Equipment fault detection method and device, equipment, storage medium and program product

Abstract

The embodiment of the invention provides an equipment fault detection method and device, equipment, a storage medium and a program product. The method comprises the steps of obtaining historical operation data of tire forming equipment to extract data features, performing dimension reduction according to correlation of the data features to obtain a feature subset, training a fault detection model according to the feature subset and the historical operation data, and applying the trained model to fault detection of the equipment. Wherein the historical operation data can comprise pressure data related to the equipment and tire detection results, deep features of the data and association between the deep features and equipment faults can be learned through the training model, so that the fault detection results of the equipment can be obtained by inputting real-time data of the equipment into the model, normal operation of the equipment does not need to be intervened, and the efficiency is improved. Automatic processing can be realized only by acquiring related data input models of the equipment under normal operation, the fault detection efficiency is high, the timeliness is strong, and the equipment maintenance difficulty can be further reduced.

Description

Technical Field

[0001] This application relates to the field of tire manufacturing, and particularly to a method, device, equipment, storage medium, and program product for detecting equipment failures. Background Art

[0002] In the tire industrial chain, tire forming equipment is a key piece of equipment for manufacturing tires, which can assemble tire components into tire carcasses. The structure of tire forming equipment is complex and its operating conditions vary, making maintenance difficult.

[0003] Currently, the maintenance strategies for tire forming equipment often involve methods such as regular maintenance and breakdown maintenance. Among them, regular maintenance delays the normal operation of the equipment, is prone to over-maintenance, and causes waste of resources; breakdown maintenance, on the other hand, interrupts production due to sudden equipment failures, resulting in significant losses. Summary of the Invention

[0004] Embodiments of this application provide a method, device, equipment, storage medium, and program product for detecting equipment failures to solve the problem of high maintenance difficulty of tire forming equipment.

[0005] In a first aspect, embodiments of this application provide a method for detecting equipment failures, including:

[0006] Obtain historical operation data of the tire forming equipment, where the historical operation data includes pressure data and tire detection results when the tire forming equipment processes tires;

[0007] Extract features from the historical operation data to obtain data features;

[0008] Perform dimensionality reduction processing on the data features according to the correlation of the data features to obtain a reduced-dimensional feature subset;

[0009] Train a fault detection model based on the feature subset and the historical operation data, and use the trained fault detection model to detect the tire forming equipment.

[0010] In a possible implementation, the data features include multiple data indicators. The step of performing dimensionality reduction processing on the data features according to the correlation of the data features to obtain a reduced-dimensional feature subset includes:

[0011] Determine the correlation of different data indicators through a preset algorithm, and perform dimensionality reduction processing on the data features according to the correlation to obtain a feature subset. The algorithm includes at least one of principal component analysis, random forest, and autoencoder.

[0012] In a possible implementation, the step of performing dimensionality reduction processing on the data features according to the correlation to obtain a feature subset includes:

[0013] Calculate the covariance matrix of the data features;

[0014] Perform eigenvalue decomposition according to the covariance matrix to obtain a plurality of eigenvalues and eigenvectors corresponding to the eigenvalues;

[0015] Select target eigenvalues with a contribution rate greater than or equal to a preset threshold from the plurality of eigenvalues, and perform a linear transformation on the data features according to the eigenvectors corresponding to the target eigenvalues to obtain a feature subset.

[0016] In a possible implementation manner, training a fault detection model based on the feature subset and the historical operation data includes:

[0017] Extract deep features in the feature subset through a deep belief network;

[0018] Determine a decision rule according to the deep features, and construct an initial model corresponding to the decision rule, where the decision rule is used to determine the fault type of the tire forming equipment;

[0019] Select a data set from the historical operation data, and use the data set as a training sample to train the initial model to obtain a trained fault detection model.

[0020] In a possible implementation manner, the historical operation data further includes identification information of the tire, and the data set is obtained through the following method:

[0021] Determine the correspondence between the pressure data and the tire detection result according to the identification information;

[0022] Label the pressure data according to the tire detection result and the correspondence, and classify the labeled pressure data into a preset data set.

