Abnormality detection method and system for intelligent manufacturing equipment

Abstract

The invention relates to the technical field of anomaly detection, in particular to an anomaly detection method and system for intelligent manufacturing equipment, and the method comprises the following steps: determining the working areas of the intelligent manufacturing equipment, and setting monitoring ports corresponding to the working areas, establishing a local area network and a corresponding network communication protocol among the monitoring port, the abnormity judgment module and the central control platform; along with the start of work of the intelligent manufacturing equipment, the central control platform synchronously sends a working state monitoring instruction to each monitoring port, and the monitoring ports obtain a real-time fusion index of the current working state in real time based on the working state monitoring instruction; the abnormity judgment module receives the real-time fusion index, executes abnormity judgment after analysis and sends an abnormal state feedback instruction to the central control platform, and the central control platform analyzes the abnormal state feedback instruction and outputs a fault code corresponding to the current working state; and timely, comprehensive, accurate and automatic abnormal condition judgment in the working process of the intelligent manufacturing equipment is realized.

Description

Technical Field

[0001] The present invention relates to the field of anomaly detection technology, and in particular to an anomaly detection method and system for intelligent manufacturing equipment. Background Art

[0002] Smart manufacturing equipment can realize the automation, intelligence and refined operation of the manufacturing process. It comprehensively applies a variety of smart manufacturing technologies such as advanced manufacturing, digital control, and intelligent sensing, improves the agility of the manufacturing process, and has significant advantages in improving product production quality and efficiency, reducing costs and energy consumption, etc.

[0003] Traditional intelligent manufacturing equipment cannot achieve real-time monitoring and instant feedback due to the lack of targeted communication network methods. Due to the lack of effective data exchange and synchronization mechanisms, abnormal information cannot be quickly and accurately transmitted to the monitoring system or operators, resulting in a slow response to abnormal event identification. The shortcomings of this communication system not only increase the difficulty of monitoring the production line, but also reduce the timeliness of equipment fault warning and processing, thereby affecting the intelligence level and automated operation of the entire manufacturing process.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for detecting abnormalities in intelligent manufacturing equipment, so as to realize timely, comprehensive, accurate and automatic determination of abnormal situations during the operation of intelligent manufacturing equipment. In order to achieve the above object, the technical solution adopted by the present invention is: a method for detecting abnormalities of intelligent manufacturing equipment, wherein the intelligent manufacturing equipment is connected to a central control platform through an abnormality determination module, and the abnormality detection method comprises the following steps: Determine the working areas of the intelligent manufacturing equipment and set monitoring ports corresponding to each of the working areas, establish a local area network and a corresponding network communication protocol between the monitoring ports, the abnormality determination module and the central control platform; As the intelligent manufacturing equipment starts working, the central control platform synchronously sends a working status monitoring instruction to each of the monitoring ports based on the local area network and the network communication protocol, and the monitoring port obtains the real-time fusion index of the current working status in the production process of the intelligent manufacturing equipment in real time based on the working status monitoring instruction; The abnormality determination module receives the real-time fusion index based on the local area network and the network communication protocol, performs abnormality determination after analysis, and sends an abnormal status feedback instruction to the central control platform. The central control platform analyzes the abnormal status feedback instruction and outputs a fault code corresponding to the current working status.

[0006] Furthermore, the monitoring port obtains the real-time fusion index of the current working state in the production process of the intelligent manufacturing equipment in real time based on the working state monitoring instruction; comprising the following steps: Determine several working areas of intelligent manufacturing equipment and set up corresponding monitoring probes for each working area; Each of the monitoring ports receives the working status monitoring instruction, and based on the working status monitoring instruction, performs real-time video monitoring of the current working area of ​​the intelligent manufacturing equipment through the corresponding monitoring probe to obtain a real-time video record of the current working status; Intercepting continuous valid video frames in the real-time video record, and performing feature separation on the continuous valid video frames to obtain image information frames and audio information frames; Extracting features of the image information frame and the audio information frame respectively to obtain corresponding image feature vectors and audio feature vectors; The image feature vector and the audio feature vector are spliced ​​and fused through a fusion network model to obtain the real-time fusion index of the current working state.

[0007] Furthermore, the image feature vector and the audio feature vector are spliced ​​and fused by a fusion network model to obtain the real-time fusion index of the current working state; comprising the following steps: splicing the image feature vector and the audio feature vector in a feature dimension to form a splicing feature; Determine a weight matrix and a bias vector corresponding to the splicing feature based on the current working state, and construct a fusion network model according to the weight matrix and the bias vector; The splicing features are fused and processed through the fusion network model to form the real-time fusion index.

