Power line on-line production monitoring method and system

Abstract

The invention provides a power line on-line production monitoring method and system, and the method comprises the steps: obtaining sample production parameter state path data which carries or does not carry abnormal state chain label data, and training and optimizing an abnormal state chain prediction network through the sample production parameter state path data; the abnormal state chain in the production line can be effectively identified, and the accuracy and reliability of abnormal state prediction are improved by introducing and guiding the machine learning network to generate confidence evaluation. The optimized abnormal state chain prediction network can accurately perform abnormal state chain prediction on the target production parameter state path data, so that real-time monitoring and fault early warning of the power line production process are realized, and the production efficiency and the product quality are remarkably improved.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent systems for power cord production detection, and more specifically, to a method and system for online production monitoring of power cords. Background Art

[0002] During the production process of power cords, various factors, such as equipment failure, raw material problems, or improper operation, may cause quality problems or abnormal conditions in the produced power cords. These abnormal conditions not only affect the performance and service life of the power cords, but may also pose a threat to the safety of users. Therefore, it is particularly important to monitor the power cord production process in real time and identify abnormal conditions.

[0003] Traditional power cord production monitoring methods mainly rely on manual inspections and sampling tests, which are not only inefficient but also difficult to fully cover all aspects of the production process. Machine learning technology has gradually been introduced into production monitoring to improve the accuracy and efficiency of monitoring. However, existing machine learning algorithms often face problems such as difficulty in data labeling and low accuracy in abnormal state recognition when dealing with power cord production monitoring tasks.

[0004] Specifically, since there are many types of abnormal conditions in the power cord production process, and many of them are rare or difficult to reproduce, it is difficult to obtain a large amount of training data with abnormal state labels. This results in the existing machine learning algorithms lacking sufficient supervision information during the training process, making it difficult to learn an effective abnormal state recognition model. At the same time, due to the complexity of the power cord production process, abnormal conditions are often not caused by a single factor, but the result of the interaction of multiple factors, which makes the identification of abnormal conditions even more difficult. Summary of the invention

[0005] In view of this, the purpose of the present application is to provide a method and system for online production monitoring of power lines.

[0006] According to a first aspect of the present application, a method for online production monitoring of a power line is provided, the method comprising:

[0007] Acquire first sample production parameter state path data and second sample production parameter state path data of each reference power cord production line, wherein the first sample production parameter state path data carries abnormal state chain label data, and the second sample production parameter state path data does not carry abnormal state chain label data;

[0008] The first sample production parameter state path data and the second sample production parameter state path data are respectively loaded into an abnormal state chain prediction network as a guided machine learning network, and a first abnormal state chain prediction result of the first sample production parameter state path data and a second abnormal state chain prediction result of the second sample production parameter state path data are generated;

[0009] Loading the second sample production parameter state path data into the guiding machine learning network corresponding to the guided machine learning network, generating a third abnormal state chain prediction result of the second sample production parameter state path data, wherein the third abnormal state chain prediction result represents the confidence of the abnormal state chain and the target abnormal state label;

[0010] If the confidence is greater than a predefined threshold value corresponding to the target abnormal state label, the third abnormal state chain prediction result is determined to be a trust result, and the abnormal state chain and the target abnormal state label represented in the third abnormal state chain prediction result are used as the abnormal state chain label data corresponding to the second sample production parameter state path data, so as to optimize the abnormal state chain prediction network based on the first abnormal state chain prediction result, the second abnormal state chain prediction result, the abnormal state chain label data corresponding to the first sample production parameter state path data, and the abnormal state chain label data corresponding to the second sample production parameter state path data;

[0011] Obtain target production parameter state path data, load the target production parameter state path data into the optimized abnormal state chain prediction network, and generate an abnormal state chain prediction result output by the abnormal state chain prediction network.

[0012] According to the second aspect of the present application, a power line online production monitoring system is provided, the power line online production monitoring system comprising a processor and a readable storage medium, the readable storage medium storing a program, which implements the aforementioned power line online production monitoring method when executed by the processor.

[0013] According to a third aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are monitored to be executed, the aforementioned power line online production monitoring method is implemented.

[0014] According to any of the above aspects, the embodiment of the present application can not only effectively identify the abnormal state chain in the production line by obtaining sample production parameter state path data with and without abnormal state chain label data, and use these sample production parameter state path data to train and optimize the abnormal state chain prediction network, but also improve the accuracy and reliability of abnormal state prediction by introducing a guide machine learning network to generate confidence assessment. The optimized abnormal state chain prediction network can accurately predict the abnormal state chain of the target production parameter state path data, thereby realizing real-time monitoring and fault warning of the power cord production process, significantly improving production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A schematic diagram of a process of an online production monitoring method for power lines provided in an embodiment of the present application is shown;

[0017] Figure 2 A schematic diagram of the component structure of an online production monitoring system for power lines provided in an embodiment of the present application for implementing the above-mentioned online production monitoring method for power lines is shown. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present application.

[0019] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0020] Figure 1 The flowchart of the online production monitoring method for power lines provided in the embodiment of the present application is shown. It should be understood that in other embodiments, the order of some steps in the online production monitoring method for power lines can be interchanged according to actual needs, or some steps can be omitted or deleted. The detailed steps of the online production monitoring method for power lines are described as follows.

[0021] Step S110, obtaining first sample production parameter state path data and second sample production parameter state path data of each reference power cord production line, wherein the first sample production parameter state path data carries abnormal state chain label data, and the second sample production parameter state path data does not carry abnormal state chain label data.

[0022] In this embodiment, the server monitors the production conditions of multiple power line production lines. In this large production workshop, there are many power line production lines operating simultaneously.

[0023] For the first example, the acquisition of production parameter status path data:

[0024] The server is connected to the production management system and sensor network of each power cord production line. For example, on production line A, sensors can detect parameters such as the feed rate, temperature, pressure, and current of raw materials. These parameters change continuously as the production process progresses, forming a time series parameter state path. Assume that during a certain period of time, production line A has unqualified product quality. After investigation by engineers, it is found that the temperature is too high and the feed rate is unstable at a certain stage in the production process. This series of abnormal states is marked as abnormal state chain label data. This production parameter state path data with abnormal state chain label data becomes the first example production parameter state path data. Specifically, at a certain moment, the temperature sensor may detect that the temperature suddenly rises to 80 degrees Celsius, which is beyond the normal range (the normal range is 60-70 degrees Celsius). At the same time, the feed speed sensor shows that the feed speed suddenly drops from a stable 10 meters per minute to 5 meters. This abnormal state chain is marked with a label of "overtemperature-unstable feed speed", and this label is associated with the entire production parameter state path data, such as the temperature parameter, feed speed parameter, and the current, pressure and other related parameters at that time. The value change path during this time period constitutes the first sample production parameter state path data for production line A. The same situation also occurs in other production lines, such as production lines B, C, etc. The server collects the first sample production parameter state path data with abnormal state chain label data from these production lines.

[0025] For the second example, the acquisition of production parameter status path data:

[0026] Other production lines in the same workshop that are operating normally, such as production line D. The server obtains production parameters from the production management system and sensor network of production line D. During the entire production process, there were no product quality problems or any known abnormalities. Sensors continuously detect various parameters, such as the temperature has been stable at around 65 degrees Celsius, the feed speed has been maintained at 10 meters per minute, and parameters such as current and pressure have also fluctuated within the normal range. The state path data formed by these production parameters under normal operation over time is the second example production parameter state path data. Since there are no abnormal situations, it does not carry abnormal state chain label data. The server collects similar second example production parameter state path data from multiple normal production lines.

