A device fault diagnosis method, device, apparatus and storage medium

Abstract

The application discloses a kind of equipment fault diagnosis method, device, equipment and storage medium, the method comprises: the dial image of detection component is collected, and detection data in dial image is extracted;According to detection data, obtain the prediction data of detection data by prediction;Detection data and prediction data are respectively input into trained fault diagnosis model, respectively obtain preliminary diagnosis result and fault prediction result, compare and analyze preliminary diagnosis result and fault prediction result, determine the final equipment fault diagnosis result.The equipment fault diagnosis method disclosed in the application improves the data entry reliability and efficiency of live detection data by automatically extracting detection data, reduces labor cost, and by including preliminary diagnosis result and fault prediction result in the consideration of fault diagnosis simultaneously, improves the accuracy of fault diagnosis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power detection, and particularly relates to a device fault diagnosis method and device, equipment and a storage medium. BACKGROUND

[0002] The acquisition and utilization of power live detection data is an important link in the decision-making process in power operation and maintenance, and the standardized, reliable acquisition and effective analysis and utilization of power live detection data are of great significance to perfecting the intelligent operation and maintenance system of power equipment and power equipment state evaluation, and the realization of data portable acquisition and analysis interworking with the management and control side is the main research content of establishing a new mode of PHM intelligent operation and maintenance platform.

[0003] In the current power operation and maintenance work, the form of manually entering device live detection data and then performing fault diagnosis is often adopted, which has certain shortcomings.

[0004] 1. The data recorded by field workers is dependent on the professional quality and level of field workers in terms of completeness and correctness;

[0005] 2. Since various types of live detection instruments are applied to various types of equipment, the automaticity of live detection instruments for various types of equipment is different, and the related standards are not unified, and old equipment can only be recorded by paper, which causes non-standard data recording and low data availability;

[0006] 3. The measurement results recorded by workers need high operation and maintenance cost from paper records to archiving statistical analysis, the work efficiency is low, the data availability is not high, and there is a lack of timely and effective diagnosis and evaluation analysis, and the data application timeliness is insufficient. SUMMARY

[0007] The present application provides a device fault diagnosis method, device, equipment and storage medium to realize automatic entry of device detection data and fault diagnosis.

[0008] According to an aspect of the present application, a device fault diagnosis method is provided, comprising:

[0009] Collecting a dial image of a detection component, and extracting detection data in the dial image;

[0010] According to the detection data, prediction data of the detection data is obtained;

[0011] The detection data and the prediction data are respectively input into a trained fault diagnosis model to respectively obtain a preliminary diagnosis result and a fault prediction result, and the preliminary diagnosis result and the fault prediction result are compared and analyzed to determine a final device fault diagnosis result.

[0012] Further, the detection data in the dial image is extracted, including:

[0013] The dial image is subjected to image recognition to extract initial data values;

[0014] The initial data values are subjected to inspection to determine inspection qualified data, and the inspection qualified data is determined as the detection data.

[0015] Further, the initial data values are subjected to inspection to determine inspection qualified data, including:

[0016] According to a preset data threshold, data in the initial data values that meets the preset data threshold is determined as first qualified data, and data that does not meet the preset data threshold is determined as to-be-retested data;

[0017] The dial image corresponding to the to-be-retested data is subjected to image recognition again, and if the obtained data meets the preset data threshold, the data is determined as second qualified data;

[0018] The first qualified data and the second qualified data are determined as the inspection qualified data.

[0019] Further, prediction is performed according to the detection data to obtain prediction data of the detection data, including:

[0020] The detection data is subjected to oversampling processing to obtain prediction input data;

[0021] The prediction input data is input into a trained prediction model, and an output result of the prediction model is taken as prediction data of the detection data.

[0022] Further, the preliminary diagnosis result and the fault prediction result are compared and analyzed to determine a final equipment fault diagnosis result, including:

[0023] The preliminary diagnosis result and the fault prediction result are matched in a preset case library;

[0024] The equipment fault diagnosis result is determined according to a matching result.