[0008] Further, the abnormality determination module receives the real-time fusion index based on the local area network and the network communication protocol, performs abnormality determination after analysis, and sends an abnormal state feedback instruction to the central control platform, and the central control platform analyzes the abnormal state feedback instruction and outputs a fault code corresponding to the current working state; comprising the following steps: Determine the standard indicator interval corresponding to the current working state, and compare the real-time fusion indicator with the standard indicator interval, and determine whether the current working state of the intelligent manufacturing equipment is in a normal state according to the determination result; If it is within the standard index range, the real-time fusion index of the working status in the next production process will be compared and determined as the working process progresses until the intelligent manufacturing equipment completes a production process cycle; If it is not within the standard indicator range, an abnormal state feedback instruction is sent to the central control platform, and the central control platform parses the abnormal state feedback instruction and outputs a fault code corresponding to the current working state.

[0009] Furthermore, the production process of the intelligent manufacturing equipment includes an equipment preparation process, a raw material loading process, an equipment operation process, a product unloading process, and a waste removal process.

[0010] Furthermore, after performing abnormality detection on the working state of the equipment operation process, a product abnormality detection process is also included, and the product abnormality detection includes the following steps: Selecting the last frame image at the end of the operation process of the equipment as the target image corresponding to the product, and extracting the target feature vector of the product in the target image; Acquire a standard image of a product and a standard feature vector of the product in the standard image; Determining whether the product in the target image is qualified according to the matching degree between the target feature vector and the standard feature vector; If the current product is unqualified, an abnormal product feedback instruction is output and the waste rejection process is executed; If the current product is qualified, the product unloading process is executed.

[0011] Furthermore, the number of times the waste rejection process and the product unloading process are executed during the operation of the intelligent manufacturing equipment is counted and recorded in real time to determine the product yield information.

[0012] Furthermore, it also includes constructing an exception database, which includes a number of exception information, each of which includes a one-to-one corresponding fault code, abnormal situation and solution, and the exception database is updated in real time with the output of each fault code; the central control platform retrieves the exception database according to the fault code and outputs the abnormal situation corresponding to the current working state and its corresponding solution.

[0013] The present invention also provides an abnormality detection system for intelligent manufacturing equipment, comprising a central control platform, a monitoring port and an abnormality determination module; The central control platform is used to send a working status monitoring instruction to the monitoring port synchronously with the start of the intelligent manufacturing equipment and to determine whether to execute the real-time fusion index comparison and determination process of the next production process according to the output result of the abnormality determination module; The monitoring port is used to receive the working state monitoring instruction, and obtain the real-time fusion index of the current working state in the production process of the intelligent manufacturing equipment in real time based on the working state monitoring instruction; The abnormality determination module is used to determine the standard index interval corresponding to the current working state, and compare the real-time fusion index with the standard index interval, and determine whether the current working state of the intelligent manufacturing equipment is in a normal state according to the determination result; If it is within the standard index range, the real-time fusion index of the working status in the next production process will be compared and determined as the working process progresses until the intelligent manufacturing equipment completes a production process cycle; If it is not within the standard indicator range, an abnormal status feedback instruction is sent to the central control platform. The central control platform outputs a fault code corresponding to the current working state based on the abnormal status feedback instruction, and outputs the abnormal situation corresponding to the current working state and its corresponding solution according to the fault code.

[0014] Furthermore, it also includes a product anomaly detection module, which is used to determine whether the current product is in a qualified state after performing an abnormality detection on the working state of the equipment operation process.

[0015] Furthermore, it also includes an exception database, which includes a number of exception information, each of which includes a one-to-one corresponding fault code, abnormal situation and solution, and the exception database is updated in real time with the output of each fault code; the central control platform retrieves the exception database according to the fault code and outputs the abnormal situation corresponding to the current working state and its corresponding solution.

