Circuit breaker hydraulic mechanical fault detection method and device, electronic equipment and storage medium

Abstract

The invention discloses a circuit breaker hydraulic mechanical fault detection method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining current hydraulic oil state data of a circuit breaker; wherein the current hydraulic oil state data comprises the current hydraulic oil state data; moisture content, viscosity and pollution degree of the hydraulic oil; inputting the current hydraulic oil state data into a preset circuit breaker hydraulic mechanical fault detection model, so that the circuit breaker hydraulic mechanical fault detection model extracts a plurality of local features from the current hydraulic oil state data, and fusing the local features to obtain corresponding global features, and then identifying the breaker hydraulic mechanical fault according to the global features, and outputting a corresponding breaker hydraulic mechanical fault type. According to the invention, real-time and accurate fault diagnosis of the circuit breaker hydraulic machinery can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of circuit breaker fault detection, and particularly to a detection method, device, electronic device and storage medium for hydraulic mechanical faults of a circuit breaker. Background Art

[0002] With the development of social economy, the power consumption of various industries continues to increase, posing a severe challenge to the power supply capacity of the power grid. High-voltage circuit breakers are key components in the power system, mainly used to control and protect equipment from the destructive effects of abnormal currents. Therefore, they are indispensable complex electrical equipment in the power system.

[0003] Common types of circuit breaker mechanisms include electromagnetic, spring, pneumatic, permanent magnet, and hydraulic mechanical types. Among them, the switching operation of the hydraulic mechanical type is based on the principle of incompressibility of liquids, using hydraulic oil as the power transmission medium. Through the switching of high and low valves, the hydraulic oil transfers energy between the closed high-pressure oil storage piston and the working cylinder in the oil circuit system, thereby realizing the fast operation of the circuit breaker. The driving force characteristics of the hydraulic cylinder match well with the load characteristics of the switching element, making the operation process smooth. Compared with other mechanical systems, the hydraulic mechanical type is smaller in volume while outputting the same power, requires less energy for switch control, is easy to operate, and is easy to automate. In addition, the hydraulic drive system has a rapid response and a large output force. Therefore, most high-voltage and extra-high-voltage circuit breakers use hydraulic mechanical drive.

[0004] However, problems such as design, manufacturing, assembly, and degradation of hydraulic machinery often lead to faults in the circuit breaker, such as coolant degradation, deterioration of valve operation performance, internal pump oil leakage, and accumulator gas leakage, seriously affecting the reliability of equipment operation and increasing the maintenance difficulty and repair time. To detect faults in hydraulic machinery, current traditional detection methods, such as manual inspection and regular maintenance, have significant defects. Manual inspection relies on the experience of operators, easily overlooks early or potential faults, and often requires disassembling the equipment, increasing the maintenance risk and time cost. At the same time, traditional methods usually cannot monitor the equipment status in real time, resulting in untimely fault response and inability to achieve real-time and accurate fault diagnosis. Summary of the Invention

[0005] The present invention provides a detection method, device, electronic device and storage medium for hydraulic mechanical faults of a circuit breaker to solve the technical problems that manual inspection relies on the experience of operators, easily overlooks early or potential faults, often requires disassembling the equipment, increases the maintenance risk and time cost, and cannot monitor the equipment status in real time, resulting in untimely fault response and inability to achieve real-time and accurate fault diagnosis.

[0006] To solve the above technical problems, an embodiment of the present invention provides a method for detecting hydraulic mechanical faults of a circuit breaker, including:

[0007] Obtain the current hydraulic oil state data of the circuit breaker; wherein, the current hydraulic oil state data includes the moisture content, viscosity, and pollution degree of the hydraulic oil;

[0008] Input the current hydraulic oil state data into a preset hydraulic mechanical fault detection model of the circuit breaker, so that the hydraulic mechanical fault detection model of the circuit breaker extracts several local features from the current hydraulic oil state data, fuses the local features to obtain corresponding global features, and then identifies the hydraulic mechanical faults of the circuit breaker according to the global features, and outputs the corresponding hydraulic mechanical fault types of the circuit breaker;

[0009] Wherein, the hydraulic mechanical fault detection model of the circuit breaker is trained by using the historical hydraulic oil state data of the circuit breaker as input and the corresponding hydraulic mechanical fault types of the historical hydraulic oil state data as output for a preset neural network model.