[0023] In a possible implementation manner, training a fault detection model based on the feature subset and the historical operation data further includes:

[0024] Verify the training effect of the fault detection model through a cross-validation algorithm, and adjust the model parameters of the fault detection model according to the training effect.

[0025] In a possible implementation manner, using the trained fault detection model to detect the tire forming equipment includes:

[0026] Deploy the trained fault detection model to the monitoring system of the tire forming equipment, and the monitoring system is used to collect the operation data of the tire forming equipment;

[0027] The operation data collected by the monitoring system is preprocessed and then input into the fault detection model to obtain the detection result output by the fault detection model.

[0028] In a second aspect, an embodiment of the present application provides a device fault detection apparatus, including:

[0029] An acquisition module, configured to acquire historical operation data of a tire forming device, where the historical operation data includes pressure data and tire detection results when the tire forming device processes a tire;

[0030] A feature extraction module, configured to extract features from the historical operation data to obtain data features;

[0031] A feature processing module, configured to perform dimensionality reduction processing on the data features according to the correlation of the data features to obtain a reduced-dimensional feature subset;

[0032] A detection module, configured to train a fault detection model based on the feature subset and the historical operation data, and use the trained fault detection model to detect the tire forming device.

[0033] In a third aspect, an embodiment of the present application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0034] The memory stores computer execution instructions;

[0035] The processor executes the computer execution instructions stored in the memory, so that the processor executes the first aspect and / or various possible implementation manners of the first aspect as above.

[0036] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer execution instructions are stored, and when the computer execution instructions are executed by a processor, they are used to implement the first aspect and / or various possible implementation manners of the first aspect as above.

[0037] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the first aspect and / or various possible implementation manners of the first aspect as above.

[0038] The device fault detection method, device, equipment, storage medium and program product provided by the embodiments of the present application can extract data features by obtaining the historical operation data of the tire forming equipment. According to the correlation of the data features, dimensionality reduction can be performed to obtain a feature subset. Then, based on the feature subset and the historical operation data, a fault detection model can be trained, and the trained model can be used for device fault detection. Among them, the historical operation data can include pressure data related to the device and tire detection results. By training the model, the deep features of the data and their association with device faults can be learned. Thus, the real-time data of the device can be input into the model to obtain the device fault detection result without interfering with the normal operation of the device. Only the relevant data of the device under normal operation needs to be collected and input into the model, which can achieve automated processing, high fault detection efficiency and strong timeliness, and further reduce the difficulty of device maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present application and used together with the specification to explain the principles of the present application.

[0040] Figure 1 is a schematic flowchart of a device fault detection method provided by an embodiment of the present application;

[0041] Figure 2 is a schematic flowchart of dimensionality reduction processing for data features provided by an embodiment of the present application;

[0042] Figure 3 is a schematic flowchart of training a fault detection model provided by an embodiment of the present application;

[0043] Figure 4a is a schematic diagram of the correct rate score of a fault detection model provided by an embodiment of the present application;

[0044] Figure 4b is a schematic diagram of the correct rate score of a fault detection model provided by another embodiment of the present application;

[0045] Figure 5 is a schematic structural diagram of a device fault detection device provided by an embodiment of the present application;

[0046] Figure 6 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

[0047] Through the above accompanying drawings, the clear embodiments of the present application have been shown, and there will be more detailed descriptions hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0049] The terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, product or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, product or device. In the absence of further restrictions, it is not excluded that there are other identical or equivalent elements in the process, method, product or device including the elements. For example, if the words first, second, etc. are used to indicate names, they do not indicate any particular order.

[0050] Tire molding is an important part of the tire production process, which directly affects the quality and performance of the tire. Tire molding equipment is one of the key equipment used to manufacture tires. It can assemble various tire components (such as carcass cord layer, tread rubber, sidewall rubber, etc.) into unvulcanized tire embryos according to specific process sequences and requirements. After the tire embryo is molded, it needs to be processed through subsequent process steps such as vulcanization to obtain a finished tire.

