A method and system for monitoring the operation of intelligent production equipment management

Abstract

The present invention relates to an intelligent production equipment management and operation monitoring method and monitoring system, which implant a program supervision data packet into a target production equipment and match it with a code layer of the target production equipment; deploy the program supervision data packet on the target production equipment to generate a first monitoring vector; identify a judgmental token string in the code layer of the target production equipment, and arrange several identified judgmental token strings into an if table, and use an ETL architecture tool to perform miscellaneous optimization on the if table to obtain a monitoring target of the target production equipment; layout the monitoring target on the first monitoring vector to obtain a second monitoring vector; call out a preset neural network model, create a historical database corresponding to past production processes in the neural network model, and perform process monitoring training on the neural network model; deploy the trained neural network model on the second monitoring vector to obtain a third monitoring vector; perform data visualization processing on the third monitoring vector and feed it back to the front-end layer.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment identification and monitoring, and in particular to an intelligent production equipment management and operation monitoring method and a monitoring system. Background Art

[0002] In the intelligent manufacturing system, the intelligent management of production equipment has become a key link in improving production efficiency, ensuring product quality and reducing operating costs. However, traditional production equipment management methods mainly rely on manual inspections and simple sensor monitoring, which have problems such as incomplete monitoring, untimely response, and insufficient data analysis capabilities, making it difficult to meet the modern industrial production's demand for efficient, accurate, and intelligent management. At present, with the rapid development of technologies such as the Internet of Things, big data, and artificial intelligence, new solutions have been provided for production equipment management. By applying these advanced technologies to the field of production equipment management, real-time monitoring, data analysis, fault warning, and intelligent decision-making of production equipment can be realized, significantly improving management efficiency and level.

[0003] Referring to the current technical solution CN114626615B - A production process monitoring management method and system, it discloses a production process monitoring management method and system, which obtains instrument monitoring information through Internet of Things monitoring instruments; performs trend analysis based on instrument monitoring information to obtain instrument trend prediction information; obtains production monitoring information through image monitoring equipment; extracts features from production monitoring information to obtain production monitoring features; obtains production prediction information based on instrument trend prediction information and production monitoring features; determines whether the production prediction information meets the preset requirements; and sends the first warning information when the preset requirements are not met. It solves the technical problem that the existing technology has single monitoring data collection, low data processing reliability, and cannot meet the production process monitoring requirements. It achieves the technical effect of using instrument monitoring data and monitoring images for comprehensive analysis and prediction, and quickly responds to abnormal conditions of equipment and instruments through alarm management and equipment operation status monitoring, thereby improving the agility and intelligence level of production management.

[0004] However, in practical applications, there are still some drawbacks that cannot be ignored.

[0005] First, the system integration is complex. The system proposed by the current technical solution involves the integration of multiple monitoring devices and data analysis models. This requires enterprises to carry out large-scale transformation and upgrading of existing production lines and monitoring equipment when deploying the system, which not only increases the initial investment cost, but also affects normal production activities. In addition, there are differences in data interfaces and communication protocols between different devices, which increases the technical difficulty of system integration.

[0006] Secondly, the real-time and accuracy of data processing needs to be improved. Although the system improves the reliability of prediction results by comprehensively analyzing instrument monitoring data and monitoring image information, real-time data processing capabilities have become a key challenge in large-scale production environments. With the rapid increase in the amount of production data, how to ensure the efficiency and accuracy of data processing and avoid production decisions affected by data processing delays is an issue that the current system needs to be further optimized.

[0007] Furthermore, system maintenance and upgrades are difficult. Since the system integrates a variety of monitoring devices and data analysis modules, once a part fails or needs to be upgraded, it will affect the normal operation of the entire system. This requires companies to have a professional technical team and a complete maintenance system to ensure the stable operation and continuous upgrade of the system. However, this undoubtedly adds additional operating costs and management difficulties for many small and medium-sized enterprises. Summary of the invention

[0008] The main purpose of the present invention is to provide an intelligent production equipment management and operation monitoring method and monitoring system, aiming to solve the problems of incomplete monitoring and untimely response in traditional production equipment management methods, and to improve the intelligence level and real-time performance of production equipment management.