[0027] Step S120: respectively load the first sample production parameter state path data and the second sample production parameter state path data into the abnormal state chain prediction network serving as the guided machine learning network, and generate a first abnormal state chain prediction result of the first sample production parameter state path data and a second abnormal state chain prediction result of the second sample production parameter state path data.

[0028] In this embodiment, the server transmits the previously collected first sample production parameter state path data and second sample production parameter state path data to the abnormal state chain prediction network. The abnormal state chain prediction network is a pre-designed machine learning model, which may be built based on a neural network architecture.

[0029] Taking the first sample production parameter state path data of production line A as an example, when it is loaded into the abnormal state chain prediction network:

[0030] The abnormal state chain prediction network first parses the state path data of the production parameters of the first example and identifies the values ​​of each production parameter at different time points. Assuming that this abnormal state chain prediction network has multiple hidden layers, the data first enters the input layer and then passes through the first hidden layer. In this layer, the neurons perform weighted summation on the input data and perform nonlinear transformation through the activation function. For example, for the temperature parameter, the neuron may perform weighted processing on the temperature value according to the previously learned weights, and then combine it with the results of other parameters such as feed speed after being processed by the activation function (such as ReLU function), and continue to pass it to the deeper hidden layer. As the data propagates in the network, the network analyzes the state path data of the production parameter with the abnormal state chain label data according to the patterns learned during the previous training. Since similar abnormal situations have been encountered in the training process before, the network will predict the results related to the previously marked abnormal state chain label data based on the characteristics of the current input data. For example, the network may predict which links in this production process are most likely to be abnormal, and give the probability of each abnormal link. This result is the first abnormal state chain prediction result.

[0031] For the second sample production parameter state path data of production line D:

[0032] Similarly, the data is loaded into the abnormal state chain prediction network. The network analyzes the data according to the same processing flow. Because this data is the parameter state path data under normal production conditions, the network will determine whether the values ​​of each production parameter are within the normal range and whether the entire production process conforms to the normal mode during the analysis process. For example, the network will detect the stable state of parameters such as temperature, feed rate, and current, and based on these normal characteristics, predict that the production process is normal and there is no abnormal state chain. This result is the second abnormal state chain prediction result. At the same time, the network may also give some indicators about production stability, such as whether the fluctuation range of each parameter is within a very small normal range.

[0033] Step S130: load the second sample production parameter state path data into the guiding machine learning network corresponding to the guided machine learning network, and generate a third abnormal state chain prediction result of the second sample production parameter state path data, wherein the third abnormal state chain prediction result represents the confidence of the abnormal state chain and the target abnormal state label.

[0034] In this embodiment, the server transmits the second sample production parameter state path data obtained from production line D to the guidance machine learning network corresponding to the abnormal state chain prediction network. This guidance machine learning network is also a complex model, and its structure and parameter settings are related to the abnormal state chain prediction network.

[0035] When data enters the guided machine learning network: Assume that this guided machine learning network is a network built based on a probability model. The network first extracts features from the input production parameter state path data. For example, for temperature parameters, it analyzes the fluctuation characteristics of temperature throughout the production process, whether it is always stable near a certain value or has small periodic fluctuations. For feed rate parameters, it pays attention to whether the feed rate remains constant or changes according to a certain rule. The network performs calculations in the internal probability model based on these feature extraction results. Assume that there is a probability distribution model for the abnormal state chain in this network. For different target abnormal state labels, such as "abnormal temperature increase" and "abnormal feed rate decrease", the network will calculate the probability of each target abnormal state label according to the characteristics of the current data. This probability is the confidence that characterizes the abnormal state chain and the target abnormal state label. For example, for the target abnormal state label "abnormal temperature rise", the network finds that the temperature has been stable within the normal range based on the production parameter state path data of production line D. After calculation, it is found that the probability of the target abnormal state label is very low, such as only 0.01, which means that the network has a low confidence in the abnormal state chain of "abnormal temperature rise" on production line D. For the default target abnormal state label "normal production", the probability calculated by the network may be very high, such as 0.99, indicating that the network is very confident that production line D is in normal production and there is no abnormal state chain. This result, which includes the confidence of each target abnormal state label, is the prediction result of the third abnormal state chain.

[0036] Step S140: If the confidence level is greater than a predefined threshold value corresponding to the target abnormal state label, the third abnormal state chain prediction result is determined to be a trust result, and the abnormal state chain and the target abnormal state label represented in the third abnormal state chain prediction result are used as the abnormal state chain label data corresponding to the second sample production parameter state path data, so as to optimize the abnormal state chain prediction network based on the first abnormal state chain prediction result, the second abnormal state chain prediction result, the abnormal state chain label data corresponding to the first sample production parameter state path data, and the abnormal state chain label data corresponding to the second sample production parameter state path data.

[0037] In this embodiment, a threshold value is pre-defined in the server for each target abnormal state tag. For example, for the target abnormal state tag of "abnormal temperature rise", the threshold value is set to 0.1.

[0038] For production line D:

[0039] Previously, in step S130, the guidance machine learning network calculated that the confidence of the target abnormal state label "abnormal temperature rise" is 0.01, which is less than the threshold value of 0.1. Therefore, this result is not regarded as a trust result, and no abnormal state chain and target abnormal state label will be used as the abnormal state chain label data corresponding to the second sample production parameter state path data of production line D.

[0040] However, suppose that the situation of the second sample production parameter state path data of another production line E is different. After the data of production line E is loaded into the guiding machine learning network, the confidence calculated by the network for the target abnormal state label of "abnormal reduction in feed speed" is 0.2, and the threshold value corresponding to this target abnormal state label is set to 0.15. Since 0.2 is greater than 0.15, the prediction result of the third abnormal state chain is determined to be a trusted result. At this time, the server will use the abnormal state chain of "abnormal reduction in feed speed" represented in the third abnormal state chain prediction result and the target abnormal state label of "abnormal reduction in feed speed" as the abnormal state chain label data corresponding to the second sample production parameter state path data of production line E.

[0041] Then, based on the abnormal state chain label data corresponding to the first sample production parameter state path data (e.g., data with real abnormal state chain label data of production line A), the newly determined abnormal state chain label data of the second sample production parameter state path data of production line E, and the first abnormal state chain prediction result (prediction result for production line A, etc.) and the second abnormal state chain prediction result (prediction result for production line D, etc.) obtained previously, the abnormal state chain prediction network is optimized. For example, the server uses an algorithm based on error back propagation to adjust the neuron weights in the abnormal state chain prediction network. If in the first sample production parameter state path data, the real abnormal state chain label data indicates that the feed speed should be abnormally reduced at a certain time point, and the previous first abnormal state chain prediction result does not accurately predict this situation, then in the optimization process, the neuron weights related to the feed speed parameter processing in the network will be adjusted according to this error, so that the network can predict more accurately when encountering a similar situation next time. The same is true for the second sample production parameter state path data. If there is a difference between the newly determined abnormal state chain label data and the previous second abnormal state chain prediction result, the network weight will also be adjusted according to this difference to improve the prediction accuracy of the network.

[0042] Step S150, obtaining target production parameter state path data, loading the target production parameter state path data into the optimized abnormal state chain prediction network, and generating an abnormal state chain prediction result output by the abnormal state chain prediction network.