[0051] Tire molding equipment can include components such as molding drums, feeding devices and control devices, and the overall structure is complex. At present, the maintenance of tire molding equipment can be carried out by regular inspections or post-maintenance. Regular inspections will interfere with the normal operation of the equipment, and frequent inspections will cause waste of resources; post-maintenance refers to testing the equipment after the equipment fails to find out the cause of the failure. Not only may the tire embryo assembled in the faulty state be scrapped, but it will also affect the upstream and downstream operation processes and delay the production progress. An important purpose of maintaining tire molding equipment is to detect whether the equipment has a fault, so as to avoid the tire molding equipment from operating in a faulty state as much as possible. Fault detection will interfere with the normal operation of the equipment, and it is necessary to conduct a detailed inspection of each component of the equipment, which is difficult and time-consuming.

[0052] The inventor has discovered through research that the operating results of the tire molding equipment under a faulty state will affect the appearance of the tire (such as the smoothness of the tread, etc.), and failure in the appearance inspection may cause the tire to be scrapped. Appearance inspection is usually performed on finished tires, and there are still multiple intermediate links from tire molding to inspection of finished tires.

[0053] The inventor further found that there is a correlation between the pressure exerted by the forming drum in the tire forming equipment and the tire appearance inspection. Based on this, the inventor proposed a technical concept, collecting the historical data of the tire forming equipment, extracting the features in the data and learning the rules in the features through a training model. According to the data of the tire forming equipment in the normal state and the fault state, learn the correlation between the pressure of the forming drum, the tire appearance inspection result and the equipment fault, and obtain a model that can be used for fault detection. Through this model, fault detection can be carried out according to the real-time data of the tire forming equipment without interfering with the normal operation of the equipment. Only the relevant data of the equipment under normal operation needs to be collected and input into the model, which can realize automatic processing, with high fault detection efficiency and strong timeliness. Therefore, this detection method can be used for the maintenance of the tire forming equipment to reduce the maintenance difficulty.

[0054] The above-mentioned application scenarios are only partial examples. Those skilled in the art can expand the application according to specific scenarios, and the embodiments of the present application do not make specific restrictions on this. It should be noted that the above application scenarios are only shown for the convenience of understanding the spirit and principle of the present application, and the embodiments of the present application are not restricted in this regard. On the contrary, the embodiments of the present application can be applied to any applicable scenario.

[0055] The following uses specific embodiments to elaborate in detail on the technical solution of the present application and how the technical solution of the present application solves the above technical problems. These specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The following will describe the embodiments of the present application with reference to the drawings.

[0056] Figure 1 It is a schematic flowchart of a device fault detection method provided by an embodiment of the present application. As Figure 1 shown, the method includes:

[0057] Step S101, obtain the historical operation data of the tire forming equipment.

[0058] Among them, the historical operation data includes the pressure data when the tire forming equipment processes the tire and the tire detection result. The pressure data can be the extrusion force of the forming drum on the tire tread and the tire sidewall. The tire detection result can be the appearance detection result of the finished tire.

[0059] Exemplarily, during the production of the tire embryo by the tire forming equipment, the pressure data corresponding to the forming drum can be collected and uploaded to the cloud platform, and the detection result can also be uploaded to the cloud platform after the appearance and other quality inspections of the finished tire processed from the produced tire embryo.

[0060] Step S102, extract features from the historical operation data to obtain data features.

[0061] Taking a single device as an example, the pressure data corresponding to multiple tires processed by the device within a certain period can be selected, and features such as the mean, variance, and median of these data can be extracted.

[0062] Taking multiple devices as an example, the pressure data corresponding to multiple tires processed by multiple devices at the same moment or within the same period can be selected, and corresponding features can be extracted based on these data. Optionally, in addition to the pressure data, features such as the mean and variance can also be extracted for the tire detection results.

[0063] Step S103: Perform dimensionality reduction processing on the data features according to the correlation of the data features to obtain a reduced feature subset.

[0064] Among them, the data features can include multiple data indicators.

[0065] Specifically, after extracting the data features, a feature subset that is crucial for device fault diagnosis can be screened out through methods such as feature selection methods based on correlation analysis or principal components analysis (PCA), reducing the data dimension, decreasing the algorithm calculation amount, and improving the diagnosis accuracy.