[0009] To achieve the above object, the present invention provides a method for managing and operating intelligent production equipment, comprising the following steps:

[0010] Implanting a program supervision data package into a target production device, and matching the code layer of the target production device based on a code environment architecture preset in the program supervision data package;

[0011] Deploy the program supervision data packet on the target production device according to the code environment architecture after successful matching to generate a first monitoring vector;

[0012] Identify the judgment token string in the code layer of the target production equipment, and arrange the identified judgment token strings into an if table, and use the ETL architecture tool to perform miscellaneous optimization on the if table to obtain the monitoring target of the target production equipment;

[0013] According to the production process of the target production equipment, the monitoring target is arranged on the first monitoring vector to obtain a second monitoring vector;

[0014] Calling out a preset neural network model, and using the past production process data of the target production equipment, creating a historical database corresponding to the past production process in the neural network model, and performing process monitoring training on the neural network model based on the historical database;

[0015] Deploying the trained neural network model on the second monitoring vector to create a corresponding alarm mechanism in each monitoring target on the second monitoring vector, thereby obtaining a third monitoring vector;

[0016] Data visualization processing is performed on the third monitoring vector and the data is fed back to the front-end layer.

[0017] Furthermore, the step of matching the code environment architecture preset in the program supervision data package with the code layer of the target production device includes:

[0018] Calling the microservice pass architecture in the program supervision data packet to perform environmental identification on the code layer of the target production device;

[0019] The identified target production equipment environment is used to perform environmental adjustments on the data code in the program supervision data packet to obtain a program supervision code suitable for the target production equipment.

[0020] Further, the step of deploying the program supervision data packet on the target production device according to the code environment architecture after successful matching to generate the first monitoring vector includes:

[0021] Import the program supervision code into the preset SVM tool;

[0022] Generate two virtual vectors using an SVM tool, including a first virtual vector and a second virtual vector, wherein the first virtual vector measures a code layer in a target production device to identify a node in the code layer in the target production device where a supervisory program code is inserted, and the second virtual vector associates the node as a signal source implant interface;

[0023] After the deployment is completed, the first monitoring vector is generated.

[0024] Further, the steps of identifying the judgment token string in the code layer of the target production equipment, arranging the identified judgment token strings into an if table, and performing miscellaneous optimization on the if table using an ETL architecture tool to obtain the monitoring target of the target production equipment include:

[0025] Identify the judgment sentence pattern in the code layer of the target production equipment, and arrange the judgment token string corresponding to the judgment sentence pattern in the if table to identify the process nodes that the target production equipment needs to judge when producing products;

[0026] The ETL architecture tool is used to reflect the judgment token string in the current if table to the front end, and the signal input by the front end is received. The if table is adjusted according to the signal to formulate corresponding judgment conditions according to actual monitoring and management needs to obtain the monitoring target of the target generation device.

[0027] Furthermore, according to the production process of the target production equipment, the monitoring target is arranged on the first monitoring vector to obtain the second monitoring vector, including:

[0028] The production process is obtained by identifying the code of the target production equipment, and based on the second virtual vector as a signal source interface, the monitoring target is input into the first monitoring vector to obtain the second monitoring vector.

[0029] Further, calling out a preset neural network model, and using the past production process data of the target production equipment, creating a historical database corresponding to the past production process in the neural network model, and performing process monitoring training on the neural network model based on the historical database, includes:

[0030] Retrieving a preset LSTM neural network model, acquiring the past production process data, and implanting the past production process input into the input layer of the LSTM neural network model to create the historical database;

[0031] Based on the monitoring targets of the if table, calibrate similar items in the past production process;

[0032] Perform node training on several similar items in history to obtain monitoring error standards during product production;

[0033] The monitoring error standards corresponding to each of the monitoring calibrations are continuously implanted into the second monitoring vector at the corresponding position to complete the training of the LSTM neural network model.

[0034] Further, the step of deploying the trained neural network model on the second monitoring vector to create a corresponding alarm mechanism in each monitoring target on the second monitoring vector, thereby obtaining a third monitoring vector, includes:

[0035] The trained LSTM neural network model is deployed on the second monitoring vector, and the nodes corresponding to each monitoring target in the second monitoring vector are optimized according to the monitoring standard for each monitoring target of the production process in the LSTM neural network model, so that the judgment token string of the monitoring standard after training and optimization is obtained on each node;

[0036] Based on the monitoring standard judgment token string trained and optimized on each node, an alarm mechanism for each node is formed, thereby completing the output of the third monitoring vector.