[0043] In this embodiment, after the optimization of the previous steps, the entire power cord production workshop continues to be monitored. Now there is a new production line F, and the server obtains the target production parameter state path data from the production line F. This target production parameter state path data is similar to the production parameter state path data of the first and second examples obtained previously, and includes the value change paths of various production parameters (such as temperature, feed rate, current, pressure, etc.) during the production process.

[0044] The server loads the target production parameter state path data into the optimized abnormal state chain prediction network. The abnormal state chain prediction network processes the input target production parameter state path data according to the previously optimized neuron weights and structures. For example, when the target production parameter state path data enters the input layer of the abnormal state chain prediction network, the neurons in the network perform weighted summation of the target production parameter state path data according to their respective weights, and process it through the activation function, and then the target production parameter state path data is transmitted between the hidden layers of the abnormal state chain prediction network. The abnormal state chain prediction network will analyze the production parameter state path data of production line F according to the previously learned pattern. Assume that during the production process of production line F, the feed speed fluctuates within a certain period of time, and the temperature also has a small upward trend. After analysis by the abnormal state chain prediction network, the abnormal state chain prediction network may predict the existence of an abnormal state chain such as "unstable feed speed-rising temperature", and give the probability of each abnormal link, such as the probability of unstable feed speed is 0.6, and the probability of rising temperature is 0.4. This includes the predicted abnormal state chain and the probability of each link, which is the abnormal state chain prediction result output by the abnormal state chain prediction network. Based on this result, the server can take timely measures to avoid possible product quality problems or production line failures, such as adjusting the feeding equipment or checking the temperature control system.

[0045] Based on the above steps, the embodiment of the present application obtains sample production parameter state path data with and without abnormal state chain label data, and uses these sample production parameter state path data to train and optimize the abnormal state chain prediction network, which can not only effectively identify the abnormal state chain in the production line, but also improve the accuracy and reliability of abnormal state prediction by introducing a guidance machine learning network to generate confidence assessment. The optimized abnormal state chain prediction network can accurately predict the abnormal state chain of the target production parameter state path data, thereby realizing real-time monitoring and fault warning of the power cord production process, significantly improving production efficiency and product quality.

[0046] In a possible implementation manner, the second sample production parameter state path data is any one of a training data sequence consisting of a plurality of sample production parameter state path data that do not carry abnormal state chain label data.

[0047] The method further comprises:

[0048] Step A110, using the guiding machine learning network to obtain multiple abnormal state chain prediction results corresponding to the multiple sample production parameter state path data.

[0049] Step A120, determining the confidences respectively corresponding to the target abnormal state labels in the plurality of abnormal state chain prediction results, and the number of times the target abnormal state labels are triggered in the plurality of abnormal state chain prediction results.

[0050] Step A130, based on the confidences respectively corresponding to the target abnormal state labels in the multiple abnormal state chain prediction results and the number of triggering times of the target abnormal state labels in the multiple abnormal state chain prediction results, determine the threshold value corresponding to the target abnormal state label.

[0051] In this embodiment, among the normal operation data of many production lines, the second sample production parameter state path data is selected from the training data sequence composed of multiple sample production parameter state path data that do not carry abnormal state chain label data. For example, in the normal operation production lines such as production lines G, H, and I, the server obtains their respective production parameter state path data, which constitute the training data sequence, and any one of them, such as the data of production line G, is selected as the second sample production parameter state path data.

[0052] The server uses the guided machine learning network to process these multiple sample production parameter state path data. Taking the data of production lines G, H, and I as an example, these data are loaded into the guided machine learning network respectively. The network begins to analyze each data, just like processing a single second sample production parameter state path data, extracting the characteristics of production parameters such as temperature, feed rate, current, pressure, etc., and then calculates according to the internal probability model or algorithm rules to obtain the abnormal state chain prediction results corresponding to each data. For example, for the data of production line G, the network may predict that there is a very small probability of an abnormal state chain such as "slight fluctuation in feed rate" at a certain moment, but the overall production line is considered normal; for the data of production line H, it may predict that there is a very low probability of an abnormal state chain such as "instantaneous slight rise in temperature", etc.

[0053] After obtaining multiple abnormal state chain prediction results, the server starts to process the target abnormal state label. For example, if the target abnormal state label is "abnormal fluctuation of feed speed", the server searches for the confidence level corresponding to this label in each prediction result. In the prediction results of production line G, the confidence level of "abnormal fluctuation of feed speed" may be 0.05, indicating that the network does not believe that production line G will have this abnormal state; in the prediction results of production line H, the confidence level may be 0.03, etc. At the same time, the server counts the number of times the target abnormal state label "abnormal fluctuation of feed speed" is triggered in these prediction results. If, after analyzing the data of 10 production lines, this label is triggered only in the prediction results of production line G and another production line J, then the number of triggers is 2.

[0054] The threshold value is determined based on the confidence and trigger times corresponding to the target abnormal state labels obtained previously. For the target abnormal state label "abnormal fluctuation of feed speed", if the confidence is generally low and the trigger times are few in the prediction results of all production lines, for example, the confidence is mostly between 0.01-0.05, and the trigger times do not exceed 3 times, the server will set a relatively low threshold value, such as 0.08. This is because from the existing data, this abnormal state rarely occurs, and the network's confidence in its judgment is not high, so when judging whether new data has this abnormal state in the future, the threshold value is also correspondingly low. On the contrary, if the confidence value has a higher value and the trigger times are more, a higher threshold value will be set.

[0055] In a possible implementation, the first sample production parameter state path data is any one of a training data sequence consisting of a plurality of sample production parameter state path data each carrying abnormal state chain label data, wherein the abnormal state chain label data includes an abnormal state label and an abnormal state chain.

[0056] Step S140 includes:

[0057] Step S141, if the confidence is greater than a predefined threshold value corresponding to the target abnormal state label, then based on the abnormal state chain corresponding to the target abnormal state label in the second sample production parameter state path data, the corresponding first abnormal state chain path data is extracted from the second sample production parameter state path data.

[0058] Step S142: Load the first abnormal state chain path data into a set graph autoencoder to generate a first graph autoencoder vector.

[0059] Step S143, obtaining one or more sample production parameter state path data corresponding to the target abnormal state label in the training data sequence.

[0060] Step S144, based on the abnormal state chains respectively corresponding to the target abnormal state tags in the one or more sample production parameter state path data, respectively extract one or more corresponding second abnormal state chain path data from the one or more sample production parameter state path data.

[0061] Step S145 , loading the one or more second abnormal state chain path data into the graph autoencoder respectively to generate one or more second graph autoencoder vectors.

[0062] Step S146: determine a target second image self-encoder vector based on the one or more second image self-encoder vectors.

[0063] Step S147: If the characteristic distance between the first image auto-encoder vector and the target second image auto-encoder vector is less than a set distance, determining that the third abnormal state chain prediction result is a trust result.

[0064] In this embodiment, when the server manages the power line production line data, the first sample production parameter state path data is selected from a training data sequence consisting of multiple sample production parameter state path data that all carry abnormal state chain label data (including abnormal state labels and abnormal state chains). For example, from the production line data with abnormal situation records of production lines K, L, M, etc., the data of production line K is selected as the first sample production parameter state path data.

[0065] When it is found that the confidence level of the target abnormal status label is greater than the predefined threshold value, the subsequent operation is started. Assume that the target abnormal status label is "temperature is too high and the feed speed is abnormally reduced", and the confidence level exceeds the corresponding threshold value. The server will extract the corresponding first abnormal status chain path data from the data of production line N based on the abnormal status chain corresponding to this target abnormal status label in the second sample production parameter status path data (such as data from production line N). This process is like filtering out the part related to "temperature is too high and feed speed is abnormally reduced" from the complete production parameter status path data of production line N, and extracting this part of the data, such as extracting the corresponding temperature, feed speed, and related current, pressure and other parameters' numerical change path data when such an abnormal state is suspected to occur.