[0066] Exemplarily, the correlation of different data indicators is determined through a preset algorithm, and the data features are subjected to dimensionality reduction processing according to the correlation to obtain a feature subset. Among them, the algorithm includes at least one of principal component analysis method, random forest, and autoencoder.

[0067] Step S104: Train a fault detection model based on the feature subset and historical operation data, and use the trained fault detection model to detect the tire forming equipment.

[0068] Exemplarily, a model structure (such as the discrimination and decision of fault types, etc.) can be constructed according to the feature subset, and the data of the device in the normal state and the data in the fault state are selected from the historical operation data of the tire forming equipment, the training set, validation set, and test set are divided, the constructed model is trained using the training set, and the model is adjusted through the validation set and test set. After the model training and adjustment are completed, the real-time operation data of the device can be input into the model, and the model can determine fault information such as whether the device is faulty and the fault type according to the input data.

[0069] In the above embodiments, data features can be extracted by obtaining the historical operation data of the tire forming equipment. According to the correlation of the data features, dimensionality reduction can be performed to obtain a feature subset. Then, based on the feature subset and the historical operation data, a fault detection model can be trained, and the trained model can be used for fault detection of the equipment. Among them, the historical operation data can include pressure data related to the equipment and tire detection results. By training the model, the deep features of the data and their association with equipment faults can be learned. Thus, the real-time data of the equipment can be input into the model to obtain the fault detection result of the equipment, without interfering with the normal operation of the equipment. Only the relevant data of the equipment under normal operation needs to be collected and input into the model, which can achieve automatic processing, high fault detection efficiency, strong timeliness, and further reduce the difficulty of equipment maintenance.

[0070] In one embodiment, before feature extraction, operations such as data cleaning and data normalization can be performed on the data. For example, when collecting pressure data, data cleaning can be performed on data outside the preset range, abnormal data, and constant value data, and the data can be smoothed to reduce noise interference in the data. Also, for example, the data can be normalized and mapped to a unified numerical interval for subsequent algorithm processing.

[0071] Exemplarily, the original data can be scaled according to the maximum and minimum values of the data to achieve the purpose of data normalization, which can be expressed by the formula as y = (x - x min ) / (x max - x min ). Wherein, x is the original data, y is the normalized data, x min and x max are the minimum and maximum values of the original data respectively.

[0072] Exemplarily, the original data can be normalized according to the standard deviation of the data, which can be expressed by the formula as: y = (x - μ) / σ. Wherein, x is the original data, y is the normalized data, μ is the mean of the original data, and σ is the standard deviation of the original data.

[0073] Exemplarily, normalization processing can also be achieved through decimal scaling normalization, which can be expressed by the formula as: y = x / 10 k . Wherein, x is the original data, y is the normalized data, and k is an integer value that makes the absolute value of all data less than 1.

[0074] In one embodiment, as Figure 2 shown, dimensionality reduction processing is performed on the data features according to the correlation to obtain a feature subset, including:

[0075] Step S201, calculate the covariance matrix of the data features.

[0076] Exemplarily, the pressure data can be standardized first, and the standardization process can be expressed by the formula: . Among them, and can respectively represent the sample mean and sample standard deviation of the tire pressure. The standardized data can be denoted as x qi , x q , X, where X can represent the sequence containing the tire pressure data, q can represent the tire number, q = 1, 2, …, p; i can represent the pressure sequence of each tire number, i = 1, 2, …, n; x qi can represent the i-th value of a pressure sequence corresponding to the tire numbered q.

[0077] The covariance matrix can be calculated by the following formula:

[0078]

[0079] Among them, C represents the covariance matrix, r pp can represent the operation result of x p and x p .

[0080] Step S202, perform eigenvalue decomposition according to the covariance matrix to obtain multiple eigenvalues and eigenvectors corresponding to the eigenvalues.

[0081] Exemplarily, denote the eigenvalue as , and the eigenvector as , then there is , and this formula can be converted into an equation about , and the necessary and sufficient condition for this equation to have a non-zero solution is the determinant . This is a homogeneous linear equation system about . Solve this equation system to obtain , and sort it from largest to smallest. Then substitute into the equation about to solve the eigenvector , and the eigenvector can be unitized.