[0010] As a preferred solution, the generation of the hydraulic mechanical fault detection model of the circuit breaker includes:

[0011] Obtain several historical hydraulic oil state data of the circuit breaker and the corresponding hydraulic mechanical fault types of each historical hydraulic oil state data;

[0012] Use the historical hydraulic oil state data and the corresponding hydraulic mechanical fault types as a sample training set, and train a preset neural network model according to the sample training set and a preset loss function. If the neural network model converges during the training process, save the current model parameters, and then obtain the corresponding hydraulic mechanical fault detection model according to the model parameters.

[0013] As a preferred solution, the hydraulic mechanical fault detection model of the circuit breaker includes a convolutional layer, a pooling layer, and a fully connected output layer;

[0014] The process of extracting several local features from the current hydraulic oil state data, fusing the local features to obtain corresponding global features, and then identifying the hydraulic mechanical faults of the circuit breaker according to the global features and outputting the corresponding hydraulic mechanical fault types includes:

[0015] The convolutional layer extracts several local features from the hydraulic oil state data and generates a corresponding feature matrix according to the extracted local features;

[0016] The pooling layer converts the feature matrix into a corresponding one-dimensional representation;

[0017] The fully connected output layer fuses the one-dimensional representations to generate corresponding global features, and identifies the hydraulic mechanical faults of the circuit breaker based on the global features, and outputs the corresponding types of hydraulic mechanical faults of the circuit breaker.

[0018] As a preferred solution, local features are extracted from the current hydraulic oil state data through the following formula:

[0019] x out,nk = f cov (x in,1h × w 1(h)n(k) + x in,1(h+1)n(k) × w 1(h+1)n(k) + x in,1(h+2) × w 1(h+2)n(k) +... + b n );

[0020] Wherein, x out，nk is the output value of the k-th neuron on the n-th output feature map in the convolutional layer; x in，mk is the output value of the h-th neuron in the input feature map m; b n is the bias value of the output feature map n; f cov is an activation function that maps the input to the output to increase non-linearity.

[0021] As a preferred solution, the feature matrix is converted into a corresponding one-dimensional representation through the following formula:

[0022] t out,nl = f sub [t in,nq , t in,n(q+1) ;

[0023] In the formula, t out，nl is the output value of the l-th neuron on the n-th output feature map in the pooling layer; t in，nq is the input value of the q-th neuron on the n-th input feature map in the pooling layer; f sub is an operation of taking the average value.

[0024] Based on the above embodiments, another embodiment of the present invention provides a detection device for hydraulic mechanical faults of a circuit breaker, including:

[0025] A current hydraulic oil state data acquisition module and a hydraulic mechanical fault detection module for the circuit breaker;

[0026] The current hydraulic oil state data acquisition module is used to acquire the current hydraulic oil state data of the circuit breaker; wherein, the current hydraulic oil state data includes: the moisture content, viscosity and pollution degree of the hydraulic oil;

[0027] The circuit breaker hydraulic mechanical fault detection module is used to input the current hydraulic oil state data into a preset circuit breaker hydraulic mechanical fault detection model, so that the circuit breaker hydraulic mechanical fault detection model extracts several local features from the current hydraulic oil state data, fuses the local features to obtain corresponding global features, and then identifies the circuit breaker hydraulic mechanical fault according to the global features and outputs the corresponding circuit breaker hydraulic mechanical fault type. Among them, the circuit breaker hydraulic mechanical fault detection model is trained by using the historical hydraulic oil state data of the circuit breaker as the input and the circuit breaker hydraulic mechanical fault type corresponding to the historical hydraulic oil state data as the output to train a preset neural network model.