[0025] Further, after the final equipment fault diagnosis result is determined, further including:

[0026] A corresponding fault handling decision is determined according to the equipment fault diagnosis result.

[0027] Further, the fault diagnosis model includes a support vector machine model, and a training method of the fault diagnosis model includes:

[0028] The model parameters of the support vector machine model are determined by using the dung beetle algorithm to obtain an initial diagnosis model;

[0029] According to historical detection data, a training data set is extracted, the training data set is input into the initial diagnosis model, and parameter adjustment is performed on the initial diagnosis model according to model output until the output of the initial diagnosis model reaches a set accuracy threshold.

[0030] According to another aspect of the present application, a device fault diagnosis apparatus is provided, comprising:

[0031] A detection data extraction module is configured to collect a dial image of a detection component and extract detection data from the dial image;

[0032] A data prediction module is configured to perform prediction according to the detection data to obtain predicted data of the detection data;

[0033] A fault diagnosis module is configured to input the detection data and the predicted data into a trained fault diagnosis model to obtain preliminary diagnosis results and fault prediction results, respectively, and to compare and analyze the preliminary diagnosis results and the fault prediction results to determine a final device fault diagnosis result.

[0034] Optionally, the detection data extraction module is further configured to:

[0035] Perform image recognition on the dial image to extract initial data values;

[0036] Verify the initial data values to determine qualified verification data, and determine the qualified verification data as the detection data.

[0037] Optionally, the detection data extraction module is further configured to:

[0038] According to a preset data threshold, data in the initial data values that meet the preset data threshold are determined as first qualified data, and data that do not meet the preset data threshold are determined as to-be-retested data;

[0039] The dial image corresponding to the to-be-retested data is re-recognized, and if the obtained data meets the preset data threshold, the data is determined as second qualified data;

[0040] The first qualified data and the second qualified data are determined as the qualified verification data.

[0041] Optionally, the data prediction module is further configured to:

[0042] Perform oversampling processing on the detection data to obtain prediction input data;

[0043] inputting the predicted input data into the trained prediction model, and taking an output result of the prediction model as predicted data of the detection data.

[0044] Optionally, the fault diagnosis module is further configured to:

[0045] match the preliminary diagnosis result and the fault prediction result in a preset case library.

[0046] determine the equipment fault diagnosis result according to a matching result.

[0047] Optionally, the apparatus further comprises a fault handling decision determination module configured to determine a corresponding fault handling decision according to the equipment fault diagnosis result.

[0048] Optionally, the fault diagnosis model comprises a support vector machine model, and the apparatus further comprises a model training module configured to:

[0049] determine model parameters of the support vector machine model by using the beetle algorithm to obtain an initial diagnosis model;

[0050] extract a training data set from historical detection data, input the training data set into the initial diagnosis model, and adjust parameters of the initial diagnosis model according to a model output until an output of the initial diagnosis model reaches a set accuracy threshold.

[0051] According to another aspect of the present application, an electronic device is provided, which comprises:

[0052] at least one processor; and

[0053] a memory connected in communication with the at least one processor; wherein

[0054] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the equipment fault diagnosis method according to any one of the embodiments of the present application.

[0055] According to another aspect of the present application, a computer readable storage medium is provided, which stores computer instructions for enabling a processor to perform the equipment fault diagnosis method according to any one of the embodiments of the present application when executed by the processor.

[0056] The application discloses a device fault diagnosis method, first, the dial image of a detection component is collected, detection data in the dial image is extracted, then prediction data of the detection data is obtained according to the detection data, finally, the detection data and the prediction data are respectively input into a trained fault diagnosis model, preliminary diagnosis results and fault prediction results are respectively obtained, the preliminary diagnosis results and the fault prediction results are compared and analyzed, and final device fault diagnosis results are determined.