[0082] Step S203, select the target eigenvalues with a contribution rate greater than or equal to a preset threshold from multiple eigenvalues, and perform a linear transformation on the data features according to the eigenvectors corresponding to the target eigenvalues to obtain a feature subset.

[0083] Exemplarily, the eigenvalues with a cumulative contribution greater than or equal to 95% can be selected as the target eigenvalues, denoted as , and the corresponding unit eigenvectors are ​, then, according to the eigenvector, the original data features can be subjected to a projection transformation to obtain a transformed feature subset.

[0084] For example, denote , let , and using the linear transformation , the principal component Y can be obtained.

[0085] In the above embodiments, by using correlation analysis methods such as the principal component analysis method, the dimension of the data features can be reduced, the features that have a greater impact on fault detection can be screened out, the amount of data to be processed can be simplified, the data processing efficiency can be improved, and the progress of fault detection can be accelerated.

[0086] In one embodiment, as Figure 3 shown, training a fault detection model based on the feature subset and historical operation data includes:

[0087] Step S301, extracting deep features in the feature subset through a deep belief network.

[0088] Among them, the deep belief network (DBN, Deep Belief Network, DBN) is a deep learning model and can be composed of multiple RBMs (Restricted Boltzmann Machine, restricted Boltzmann machine).

[0089] In the embodiments of the present application, the feature subset may include features in multiple dimensions such as pressure mean, standard deviation, maximum value, and minimum value. Through the DBN, these features can be continuously reduced in dimension representation, and feature reconstruction can be performed according to the reduced-dimensional features. For example, originally the feature subset contains features in M dimensions, and the DBN can reduce its dimension representation to features in N dimensions (N is less than M), and can reconstruct the features in M dimensions according to these N-dimensional features. If the difference between the reconstructed features in M dimensions and the features in M dimensions before reconstruction is less than a preset standard, the reduced-dimensional N-dimensional features can be extracted as deep features.

[0090] Step S302, determining a decision rule according to the deep features and constructing an initial model corresponding to the decision rule.

[0091] Among them, the decision rule is used to determine the fault type of the tire forming equipment.

[0092] Specifically, a method combining a deep belief network and a decision tree can be adopted to determine the decision rule. The DBN is used to automatically extract deep features and patterns in the data, and the decision tree discriminates and makes decisions on the fault type based on the features extracted by the DBN. An initial model (i.e., the fault detection model to be trained) is constructed according to the decision rule to improve the accuracy and generalization ability of fault diagnosis.

[0093] Exemplarily, first, pre-train layer by layer. Each layer of RBM in the DBN is independently trained unsupervised. Each layer of RBM learns the representation of the data by comparing the original data and the reconstructed data. After the pre-training is completed, the DBN can be used as a whole for supervised learning to optimize the performance of specific tasks (such as fault type determination, etc.). In this process, the parameters of the DBN can be updated according to the gradient descent or other optimization algorithms.

[0094] Among them, each RBM includes a visible layer and a hidden layer, with a full connection between the layers, but there is no connection between the neurons within the layer. The energy function of the RBM can be defined as:

[0095]

[0096] Among them, v and h are the neuron states of the visible layer and the hidden layer respectively, a i and b j are bias terms, w ij is the connection weight, and the optimization goal is to minimize the energy function E(v, h).

[0097] Step S303, select a data set from the historical operation data, and use the data set as a training sample to train the initial model to obtain a trained fault detection model.

[0098] Optionally, the historical operation data also includes the identification information of the tire. The corresponding relationship between the pressure data and the tire detection result can be determined according to the identification information; the pressure data is labeled according to the tire detection result and the corresponding relationship, and the labeled pressure data is classified into a preset data set.

[0099] Among them, the identification information can be the bar code number of the tire embryo assembled by the forming drum.