[0028] As a preferred solution, the generation of the circuit breaker hydraulic mechanical fault detection model includes:

[0029] Obtain several historical hydraulic oil state data of the circuit breaker and the circuit breaker hydraulic mechanical fault type corresponding to each historical hydraulic oil state data;

[0030] Use the historical hydraulic oil state data and the corresponding circuit breaker hydraulic mechanical fault type as a sample training set, and train a preset neural network model according to the sample training set and a preset loss function. If the neural network model converges during the training process, save the current model parameters, and then obtain the corresponding circuit breaker hydraulic mechanical fault detection model according to the model parameters.

[0031] As a preferred solution, the circuit breaker hydraulic mechanical fault detection model includes: a convolutional layer, a pooling layer, and a fully connected output layer;

[0032] The process of extracting several local features from the current hydraulic oil state data, fusing the local features to obtain corresponding global features, and then identifying the circuit breaker hydraulic mechanical fault according to the global features and outputting the corresponding circuit breaker hydraulic mechanical fault type includes:

[0033] The convolutional layer extracts several local features from the hydraulic oil state data and generates a corresponding feature matrix according to the extracted local features;

[0034] The pooling layer converts the feature matrix into a corresponding one-dimensional representation;

[0035] The fully connected output layer fuses the one-dimensional representation to generate corresponding global features, and identifies the circuit breaker hydraulic mechanical fault according to the global features and outputs the corresponding circuit breaker hydraulic mechanical fault type.

[0036] Based on the above embodiments, another embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method for detecting the hydraulic mechanical fault of the circuit breaker described in the above embodiments of the present invention.

[0037] Based on the above embodiments, another embodiment of the present invention provides a storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the storage medium is located to execute the method for detecting the hydraulic mechanical fault of the circuit breaker described in the above embodiments of the present invention.

[0038] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0039] The present invention provides a method for detecting the hydraulic mechanical fault of a circuit breaker, which obtains the current hydraulic oil state data of the circuit breaker. Among them, the current hydraulic oil state data includes the moisture content, viscosity, and pollution degree of the hydraulic oil. The current hydraulic oil state data is input into a preset circuit breaker hydraulic mechanical fault detection model, so that the circuit breaker hydraulic mechanical fault detection model extracts several local features from the current hydraulic oil state data, fuses the local features to obtain corresponding global features, and then identifies the hydraulic mechanical fault of the circuit breaker according to the global features and outputs the corresponding circuit breaker hydraulic mechanical fault type. Among them, the circuit breaker hydraulic mechanical fault detection model is trained by using the historical hydraulic oil state data of the circuit breaker as the input and the corresponding circuit breaker hydraulic mechanical fault type of the historical hydraulic oil state data as the output for a preset neural network model. Through the present invention, the hydraulic mechanical fault of the circuit breaker can be automatically detected in real time according to the current hydraulic oil state data of the circuit breaker and the pre-trained circuit breaker hydraulic mechanical fault detection model, realizing real-time and accurate fault diagnosis of the circuit breaker hydraulic machinery. Description of the Drawings

[0040] Figure 1 is a schematic flowchart of a method for detecting the hydraulic mechanical fault of a circuit breaker provided by an embodiment of the present invention;

[0041] Figure 2 is an overall schematic diagram of the detection of physical and chemical indicators of hydraulic oil;

[0042] Figure 3 is a schematic structural diagram of a CNN hydraulic oil state diagnosis model;

[0043] Figure 4 is a method block diagram of the fault diagnosis test process of CNN;

[0044] Figure 5It is a schematic structural diagram of a detection device for hydraulic mechanical faults of a circuit breaker provided by an embodiment of the present invention. Detailed implementation manners

[0045] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the scope of protection of the present application.

[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above drawings are intended to cover non-exclusive inclusion.

[0047] In the description of the embodiments of the present application, technical terms such as "first" and "second" are only used to distinguish different objects and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity, specific order or primary-secondary relationship of the indicated technical features. In the description of the embodiments of the present application, "a plurality of" means two or more, unless otherwise specifically defined.

[0048] Referring to "embodiments" herein means that specific features, structures or characteristics described in connection with the embodiments may be included in at least one embodiment of the present application. The phrase does not necessarily refer to the same embodiment at every occurrence in the specification, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.