[0057] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the application, nor is it used to limit the scope of the application. Other features of the application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0059] Figure 1 It is a flow chart of a device fault diagnosis method provided by the first embodiment of the application;

[0060] Figure 2 It is a flow chart of a device fault diagnosis method provided by the second embodiment of the application;

[0061] Figure 3 It is a structural schematic diagram of a device fault diagnosis device provided by the third embodiment of the application;

[0062] Figure 4 It is a structural schematic diagram of an electronic device for realizing the device fault diagnosis method of the fourth embodiment of the application. DETAILED DESCRIPTION

[0063] In order to make the person in the art better understand the application scheme, the technical solutions in the embodiments of the application will be described clearly and completely by combining the drawings in the embodiments of the application. Obviously, the described embodiments are only some embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor should belong to the scope of protection of the application.

[0064] It should be noted that the terms "first", "second", and the like in the description and in the claims of the present application and above-mentioned drawings are intended to distinguish similar objects and not necessarily describe a particular sequential or chronological order. It should be understood that the data thus used can be interchanged under appropriate circumstances so that the embodiments of the application described herein can be implemented in orders other than those illustrated or described herein. In addition, the terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0065] Embodiment one

[0066] Figure 1 A flow chart of a device fault diagnosis method provided by the first embodiment of the present application, the embodiment can be applicable to the detection and fault diagnosis of power equipment. The method can be applied to a device fault diagnosis apparatus, which can be realized in the form of hardware and / or software, and can be configured in an electronic device. As shown in the figure, the method comprises: Figure 1

[0067] S110, collecting the dial image of the detection component and extracting the detection data in the dial image.

[0068] The detection component can be an instrument or device for detecting various state data during the operation of the power equipment, such as a sensor, etc. The detection data can be the data displayed on the dial after the detection component performs the corresponding detection task.

[0069] In this embodiment, the detection data in the dial image can be extracted by using an image acquisition device, such as a camera, to take a picture of the dial of the detection component, and then converting the image into corresponding digital data through image recognition. By automatically collecting the dial image of the detection component, the dependence on manual data recording on site can be reduced in the case of live detection.

[0070] Preferably, the image acquisition device can be a smart terminal with a photographing function. After completing the shooting in response to the image acquisition instruction, the image recognition can be automatically performed, and the recognized detection data can be uploaded to the fault diagnosis platform.

[0071] ​Specifically, for the image collection intelligent terminal, the live detection task from the fault diagnosis platform can be received, and the dial plate image of the detection component can be collected in response to the operation of the operator of the intelligent terminal on the terminal. Then, the collected image is recognized and converted into corresponding numbers, i.e., detection data, which is filled into the preset blank detection data column. Preferably, the intelligent terminal can be a small portable device, which can be built-in with a live detection task report related to the live detection task for guiding the operation of the operator.

[0072] S120, predicting according to the detection data to obtain prediction data of the detection data.

[0073] The detection data is the measured data obtained from the detection component, and the prediction data is the predicted value in a period of time obtained by predicting the current detection data.

[0074] In this embodiment, the prediction method of the detection data can be that the obtained detection data is used as the model input by using the trained prediction model, and the output result of the prediction model, i.e., the prediction data of the detection data, is obtained.

[0075] Optionally, when training the prediction model, the historical measured data can be selected as the data set for model training, and the parameters of the prediction model are adjusted according to the difference between the predicted value and the true value output by the model until the accuracy of the predicted value reaches the set threshold. After the model is trained, the detection data obtained in the previous step is input into the trained prediction model to obtain the prediction data.

[0076] S130, the detection data and the prediction data are input into the trained fault diagnosis model respectively to obtain preliminary diagnosis results and fault prediction results respectively, and the preliminary diagnosis results and the fault prediction results are compared and analyzed to determine the final equipment fault diagnosis result.

[0077] In this embodiment, after obtaining the detection data and the prediction data, they can be input into the trained fault diagnosis model respectively, and correspondingly, the outputs of the fault diagnosis model are the preliminary diagnosis results and the fault prediction results respectively.