[0037] Furthermore, the step of performing data visualization processing on the third monitoring vector and feeding back the data to the front-end layer includes:

[0038] Using data visualization tools, identify the third monitoring vector and output the current production process of the target production equipment;

[0039] The current production process is output visually.

[0040] Furthermore, before the step of implanting the program supervision data packet into the target production device, the program supervision data packet is placed in a USB flash drive, and when the USB flash drive is inserted into the USB interface of the target production device, the step of implanting the program supervision data packet into the target production device is executed.

[0041] The present invention also proposes an intelligent production equipment management and operation monitoring system, comprising:

[0042] An implantation unit, used for implanting a program supervision data package into a target production device, and matching the code layer of the target production device based on a code environment architecture preset in the program supervision data package;

[0043] A matching unit, configured to deploy the program supervision data packet on a target production device according to a code environment architecture after successful matching, and generate a first monitoring vector;

[0044] An identification unit, used to identify the judgment token string in the code layer of the target production equipment, and arrange the identified judgment token strings into an if table, and perform miscellaneous optimization on the if table using an ETL architecture tool to obtain the monitoring target of the target production equipment;

[0045] A process unit, configured to place the monitoring target on the first monitoring vector according to the production process of the target production equipment to obtain a second monitoring vector;

[0046] A model unit, used to call out a preset neural network model, and create a historical database corresponding to the past production process in the neural network model through the past production process data of the target production equipment, and perform process monitoring training on the neural network model based on the historical database;

[0047] an alarm unit, configured to deploy the trained neural network model on the second monitoring vector, so as to create a corresponding alarm mechanism in each monitoring target on the second monitoring vector, thereby obtaining a third monitoring vector;

[0048] The monitoring unit is used to perform data visualization processing on the third monitoring vector and feed back the data to the front-end layer.

[0049] The intelligent production equipment management and operation monitoring method and monitoring system provided by the present invention have the following beneficial effects:

[0050] By matching and deploying program supervision data packages with the code layer, seamless integration and precise monitoring can be achieved. By implanting program supervision data packages into target production equipment and accurately matching and deploying them with the equipment code layer, seamless integration and all-round precise monitoring of production equipment are achieved, avoiding problems such as data interface mismatch and monitoring blind spots in traditional monitoring methods, and improving the accuracy and reliability of monitoring.

[0051] By optimizing the if table through the ETL architecture tool, flexible configuration and efficient management can be achieved. By using the ETL architecture tool to optimize and adjust the judgment token string in the if table, judgment conditions can be quickly formulated and modified according to actual monitoring and management needs without making extensive modifications to the device code, thereby improving the flexibility and configurability of the monitoring system and reducing management costs.

[0052] Through LSTM neural network model training, the effect of intelligent prediction and alarm can be achieved. The LSTM neural network model is called and trained using historical production process data. A historical database corresponding to past production processes is created. The model is then trained for process monitoring based on this database, realizing intelligent prediction of the production process and abnormal alarm. Compared with traditional methods, this improves the accuracy and timeliness of alarms and effectively prevents the occurrence of production accidents.

[0053] Through data visualization, the effects of intuitive display and rapid response are achieved. By visualizing the monitoring data, the current production process of the target production equipment can be intuitively displayed on the front-end layer, enabling production managers to quickly grasp the production status, make decisions and adjustments in a timely manner, and improve the agility and response speed of production management.

[0054] By implanting the program supervision data package into a USB flash drive, convenient deployment and wide applicability can be achieved. The program supervision data package is placed in a USB flash drive and deployment can be completed by simply inserting the USB flash drive into the USB interface of the target production equipment. This method does not require a complicated installation and debugging process, simplifies the deployment process, reduces the difficulty of deployment, and enables the solution to be widely used in production equipment of different types and sizes. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a flow chart of a method for managing and monitoring operation of intelligent production equipment in one embodiment of the present invention;

[0056] Figure 2 It is a structural block diagram of an intelligent production equipment management and operation monitoring system in one embodiment of the present invention.