[0049] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " herein generally represents an "or" relationship between the associated objects before and after.

[0050] In the description of the embodiments of the present application, the term "a plurality of" refers to two or more (including two). Similarly, "a plurality of groups" refers to two or more groups (including two groups), and "a plurality of pieces" refers to two or more pieces (including two pieces).

[0051] In the description of the embodiments of the present application, unless otherwise clearly defined and limited, technical terms such as "installation", "connection", "connection", "fixation", etc. should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral body; it can also be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, and can be the communication inside two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to specific circumstances.

[0052] Embodiment 1

[0053] Please refer to Figure 1 , which is a schematic flow chart of a method for detecting hydraulic mechanical faults of a circuit breaker provided by an embodiment of the present invention, and includes the following specific steps:

[0054] S1. Obtain the current hydraulic oil state data of the circuit breaker; wherein, the current hydraulic oil state data includes: the moisture content, viscosity, and pollution degree of the hydraulic oil;

[0055] Hydraulic oil is regarded as the "blood" of the hydraulic system. In addition to being an important power transmission medium, it also functions to cool the hydraulic system, protect hydraulic components from corrosion, reduce wear between friction pairs, and provide a buffering effect. The moisture content, viscosity, and pollution degree of hydraulic oil are closely related to the occurrence of faults. Therefore, the present invention monitors and analyzes the state of the oil in real time and combines intelligent diagnosis technology to diagnose hydraulic mechanical faults of the circuit breaker.

[0056] Before identifying mechanical faults, the physical and chemical properties of the hydraulic oil should be effectively detected. Please refer to Figure 2 , which is an overall schematic diagram of the detection of physical and chemical indicators of hydraulic oil. Through humidity, viscosity, and pollutant sensors, the physical and chemical indicators such as the moisture, viscosity, and pollutant content of the hydraulic oil are monitored in real time, providing reliable basic data for subsequent hydraulic mechanical fault diagnosis.

[0057] The user can select different liquid sources, such as clean liquid or hydraulic oil, through a selection valve. First, drive the motor to operate the piston pump to circulate the liquid through the pipeline. The two-way design allows the liquid to flow in both directions, facilitating multiple tests or reuse. The system has a two-way flow design, supporting the two-way flow of the liquid in the pipeline, facilitating multiple tests or reuse, and increasing the flexibility and practicality of the system. This design can ensure the selection of different liquid sources (such as clean liquid or hydraulic oil) during the detection process and ensure that the liquid can be cleaned after passing through the test channel to prevent cross-contamination.

[0058] After use, the test channel can be cleaned with a cleaning solution to ensure the accuracy of the next detection. The configured humidity, viscosity, and contaminant sensors measure the performance indicators of the test oil sample. The data conversion and processing module collects and processes data from each sensor. This module converts analog signals into digital signals and performs necessary data analysis and calibration. Finally, the processed data is displayed on the device screen in real time, facilitating the user to immediately view the test results. The system converts analog signals into digital signals through the data conversion and processing module, performs data analysis and calibration, and finally displays it on the device screen in real time, facilitating the user to view the detection results at any time and improving the operation convenience.

[0059] S2. Input the current hydraulic oil state data into a preset circuit breaker hydraulic mechanical fault detection model, so that the circuit breaker hydraulic mechanical fault detection model extracts several local features from the current hydraulic oil state data, fuses the local features to obtain corresponding global features, and then identifies the circuit breaker hydraulic mechanical fault according to the global features and outputs the corresponding circuit breaker hydraulic mechanical fault type.

[0060] Among them, the circuit breaker hydraulic mechanical fault detection model is obtained by training a preset neural network model with the historical hydraulic oil state data of the circuit breaker as the input and the circuit breaker hydraulic mechanical fault type corresponding to the historical hydraulic oil state data as the output.