[0078] Optionally, the fault diagnosis model includes a support vector machine model, and the training method of the fault diagnosis model can be: using the scarab beetle algorithm to determine the model parameters of the support vector machine model to obtain an initial diagnosis model; extracting a training data set according to the historical detection data, inputting the training data set into the initial diagnosis model, and adjusting the parameters of the initial diagnosis model according to the model output until the output of the initial diagnosis model reaches the set accuracy threshold.

[0079] The support vector machine (SVM) is a kind of generalized linear classifier for binary classification of data in a supervised learning manner. In the SVM model, the kernel parameter and the penalty factor are key factors for determining the performance of the SVM model. In order to improve the performance of the SVM model, the Dung Beetle Optimizer (DBO) can be used to optimize the parameters of the SVM model to determine the optimal kernel parameter and penalty factor, establish an initial diagnosis model, and then train the initial diagnosis model according to the model training method using historical detection data to obtain a trained fault diagnosis model.

[0080] Specifically, in the dung beetle algorithm, it is assumed that there are four roles in the dung beetle population: rolling dung beetle, foraging dung beetle, ovipositing dung beetle and thief dung beetle. Among them, the rolling dung beetle navigates through celestial clues. It is assumed that the light intensity will affect its path during the rolling process. Therefore, the position update of the dung beetle during the rolling process is as follows:

[0081] x i (t+1)=x i (t)+a×k×x i (t-1)+b×Δx,

[0082] Δx=|x i (t)-X w |,

[0083] Where t represents the current iteration number, x i (t) represents the position information of the i-th dung beetle at the t-th iteration, k is the deflection coefficient, a is the natural coefficient, b is the constant, and Δx simulates the change of light intensity. X w represents the global worst position. Under the premise of ensuring search performance, the light intensity factor Δx is introduced to control the search range, which prevents falling into a local optimal solution. The natural coefficient a is introduced to ensure that the dung beetle can determine the position again by dancing when it encounters an obstacle and cannot move forward, and a new route is obtained.

[0084] In order to simulate the dancing behavior of the dung beetle, a tangent function is used to obtain a new rolling direction. Once the dung beetle successfully determines a new direction, it should continue to roll the ball backward. The position update of the dung beetle is defined as follows:

[0085] x i (t+1)=x i (t)+tan(θ)|x i (t)-x i (t-1)|,

[0086] Where θ is the deflection angle, and θ∈[0,π].

[0087] The foraging process of the Atta sexdens in nature, the boundaries of the optimal foraging region are as follows:

[0088] Lb b = max(X b (1-R), Lb),

[0089] Ub b = min(X b (1+R), Ub),

[0090] where Lb b and Ub b denote the upper and lower bounds of the optimal foraging region, X b denotes the global optimal position, Lb and Ub denote the upper and lower bounds of the optimization problem, R = 1-t / T max , T max denotes the maximum number of iterations.

[0091] The position update of the Atta sexdens is as follows:

[0092] B i (t+1) = X * +b1×(B i (t)-Lb * )+b2×(B i (t)-Ub * ),

[0093] where B i (t) is the position of the i-th egg ball at the t-th iteration, b1 and b2 are two independent 1×D random vectors, D is the dimension of the optimization problem, X * denotes the current local optimal position, Lb * and Ub * denote the upper and lower bounds of the foraging region, and are as follows:

[0094] Lb * = max(X * (1-R), Lb),

[0095] Ub * = min(X * (1+R), Ub).

[0096] On the other hand, some of the roles of the Atta sexdens are defined as thieves, who will steal the fecal balls from other Atta sexdens, and the positions of the thieves are also updated during the iteration process, which can be described as:

[0097] C i (t+1) = X b +S×t×(|C i (t)-X* |+|C i (t)-X b |),

[0098] where C i (t) is the position of the i-th egg at the t-th iteration, S is a constant, and g is a random vector of size 1 x D that follows a normal distribution.