[0057] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] Reference Figure 1 , which is a flow chart of a method for intelligent production equipment management and operation monitoring proposed by the present invention, comprising the following steps:

[0060] S1, implanting a program supervision data package into a target production device, and matching the code environment architecture preset in the program supervision data package with the code layer of the target production device;

[0061] S2, deploying the program supervision data packet on the target production device according to the code environment architecture after successful matching, and generating a first monitoring vector;

[0062] S3, identifying the judgment token string in the code layer of the target production equipment, and arranging the identified judgment token strings into an if table, and performing miscellaneous optimization on the if table using an ETL architecture tool to obtain the monitoring target of the target production equipment;

[0063] S4, according to the production process of the target production equipment, the monitoring target is arranged on the first monitoring vector to obtain a second monitoring vector;

[0064] S5, calling out a preset neural network model, and using the past production process data of the target production equipment, creating a historical database corresponding to the past production process in the neural network model, and performing process monitoring training on the neural network model based on the historical database;

[0065] S6, deploying the trained neural network model on the second monitoring vector to create a corresponding alarm mechanism in each monitoring target on the second monitoring vector, thereby obtaining a third monitoring vector;

[0066] S7: Perform data visualization processing on the third monitoring vector and feed it back to the front-end layer.

[0067] In the process of specific implementation,

[0068] For S1, the microservice pass architecture in the program supervision data packet is called to perform environmental identification on the code layer of the target production equipment; the identified target production equipment environment is used to adjust the data code in the program supervision data packet to obtain a program supervision code suitable for the target production equipment.

[0069] In step S1, after successfully implanting the program supervision data package into the target production device and matching the code environment architecture, the next key step is to call the microservice PASS (Platform as a Service, Software as a Service) architecture in the program supervision data package for further operation.

[0070] Specifically:

[0071] The program supervision data package contains a set of microservice components, which communicate through predefined interfaces and protocols to support the operation of the entire monitoring system. The environment identification module in the microservice will conduct an in-depth analysis of the code layer of the target production equipment to identify the key information of the device's operating system, hardware configuration, software version, and existing application architecture. The environment identification process involves scanning, parsing, and comparing the device code layer to ensure that the acquired environment information is accurate and comprehensive. After identifying the environment of the target production device, the system will make corresponding environmental adjustments to the data code in the program supervision data package based on this information. This includes modifying the data format, adjusting the communication protocol, optimizing the algorithm parameters, etc., to ensure that the program supervision code can run smoothly on the target device and seamlessly integrate with the existing functions of the device. During the environment adjustment process, the system will generate some intermediate files or configuration files to record the adjusted code status and environment configuration information for subsequent management and maintenance. After the adjustment is completed, the system will perform a verification test to ensure that the adjusted program supervision code can run normally on the target device and will not interfere with the normal production activities of the device. After the verification is passed, the adjusted program supervision code will be officially deployed to the target device to start the monitoring task. At this point, the intelligent production equipment management and operation monitoring system has already possessed core capabilities such as real-time monitoring, data analysis, and abnormal alarm.

[0072] For S2, the program supervision code is imported into the preset SVM tool; two virtual vectors are generated using the SVM tool, including a first virtual vector and a second virtual vector, wherein the first virtual vector measures the code layer in the target production equipment to identify the node in the code layer in the target production equipment where the supervision program code is inserted, and the second virtual vector is associated with the node as a signal source implant interface; after deployment, the first monitoring vector is generated.

[0073] Specifically:

[0074] In the intelligent production equipment management and operation monitoring system, the program supervision code that has been adjusted for the environment is imported into the preset support vector machine (SVM) tool, and the special capabilities of the SVM tool are used to further process and analyze the code layer to ensure that the supervision program code can be accurately inserted into the target production equipment and effectively monitored.