[0061] Preferably, the generation of the circuit breaker hydraulic mechanical fault detection model includes: obtaining several historical hydraulic oil state data of the circuit breaker and the circuit breaker hydraulic mechanical fault type corresponding to each historical hydraulic oil state data; using the historical hydraulic oil state data and the corresponding circuit breaker hydraulic mechanical fault type as a sample training set, and training a preset neural network model according to the sample training set and a preset loss function. If the neural network model converges during the training process, save the current model parameters, and then obtain the corresponding circuit breaker hydraulic mechanical fault detection model according to the model parameters.

[0062] Preferably, the circuit breaker hydraulic mechanical fault detection model includes: a convolutional layer, a pooling layer, and a fully connected output layer; extracting a plurality of local features from the current hydraulic oil state data, and fusing the local features to obtain corresponding global features, and then identifying the circuit breaker hydraulic mechanical fault according to the global features, and outputting the corresponding circuit breaker hydraulic mechanical fault type, including the convolutional layer extracting a plurality of local features from the hydraulic oil state data and generating a corresponding feature matrix according to the extracted local features; the pooling layer converting the feature matrix into a corresponding one-dimensional representation; the fully connected output layer fusing the one-dimensional representation to generate a corresponding global feature, and identifying the circuit breaker hydraulic mechanical fault according to the global feature, and outputting the corresponding circuit breaker hydraulic mechanical fault type.

[0063] Preferably, the following formula is used to extract local features from the current hydraulic oil state data:

[0064] x out,nk = f cov (x in,1h × w 1(h)n(k) + x in,1(h+1)n(k) × w 1(h+1)n(k) + x in,1(h+2) × w 1(h+2)n(k) +... + b n );

[0065] where x out，nk is the output value of the k-th neuron on the n-th output feature map in the convolutional layer; x in，mk is the output value of the h-th neuron in the input feature map m; b n is the bias value of the output feature map n; f cov is an activation function that maps the input to the output to increase non-linearity.

[0066] Preferably, the following formula is used to convert the feature matrix into a corresponding one-dimensional representation:

[0067] t out,nl = f sub [t in,nq , t in,n(q+1) ;

[0068] In the formula, t out，nl is the output value of the l-th neuron on the n-th output feature map in the pooling layer; t in，nq is the input value of the q-th neuron on the n-th input feature map in the pooling layer; f sub is an operation of taking the average value.

[0069] As one of the most effective algorithms in deep learning, the Convolutional Neural Network (CNN) is good at automatically learning and extracting features from raw data, so it is very suitable for classification, regression, and prediction tasks. In this invention, the three physicochemical indexes of the above hydraulic oil are selected as the input of the CNN, a mapping model between the hydraulic oil state data and the hydraulic system faults is constructed, and a data-driven CNN model is established for the fault diagnosis of the hydraulic system. The convolutional neural network includes a convolutional layer, a pooling layer, and a fully connected output layer. It can extract local features at a lower level and combine these features into more abstract features at a higher level.

[0070] Among them, the convolutional layer uses a set of learnable convolutional kernels to extract multiple feature matrices from the input. The expression of the convolution process is as follows:

[0071] x out,nk =f cov (x in,1h ×w 1(h)n(k) +x in,1(h+1)n(k) ×w 1(h+1)n(k) +x in,1(h+2) ×w 1(h+2)n(k) +…+b n ) (1)

[0072] In the formula, x out，nk is the output value of the k-th neuron on the n-th output feature map in the convolutional layer; x in，mk is the output value of the h-th neuron in the input feature map m; b n is the bias value of the output feature map n; f cov is the activation function, which maps the input to the output to increase the non-linearity.

[0073] The pooling layer usually receives the output from the convolutional layer as the input and improves the efficiency of the model by reducing the size of the feature map. The pooling operation can be expressed by the following formula (see formula 2):

[0074] t out,nl =f sub [t in,nq ,t in,n(q+1) (2)

[0075] In the formula, t out，nl is the output value of the l-th neuron on the n-th output feature map in the pooling layer; t in，nq is the input value of the q-th neuron on the n-th input feature map in the pooling layer; f sub is the operation of taking the average value.