[0099] Each termite population is composed of the above four different roles, the sum of the number of the four kinds of termites should be equal to the set population number, the position of the four kinds of termites is updated constantly in the iteration process, the current optimal solution and its fitness value are updated by judging whether their positions exceed the boundary value, and the process is repeated constantly, and finally the global optimal solution and its fitness value are output. The global optimal region is determined by the constant updating and iteration of the positions of the four kinds of termites in the algorithm, and finally the kernel parameter and the penalty factor of the SVM model are determined by using the optimal parameter combination obtained by the termite optimization, so as to create a DBO-SVM diagnosis model, that is, an initial diagnosis model. Further, after obtaining the initial diagnosis model, the training data set can be extracted from the historical detection data, the training data set is input into the initial diagnosis model, and the parameters of the initial diagnosis model are adjusted according to the model output until the output of the initial diagnosis model reaches the set accuracy threshold.

[0100] Further, the method for comparing and analyzing the preliminary diagnosis result and the fault prediction result can be that, in the fault diagnosis platform, the preliminary diagnosis result and the fault prediction result are matched with the pre-stored decision library and case library in the platform to determine the matched cases, so as to obtain the final equipment fault diagnosis result.

[0101] For example, if the preliminary diagnosis result is that the current power equipment has failed, appropriate measures should be taken to prevent the accident from evolving; if the preliminary diagnosis result is normal but the fault prediction result indicates that the transformer may fail in a short period of time in the future, appropriate operation and maintenance measures should also be taken at this time. The former may be designed to be powered off, and the latter may take measures according to the suggestions given by the high-level application layer case matching, and the measures may be relatively conservative.

[0102] The application discloses a device fault diagnosis method, which comprises the following steps: firstly, collecting a dial image of a detection component, extracting detection data in the dial image, then making a prediction according to the detection data to obtain predicted data of the detection data, finally inputting the detection data and the predicted data into a trained fault diagnosis model to obtain a preliminary diagnosis result and a fault prediction result respectively, comparing and analyzing the preliminary diagnosis result and the fault prediction result to determine a final device fault diagnosis result. The device fault diagnosis method disclosed by the application can automatically extract detection data, improve the data entry reliability and efficiency of live detection data, reduce labor cost, and improve the accuracy of fault diagnosis by taking the preliminary diagnosis result and the fault prediction result into consideration simultaneously.

[0103] Embodiment two

[0104] Figure 2 A flowchart of a device fault diagnosis method provided by the application for embodiment two, which can be applied to a device fault diagnosis device, and is a refinement of the above-mentioned embodiment. As shown in the figure, the method comprises the following steps: Figure 2

[0105] S210, collecting a dial image of a detection component, performing image recognition on the dial image, and extracting an initial data value.

[0106] In this embodiment, a smart terminal or other device with a photographing function can be used to take a picture of the dial of the detection component, and then the image can be converted into corresponding numbers through image recognition, and the detection data can be extracted from the dial image.

[0107] Optionally, the smart terminal can be an industrial tablet computer, which can take pictures of the live detection state quantity data required by each power device through a camera. For general live detection sensors, the data is displayed through a dial. Preferably, after the camera collects the dial image of the data measured by the sensor, the image containing digital information collected can be converted into corresponding digital form through light symbol recognition of an SDK card, and the number can be filled into the corresponding detection data column in the detection report as an initial data value.

[0108] S220, verifying the initial data value, determining the qualified data in the initial data value, and determining the qualified data as detection data.

[0109] In this embodiment, the identified initial data value can have a deviation from the actual value. In order to eliminate the influence of the deviation on the subsequent fault diagnosis, the initial data value can be verified, and the verified initial data value can be used as detection data for fault diagnosis.

[0110] ​Optionally, the way of testing the initial data value to determine the test qualified data can be: according to the preset data threshold, the data in the initial data value that meets the preset data threshold is determined as the first qualified data, and the data that does not meet the preset data threshold is determined as the to-be-retested data; the dial image corresponding to the to-be-retested data is re-identified, and if the obtained data meets the preset data threshold, it is determined as the second qualified data; the first qualified data and the second qualified data are determined as the test qualified data.