[0100] The structure and operation of the tire forming equipment are extremely complex. During the processing and forming of a single tire by the forming drum, multiple pressure values will be involved, such as the pressure values of different parts of the tire embryo and the pressure values when the embryo undergoes different processes, etc. The inventor found that the pressure data involved in the training and use of the fault detection model has the problem of complex feature types. The more complex the features, the more difficult it is to build a suitable model, or the built model is too complex with too many parameters and poor generalization ability.

[0101] In the above embodiments, the Deep Belief Network (DBN) can be used to reconstruct the features in the feature subset. Based on the feature reconstruction, deeper features that can better represent the pressure data can be extracted. Compared with the features in the feature subset, the deeper features extracted by the DBN have a lower dimension and do not lose the intrinsic information of the pressure data. According to the decision rules determined by the low-dimensional deeper features, the structure of the model to be trained can be simplified. As a result, not only can the number of model parameters and the training difficulty be reduced, but also overfitting can be alleviated, and the generalization ability and prediction accuracy of the fault detection model for different pressure data can be improved.

[0102] In some possible implementation manners, the training effect of the fault detection model can also be verified by a cross-validation algorithm, and the model parameters of the fault detection model can be adjusted according to the training effect.

[0103] Exemplarily, for the training of the fault detection model, cross-validation techniques (such as K-fold cross-validation) can be used to monitor the training effect of the model and prevent overfitting. Using K-fold cross-validation can make full use of the data set under limited data, improving the model accuracy and generalization ability. K-fold cross-validation divides the data set into K non-overlapping subsets. Each time, one of the subsets is selected as the test set, and the remaining K - 1 subsets are used as the training set. Then, the error of the model on the test set is calculated, and this process is repeated K times. Finally, the average of the K error values is obtained as the performance metric of the model.

[0104] By setting the value of K, the model accuracy scores for different values of K can be obtained. For example, Figure 4a FIG. shows the accuracy score diagram of the fault detection model when K is 9. Figure 4b FIG. shows the accuracy score diagram of the fault detection model when K is 14.

[0105] In the above embodiments, dividing the pressure data for training into K different subsets and successively selecting one subset as the training set can reduce the evaluation bias caused by different data set divisions, improve the utilization rate of the pressure data, and maximize the generalization ability of the model.

[0106] In one embodiment, using the trained fault detection model to detect the tire forming equipment further includes:

[0107] Deploying the trained fault detection model to the monitoring system of the tire forming equipment; preprocessing the operation data collected by the monitoring system and then inputting it into the fault detection model to obtain the detection result output by the fault detection model.

[0108] Among them, the monitoring system is used to collect the operation data of the tire forming equipment.

[0109] Exemplarily, after the fault detection model is trained, the model can be deployed to the online monitoring system of the tire forming equipment to collect the operation data of the equipment in real time. The collected data can be normalized and feature-extracted in the same way as in the model training process and then input into the model. The model can perform fault detection based on the input data and output the results.

[0110] In some possible implementation manners, corresponding warning information can also be generated according to the fault detection result of the model for warning. For example, when the model detects that the equipment may have a fault, the detection result can be transmitted to the monitoring system, and the monitoring system can generate corresponding warning information according to the severity and occurrence probability of the fault. The warning information can be transmitted to the equipment maintenance personnel in various ways, such as on-site equipment sound and light alarm, key person SMS notification, and display of a detailed fault prediction report on the monitoring system interface, so that the maintenance personnel can take preventive maintenance measures in time.

[0111] Figure 5 The following is a schematic structural diagram of a device fault detection device provided by an embodiment of the present application. As Figure 5 shown, the device fault detection device 500 may include:

[0112] An acquisition module, configured to acquire historical operation data of the tire forming equipment, where the historical operation data includes pressure data and tire detection results 501 when the tire forming equipment processes a tire.

[0113] A feature extraction module 502, configured to extract features from the historical operation data to obtain data features.

[0114] A feature processing module 503, configured to perform dimensionality reduction processing on the data features according to the correlation of the data features to obtain a reduced-dimensional feature subset.

[0115] A detection module 504, configured to train a fault detection model based on the feature subset and the historical operation data, and detect the tire forming equipment by using the trained fault detection model.