[0075] First, the program supervision code that has been environmentally adjusted and ready for monitoring is imported into the preset SVM tool. Using the SVM tool, two key virtual vectors are generated: the first virtual vector and the second virtual vector. Among them, the first virtual vector is used to measure the code layer in the target production equipment, especially to identify the best node or position in the code layer for inserting the supervision program code. This step is similar to finding an "insertion point" in a complex code structure so that the supervision code can run effectively without affecting the original production process. The second virtual vector is associated with the node identified by the first virtual vector and serves as a signal source implantation interface. This means that the second virtual vector not only marks the location where the supervision code should be inserted, but also defines how to communicate and interact with the supervision code through this location for subsequent monitoring and management. After determining the insertion point and the signal source implantation interface, the system begins to deploy the program supervision code. The supervision code is accurately inserted into the node identified by the first virtual vector, and the second virtual vector ensures the effectiveness and reliability of this insertion process. After the deployment is completed, the system will generate the first monitoring vector. The first monitoring vector is a collection of data structures used to collect and analyze the operating status data of the target production equipment in real time after the supervision code is inserted. The first monitoring vector is the basis for subsequent production process monitoring, data analysis, abnormal alarm and other operations.

[0076] For S3, identify the judgment sentence pattern in the code layer of the target production equipment, and arrange the judgment token string corresponding to the judgment sentence pattern in the if table to identify the process nodes that the target production equipment needs to judge when producing products; use the ETL architecture tool to reflect the judgment token string in the current if table to the front end, and receive the signal input by the front end, adjust the if table according to the signal, so as to formulate corresponding judgment conditions according to the actual monitoring and management needs, and obtain the monitoring target of the target production equipment.

[0077] Specifically:

[0078] The system first goes deep into the code layer of the target production equipment and identifies the judgment sentences therein. These judgment sentences are an important part of the code logic and are used to control the branches and decisions in the production process. During the identification process, the system pays special attention to the key judgment points that affect the quality of the production products, such as raw material inspection, process parameter setting, product quality inspection, etc. Arrange judgment token strings: For each judgment sentence identified, the system extracts the judgment token string (i.e., the keyword or variable used for conditional judgment). These judgment token strings are arranged in the data structure of the "if table". The if table is a table used to manage and maintain judgment logic. It records each judgment point, the corresponding judgment condition, and the result branch. Identify process nodes: By analyzing the content of the if table, the system can identify the process nodes that the target production equipment needs to pay special attention to when producing products. These nodes are usually key points in the production process, such as raw material feeding, temperature control, pressure adjustment, etc. Accurate identification of these nodes helps the system to monitor and manage them in a targeted manner in the future. Use ETL architecture tools: The system uses ETL (Extract, Transform, Load) architecture tools to reflect the judgment token strings in the current if table to the front-end interface. ETL tool is a commonly used data processing tool that can efficiently extract data from data sources, perform data conversion, and load it into the target system. ETL tool displays the data in the if table to the front-end user in a visual way, which is convenient for users to understand and operate. The front-end user inputs corresponding signals or instructions through the front-end interface according to the actual production situation and monitoring management needs. These signals include modifying judgment conditions, adding new monitoring points, adjusting monitoring parameters, etc. After receiving the signals input by the front-end, the system adjusts the if table accordingly according to these signals. Adjustment involves operations such as modifying judgment conditions, updating judgment results, adding or deleting monitoring points, etc. Through this step, the system can dynamically adjust the judgment logic according to the actual monitoring management needs to ensure the accuracy and effectiveness of the monitoring target. Obtaining the monitoring target: After adjustment and optimization of the above steps, the system finally determines the monitoring target of the target production equipment. These monitoring targets are formulated based on the actual production process, judgment logic and monitoring management needs, and can comprehensively and accurately reflect the operating status and product quality of the production equipment.

[0079] For S4, the production process is obtained by identifying the code of the target production equipment, and based on the second virtual vector as a signal source interface, the monitoring target is input into the first monitoring vector to obtain the second monitoring vector.

[0080] For S5, call the preset LSTM neural network model, obtain the past production process data, and implant the past production process input into the input layer of the LSTM neural network model to create the historical database; based on the monitoring target of the if table, calibrate the similar items in the past production process; perform node training on several similar items in the history to obtain the monitoring error standard during product production; continuously implant the monitoring error standard corresponding to each monitoring calibration into the second monitoring vector at the corresponding position to complete the training of the LSTM neural network model.