[0016] The beneficial effects of the present invention are as follows: the present invention realizes efficient data exchange and synchronization between the monitoring ports of each working area of ​​the intelligent manufacturing equipment and the central control platform by establishing a local area network and a corresponding network communication protocol, thereby improving the reliability and real-time performance of data transmission. By setting up a central control platform, the working status of the intelligent manufacturing equipment in the production process is acquired in real time. The abnormality judgment module performs abnormality judgment based on real-time fusion indicators, which reduces the dependence on manual monitoring, improves the automation level of abnormality handling, and reduces the possibility of human error, thereby realizing real-time monitoring and automated abnormality judgment, and can quickly identify abnormal situations of intelligent manufacturing equipment in the production process, making fault handling more standardized and systematic, thereby improving production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 Schematic diagram of a process for detecting abnormality of intelligent manufacturing equipment according to an embodiment of the present invention; Figure 2 A schematic diagram of a process for obtaining a real-time fusion index in an embodiment of the present invention; Figure 3 Schematic diagram of the process of splicing and fusing image feature vectors and audio feature vectors in an embodiment of the present invention; Figure 4 Schematic diagram of the process of abnormal state determination in an embodiment of the present invention; Figure 5 Schematic diagram of the process flow of product abnormality detection process in an embodiment of the present invention; Figure 6 Schematic diagram of the structure of the intelligent manufacturing equipment abnormality detection system in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0020] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only and do not represent the only implementation method.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0022] like Figures 1 to 5The abnormality detection method of the intelligent manufacturing equipment shown in the figure, the intelligent manufacturing equipment is connected to the central control platform through the abnormality determination module, and the abnormality detection method includes the following steps: Determine the working areas of the intelligent manufacturing equipment and set monitoring ports corresponding to each of the working areas, establish a local area network and a corresponding network communication protocol between the monitoring ports, the abnormality determination module and the central control platform; As the intelligent manufacturing equipment starts working, the central control platform sends a working status monitoring instruction to each monitoring port synchronously based on the local area network and the network communication protocol, and the monitoring port obtains the real-time fusion index of the current working status in the production process of the intelligent manufacturing equipment in real time based on the working status monitoring instruction; The abnormality determination module receives the real-time fusion index based on the local area network and the network communication protocol, performs abnormality determination after analysis, and sends an abnormal status feedback instruction to the central control platform. The central control platform analyzes the abnormal status feedback instruction and outputs a fault code corresponding to the current working status.

[0023] The present invention realizes efficient data exchange and synchronization between the monitoring ports of each working area of ​​the intelligent manufacturing equipment and the central control platform by establishing a local area network and a corresponding network communication protocol, thereby improving the reliability and real-time performance of data transmission. By setting up a central control platform, the working status of the intelligent manufacturing equipment in the production process is acquired in real time. The abnormality judgment module performs abnormality judgment based on real-time fusion indicators, which reduces the dependence on manual monitoring, improves the automation level of abnormality handling, and reduces the possibility of human error, thereby realizing real-time monitoring and automated abnormality judgment, and can quickly identify abnormal situations of intelligent manufacturing equipment in the production process, making fault handling more standardized and systematic, thereby improving production efficiency and product quality.

[0024] It should be noted that the start, stop and progress of the working process of intelligent manufacturing equipment can also be uniformly scheduled and controlled by the central control platform to ensure the synchronization between the anomaly detection process and the production process of intelligent manufacturing equipment and reduce response errors.

[0025] LANs usually have a small coverage area and are relatively simple to manage and maintain. Therefore, they can provide high-bandwidth and low-latency data transmission, and perform network monitoring and maintenance more efficiently, reducing interruptions in working status monitoring due to network problems; they can quickly and reliably transmit large amounts of data when performing abnormality detection on smart manufacturing equipment, thereby improving production efficiency; network communication protocols are a set of regulations and conventions that manage and control network communications to ensure that data transmission in the network is orderly and reliable. Common protocols include Ethernet protocols, transmission control protocols / Internet protocols, etc., which are not specifically limited here; network communication protocols ensure effective communication between monitoring ports in different areas of smart manufacturing equipment and between smart manufacturing equipment and the central control platform; dividing and integrating different working areas of smart manufacturing equipment into the LAN can not only improve network management efficiency and security, but also optimize network performance, reduce costs, and provide higher flexibility and scalability to ensure efficient, stable and safe operation of the system.