[0116] In one embodiment, the data features include multiple data indicators, and the feature processing module 503 is further configured to: determine the correlation of different data indicators through a preset algorithm, and perform dimensionality reduction processing on the data features according to the correlation to obtain a feature subset, where the algorithm includes at least one of a principal component analysis method, a random forest, and an autoencoder.

[0117] In one embodiment, the feature processing module 503 is further configured to: calculate the covariance matrix of the data features; perform eigenvalue decomposition on the covariance matrix to obtain a plurality of eigenvalues and the eigenvectors corresponding to the eigenvalues; select target eigenvalues with a contribution rate greater than or equal to a preset threshold from the plurality of eigenvalues, and perform a linear transformation on the data features according to the eigenvectors corresponding to the target eigenvalues to obtain a feature subset.

[0118] In one embodiment, the detection module 504 is further configured to: extract deep features from the feature subset through a deep belief network; determine a decision rule according to the deep features, and construct an initial model corresponding to the decision rule, where the decision rule is used to determine the fault type of the tire forming device;

[0119] Select a data set from the historical operation data, and use the data set as a training sample to train the initial model to obtain a trained fault detection model.

[0120] In one embodiment, the detection module 504 is further configured to: determine the correspondence between the pressure data and the tire detection result according to the identification information; label the pressure data according to the tire detection result and the correspondence, and classify the labeled pressure data into a preset data set.

[0121] In one embodiment, the detection module 504 is further configured to: verify the training effect of the fault detection model through a cross-validation algorithm, and adjust the model parameters of the fault detection model according to the training effect.

[0122] In one embodiment, the detection module 504 is further configured to: deploy the trained fault detection model to the monitoring system of the tire forming device, where the monitoring system is used to collect the operation data of the tire forming device; preprocess the operation data collected by the monitoring system and then input it into the fault detection model to obtain the detection result output by the fault detection model.

[0123] The device fault detection device provided in this embodiment can be used to execute the technical solutions in any of the foregoing method embodiments, and its implementation principle and technical effects are similar, and will not be described in detail here.

[0124] It should be understood that the above device embodiments are illustrative, and the devices of the present application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0125] In addition, unless otherwise specified, each functional unit / module in the embodiments of the present application may be integrated into one unit / module, may exist physically as individual units / modules, or may be integrated together with two or more units / modules. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of a software program module.

[0126] Figure 6 FIG. is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As Figure 6 shown, the electronic device 60 may include:

[0127] a processor 61, a memory 62, and a communication interface 63;

[0128] The memory 62 is used to store executable instructions of the processor 61; the executable instructions may be computer execution instructions;

[0129] Wherein, the processor 61 is configured to execute the technical solutions in any of the foregoing method embodiments by executing the executable instructions.

[0130] Optionally, the memory 62 may be either independent or integrated with the processor 61.

[0131] Optionally, when the memory 62 is a device independent of the processor 61, the electronic device 60 may further include:

[0132] a bus 64, and the memory 62 and the communication interface 63 are connected to the processor 61 through the bus 64 to complete mutual communication, and the communication interface 63 is used to communicate with other devices.

[0133] Optionally, the communication interface 63 may be specifically implemented by a transceiver. The communication interface is used to implement communication between the database access device and other devices (such as clients, read-write libraries, and read-only libraries). The memory may include a random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.

[0134] The bus 64 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, only one line is shown in the figure, but it does not mean that there is only one bus or one type of bus.

[0135] The foregoing processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0136] The electronic device is used to execute the technical solutions in any of the foregoing method embodiments, and its implementation principle and technical effects are similar, which will not be elaborated here.

[0137] This application embodiment also provides a readable storage medium, which may be a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the technical solutions provided in any of the foregoing method embodiments.

[0138] This application embodiment also provides a computer program product, including a computer program, which is used to implement the technical solutions provided in any of the foregoing method embodiments when executed by a processor.

[0139] In the foregoing embodiments, the descriptions of the respective embodiments have their own focuses. For parts not elaborated in a certain embodiment, reference may be made to the relevant descriptions of other embodiments. The technical features of the foregoing embodiments may be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the foregoing embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0140] In addition, in each embodiment of this application, each functional unit may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.

