A smart manufacturing system for large, complex, thin-walled parts, characterized in that: This includes a process database, intelligent control system, machining system, measurement feedback system, and monitoring system. The process database is used to collect and store process data and process experience data during the manufacturing process, perform knowledge modeling on the process data and process experience data to form an expert knowledge system, and output process routes and processing parameters by combining the input workpiece information. The intelligent control system is used to receive process routes and processing parameters from the process database, generate CNC programs, and intelligently schedule production plans. After receiving instructions from the intelligent control system, the machining system completes the machining and heat treatment processes according to the CNC program and process specifications. The monitoring system uses vibration sensors, sound sensors, temperature sensors, and surveillance cameras to monitor the operation of the equipment, and the obtained equipment operation information is fed back to the intelligent control system in real time. The measurement feedback system uses computer vision and point cloud technology to reconstruct the workpiece in three dimensions, detect its deformation in real time, and feed it back to the process database. The process data includes machine tool speed, feed rate, depth of cut, and corresponding critical dimensions of the workpiece to be machined; the process experience data includes process routes, machining allowances for each process, and heat treatment nodes; the process database establishes a dynamic multi-objective optimization algorithm for product manufacturability. Based on the process route proposed by the expert knowledge system, the dynamic multi-objective optimization algorithm for product manufacturability is used to obtain the most suitable machining parameters for each process. The model of the dynamic multi-objective optimization algorithm is as follows: Where x represents the three machining elements, critical dimensions, and machining time; X represents the actual three machining elements, critical dimensions, and machining time; μ represents the three machining elements, critical dimensions, and machining time in the database; n represents the number of data sets in the process database; and R represents the variation limit of the three machining elements, critical dimensions, and machining time.

2. The intelligent manufacturing system for large, complex, thin-walled parts according to claim 1, characterized in that: The process database adjusts the machining parameters and process routes based on the machining deformation information fed back by the measurement feedback system.

3. The intelligent manufacturing system for large, complex, thin-walled parts according to claim 1, characterized in that: The monitoring system integrates a fault diagnosis neural network algorithm. The algorithm takes machine tool temperature (T1, T2...Tm), sound signal (H1, H2...Hk), and vibration signal (f1, f2...fn) as inputs and outputs the probability of equipment failure.

Images