[0026] Specifically, in different production processes of intelligent manufacturing equipment, due to the influence of factors such as processing procedures and processed products, the working status of intelligent manufacturing equipment is not the same. Therefore, in the monitoring process, by dividing the intelligent manufacturing equipment into different working areas and integrating them into the local area network, the monitoring port extracts the status of the positions of several working areas corresponding to different production processes and obtains real-time fusion indicators, which can reflect the current working status, and then implement specific network management strategies for each area, thereby improving the efficiency and pertinence of network management; in the process of obtaining real-time fusion indicators, since each working area is an independent grid unit, the mutual reading and influence of data between different areas is greatly reduced, and the accuracy of the monitoring results is ensured. Through the constructed local area network and network communication protocol, the timeliness and accuracy of the transmission of working status monitoring instructions and abnormal status feedback instructions between the central control platform and each monitoring port can be achieved, which is also convenient for subsequent fault location and troubleshooting, reducing the complexity and time of troubleshooting.

[0027] On the basis of the above embodiment, the monitoring port obtains the real-time fusion index of the current working state in the production process of the intelligent manufacturing equipment in real time based on the working state monitoring instruction; including the following steps: Determine several working areas of intelligent manufacturing equipment and set corresponding monitoring probes for each working area; accurately monitor the working status of intelligent manufacturing equipment in each process to ensure the pertinence and effectiveness of monitoring; Each of the monitoring ports receives the working status monitoring instruction, and based on the working status monitoring instruction, performs real-time video monitoring of the current working area of ​​the intelligent manufacturing equipment through the corresponding monitoring probe to obtain a real-time video record of the current working status; this helps to capture the working details of the intelligent manufacturing equipment in the working area of ​​each production process in a refined manner, so as to facilitate the subsequent timely and accurate discovery of abnormal situations; Intercept continuous valid video frames in real-time video recording, and perform feature separation on continuous valid video frames to obtain image information frames and audio information frames; by selecting continuous valid video frames, it is possible to ensure that key moments in the current working state of the device are captured, improve the consistency and accuracy of data, and ensure the clarity of images and audio in the subsequent feature separation process. The separated image information frames and audio information frames can clearly and quickly distinguish different types of data, which is convenient for subsequent targeted extraction of corresponding features; The features of the image information frame and the audio information frame are extracted respectively to obtain the corresponding image feature vector and audio feature vector, which can reflect the working state of the intelligent manufacturing equipment from the two aspects of visual state and sound feature, respectively, and ensure the comprehensiveness of feature extraction; specifically, in different production processes of intelligent manufacturing equipment, the working state reflected by the equipment is also different, which is generally expressed through the running track and the sound generated by the processing. By separating and extracting the image information frame and the audio information frame, it is convenient to separate different types of data information, so as to comprehensively identify and judge the occurrence of abnormal situations by combining the information of the two parts of image and audio, and reduce the misjudgment caused by a single data source. In addition, the feature extraction fusion method can ensure the accuracy of intelligent monitoring of abnormal situations in the face of noise or interference, and improve the reliability of overall abnormal monitoring; In addition, the extracted image information frames and audio information frames provide feedback and optimization for the division of the working area in the subsequent working status monitoring process to a certain extent, and feature extraction is performed on the image information frames and audio information frames, including features such as shape, texture, color in the image, and features such as frequency and rhythm in the audio. After machine learning, the features can be finely identified, thereby greatly improving the accuracy of abnormal monitoring; and after iterative reading of the features on the separated image information frames and audio information frames for multiple times, monitoring noise points or abnormal frequent areas in the current working area can be found, so that the working area can be re-divided to form targeted monitoring or elimination of noise areas. According to the above-mentioned feedback iteration mechanism, the abnormal judgment module can continuously capture abnormal events in the image information frames and audio information frames, and further realize the iterative optimization of the division of different working areas through the feedback mechanism; By concatenating and fusing the image feature vector and the audio feature vector through a fusion network model, the real-time fusion index of the current working status is obtained. By constructing a fusion network model, the data from both the visual status and the sound features are effectively integrated to form a real-time fusion index, which can more accurately and comprehensively reflect the overall working status of the equipment.

[0028] Specifically, by setting corresponding monitoring probes in several working areas of intelligent manufacturing equipment, when the monitoring port receives a working status monitoring instruction, the working status of the current working area can be recorded in real time through the corresponding monitoring probe, and provided for subsequent automated feature extraction and fusion processing, so that a quick response can be made when an abnormal state occurs, reducing dependence on manual monitoring; and the real-time fusion indicators obtained provide data decision support for subsequent state judgment, so that the central control platform and management personnel can make scientific decisions based on the real-time fusion indicators of the current working state.