[0081] The system first calls the preset long short-term memory network (LSTM) neural network model. LSTM is a special recurrent neural network (RNN) that can effectively handle long-term dependency problems in sequence data and is very suitable for analysis and prediction of production process data. The system obtains a large amount of past production process data from the historical database. These data record the operating status, process parameters, product quality and other information of the production equipment in the past period of time, and are the basis for model training. The obtained past production process data is input into the input layer of the LSTM neural network model. The LSTM model processes and analyzes these data to extract key features and patterns. If the historical database is not yet complete, the system will create or update the historical database based on the input production process data in this step. The historical database is an important tool for storing and analyzing past production data, providing support for subsequent data mining and model training. The system calibrates similar items in the past production process data according to the if table (including the monitoring target) determined in step S3. Similar items refer to production process fragments that are related to the monitoring target and have similar characteristics or behavior patterns. By calibrating similar items, the system can focus more accurately on the data part related to the target monitoring task. Several similar items in the history are input as training samples into the LSTM neural network model for node training. Node training refers to the model learning and fitting the features and patterns in the training samples through an iterative optimization process to improve the prediction accuracy. Through node training, the system can obtain the monitoring error standard during product production. The monitoring error standard refers to the error range that the model can accept when predicting the production process status or product quality. It is an important indicator for measuring the prediction accuracy and reliability of the model. The system continuously implants the monitoring error standards corresponding to each monitoring calibration into the second monitoring vector at the corresponding position. The second monitoring vector is a data structure used to collect and analyze production process data in real time. It combines the prediction results of the LSTM model and the actual monitoring error standard, and can more comprehensively reflect the operating status of the production equipment and the product quality. By implanting the monitoring error standard into the second monitoring vector, the system can evaluate the prediction accuracy of the model in real time during the subsequent monitoring process, and intervene and adjust when necessary.

[0082] For S6, the trained LSTM neural network model is deployed in the second monitoring vector, and the nodes corresponding to each monitoring target in the second monitoring vector are optimized according to the monitoring standards for each monitoring target in the production process in the LSTM neural network model, so that each node can obtain a trained and optimized monitoring standard judgment token string; based on the trained and optimized monitoring standard judgment token string on each node, an alarm mechanism for each node is formed, thereby completing the output of the third monitoring vector.

[0083] Specifically:

[0084] The LSTM model that has been trained on historical data or other training data sets is applied to the real-time production monitoring system, specifically processing the second monitoring vector. LSTM is often used for time series prediction or anomaly detection in the field of production monitoring because of its ability to process sequence data and capture long-term dependencies. "Node" refers to different monitoring points or data input points in the production process. The LSTM model learns the normal behavior pattern or threshold range (i.e., monitoring standard) of each monitoring target in the production process through training. After the model is deployed, by processing the real-time data of each monitoring target in the second monitoring vector, the model can output the degree of deviation between the current state of each node (monitoring point) and the normal state or the judgment of whether it is abnormal. Data preprocessing: Format the data of the second monitoring vector into a format that can be processed by the LSTM model. Model reasoning: Input the preprocessed data into the LSTM model to obtain the prediction result or anomaly score of each monitoring target. Threshold comparison: Compare the prediction result or anomaly score with the preset threshold to determine whether the current state is abnormal. Optimize token strings: Generate monitoring standard judgment token strings on each node based on the comparison results. These token strings contain information such as whether the current state is abnormal and the degree of abnormality. Form an alarm mechanism and trigger the alarm: When the token string of a node indicates that the current state is abnormal, the alarm mechanism is triggered. Alarm information generation: Generate specific alarm information based on the degree of abnormality and the preset alarm rules, including abnormal type, abnormal location, abnormal time, etc. Alarm notification: Notify relevant personnel of the alarm information through appropriate channels (such as system messages, SMS, emails, etc.).

[0085] For S7, a data visualization tool is used to identify the third monitoring vector and output the current production process of the target production equipment; and the current production process is visualized and output.

[0086] In one embodiment, before the step of implanting the program supervision data packet into the target production device, the program supervision data packet is placed in a USB flash drive, and when the USB flash drive is inserted into the USB interface of the target production device, the step of implanting the program supervision data packet into the target production device is executed.