[0076] The fully connected output layers in a convolutional neural network are closely connected. Each neuron in the fully connected output layer is fully connected to all neurons in the previous layer, and the Softmax function is used to activate the output, converting the final output of the network into a probability distribution. Please refer to Figure 3 , which is a schematic diagram of the structure of the CNN hydraulic oil condition diagnosis model. This model extracts local features from the hydraulic oil condition detection data through multiple layers of convolution and pooling, reduces the multi-dimensional feature tensor and converts it into a one-dimensional representation. Then, the fully connected layer combines the local features with the global information. Finally, the Softmax classifier classifies the data and outputs the corresponding results. This method effectively captures the local information in the data and integrates it with the global information, thereby improving the accuracy of the model and facilitating the accurate diagnosis of the hydraulic oil condition.

[0077] The present invention takes three physicochemical indexes of the moisture, viscosity and pollutants of the hydraulic oil as the input of the CNN, and constructs a mapping model between the hydraulic oil condition data and the faults of the hydraulic system. Through automatic feature learning by the CNN, the fault diagnosis is completed. The convolutional layer and the pooling layer are used to extract the local features of the hydraulic oil condition detection data, and through multiple layers of convolution and pooling operations, more abstract and discriminative features are obtained, improving the diagnostic accuracy and robustness of the model. The CNN model combines the local features with the global information through the fully connected layer, and converts the final output into a probability distribution of the fault types through the Softmax classifier, automatically classifying the faults of the hydraulic machinery.

[0078] After the CNN model is constructed, it is used to train and test the physicochemical properties of the hydraulic oil to obtain the fault information of the hydraulic machinery. Please refer to Figure 4 , which is a method block diagram of the fault diagnosis test process of the CNN. The specific operation steps are as Figure 4 shown:

[0079] Step 1: Use the hydraulic system test equipment to collect the viscosity, moisture and pollutant data of the hydraulic oil samples, and record the fault types corresponding to each sample.

[0080] Step 2: Construct a sample data set and divide it into a training set and a test set according to the specified ratio.

[0081] Step 3: Design the model structure and initialize the parameters, and then use the training set to train the model.

[0082] Step 4: After the model training is completed, save the model parameters to obtain the fault diagnosis model. Calculate the loss between the actual data label and the label obtained from the model training to judge whether the CNN model has converged.

[0083] Step 5: If the model has converged, complete the training and save the model parameters; if not, optimize the model parameters until the CNN model training converges.

[0084] Step 6: Input the test set into the trained model, calculate the classification accuracy, and output the fault diagnosis result.

[0085] The present invention collects and labels the viscosity, moisture, and contaminant data of hydraulic oil samples, constructs a sample data set, and divides the data into a training set and a test set according to a predetermined ratio. Through this data division, it is ensured that the CNN model can be effectively verified after sufficient training. During the training process, by calculating the loss between the actual data label and the output label of the CNN model, it is judged whether the model has converged. If not, the accuracy of the model is further improved by optimizing the parameters until the model converges. The trained CNN model can classify the test set and output the diagnosis result of hydraulic machinery faults, with a high classification accuracy, providing a reliable tool for the diagnosis of hydraulic machinery faults.

[0086] It can be seen that the present invention provides a method for detecting faults in the hydraulic machinery of a circuit breaker. Through an automated detection process (from the collection of hydraulic oil samples to data processing and fault diagnosis), the intervention of manual operations is greatly reduced, the efficiency and accuracy of detection are improved, and at the same time, the dependence on professional technicians is reduced. At the same time, automated fault diagnosis not only improves the detection efficiency but also simplifies the operation process, enabling maintenance personnel to quickly and accurately diagnose faults, and optimizing the maintenance and management process of the hydraulic machinery of the circuit breaker. The following beneficial effects can be achieved through the present invention:

[0087] (1) Improved detection and early warning capabilities: By establishing a mapping model between the hydraulic oil state data and the faults of the hydraulic system and combining convolutional neural networks (CNNs) for data-driven fault diagnosis, the problems of lagging response and insufficient diagnostic accuracy of traditional detection methods are overcome, enabling potential faults to be detected earlier, thus enabling early maintenance and reducing the occurrence of equipment failures;