[0029] On the basis of the above embodiment, the image feature vector and the audio feature vector are spliced ​​and fused by a fusion network model to obtain a real-time fusion index of the current working state; the following steps are included: The image feature vector and the audio feature vector are concatenated in the feature dimension to form a concatenated feature; the concatenated feature can integrate information from different modalities and provide more comprehensive input data for subsequent analysis; Based on the current working state, the weight matrix and bias vector corresponding to the splicing feature are determined, and the fusion network model is constructed according to the weight matrix and bias vector; by assigning different weights to different features, the important features in the current working state can be highlighted, and the recognition ability of the fusion network model for abnormal states can be improved; The splicing features are fused through the fusion network model to form a real-time fusion index.

[0030] Specifically, the weight matrix and bias vector give the fusion network model the ability to learn and predict, so that it can combine the image feature vector and the audio feature vector according to different working states to improve the monitoring accuracy of the working state of intelligent manufacturing equipment. The introduction of the weight matrix enables the fusion network model to learn different feature readings during the continuous input of splicing features, such as the edge, texture, shape of the image, the pitch, timbre, frequency and other information of the sound, thereby enhancing the accuracy of the real-time fusion index generated by the fusion network model; the introduction of the bias vector can adjust the flexibility and bias of the fusion network model, so that it has better fitting ability when facing different working states of intelligent manufacturing equipment, and is more in line with the current needs of real-time fusion index extraction, thereby ensuring that the formed real-time fusion index can accurately reflect the current working state, and further ensure the real-time and accuracy of subsequent abnormal state detection results.

[0031] On the basis of the above embodiment, the abnormality determination module receives the real-time fusion index based on the local area network and the network communication protocol, performs abnormality determination after analysis, and sends an abnormal state feedback instruction to the central control platform, and the central control platform analyzes the abnormal state feedback instruction and outputs the fault code corresponding to the current working state; including the following steps: Determine the standard indicator range corresponding to the current working state, and compare the real-time fusion indicator with the standard indicator range, and determine whether the current working state of the intelligent manufacturing equipment is normal according to the judgment result; If it is within the standard index range, the real-time fusion index of the working status in the next production process will be compared and determined as the working process progresses until the intelligent manufacturing equipment completes a production process cycle; If it is not within the standard indicator range, an abnormal state feedback instruction is sent to the central control platform, and the central control platform parses the abnormal state feedback instruction and outputs the fault code corresponding to the current working state.

[0032] Specifically, the standard indicator interval provides a range of the intelligent manufacturing equipment in normal working state, and the corresponding standard indicator interval is determined according to different working states to ensure the accuracy of the subsequent abnormal judgment of the working state; when the real-time fusion indicator meets the standard indicator interval range, it indicates that the intelligent manufacturing equipment is currently in normal working state, and can enter the next production process, and the real-time fusion indicator of the next production process is compared and judged again, thereby realizing the process of automated inspection, greatly improving the comprehensiveness and efficiency of equipment status evaluation; when the real-time fusion indicator does not meet the standard indicator interval range, it indicates that the intelligent manufacturing equipment is currently in an abnormal working state, such as jamming, over-processing, processing station error, etc. At this time, an abnormal state feedback instruction is sent to the central control platform, and the central control platform can analyze the abnormal state of the intelligent manufacturing equipment according to the abnormal state feedback instruction, and form the corresponding fault code output, and at the same time, it can also match the corresponding fault code to obtain the abnormal situation corresponding to the current working state and its corresponding solution, so that the supervisor can refer to it and take quick action to reduce the time of equipment downtime maintenance, maintain the stability of the production process, and further improve product quality and reduce energy consumption.

[0033] On the basis of the above-mentioned embodiment, the production process of the intelligent manufacturing equipment includes an equipment preparation process, a raw material loading process, an equipment operation process, a product unloading process, and a waste removal process.

[0034] Specifically, the equipment preparation process is the preparation process from the start of the equipment to the production. By real-time abnormal monitoring of the equipment preparation process, it is ensured that the intelligent manufacturing equipment is in the best state and production delays or failures caused by equipment problems can be avoided; the raw material loading process is the process in which several kinds of raw materials required for processing products are transported to the processing station on the intelligent manufacturing equipment through the corresponding transmission mechanism. In this process, there may be abnormal situations such as raw materials not being delivered in place and raw materials shortage. By real-time abnormal monitoring of the raw material loading process, the continuity and accuracy of raw material feeding supply can be ensured, production interruptions can be reduced and the defective rate can be reduced; the equipment operation process is the process in which the intelligent manufacturing equipment processes several kinds of raw materials transported to the processing station to obtain products. In this process, there may be equipment failures, processing abnormalities, etc. By real-time abnormal monitoring of the equipment operation process, equipment abnormalities or failures can be discovered in time, and unexpected downtime can be reduced, thereby ensuring the continuity and efficiency of production. Among them, the equipment operation process may have multiple processing steps. By monitoring each processing step for abnormalities in turn, the subdivision monitoring of the equipment operation process can be achieved, and the accuracy of abnormality identification can be further improved.