[0111] Specifically, for each item of data detected by the detection component, such as temperature, current, voltage, etc., there is a corresponding reasonable range. According to the reasonable range of each item of data, a corresponding data threshold can be preset in the dial image acquisition device. If the initial data value identified meets the corresponding data threshold, it is qualified data, otherwise it needs to be retested, that is, image recognition is performed again and it is judged whether the identification result meets the corresponding data threshold. If the retest result meets, it is determined as qualified data, otherwise it is determined as unqualified data and discarded.

[0112] S230, oversampling the detection data to obtain prediction input data.

[0113] Among them, oversampling refers to the process of sampling far higher than twice the signal bandwidth or the highest frequency thereof. Generally, it refers to a sampling frequency higher than twice the highest frequency of the signal.

[0114] In this embodiment, after obtaining the detection data, a part of the data can be directly used for fault diagnosis, and another part of the data can be first subjected to oversampling processing, and then prediction is performed according to the oversampling processing result, and the prediction data is used for fault diagnosis, so as to obtain more comprehensive fault diagnosis result.

[0115] Preferably, when the detection data is subjected to oversampling processing, SMOTE algorithm, i.e. synthetic minority over-sampling technique, can be used. It is an improved scheme based on random oversampling algorithm. Since random oversampling takes the strategy of simply copying samples to increase minority class samples, it is easy to produce the problem of model overfitting, that is, the information learned by the model is too special and not general enough. The basic idea of SMOTE algorithm is to analyze the minority class samples and artificially synthesize new samples to add to the data set, which is in line with the characteristics of small amount of on-site sampling data. The data after oversampling processing can be used as the prediction input data of the data prediction model. Oversampling processing can increase the sample capacity, which is beneficial to improve the performance of the prediction model.

[0116] S240, inputting the prediction input data into the trained prediction model, and taking the output result of the prediction model as the prediction data of the detection data.

[0117] In this embodiment, after obtaining the prediction input data, the prediction input data can be input into the trained prediction model, and the output result of the prediction model is the prediction data.

[0118] Preferably, the prediction model can be a BP neural network model. The BP (back propagation) neural network is a kind of multi-layer feedforward neural network trained according to the error back propagation algorithm, and is one of the most widely used neural network models.

[0119] Further, the BP neural network model can be optimized by using a sparrow optimization algorithm. The sparrow search algorithm (SSA) is a new type of swarm intelligence optimization algorithm, which is mainly inspired by the foraging behavior and anti-predation behavior of sparrows. After the BP neural network is optimized by the SSA algorithm, an SSA-BP prediction model is established and trained, and then the prediction input data obtained in the above steps is input into the trained SSA-BP prediction model, so that the prediction value of the current data set in a relatively small period of time, i.e., the prediction data of the detection data, is obtained.

[0120] S250, input the detection data and the prediction data into the trained fault diagnosis model respectively to obtain a preliminary diagnosis result and a fault prediction result respectively, match the preliminary diagnosis result and the fault prediction result in a preset case library, and determine the equipment fault diagnosis result according to a matching result.

[0121] In this embodiment, after obtaining the detection data and the prediction data, they can be input into the trained fault diagnosis model respectively, and the outputs of the fault diagnosis model are the preliminary diagnosis result and the fault prediction result respectively. In the fault diagnosis platform, the preliminary diagnosis result and the fault prediction result can be matched with the decision library and the case library pre-stored in the platform to determine the corresponding case, so that the final equipment fault diagnosis result is obtained.

[0122] Further, after determining the final equipment fault diagnosis result, the corresponding fault handling decision can be determined according to the equipment fault diagnosis result.

[0123] Optionally, the corresponding fault handling decision can be determined according to the final equipment fault diagnosis result, for example, the power equipment that has occurred a fault is powered off, or the running power of the power equipment is reduced, etc.