[0066] Then, the extracted first abnormal state chain path data is loaded into the set graph autoencoder. The graph autoencoder processes this data and converts it into a specific representation form, that is, generates a first graph autoencoder vector. This vector contains the feature information of the first abnormal state chain path data encoded by the graph autoencoder.

[0067] Then the server obtains one or more sample production parameter state path data corresponding to the target abnormal state label "temperature is too high and feed speed is abnormally reduced" in the training data sequence. For example, from the previously collected data of production lines K, L, M, etc. with abnormal conditions, find the data related to this target abnormal state label. Based on the abnormal state chains corresponding to this target abnormal state label in these data (such as the data of production line K), the corresponding second abnormal state chain path data is extracted from the data of production line K. This data is also part of the data related to "temperature is too high and feed speed is abnormally reduced", such as the path of parameter changes such as temperature and feed speed within a specific time period. Similarly, this operation is performed on the data of other related production lines (such as production lines L and M) to obtain one or more second abnormal state chain path data.

[0068] The one or more second abnormal state chain path data are loaded into the graph autoencoder respectively, and the graph autoencoder generates a corresponding second graph autoencoder vector for each data. These second graph autoencoder vectors all contain the feature information of the respective second abnormal state chain path data.

[0069] Then, a target second image self-encoding vector is determined based on the one or more second image self-encoding vectors. For example, the most representative one can be selected from the vectors as the target second image self-encoding vector by comparing certain similarities between the vectors or according to a preset rule.

[0070] Finally, if the characteristic distance between the first image autoencoder vector and the target second image autoencoder vector is less than the set distance, the third abnormal state chain prediction result is determined to be a trusted result. The characteristic distance here is calculated by a certain distance measurement method (such as Euclidean distance, etc.). If this distance is less than the set distance, it means that the first abnormal state chain path data extracted from production line N is very similar in characteristics to the representative second abnormal state chain path data extracted from the training data sequence (such as production lines K, L, M, etc.), then the third abnormal state chain prediction result can be trusted, that is, it is believed that production line N may indeed have an abnormal situation related to the target abnormal state label.

[0071] In a possible implementation, step S140 further includes:

[0072] A first training error parameter is determined based on the first abnormal state chain prediction result and the abnormal state chain label data corresponding to the first sample production parameter state path data.

[0073] A second training error parameter is determined based on the second abnormal state chain prediction result and the abnormal state chain label data corresponding to the second sample production parameter state path data.

[0074] A global training error parameter is determined based on the first training error parameter and the second training error parameter to optimize the neuron weight information of the abnormal state chain prediction network based on the global training error parameter. Wherein, the neuron weight information of the guidance machine learning network is optimized based on the neuron weight information of the abnormal state chain prediction network.

[0075] In this embodiment, when processing the power line production line data, in order to optimize the abnormal state chain prediction network, it is necessary to perform operations based on previously obtained related results.

[0076] First, the first training error parameter is determined based on the first abnormal state chain prediction result and the abnormal state chain label data corresponding to the first sample production parameter state path data. For example, in the case of production line K, the abnormal state chain label data marked in the first sample production parameter state path data indicates that there is a situation of "temperature is too high and the feed speed is abnormally reduced" in a certain period of time, and clarifies the specific numerical range or time node and other detailed information corresponding to each link in this abnormal state chain. The first abnormal state chain prediction result may predict this abnormal state chain, and the probability of predicting the link of "temperature is too high" in the prediction result may be 0.8, and the probability of predicting the link of "feed speed is abnormally reduced" may be 0.7, but these two links are bound to occur in the actual abnormal state chain label data (probability is 1). The first training error parameter is determined by calculating the difference between the predicted result and the actual label data. For example, the mean square error (MSE) calculation method can be used to add the square of the difference between the predicted probability of the "temperature is too high" link and the actual probability (1) to the square of the difference between the predicted probability of the "abnormally reduced feed speed" link and the actual probability (1) to obtain a part of the first training error parameter, and then the error calculation of other related parameters is combined to finally obtain the first training error parameter.

[0077] Next, the second training error parameter is determined based on the second abnormal state chain prediction result and the abnormal state chain label data corresponding to the second sample production parameter state path data. Taking the data of production line N as an example, the second sample production parameter state path data was assigned abnormal state chain label data in the previous operation (assuming that an abnormal situation is considered to exist and marked after the previous step), such as an abnormal state chain marked as "slight temperature fluctuation and slightly unstable feed speed". The second abnormal state chain prediction result predicts that the probability of "slight temperature fluctuation" for this abnormal state chain is 0.1, and the probability of "slightly unstable feed speed" is 0.2, but the actual probability of "slight temperature fluctuation" in the actual marked abnormal state chain label data may be 0.3, and the actual probability of "slightly unstable feed speed" is 0.4. A similar mean square error calculation method is also used to calculate the sum of the squares of the difference between the predicted probability and the actual probability, thereby obtaining the second training error parameter.

[0078] Finally, the global training error parameter is determined based on the first training error parameter and the second training error parameter. For example, the global training error parameter can be obtained by simply adding the first training error parameter and the second training error parameter, or different weights can be assigned to the two error parameters according to their importance in the entire training process, and then the weighted sum is obtained to obtain the global training error parameter. After obtaining the global training error parameter, the neuron weight information of the abnormal state chain prediction network can be optimized based on this parameter. The server can use an optimization algorithm based on gradient descent to calculate the gradient of each neuron weight according to the global training error parameter, and then update the weight according to a certain learning rate, so that the network can more accurately predict the abnormal state chain in the next prediction. For example, if a neuron is related to the processing of the "temperature" parameter, and it is found that the prediction error related to temperature is large when calculating the global training error parameter, then when updating the weight, the weight of the neuron will be adjusted in the direction of reducing this error, thereby improving the accuracy of the abnormal state chain prediction network for the abnormal state chain prediction.

[0079] In a possible implementation manner, the step of loading the first sample production parameter state path data and the second sample production parameter state path data into an abnormal state chain prediction network as a guided machine learning network, respectively, includes:

[0080] The first sample production parameter state path data and the second sample production parameter state path data are respectively subjected to shallow path derivation reinforcement processing, and the first sample production parameter state path data after the path derivation reinforcement processing and the second sample production parameter state path data after the path derivation reinforcement processing are respectively loaded into the abnormal state chain prediction network which serves as the guided machine learning network.

[0081] In this embodiment, when processing the data of the power cord production line, for the operation of loading the first sample production parameter state path data and the second sample production parameter state path data into the abnormal state chain prediction network, it is necessary to first perform shallow path derivation and enhancement processing.

[0082] Take the first sample production parameter state path data of production line K as an example. This data contains the value change paths of various production parameters such as temperature, feed rate, current, pressure, etc. during the production process, and carries abnormal state chain label data. When performing shallow path derivative enhancement processing, the server will perform some simple feature derivative operations on these production parameters. For temperature parameters, the server may calculate the rate of change of temperature within a certain time window, such as calculating the average rate of change of temperature every 5 minutes, and obtain a new temperature change rate sequence. This sequence is the shallow derivative feature of the temperature parameter. For feed rate, the server can calculate the difference between feed rates in adjacent time periods to form a feed rate difference sequence. Similarly, similar operations are performed on current and pressure parameters, such as calculating derivative features such as the fluctuation amplitude of current and the relative stable value of pressure. Combining these derivative features with the original production parameter values, the shallow path derivative enhancement processing of the first sample production parameter state path data of production line K is completed.