[0087] In summary, a program supervision data packet is implanted into the target production equipment, and the code layer of the target production equipment is matched based on the preset code environment architecture in the program supervision data packet; the program supervision data packet is deployed on the target production equipment according to the code environment architecture after successful matching to generate a first monitoring vector; the judgmental token string in the code layer of the target production equipment is identified, and several judgmental token strings identified are arranged in an if table, and the if table is miscellaneously optimized using an ETL architecture tool to obtain the monitoring target of the target production equipment; according to the production process of the target production equipment, the monitoring target is laid out on the first monitoring vector to obtain a second monitoring vector; the preset neural network model is called out, and through the past production process data of the target production equipment, a historical database corresponding to the past production process is created in the neural network model, and the neural network model is trained for process monitoring based on the historical database; the trained neural network model is deployed on the second monitoring vector to create a corresponding alarm mechanism in each monitoring target on the second monitoring vector, thereby obtaining a third monitoring vector; the third monitoring vector is subjected to data visualization processing and fed back to the front-end layer.

[0088] like Figure 2 The structural block diagram of an intelligent production equipment management and operation monitoring system proposed by the present invention includes:

[0089] An implantation unit, used for implanting a program supervision data package into a target production device, and matching the code layer of the target production device based on a code environment architecture preset in the program supervision data package;

[0090] A matching unit, configured to deploy the program supervision data packet on a target production device according to a code environment architecture after successful matching, and generate a first monitoring vector;

[0091] An identification unit, used to identify the judgment token string in the code layer of the target production equipment, and arrange the identified judgment token strings into an if table, and perform miscellaneous optimization on the if table using an ETL architecture tool to obtain the monitoring target of the target production equipment;

[0092] A process unit, configured to place the monitoring target on the first monitoring vector according to the production process of the target production equipment to obtain a second monitoring vector;

[0093] A model unit, used to call out a preset neural network model, and create a historical database corresponding to the past production process in the neural network model through the past production process data of the target production equipment, and perform process monitoring training on the neural network model based on the historical database;

[0094] an alarm unit, configured to deploy the trained neural network model on the second monitoring vector, so as to create a corresponding alarm mechanism in each monitoring target on the second monitoring vector, thereby obtaining a third monitoring vector;

[0095] The monitoring unit is used to perform data visualization processing on the third monitoring vector and feed back the data to the front-end layer.

[0096] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0097] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for monitoring the operation of intelligent production equipment, characterized in that: The following steps are involved: Implanting a program supervision data package into a target production device, and matching the code layer of the target production device based on a code environment architecture preset in the program supervision data package; Deploy the program supervision data packet on the target production device according to the code environment architecture after successful matching to generate a first monitoring vector; Identify the judgment token string in the code layer of the target production equipment, and arrange the identified judgment token strings into an if table, and use the ETL architecture tool to perform miscellaneous optimization on the if table to obtain the monitoring target of the target production equipment; According to the production process of the target production equipment, the monitoring target is arranged on the first monitoring vector to obtain a second monitoring vector; Calling out a preset neural network model, and using the past production process data of the target production equipment, creating a historical database corresponding to the past production process in the neural network model, and performing process monitoring training on the neural network model based on the historical database; Deploying the trained neural network model on the second monitoring vector to create a corresponding alarm mechanism in each monitoring target on the second monitoring vector, thereby obtaining a third monitoring vector; Data visualization processing is performed on the third monitoring vector and the data is fed back to the front-end layer.

2. The intelligent production equipment management and operation monitoring method according to claim 1 is characterized in that: The step of matching the code environment architecture preset in the program supervision data package with the code layer of the target production device includes: Calling the microservice pass architecture in the program supervision data packet to perform environmental identification on the code layer of the target production device; The identified target production equipment environment is used to perform environmental adjustments on the data code in the program supervision data packet to obtain a program supervision code suitable for the target production equipment.

3. The intelligent production equipment management and operation monitoring method according to claim 2 is characterized in that: The step of deploying the program supervision data packet on the target production device according to the code environment architecture after successful matching to generate a first monitoring vector includes: Import the program supervision code into the preset SVM tool; Generate two virtual vectors using an SVM tool, including a first virtual vector and a second virtual vector, wherein the first virtual vector measures a code layer in a target production device to identify a node in the code layer in the target production device where a supervisory program code is inserted, and the second virtual vector associates the node as a signal source implant interface; After the deployment is completed, the first monitoring vector is generated.