[0088] (2) Non-destructive detection: By obtaining fault information through the detection of hydraulic oil, there is no need to stop the machine or disassemble the equipment, avoiding the problems of damage to the equipment or interruption of operation in traditional detection methods. This non-destructive detection method not only improves the operating efficiency of the system but also reduces the risk of equipment maintenance and downtime;

[0089] (3) Reduce maintenance costs and technical complexity: The proposed detection technology based on data fusion convolutional neural network (CNN) can accurately identify multiple fault types through automated analysis and fault diagnosis, greatly reducing the dependence on professional technicians and reducing the complexity of manual operation and analysis. In addition, this method has a high degree of automation and simple operation, significantly optimizing the fault diagnosis process of the equipment and providing strong support for the efficient operation of hydraulic machinery.

[0090] Embodiment 2

[0091] Please refer to Figure 5 , which is a schematic structural diagram of a detection device for circuit breaker hydraulic machinery faults provided by an embodiment of the present invention. The device includes: a current hydraulic oil state data acquisition module and a circuit breaker hydraulic machinery fault detection module;

[0092] The current hydraulic oil state data acquisition module is used to acquire the current hydraulic oil state data of the circuit breaker; wherein, the current hydraulic oil state data includes: the moisture content, viscosity, and pollution degree of the hydraulic oil;

[0093] The circuit breaker hydraulic machinery fault detection module is used to input the current hydraulic oil state data into a preset circuit breaker hydraulic machinery fault detection model, so that the circuit breaker hydraulic machinery fault detection model extracts several local features from the current hydraulic oil state data, fuses the local features to obtain corresponding global features, and then identifies the circuit breaker hydraulic machinery faults according to the global features and outputs the corresponding circuit breaker hydraulic machinery fault types; wherein, the circuit breaker hydraulic machinery fault detection model is trained by using the historical hydraulic oil state data of the circuit breaker as the input and the circuit breaker hydraulic machinery fault types corresponding to the historical hydraulic oil state data as the output for a preset neural network model.

[0094] Preferably, the generation of the circuit breaker hydraulic machinery fault detection model includes: acquiring several historical hydraulic oil state data of the circuit breaker and the circuit breaker hydraulic machinery fault types corresponding to each historical hydraulic oil state data; using the historical hydraulic oil state data and the corresponding circuit breaker hydraulic machinery fault types as a sample training set, training a preset neural network model according to the sample training set and a preset loss function. If the neural network model converges during the training process, save the current model parameters, and then obtain the corresponding circuit breaker hydraulic machinery fault detection model according to the model parameters.

[0095] Preferably, the circuit breaker hydraulic mechanical fault detection model includes: a convolutional layer, a pooling layer, and a fully connected output layer; extracting several local features from the current hydraulic oil state data, fusing the local features to obtain corresponding global features, and then identifying the circuit breaker hydraulic mechanical fault according to the global features, and outputting the corresponding circuit breaker hydraulic mechanical fault type, including: the convolutional layer extracts several local features from the hydraulic oil state data and generates a corresponding feature matrix according to the extracted local features; the pooling layer converts the feature matrix into a corresponding one-dimensional representation; the fully connected output layer fuses the one-dimensional representation, generates a corresponding global feature, and identifies the circuit breaker hydraulic mechanical fault according to the global feature, and outputs the corresponding circuit breaker hydraulic mechanical fault type.

[0096] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement without creative efforts.

[0097] Those skilled in the art can clearly understand that for the convenience and simplicity, the specific working process of the device described above can refer to the corresponding process in the foregoing method embodiments, and will not be elaborated here.

[0098] Embodiment III

[0099] Correspondingly, an embodiment of the present invention provides an electronic device, the device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the detection method of the circuit breaker hydraulic mechanical fault described in the foregoing embodiments of the present invention.

[0100] The electronic device may be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The device may include, but is not limited to, a processor and a memory.

[0101] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the device and connects various parts of the entire device through various interfaces and lines.