[0083] For the second sample production parameter state path data of production line N, it is data under normal operating conditions, and there is no abnormal state chain label data. Similar operations are also used when performing shallow path derivation and strengthening processing. For example, for temperature parameters, the average temperature change rate every 10 minutes is calculated to obtain the temperature change rate sequence; for feed speed, the difference between the maximum and minimum feed speeds within a certain time interval is calculated to form a feed speed difference feature. Related feature derivations are also performed on current and pressure parameters, such as calculating the average fluctuation period of current, the maximum stable duration of pressure, etc. In this way, the second sample production parameter state path data after shallow path derivation and strengthening processing is obtained.

[0084] Then, the server loads the first sample production parameter state path data (such as the data of production line K) after shallow path derivation and reinforcement processing and the second sample production parameter state path data (such as the data of production line N) after path derivation and reinforcement processing into the abnormal state chain prediction network as the guided machine learning network. After receiving these data, the network can use the original production parameter values ​​and newly derived features contained therein to perform predictive analysis of the abnormal state chain. For example, the neurons in the network can analyze the data more comprehensively based on features such as temperature change rate and feed speed difference, combined with the original temperature and feed speed values, to determine whether there is an abnormal state chain and its possible situation.

[0085] In a possible implementation manner, the first sample production parameter state path data and the second sample production parameter state path data are respectively loaded into an abnormal state chain prediction network as a guided machine learning network to generate a first abnormal state chain prediction result of the first sample production parameter state path data and a second abnormal state chain prediction result of the second sample production parameter state path data, including:

[0086] The first sample production parameter state path data is subjected to shallow path derivation and enhancement processing to generate third sample production parameter state path data, wherein the third sample production parameter state path data carries the abnormal state chain label data corresponding to the first sample production parameter state path data.

[0087] The second sample production parameter state path data is subjected to shallow path derivation and enhancement processing to generate fourth sample production parameter state path data.

[0088] The second sample production parameter state path data is subjected to deep path derivation and enhancement processing to generate the fifth sample production parameter state path data.

[0089] The fifth sample production parameter state path data is subjected to self-attention processing to generate the sixth sample production parameter state path data.

[0090] The third sample production parameter state path data, the fourth sample production parameter state path data, the fifth sample production parameter state path data and the sixth sample production parameter state path data are respectively loaded into the abnormal state chain prediction network as the guided machine learning network to generate a first abnormal state chain prediction result of the third sample production parameter state path data, a second abnormal state chain prediction result of the fourth sample production parameter state path data, state path segments of multiple state spans of the fourth sample production parameter state path data, state path segments of multiple state spans of the fifth sample production parameter state path data and state path segments of multiple state spans of the sixth sample production parameter state path data.

[0091] In this embodiment, when processing the power line production line data, the first sample production parameter state path data and the second sample production parameter state path data are loaded into the abnormal state chain prediction network and relevant results are generated.

[0092] For the first sample production parameter state path data, take the data of production line K as an example. First, perform shallow path derivation and enhancement processing on it to generate the third sample production parameter state path data and carry the original abnormal state chain label data. In the shallow path derivation and enhancement process, for the temperature parameters of production line K, in addition to the original temperature value, the server calculates the slope of the temperature in each production stage, such as the slope of the temperature rise in the initial feeding stage, the slope of the temperature stability in the intermediate processing stage is 0, and the slope of the temperature drop in the final stage; for the feed speed, calculate the change ratio of the feed speed in adjacent time periods, such as the feed speed increased by 20% from the first minute to the second minute. At the same time, similar operations are performed on parameters such as current and pressure, such as calculating the fluctuation frequency of the current and the rate of change of pressure in different stages. Integrating these newly derived data with the original data will obtain the third sample production parameter state path data.

[0093] For the second sample production parameter state path data, such as the data of production line N. First, perform shallow path derivative strengthening processing to generate the fourth sample production parameter state path data. For temperature parameters, calculate the average temperature change in each short period of time, such as how many degrees the temperature rises or falls on average every 3 minutes; for feed speed, calculate the maximum change in feed speed in each time period, for example, the maximum change in feed speed in a certain 5 minutes is 5 meters / minute; for current, calculate the difference between the average current and the standard current, and for pressure, calculate the fluctuation range, etc.

[0094] Then, the second sample production parameter state path data of production line N is subjected to deep path derivation and enhancement processing to generate the fifth sample production parameter state path data. This process will dig deeper relationships. For example, the server will analyze the delay relationship between temperature change and feed speed change, assuming how long the feed speed will change accordingly after the temperature rises, or whether there is a periodic correlation between current change and pressure change. These hidden relationships are found through complex algorithms and models, and new data features are generated and added to the fifth sample production parameter state path data.

[0095] Then, the fifth sample production parameter state path data is processed by self-attention to generate the sixth sample production parameter state path data. The self-attention mechanism allows each element in the data to pay attention to the importance of other related elements. For example, among the parameters such as temperature, feed rate, current, and pressure, the self-attention process may determine that the temperature has a greater impact on the entire production state at a certain production stage, and then the data representation will be adjusted accordingly to highlight the importance of the temperature parameter at this stage. These adjusted results are integrated to obtain the sixth sample production parameter state path data.

[0096] Finally, the server loads the third sample production parameter state path data (data processed by production line K), the fourth sample production parameter state path data (data processed by production line N shallowly), the fifth sample production parameter state path data (data processed by production line N deeply), and the sixth sample production parameter state path data (data processed by production line N self-attention) into the abnormal state chain prediction network respectively.

[0097] For the third sample production parameter state path data, the network analyzes the original data and derived data contained therein, as well as the abnormal state chain label data carried. For example, based on the temperature change slope of production line K, the feed speed change ratio and other data, combined with the known abnormal state chain labels (such as excessive temperature and abnormally reduced feed speed), it predicts at which production stages this abnormal state chain is most likely to occur, as well as the severity of the abnormal state at each stage, etc. This is the prediction result of the first abnormal state chain.

[0098] For the fourth sample production parameter state path data, the network analyzes the derivative characteristics of various production parameters, determines whether production line N is in a normal production state, and predicts whether various production parameters fluctuate within a normal range. This is the second abnormal state chain prediction result. At the same time, the network will segment the fourth sample production parameter state path data according to different production stages or different time spans to obtain state path segments of multiple state spans. For example, the entire production process is segmented into a span of 10 minutes to obtain state path segments of parameters such as temperature and feed speed in each span.

[0099] For the fifth sample production parameter state path data, the network will also analyze in a similar way to obtain state path fragments of multiple state spans. Due to the deep path derivation and strengthening processing, these state path fragments contain more information about deep relationships, such as the impact of the correlation between temperature and feed rate on the production state within a certain state span.

[0100] For the sixth sample production parameter state path data, the network analyzes the important information highlighted after self-attention processing to obtain state path segments of multiple state spans. The weights of each parameter in these state path segments are adjusted by self-attention, which can better reflect the key information in the production process. For example, in a certain key production stage, the temperature parameter is highlighted due to the self-attention mechanism, and the slight changes in temperature in the corresponding state path segment can be accurately analyzed, which helps to more accurately determine whether the production status is normal or whether there is an abnormal state chain.

[0101] In a possible implementation, step S130 includes:

[0102] The fourth sample production parameter state path data is loaded into the guiding machine learning network corresponding to the guided machine learning network to generate a third abnormal state chain prediction result of the fourth sample production parameter state path data.