4. The intelligent production equipment management and operation monitoring method according to claim 3 is characterized in that: The steps of identifying the judgment token string in the code layer of the target production equipment, arranging the identified judgment token strings into an if table, and performing miscellaneous optimization on the if table using an ETL architecture tool to obtain the monitoring target of the target production equipment include: Identify the judgment sentence pattern in the code layer of the target production equipment, and arrange the judgment token string corresponding to the judgment sentence pattern in the if table to identify the process nodes that the target production equipment needs to judge when producing products; The ETL architecture tool is used to reflect the judgment token string in the current if table to the front end, and the signal input by the front end is received. The if table is adjusted according to the signal to formulate corresponding judgment conditions according to actual monitoring and management needs to obtain the monitoring target of the target generation device.

5. The intelligent production equipment management and operation monitoring method according to claim 4 is characterized in that: According to the production process of the target production equipment, the monitoring target is placed on the first monitoring vector to obtain the second monitoring vector, including: The production process is obtained by identifying the code of the target production equipment, and based on the second virtual vector as a signal source interface, the monitoring target is input into the first monitoring vector to obtain the second monitoring vector.

6. The intelligent production equipment management and operation monitoring method according to claim 5 is characterized in that: The steps of calling a preset neural network model, creating a historical database corresponding to the past production process in the neural network model through the past production process data of the target production equipment, and performing process monitoring training on the neural network model based on the historical database include: Retrieving a preset LSTM neural network model, acquiring the past production process data, and implanting the past production process input into the input layer of the LSTM neural network model to create the historical database; Based on the monitoring targets of the if table, calibrate similar items in the past production process; Perform node training on several similar items in history to obtain monitoring error standards during product production; The monitoring error standards corresponding to each of the monitoring calibrations are continuously implanted into the second monitoring vector at the corresponding position to complete the training of the LSTM neural network model.

7. The intelligent production equipment management and operation monitoring method according to claim 6 is characterized in that: The step of deploying the trained neural network model on the second monitoring vector to create a corresponding alarm mechanism in each monitoring target on the second monitoring vector, thereby obtaining a third monitoring vector, includes: The trained LSTM neural network model is deployed on the second monitoring vector, and the nodes corresponding to each monitoring target in the second monitoring vector are optimized according to the monitoring standard for each monitoring target of the production process in the LSTM neural network model, so that the judgment token string of the monitoring standard after training and optimization is obtained on each node; Based on the monitoring standard judgment token string trained and optimized on each node, an alarm mechanism for each node is formed, thereby completing the output of the third monitoring vector.

8. The intelligent production equipment management and operation monitoring method according to any one of claims 1 to 7, characterized in that: The step of performing data visualization processing on the third monitoring vector and feeding back the data to the front-end layer includes: Using data visualization tools, identify the third monitoring vector and output the current production process of the target production equipment; The current production process is output visually.

9. The intelligent production equipment management and operation monitoring method according to claim 1 is characterized in that: Before the step of implanting the program supervision data packet into the target production device, the program supervision data packet is placed in a U disk. When the U disk is inserted into the USB interface of the target production device, the step of implanting the program supervision data packet into the target production device is executed.

10. An intelligent production equipment management and operation monitoring system, characterized in that: include: An implantation unit, used for implanting a program supervision data package into a target production device, and matching the code layer of the target production device based on a code environment architecture preset in the program supervision data package; A matching unit, configured to deploy the program supervision data packet on a target production device according to a code environment architecture after successful matching, and generate a first monitoring vector; An identification unit, used to identify the judgment token string in the code layer of the target production equipment, and arrange the identified judgment token strings into an if table, and perform miscellaneous optimization on the if table using an ETL architecture tool to obtain the monitoring target of the target production equipment; A process unit, configured to place the monitoring target on the first monitoring vector according to the production process of the target production equipment to obtain a second monitoring vector; A model unit, used to call out a preset neural network model, and create a historical database corresponding to the past production process in the neural network model through the past production process data of the target production equipment, and perform process monitoring training on the neural network model based on the historical database; an alarm unit, configured to deploy the trained neural network model on the second monitoring vector, so as to create a corresponding alarm mechanism in each monitoring target on the second monitoring vector, thereby obtaining a third monitoring vector; The monitoring unit is used to perform data visualization processing on the third monitoring vector and feed back the data to the front-end layer.

Images