[0124] The application discloses a device fault diagnosis method, first, a dial image of a detection component is collected, image recognition is performed on the dial image, initial data values are extracted, then the initial data values are tested, qualified test data is determined, the qualified test data is determined as detection data, oversampling processing is performed on the detection data, prediction input data is obtained, the prediction input data is input into a trained prediction model, output results of the prediction model are taken as prediction data of the detection data, finally, the detection data and the prediction data are input into a trained fault diagnosis model, preliminary diagnosis results and fault prediction results are obtained, the preliminary diagnosis results and the fault prediction results are matched in a preset case library, and device fault diagnosis results are determined according to a matching result. The device fault diagnosis method disclosed by the application automatically extracts detection data, improves data entry reliability and efficiency of live detection data, reduces labor cost, and simultaneously considers the preliminary diagnosis results and the fault prediction results in fault diagnosis, thereby improving fault diagnosis accuracy.

[0125] Embodiment three

[0126] Figure 3 A structural schematic diagram of a device fault diagnosis device provided for the third embodiment of the application is shown in the figure. Figure 3 As shown in the figure, the device comprises a detection data extraction module 310, a data prediction module 320 and a fault diagnosis module 330.

[0127] The detection data extraction module 310 is used for collecting a dial image of a detection component and extracting detection data in the dial image.

[0128] The data prediction module 320 is used for predicting according to the detection data and obtaining prediction data of the detection data.

[0129] The fault diagnosis module 330 is used for inputting the detection data and the prediction data into a trained fault diagnosis model respectively, obtaining preliminary diagnosis results and fault prediction results respectively, comparing and analyzing the preliminary diagnosis results and the fault prediction results, and determining final device fault diagnosis results.

[0130] Optionally, the detection data extraction module 310 is further used for:

[0131] performing image recognition on the dial image, extracting initial data values, testing the initial data values, determining qualified test data, and determining the qualified test data as the detection data.

[0132] Optionally, the detection data extraction module 310 is further used for:

[0133] According to the preset data threshold, data in the initial data value meeting the preset data threshold is determined as first qualified data, and data not meeting the preset data threshold is determined as to-be-retested data; the dial image corresponding to the to-be-retested data is re-identified, and if the obtained data meets the preset data threshold, the data is determined as second qualified data; and the first qualified data and the second qualified data are determined as inspection qualified data.

[0134] Optionally, the data prediction module 320 is further configured to:

[0135] The detection data is oversampled to obtain prediction input data; and the prediction input data is input into the trained prediction model, and an output result of the prediction model is taken as prediction data of the detection data.

[0136] Optionally, the fault diagnosis module 330 is further configured to:

[0137] The preliminary diagnosis result and the fault prediction result are matched in a preset case library; and a device fault diagnosis result is determined according to a matching result.

[0138] Optionally, the apparatus further comprises a fault handling decision determination module 340 configured to determine a corresponding fault handling decision according to the device fault diagnosis result.

[0139] Optionally, the fault diagnosis model comprises a support vector machine model, and the apparatus further comprises a model training module 350 configured to:

[0140] Model parameters of the support vector machine model are determined by using the beetle algorithm to obtain an initial diagnosis model; a training data set is extracted according to historical detection data, the training data set is input into the initial diagnosis model, and the initial diagnosis model is adjusted in parameters according to a model output until an output of the initial diagnosis model reaches a set accuracy threshold.

[0141] The device fault diagnosis apparatus provided in the embodiments of the present application can execute the device fault diagnosis method provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0142] Embodiment Four

[0143] Figure 4A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0144] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0145] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0146] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as device fault diagnosis methods.

[0147] In some embodiments, the device fault diagnosis method can be implemented as a computer program tangibly embodied in a computer readable storage medium, e.g., storage unit 18. In some embodiments, parts or all of the computer program can be loaded and / or installed onto electronic device 10 via, e.g., ROM 12 and / or communication unit 19. When the computer program is loaded onto RAM 13 and executed by processor 11, one or more of the steps of the device fault diagnosis described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the device fault diagnosis method by other means, e.g., with the aid of firmware.