[0103] The method uses the abnormal state chain and the target abnormal state label represented in the third abnormal state chain prediction result as the abnormal state chain label data corresponding to the second sample production parameter state path data, so as to optimize the abnormal state chain prediction network based on the first abnormal state chain prediction result, the second abnormal state chain prediction result, the abnormal state chain label data corresponding to the first sample production parameter state path data, and the abnormal state chain label data corresponding to the second sample production parameter state path data, including:

[0104] The abnormal state chain and the target abnormal state label represented in the third abnormal state chain prediction result are used as the abnormal state chain label data corresponding to the fourth sample production parameter state path data, so as to optimize the abnormal state chain prediction network based on the first abnormal state chain prediction result and the corresponding abnormal state chain label data of the third sample production parameter state path data, the second abnormal state chain prediction result and the corresponding abnormal state chain label data of the fourth sample production parameter state path data, and the state path segments of multiple state spans of the fourth sample production parameter state path data, the state path segments of multiple state spans of the fifth sample production parameter state path data, and the state path segments of multiple state spans of the sixth sample production parameter state path data.

[0105] In this embodiment, when the second sample production parameter state path data is loaded into the guiding machine learning network to generate the third abnormal state chain prediction result, the specific operation is as follows. Taking the previously mentioned production line N as an example, the server loads the fourth sample production parameter state path data obtained after shallow path derivation and reinforcement processing into the guiding machine learning network corresponding to the guided machine learning network. This guiding machine learning network has its own algorithm and model structure. When the fourth sample production parameter state path data enters the network, the network will analyze the various production parameter characteristics therein. For example, for temperature parameters, it will check the average change amplitude and maximum change amount of temperature at different stages; for feed speed, it will pay attention to the maximum change amount of feed speed and the change trend in different time periods. Based on these characteristics, the network predicts possible abnormal state chains based on internally set rules and previously learned patterns. For example, based on the sudden increase in feed rate within a certain period of time and the small fluctuation in temperature, the network may predict the existence of an abnormal state chain such as "abnormal fluctuation in feed rate-instability in temperature", and give a corresponding confidence level to each possible abnormal state (such as abnormal fluctuation in feed rate, unstable temperature). The result including the abnormal state chain and the confidence level of each state is the prediction result of the third abnormal state chain of the fourth sample production parameter state path data.

[0106] Next, when the abnormal state chain and the target abnormal state label represented in the third abnormal state chain prediction result are used as the abnormal state chain label data corresponding to the fourth sample production parameter state path data, and then the abnormal state chain prediction network is optimized. The server regards the abnormal state chain of "abnormal fluctuation of feed speed-instability of temperature" mentioned in the third abnormal state chain prediction result obtained previously and the target abnormal state label therein (such as abnormal fluctuation of feed speed, instability of temperature) as the new abnormal state chain label data of the fourth sample production parameter state path data of production line N. Then the abnormal state chain prediction network is optimized based on multiple results. First, the first abnormal state chain prediction result of the third sample production parameter state path data (from production line K) and its corresponding abnormal state chain label data, for example, the first abnormal state chain prediction result of production line K shows that there is a situation of "too high temperature and abnormal reduction of feed speed" at a certain stage and gives the prediction of each link, and the corresponding abnormal state chain label data clarifies that this abnormal state chain does exist, and there is a certain difference between the two, and this difference can be used to adjust the network. The second is the second abnormal state chain prediction result of the fourth sample production parameter state path data (production line N) and its newly determined abnormal state chain label data. For example, the second abnormal state chain prediction result may originally believe that production line N is in normal production, but the new abnormal state chain label data indicates that there is a situation of "abnormal fluctuation of feed speed-instability of temperature". The difference between them is also the basis for optimizing the network. In addition, there are state path fragments of multiple state spans of the fourth sample production parameter state path data. For example, after the production process of production line N is divided into a span of 10 minutes, the state path fragments of parameters such as temperature and feed speed in each span. The parameter changes and mutual relationships in these fragments can provide more detailed information about the production status. The state path fragments of multiple state spans of the fifth sample production parameter state path data (production line N after deep path derivative reinforcement processing) contain deeper production parameter relationships, such as the impact of the deep correlation between temperature and feed speed in a span on the production status. The state path segments of multiple state spans of the sixth sample production parameter state path data (production line N after self-attention processing) highlight the important parameters of the key production stage. For example, the temperature parameter is focused on in a key stage due to the self-attention mechanism, and its changes in the state path segments can better reflect the production status. The server integrates this information and uses appropriate algorithms (such as algorithms based on error back propagation) to adjust parameters such as neuron weights in the abnormal state chain prediction network.For example, if the prediction of excessive temperature in the third sample production parameter state path data deviates from the actual label data, and if the relationship between temperature and feed speed is found to be inconsistent with the prediction in the state path segment of a certain state span of the fourth sample production parameter state path data, the server will adjust the neuron weights related to temperature and feed speed processing in the network according to these situations, thereby optimizing the abnormal state chain prediction network so that it can more accurately judge the abnormal state chain in subsequent predictions.

[0107] In a possible implementation, the first abnormal state chain prediction result based on the third sample production parameter state path data and the corresponding abnormal state chain label data, the second abnormal state chain prediction result based on the fourth sample production parameter state path data and the corresponding abnormal state chain label data, and the state path segments of multiple state spans of the fourth sample production parameter state path data, the state path segments of multiple state spans of the fifth sample production parameter state path data, and the state path segments of multiple state spans of the sixth sample production parameter state path data, optimize the abnormal state chain prediction network, including:

[0108] Based on the first abnormal state chain prediction result of the third sample production parameter state path data and the corresponding abnormal state chain label data, a first training error parameter is determined.

[0109] Based on the second abnormal state chain prediction result of the fourth sample production parameter state path data and the corresponding abnormal state chain label data, a second training error parameter is determined.

[0110] For the state path segments of multiple state spans corresponding to the fourth sample production parameter state path data and the fifth sample production parameter state path data, a fusion calculation result of the characteristic distance training error parameter between the state path segments of the same state span is determined as the third training error parameter.

[0111] For the state path segments of multiple state spans corresponding to the fifth sample production parameter state path data and the sixth sample production parameter state path data, a fusion calculation result of the characteristic distance training error parameter between the state path segments of the same state span is determined as the fourth training error parameter.

[0112] A global training error parameter is determined based on the first training error parameter, the second training error parameter, the third training error parameter, and the fourth training error parameter to optimize neuron weight information of the abnormal state chain prediction network based on the global training error parameter.

[0113] In this embodiment, first, the first training error parameter is determined based on the first abnormal state chain prediction result of the third sample production parameter state path data and the corresponding abnormal state chain label data. Taking the case of production line K as an example, the corresponding abnormal state chain label data in the third sample production parameter state path data indicates that there is an abnormal state chain of "temperature is too high and feed speed is abnormally reduced" at a certain stage, and the accurate situation of each state is clarified. The first abnormal state chain prediction result predicts this abnormal state chain. Assuming that the probability of "temperature is too high" in the prediction is 0.8, and this state is actually bound to occur in the label data (probability is 1), the predicted probability of "feed speed is abnormally reduced" is 0.7, and the actual probability is 1. The first training error parameter is determined by calculating the difference between the prediction result and the label data. The mean square error (MSE) calculation method can be used to calculate the square of the difference between the predicted probability and the actual probability in the "temperature is too high" state, and add the square of the difference between the predicted probability and the actual probability in the "feed speed is abnormally reduced" state to obtain a part of the first training error parameter, and the first training error parameter is finally determined after comprehensively considering the errors of other related states.