[0035] After processing is completed, qualified products enter the subsequent process through the product unloading process, such as packaging or further processing. Real-time abnormal monitoring of the product unloading process can ensure that the products can be smoothly moved out of the processing station, prevent damage to the products, and ensure product quality. Unqualified waste products are automatically removed through the waste removal process to avoid mixing with qualified products and affecting subsequent processes, reducing the cost of manual removal and improving the efficiency of waste removal.

[0036] On the basis of the above embodiment, after the abnormality detection is performed on the working state of the equipment operation process, a product abnormality detection process is also included, and the product abnormality detection includes the following steps: The last frame of image at the end of the equipment operation process is selected as the target image corresponding to the product, and the target feature vector of the product in the target image is extracted; the target feature vector can reflect the key features of the product such as appearance and shape, and perform quantitative analysis to provide accurate data support for subsequent quality inspection; Obtain a standard image of the product and a standard feature vector of the product in the standard image; the standard feature vector provides a benchmark for product quality, making the quality inspection process comparable and consistent, and helping to ensure the stability and reliability of product quality; Determine whether the product in the target image is qualified based on the matching degree between the target feature vector and the standard feature vector; by comparing the target feature vector and the standard feature vector, it is possible to automatically determine whether the product meets the processing requirements and quality standards, reduce the interference of human factors, and improve the accuracy and efficiency of detection; If the current product is unqualified, an abnormal product feedback instruction is output and the waste removal process is executed; this enables the central control platform to record the specific circumstances of each abnormal product processing, which is convenient for subsequent statistics and analysis; and the automated abnormal product feedback and waste removal process can eliminate unqualified products in a timely manner, prevent defective products from entering the subsequent process, and reduce resource waste.

[0037] If the current product is qualified, the product unloading process will be executed; the automated unloading process can improve production efficiency, reduce manual operations, and reduce production costs.

[0038] Specifically, by recording the test results and processing process of each product, the transparency and traceability of production can be improved, which is helpful for quality control and subsequent problem analysis. The automated product anomaly detection process reduces the reliance on manual inspection, thereby reducing maintenance costs and human resource requirements, optimizing the production process, improving overall production efficiency, and ensuring the consistency and reliability of product quality.

[0039] On the basis of the above-mentioned embodiment, the number of times the waste removal process and the product unloading process are executed during the operation of the intelligent manufacturing equipment is counted and recorded in real time to determine the product yield information; specifically, the yield rate can be automatically calculated by the ratio between the number of times the waste removal process is executed and the total number of times the waste removal process is executed and the product unloading process is executed. Through real-time monitoring and statistical analysis, quality inspection can be carried out on each product in the production process, and data support can be provided for subsequent process optimization, equipment maintenance, quality monitoring and other optimization measures to improve the yield rate.

[0040] On the basis of the above embodiment, it also includes building an exception database, which includes a number of exception information, each exception information includes a one-to-one corresponding fault code, abnormal situation and solution, and the exception database is updated in real time with the output of each fault code; the central control platform retrieves the exception database according to the fault code and outputs the abnormal situation corresponding to the current working state and its corresponding solution; the timeliness and accuracy of the database are ensured, so that anomaly detection and fault diagnosis can be performed based on the latest data source, which can not only quickly respond to the negative impact caused by the occurrence of anomalies, but also facilitate subsequent analysis and research.