[0148] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0149] Computer programs used to implement the methods of the present application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program, when executed by the processor of the machine, implements the functions / acts specified in the flowcharts and / or block diagrams. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0150] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer-readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium will include one or more lines of a program of instructions in a transitory signal, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0151] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0152] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0153] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

[0154] It should be understood that the various forms of flow shown above can be reordered, added to, or have steps deleted. For example, the steps described in the present application can be performed in parallel, in series, or in a different order, as long as the desired results of the technical solutions of the present application can be achieved, which are not limited herein.

[0155] The above detailed description does not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for diagnosing equipment faults, characterized in that, include: The dial image of the detection component is acquired, image recognition is performed on the dial image, initial data values ​​are extracted, the initial data values ​​are checked, qualified data are determined, and the qualified data are determined as the detection data. The detection data is oversampled to obtain predicted input data. The predicted input data is then input into a trained prediction model, and the output of the prediction model is used as the predicted data for the detection data. The prediction model includes a neural network model, and the training method of the prediction model includes: using historical measured data as the training dataset for the prediction model, and adjusting the parameters of the prediction model according to the difference between the predicted value and the true value output by the prediction model until the accuracy of the predicted value reaches a set threshold. The detection data and the predicted data are respectively input into the trained fault diagnosis model to obtain preliminary diagnosis results and fault prediction results. The preliminary diagnosis results and the fault prediction results are compared and analyzed to determine the final equipment fault diagnosis result. The fault diagnosis model includes a support vector machine model. The training method of the fault diagnosis model includes: using the dung beetle algorithm to determine the model parameters of the support vector machine model to obtain an initial diagnosis model; extracting a training dataset based on historical detection data; inputting the training dataset into the initial diagnosis model; and adjusting the parameters of the initial diagnosis model based on the model output until the output of the initial diagnosis model reaches a set accuracy threshold.

2. The method according to claim 1, characterized in that, The initial data values ​​are inspected to determine the qualified data, including: According to the preset data threshold, the data in the initial data values ​​that meet the preset data threshold are determined as the first qualified data, and the data that do not meet the preset data threshold are determined as data to be retested; The dial image corresponding to the data to be retested is re-identified. If the obtained data meets the preset data threshold, it is determined to be the second qualified data. The first qualified data and the second qualified data are determined as the inspection qualified data.

3. The method according to claim 1, characterized in that, The preliminary diagnostic results and the fault prediction results are compared and analyzed to determine the final equipment fault diagnosis results, including: The preliminary diagnostic results and the fault prediction results are matched in a preset case library; The fault diagnosis result of the equipment is determined based on the matching result.

4. The method according to claim 3, characterized in that, After determining the final equipment fault diagnosis results, the following is also included: Based on the equipment fault diagnosis results, the corresponding fault handling decision is determined.

5. A device for diagnosing equipment faults, characterized in that, include: The detection data extraction module is used to acquire the dial image of the detection component, perform image recognition on the dial image, extract initial data values, verify the initial data values, determine the qualified data, and identify the qualified data as the detection data. A data prediction module is used to oversample the detection data to obtain prediction input data, input the prediction input data into a trained prediction model, and use the output of the prediction model as the prediction data of the detection data; wherein, the prediction model includes a neural network model, and the training method of the prediction model includes: using historical measured data as the dataset for training the prediction model, adjusting the parameters of the prediction model according to the difference between the predicted value and the true value output by the prediction model, until the accuracy of the prediction value reaches a set threshold; The fault diagnosis module is used to input the detection data and the predicted data into a trained fault diagnosis model, respectively, to obtain preliminary diagnosis results and fault prediction results. The preliminary diagnosis results and the fault prediction results are then compared and analyzed to determine the final equipment fault diagnosis result. The fault diagnosis model includes a support vector machine (SVM) model. The training method for the fault diagnosis model includes: using a dung beetle algorithm to determine the model parameters of the SVM model to obtain an initial diagnosis model; extracting a training dataset based on historical detection data; inputting the training dataset into the initial diagnosis model; and adjusting the parameters of the initial diagnosis model based on the model output until the output of the initial diagnosis model reaches a set accuracy threshold.

6. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the device fault diagnosis method according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the device fault diagnosis method according to any one of claims 1-4.

Images