[0141] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this application. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program codes.

[0142] Those of ordinary skill in the art can understand that all or part of the steps to implement the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps including those of the above method embodiments; and the foregoing storage medium includes various media that can store program codes, such as ROM, RAM, magnetic disks, or optical discs.

[0143] Finally, it should be noted that: After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily think of other implementation manners of the present invention. The present invention is intended to cover any variations, uses, or adaptations of the present invention, and these variations, uses, or adaptations follow the general principles of the present invention and include common general knowledge or conventional technical means in the technical field not disclosed in the present invention. It is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A device fault detection method, characterized in that: include: Acquire historical operation data of the tire molding equipment, wherein the historical operation data includes pressure data when the tire molding equipment processes the tire and tire detection results; Extracting features from the historical operation data to obtain data features; Performing dimensionality reduction processing on the data features according to the correlation of the data features to obtain a feature subset after dimensionality reduction; A fault detection model is trained based on the feature subset and the historical operation data, and the tire forming equipment is detected using the trained fault detection model.

2. The device fault detection method according to claim 1, characterized in that: The data features include multiple data indicators, and the data features are subjected to dimensionality reduction processing according to the correlation of the data features to obtain a feature subset after dimensionality reduction, including: The correlation between different data indicators is determined by a preset algorithm, and the data features are subjected to dimensionality reduction processing according to the correlation to obtain a feature subset, wherein the algorithm includes at least one of principal component analysis, random forest and autoencoder.

3. The device fault detection method according to claim 2, characterized in that: The performing dimensionality reduction processing on the data features according to the correlation to obtain a feature subset includes: Calculating a covariance matrix of the data features; Performing eigenvalue decomposition according to the covariance matrix to obtain a plurality of eigenvalues ​​and eigenvectors corresponding to the eigenvalues; A target eigenvalue whose contribution rate is greater than or equal to a preset threshold is selected from the multiple eigenvalues, and a linear transformation is performed on the data feature according to a eigenvector corresponding to the target eigenvalue to obtain a feature subset.

4. The device failure detection method according to any one of claims 1 to 3, characterized in that: The training of the fault detection model based on the feature subset and the historical operation data comprises: Extracting deep features from the feature subset through a deep belief network; Determining a decision rule according to the deep features and constructing an initial model corresponding to the decision rule, wherein the decision rule is used to determine the fault type of the tire forming equipment; A data set is selected from the historical operation data as a training sample to train the initial model to obtain a trained fault detection model.

5. The device fault detection method according to claim 4, characterized in that: The historical operation data also includes identification information of the tire, and the data set is obtained in the following manner: Determine the corresponding relationship between the pressure data and the tire detection result according to the identification information; The pressure data is labeled according to the tire detection result and the corresponding relationship, and the labeled pressure data is classified into a preset data set.

6. The device failure detection method according to any one of claims 1 to 3, characterized in that: The training of the fault detection model based on the feature subset and the historical operation data further includes: The training effect of the fault detection model is verified by a cross-validation algorithm, and the model parameters of the fault detection model are adjusted according to the training effect.

7. The device fault detection method according to any one of claims 1 to 3, characterized in that: The method of using the trained fault detection model to detect the tire building equipment includes: Deploy the trained fault detection model to a monitoring system of the tire molding equipment, wherein the monitoring system is used to collect operating data of the tire molding equipment; The operating data collected by the monitoring system is pre-processed and then input into the fault detection model to obtain the detection result output by the fault detection model.

8. An equipment fault detection device, characterized in that: include: An acquisition module, used for acquiring historical operation data of the tire forming equipment, wherein the historical operation data includes pressure data when the tire forming equipment processes the tire and tire detection results; A feature extraction module, used to extract features from the historical operation data to obtain data features; A feature processing module, used for performing dimension reduction processing on the data features according to the correlation of the data features to obtain a feature subset after dimension reduction; A detection module is used to train a fault detection model based on the feature subset and the historical operation data, and use the trained fault detection model to detect the tire forming equipment.

9. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

11. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 7 when being executed by a processor.

Images