[0041] like Figure 6 As shown, the present invention also provides an abnormality detection system for intelligent manufacturing equipment, including a central control platform, a monitoring port and an abnormality determination module; The central control platform is used to send working status monitoring instructions to the monitoring port synchronously with the start of the intelligent manufacturing equipment, and to determine whether to execute the next production process in real time based on the output results of the abnormality determination module. The monitoring port is used to receive working status monitoring instructions, and obtain the real-time fusion indicators of the current working status in the production process of the intelligent manufacturing equipment in real time based on the working status monitoring instructions; The abnormality determination module is used to determine the standard index interval corresponding to the current working state, and compare the real-time fusion index with the standard index interval, and determine whether the current working state of the intelligent manufacturing equipment is normal according to the determination result; If it is within the standard index range, the real-time fusion index of the working status in the next production process is compared and determined as the working process progresses until the intelligent manufacturing equipment completes a production process cycle; If it is not within the standard indicator range, an abnormal status feedback instruction is sent to the central control platform. The central control platform outputs the fault code corresponding to the current working state based on the abnormal status feedback instruction, and outputs the abnormal situation corresponding to the current working state and its corresponding solution according to the fault code.

[0042] The present invention realizes real-time monitoring and abnormal management of intelligent manufacturing equipment during its operation by constructing a central control platform, a monitoring port, and an abnormality determination module; the collaborative work of the central control platform and the monitoring port ensures real-time monitoring of the equipment status, and the abnormality determination module quickly identifies whether the equipment is in a normal working state by analyzing real-time data; once an abnormality is found, the system will use the information in the abnormality database to diagnose and process the fault to ensure production continuity and product quality.

[0043] On the basis of the above embodiment, it also includes a product anomaly detection module. The product anomaly detection module is used to determine whether the current product is in a qualified state after performing an abnormality detection on the working state of the equipment operation process; if the current product is unqualified, an abnormal product feedback instruction is output, and a waste product removal process is executed; if the current product is qualified, a product unloading process is executed, which can realize comprehensive monitoring of product quality, realize automatic separation of qualified products and waste products, and improve production efficiency and product quality.

[0044] On the basis of the above embodiment, it also includes an exception database including a number of exception information, each exception information includes a one-to-one corresponding fault code, abnormal situation and solution, and the exception database is updated in real time with the output of each fault code, and the central control platform retrieves the exception database according to the fault code to output the abnormal situation corresponding to the current working state and its corresponding solution; quickly respond to the negative impact caused by the occurrence of abnormalities, and ensure the timeliness and accuracy of the database; and the dynamic update mechanism of the exception database ensures that the system can adapt to new fault modes and solutions, thereby improving the adaptability and reliability of the system.

[0045] The technical effect achieved by the working process of the above-mentioned abnormality detection system is the same as that in the above-mentioned embodiment, and will not be repeated here.

[0046] Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for detecting abnormality of intelligent manufacturing equipment, characterized in that: The intelligent manufacturing equipment is connected to the central control platform through the abnormality determination module. The abnormality detection method includes the following steps: Determine the working areas of the intelligent manufacturing equipment and set monitoring ports corresponding to each of the working areas, establish a local area network and a corresponding network communication protocol between the monitoring ports, the abnormality determination module and the central control platform; As the intelligent manufacturing equipment starts working, the central control platform synchronously sends a working status monitoring instruction to each of the monitoring ports based on the local area network and the network communication protocol, and the monitoring port obtains the real-time fusion index of the current working status in the production process of the intelligent manufacturing equipment in real time based on the working status monitoring instruction; The abnormality determination module receives the real-time fusion index based on the local area network and the network communication protocol, performs abnormality determination after analysis, and sends an abnormal status feedback instruction to the central control platform. The central control platform analyzes the abnormal status feedback instruction and outputs a fault code corresponding to the current working status.

2. The method for detecting abnormality of intelligent manufacturing equipment according to claim 1, characterized in that: The monitoring port acquires the real-time fusion index of the current working state in the production process of the intelligent manufacturing equipment in real time based on the working state monitoring instruction; comprising the following steps: Determine several working areas of intelligent manufacturing equipment and set up corresponding monitoring probes for each working area; Each of the monitoring ports receives the working status monitoring instruction, and based on the working status monitoring instruction, performs real-time video monitoring of the current working area of ​​the intelligent manufacturing equipment through the corresponding monitoring probe to obtain a real-time video record of the current working status; Intercepting continuous valid video frames in the real-time video record, and performing feature separation on the continuous valid video frames to obtain image information frames and audio information frames; Extracting features of the image information frame and the audio information frame respectively to obtain corresponding image feature vectors and audio feature vectors; The image feature vector and the audio feature vector are spliced ​​and fused through a fusion network model to obtain the real-time fusion index of the current working state.