[0114] Next, the second training error parameter is determined based on the second abnormal state chain prediction result of the fourth sample production parameter state path data and the corresponding abnormal state chain label data. For the fourth sample production parameter state path data of production line N, it is assumed that the newly determined abnormal state chain label data indicates that there is a situation of "small abnormal fluctuation in feed speed and slight instability in temperature", and the predicted probability of "small abnormal fluctuation in feed speed" in the second abnormal state chain prediction result is 0.1, and the actual probability may be 0.3, and the predicted probability of "slightly unstable temperature" is 0.2, and the actual probability is 0.4. The mean square error calculation method is also used to calculate the sum of the squares of the difference between the predicted probability and the actual probability under these two states, and then consider other related states to determine the second training error parameter.

[0115] Then, for the state path segments of the multiple state spans corresponding to the fourth sample production parameter state path data and the fifth sample production parameter state path data, a third training error parameter is determined. For example, the production process of production line N is divided into a state span of 10 minutes each, and respective state path segments are obtained. For a certain same state span, such as the third 10-minute state span, in the state path segment of the fourth sample production parameter state path data, the numerical change of the temperature parameter, the numerical value of the feed speed, and the simple relationship between them are one case, and in the state path segment of the fifth sample production parameter state path data, due to the deep path derivative reinforcement processing, in addition to the numerical change of the temperature and the feed speed, it also contains more features such as a deeper correlation relationship between the temperature and the feed speed. By calculating the feature distance between the two state path segments, for example, using the Euclidean distance or other suitable distance measurement method, the distance of the temperature parameter numerical change, the distance of the feed speed numerical change, and the distance between other features are calculated, and then these distances are fused and calculated according to a certain weight (this weight can be determined according to actual conditions or experience), the feature distance training error parameter under this state span is obtained. This calculation is performed for all identical state spans, and then the results are combined as the third training error parameter.

[0116] The process of determining the fourth training error parameter for the state path segments of the multiple state spans corresponding to the fifth sample production parameter state path data and the sixth sample production parameter state path data is similar. For example, in the fifth 10-minute state span, the state path segment of the fifth sample production parameter state path data contains the production parameter relationship after deep path derivative reinforcement processing, and the state path segment of the sixth sample production parameter state path data contains the important parameter relationship highlighted after self-attention processing. Calculate the characteristic distance between the two state path segments, such as the difference in the importance change of temperature after self-attention processing and the relationship difference under deep path derivative reinforcement processing, the characteristic difference of feed speed in two cases, etc., fuse these distance calculation results according to certain weights, perform such operations on all the same state spans, and then fuse to obtain the fourth training error parameter.

[0117] Finally, the global training error parameter is determined based on the first training error parameter, the second training error parameter, the third training error parameter and the fourth training error parameter. For example, the global training error parameter can be obtained by simply adding the four training error parameters, or different weights can be assigned according to their importance in the entire training process, and then the weighted sum is obtained to obtain the global training error parameter. After obtaining the global training error parameter, the server uses this parameter to optimize the neuron weight information of the abnormal state chain prediction network. An optimization algorithm based on gradient descent can be used to calculate the gradient of each neuron weight according to the global training error parameter, and then the weight is updated according to a certain learning rate, so that the network can more accurately predict the abnormal state chain in the next prediction. For example, if a neuron is related to the processing of temperature parameters, and it is found that the error related to temperature is large when calculating the global training error parameter, then when updating the weight, the weight of the neuron will be adjusted in the direction of reducing this error, thereby improving the accuracy of the abnormal state chain prediction network for the abnormal state chain prediction.

[0118] Further, Figure 2 FIG. 1 is a schematic diagram showing the hardware structure of a power line online production monitoring system 100 for implementing the method provided in an embodiment of the present application. Figure 2 As shown, the power line online production monitoring system 100 may include at least one processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 and a controller 108 for communication functions. It can be understood by those skilled in the art that Figure 2 The structure shown is only for illustration and does not limit the structure of the power line online production monitoring system 100. For example, the power line online production monitoring system 100 may also include Figure 2 More or fewer components as shown, or with Figure 2 Different configurations shown.

[0119] The memory 104 can be used to store software programs and modules of application software, such as program instructions corresponding to the above-mentioned method embodiments in the embodiments of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, to implement the above-mentioned power line online production monitoring method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the power line online production monitoring system 100 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0120] The transmission device 106 is used to obtain or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the power line online production monitoring system 100. In one example, the transmission device 106 includes a network adapter, which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency module, which is used to communicate with the Internet wirelessly.

[0121] It should be noted that the order of the above embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. The above describes specific embodiments of the embodiments of the present application. Other embodiments are within the scope of the attached claims. In some cases, the exceptions or steps recorded in the claims can be performed according to an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or can be advantageous.

[0122] Each embodiment in the embodiments of the present application is described in a progressive manner, and the parts that are consistent and similar between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the above different embodiments, since they are basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0123] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program. The above program may be stored in a computer-readable storage medium, and the above storage medium may be a read-only memory, a disk or an optical disk, etc.

Claims

1. A method for online production monitoring of power lines, characterized in that: The method comprises: Acquire first sample production parameter state path data and second sample production parameter state path data of each reference power cord production line, wherein the first sample production parameter state path data carries abnormal state chain label data, and the second sample production parameter state path data does not carry abnormal state chain label data; The first sample production parameter state path data and the second sample production parameter state path data are respectively loaded into an abnormal state chain prediction network as a guided machine learning network, and a first abnormal state chain prediction result of the first sample production parameter state path data and a second abnormal state chain prediction result of the second sample production parameter state path data are generated; Loading the second sample production parameter state path data into the guiding machine learning network corresponding to the guided machine learning network, generating a third abnormal state chain prediction result of the second sample production parameter state path data, wherein the third abnormal state chain prediction result represents the confidence of the abnormal state chain and the target abnormal state label; If the confidence is greater than a predefined threshold value corresponding to the target abnormal state label, the third abnormal state chain prediction result is determined to be a trust result, and the abnormal state chain and the target abnormal state label represented in the third abnormal state chain prediction result are used as the abnormal state chain label data corresponding to the second sample production parameter state path data, so as to optimize the abnormal state chain prediction network based on the first abnormal state chain prediction result, the second abnormal state chain prediction result, the abnormal state chain label data corresponding to the first sample production parameter state path data, and the abnormal state chain label data corresponding to the second sample production parameter state path data; Obtain target production parameter state path data, load the target production parameter state path data into the optimized abnormal state chain prediction network, and generate an abnormal state chain prediction result output by the abnormal state chain prediction network.

2. The method for online production monitoring of power lines according to claim 1, characterized in that: The second sample production parameter state path data is any one of a training data sequence consisting of a plurality of sample production parameter state path data that do not carry abnormal state chain label data; The method further comprises: Using the guiding machine learning network to obtain a plurality of abnormal state chain prediction results corresponding to the plurality of sample production parameter state path data; Determine the confidences respectively corresponding to the target abnormal state labels in the multiple abnormal state chain prediction results, and the number of triggering times of triggering the target abnormal state labels in the multiple abnormal state chain prediction results; The threshold value corresponding to the target abnormal state label is determined based on the confidences respectively corresponding to the target abnormal state labels in the multiple abnormal state chain prediction results and the number of triggering times of triggering the target abnormal state label in the multiple abnormal state chain prediction results.

3. The method for online production monitoring of power lines according to claim 1, characterized in that: The first sample production parameter state path data is any one of a training data sequence consisting of a plurality of sample production parameter state path data each carrying abnormal state chain label data, wherein the abnormal state chain label data includes an abnormal state label and an abnormal state chain; If the confidence is greater than a predefined threshold value corresponding to the target abnormal state label, determining that the third abnormal state chain prediction result is a trust result includes: If the confidence is greater than a predefined threshold value corresponding to the target abnormal state label, extracting corresponding first abnormal state chain path data from the second sample production parameter state path data based on the abnormal state chain corresponding to the target abnormal state label in the second sample production parameter state path data; Loading the first abnormal state chain path data into a set graph autoencoder to generate a first graph autoencoder vector; Acquire one or more sample production parameter state path data corresponding to the target abnormal state label in the training data sequence; Based on the abnormal state chains respectively corresponding to the target abnormal state tags in the one or more sample production parameter state path data, extracting corresponding one or more second abnormal state chain path data from the one or more sample production parameter state path data respectively; Loading the one or more second abnormal state chain path data into the graph autoencoder respectively to generate one or more second graph autoencoder vectors; determining a target second graph autoencoder vector based on the one or more second graph autoencoder vectors; If the characteristic distance between the first image auto-encoder vector and the target second image auto-encoder vector is less than a set distance, the third abnormal state chain prediction result is determined to be a trust result.

4. The method for online production monitoring of power lines according to claim 1, characterized in that: The optimizing the abnormal state chain prediction network based on the first abnormal state chain prediction result, the second abnormal state chain prediction result, the abnormal state chain label data corresponding to the first sample production parameter state path data, and the abnormal state chain label data corresponding to the second sample production parameter state path data includes: Determining a first training error parameter based on the first abnormal state chain prediction result and the abnormal state chain label data corresponding to the first sample production parameter state path data; Determining a second training error parameter based on the second abnormal state chain prediction result and the abnormal state chain label data corresponding to the second sample production parameter state path data; A global training error parameter is determined based on the first training error parameter and the second training error parameter to optimize neuron weight information of the abnormal state chain prediction network based on the global training error parameter.

5. The method for online production monitoring of power lines according to claim 4, characterized in that: The method further comprises: Based on the neuron weight information of the abnormal state chain prediction network, the neuron weight information of the guidance machine learning network is optimized.

6. The method for online production monitoring of power lines according to claim 1, characterized in that: The step of loading the first sample production parameter state path data and the second sample production parameter state path data into the abnormal state chain prediction network as the guided machine learning network respectively includes: The first sample production parameter state path data and the second sample production parameter state path data are respectively subjected to shallow path derivation reinforcement processing, and the first sample production parameter state path data after the path derivation reinforcement processing and the second sample production parameter state path data after the path derivation reinforcement processing are respectively loaded into the abnormal state chain prediction network which serves as the guided machine learning network.

7. The method for online production monitoring of power lines according to claim 1, characterized in that: The first sample production parameter state path data and the second sample production parameter state path data are respectively loaded into the abnormal state chain prediction network as the guided machine learning network, and a first abnormal state chain prediction result of the first sample production parameter state path data and a second abnormal state chain prediction result of the second sample production parameter state path data are generated, including: Performing shallow path derivation and enhancement processing on the first sample production parameter state path data to generate third sample production parameter state path data, wherein the third sample production parameter state path data carries abnormal state chain label data corresponding to the first sample production parameter state path data; Performing shallow path derivation and enhancement processing on the second sample production parameter state path data to generate fourth sample production parameter state path data; Performing deep path derivation and enhancement processing on the second sample production parameter state path data to generate fifth sample production parameter state path data; Performing self-attention processing on the fifth sample production parameter state path data to generate sixth sample production parameter state path data; The third sample production parameter state path data, the fourth sample production parameter state path data, the fifth sample production parameter state path data and the sixth sample production parameter state path data are respectively loaded into the abnormal state chain prediction network as the guided machine learning network to generate a first abnormal state chain prediction result of the third sample production parameter state path data, a second abnormal state chain prediction result of the fourth sample production parameter state path data, state path segments of multiple state spans of the fourth sample production parameter state path data, state path segments of multiple state spans of the fifth sample production parameter state path data and state path segments of multiple state spans of the sixth sample production parameter state path data.

8. The method for online production monitoring of power lines according to claim 7, characterized in that: The step of loading the second sample production parameter state path data into the guiding machine learning network corresponding to the guided machine learning network to generate a third abnormal state chain prediction result of the second sample production parameter state path data includes: Loading the fourth sample production parameter state path data into a guiding machine learning network corresponding to the guided machine learning network, and generating a third abnormal state chain prediction result of the fourth sample production parameter state path data; The method uses the abnormal state chain and the target abnormal state label represented in the third abnormal state chain prediction result as the abnormal state chain label data corresponding to the second sample production parameter state path data, so as to optimize the abnormal state chain prediction network based on the first abnormal state chain prediction result, the second abnormal state chain prediction result, the abnormal state chain label data corresponding to the first sample production parameter state path data, and the abnormal state chain label data corresponding to the second sample production parameter state path data, including: The abnormal state chain and the target abnormal state label represented in the third abnormal state chain prediction result are used as the abnormal state chain label data corresponding to the fourth sample production parameter state path data, so as to optimize the abnormal state chain prediction network based on the first abnormal state chain prediction result and the corresponding abnormal state chain label data of the third sample production parameter state path data, the second abnormal state chain prediction result and the corresponding abnormal state chain label data of the fourth sample production parameter state path data, and the state path segments of multiple state spans of the fourth sample production parameter state path data, the state path segments of multiple state spans of the fifth sample production parameter state path data, and the state path segments of multiple state spans of the sixth sample production parameter state path data.

9. The method for online production monitoring of power lines according to claim 8, characterized in that: The method optimizes the abnormal state chain prediction network based on the first abnormal state chain prediction result of the third sample production parameter state path data and the corresponding abnormal state chain label data, the second abnormal state chain prediction result of the fourth sample production parameter state path data and the corresponding abnormal state chain label data, and the state path segments of multiple state spans of the fourth sample production parameter state path data, the state path segments of multiple state spans of the fifth sample production parameter state path data, and the state path segments of multiple state spans of the sixth sample production parameter state path data, including: Determine a first training error parameter based on the first abnormal state chain prediction result of the third sample production parameter state path data and the corresponding abnormal state chain label data; Determine a second training error parameter based on the second abnormal state chain prediction result of the fourth sample production parameter state path data and the corresponding abnormal state chain label data; For the state path segments of the plurality of state spans respectively corresponding to the fourth sample production parameter state path data and the fifth sample production parameter state path data, determining a fusion calculation result of the characteristic distance training error parameter between the state path segments of the same state span as the third training error parameter; For the state path segments of the multiple state spans respectively corresponding to the fifth sample production parameter state path data and the sixth sample production parameter state path data, determine a fusion calculation result of the characteristic distance training error parameter between the state path segments of the same state span as a fourth training error parameter; A global training error parameter is determined based on the first training error parameter, the second training error parameter, the third training error parameter, and the fourth training error parameter to optimize neuron weight information of the abnormal state chain prediction network based on the global training error parameter.

10. A power line online production monitoring system, characterized in that: The power line online production monitoring system includes a processor and a readable storage medium, wherein the readable storage medium stores a program, and when the program is executed by the processor, the power line online production monitoring method according to any one of claims 1 to 9 is implemented.

Images