3. The method for detecting abnormality of intelligent manufacturing equipment according to claim 2, characterized in that: The process of splicing and fusing the image feature vector and the audio feature vector through a fusion network model to obtain the real-time fusion index of the current working state comprises the following steps: splicing the image feature vector and the audio feature vector in a feature dimension to form a splicing feature; Determine a weight matrix and a bias vector corresponding to the splicing feature based on the current working state, and construct a fusion network model according to the weight matrix and the bias vector; The splicing features are fused and processed through the fusion network model to form the real-time fusion index.

4. The method for detecting abnormality of intelligent manufacturing equipment according to claim 2, characterized in that: The abnormality determination module receives the real-time fusion index based on the local area network and the network communication protocol, performs abnormality determination after analysis, and sends an abnormal state feedback instruction to the central control platform, and the central control platform analyzes the abnormal state feedback instruction and outputs a fault code corresponding to the current working state; comprising the following steps: Determine the standard indicator interval corresponding to the current working state, and compare the real-time fusion indicator with the standard indicator interval, and determine whether the current working state of the intelligent manufacturing equipment is in a normal state according to the determination result; If it is within the standard index range, the real-time fusion index of the working status in the next production process will be compared and determined as the working process progresses until the intelligent manufacturing equipment completes a production process cycle; If it is not within the standard indicator range, an abnormal state feedback instruction is sent to the central control platform, and the central control platform parses the abnormal state feedback instruction and outputs a fault code corresponding to the current working state.

5. The method for detecting abnormality of intelligent manufacturing equipment according to claim 3, characterized in that: The production process of the intelligent manufacturing equipment includes an equipment preparation process, a raw material loading process, an equipment operation process, a product unloading process, and a waste removal process.

6. The method for detecting abnormality of intelligent manufacturing equipment according to claim 5, characterized in that: After the working state of the equipment operation process is detected for abnormality, a product abnormality detection process is also included, and the product abnormality detection includes the following steps: Selecting the last frame image at the end of the operation process of the equipment as the target image corresponding to the product, and extracting the target feature vector of the product in the target image; Acquire a standard image of a product and a standard feature vector of the product in the standard image; Determining whether the product in the target image is qualified according to the matching degree between the target feature vector and the standard feature vector; If the current product is unqualified, an abnormal product feedback instruction is output and the waste product rejection process is executed; If the current product is qualified, the product unloading process is executed.

7. The method for detecting abnormality of intelligent manufacturing equipment according to claim 6, characterized in that: The number of times the waste rejection process and the product unloading process are executed during the operation of the intelligent manufacturing equipment is counted and recorded in real time to determine the product yield information.

8. The method for detecting abnormality of intelligent manufacturing equipment according to claim 1, characterized in that: It also includes building an abnormality database, the abnormality database includes a plurality of abnormal information, each of the abnormal information includes a one-to-one corresponding fault code, abnormal situation and solution, and the abnormality database is updated in real time with the output of each fault code; The central control platform retrieves the abnormality database according to the fault code and outputs the abnormality corresponding to the current working state and the corresponding solution.

9. An intelligent manufacturing equipment anomaly detection system, characterized in that: Including central control platform, monitoring port and abnormality determination module; The central control platform is used to send a working status monitoring instruction to the monitoring port synchronously with the start of the intelligent manufacturing equipment and to determine whether to execute the real-time fusion index comparison and determination process of the next production process according to the output result of the abnormality determination module; The monitoring port is used to receive the working state monitoring instruction, and obtain the real-time fusion index of the current working state in the production process of the intelligent manufacturing equipment in real time based on the working state monitoring instruction; The abnormality determination module is used to determine the standard index interval corresponding to the current working state, and compare the real-time fusion index with the standard index interval, and determine whether the current working state of the intelligent manufacturing equipment is in a normal state according to the determination result; If it is within the standard index range, the real-time fusion index of the working status in the next production process will be compared and determined as the working process progresses until the intelligent manufacturing equipment completes a production process cycle; If it is not within the standard indicator range, an abnormal status feedback instruction is sent to the central control platform. The central control platform outputs a fault code corresponding to the current working state based on the abnormal status feedback instruction, and outputs the abnormal situation corresponding to the current working state and its corresponding solution according to the fault code.

10. The intelligent manufacturing equipment abnormality detection system according to claim 9, characterized in that: It also includes a product anomaly detection module, which is used to determine whether the current product is in a qualified state after performing an abnormality detection on the working state of the equipment operation process.

Images