[0102] Embodiment 4

[0103] Correspondingly, an embodiment of the present invention provides a storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the storage medium is located to execute the detection method for the hydraulic mechanical fault of the circuit breaker described in the above-mentioned embodiment of the invention.

[0104] The memory can be used to store the computer program. The processor realizes various functions of the device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, FlashCard, at least one magnetic disk storage device, flash device, or other volatile solid-state storage devices.

[0105] The storage medium is a computer-readable storage medium, and the computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice within the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0106] The above is the preferred embodiment of the present invention. It should be pointed out that for those of ordinary skill in the art of the present technology, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A method for detecting hydraulic mechanical failure of a circuit breaker, characterized in that: include: Acquire the current hydraulic oil status data of the circuit breaker; wherein the current hydraulic oil status data includes: the water content, viscosity and contamination degree of the hydraulic oil; Inputting the current hydraulic oil state data into a preset circuit breaker hydraulic machinery fault detection model, so that the circuit breaker hydraulic machinery fault detection model extracts a number of local features from the current hydraulic oil state data, and fuses the local features to obtain corresponding global features, and then identifies the circuit breaker hydraulic machinery fault according to the global features, and outputs the corresponding circuit breaker hydraulic machinery fault type; The circuit breaker hydraulic mechanical fault detection model is obtained by training a preset neural network model with historical hydraulic oil state data of the circuit breaker as input and the circuit breaker hydraulic mechanical fault type corresponding to the historical hydraulic oil state data as output.

2. The method for detecting hydraulic mechanical failure of a circuit breaker according to claim 1, characterized in that: The generation of the circuit breaker hydraulic mechanical fault detection model includes: Acquire several historical hydraulic oil status data of the circuit breaker and the hydraulic mechanical fault type of the circuit breaker corresponding to each historical hydraulic oil status data; The historical hydraulic oil status data and the corresponding circuit breaker hydraulic mechanical fault types are used as sample training sets. A preset neural network model is trained according to the sample training set and a preset loss function. If the neural network model converges during the training process, the current model parameters are saved, and then the corresponding circuit breaker hydraulic mechanical fault detection model is obtained according to the model parameters.

3. The method for detecting hydraulic mechanical failure of a circuit breaker according to claim 1, characterized in that: The circuit breaker hydraulic mechanical fault detection model includes: a convolution layer, a pooling layer and a fully connected output layer; The extracting of a number of local features from the current hydraulic oil state data and fusing the local features to obtain corresponding global features, and then identifying the circuit breaker hydraulic mechanical fault according to the global features and outputting the corresponding circuit breaker hydraulic mechanical fault type, including: The convolution layer extracts a number of local features from the hydraulic oil state data, and generates a corresponding feature matrix according to the extracted local features; The pooling layer converts the feature matrix into a corresponding one-dimensional representation; The fully connected output layer fuses the one-dimensional representations to generate corresponding global features, identifies the circuit breaker hydraulic mechanical fault according to the global features, and outputs the corresponding circuit breaker hydraulic mechanical fault type.

4. The method for detecting hydraulic mechanical failure of a circuit breaker according to claim 3, characterized in that: The local features are extracted from the current hydraulic oil state data by the following formula: x out，nk =f cov (x in，1h ×w 1(h)n(k) +x in，1(h+1)n(k) ×w 1(h+1)n(k) +x in，1(h+2) ×w 1(h+2)n(k) +...+b n ); Among them, x out，nk is the output value of the kth neuron on the nth output feature map in the convolutional layer; x in，mk is the output value of the hth neuron in the input feature map m; b n is the bias value of the output feature map n; f cov It is an activation function that maps input to output to add non-linearity.

5. The method for detecting hydraulic mechanical failure of a circuit breaker according to claim 4, characterized in that: The feature matrix is ​​converted into the corresponding one-dimensional representation by the following formula: t out，nl =f sub [t in，nq ,t in，n(q+1) ]; Where, t out，nl is the output value of the Ith neuron on the nth output feature map in the pooling layer; t in，nq is the input value of the qth neuron on the nth input feature map in the pooling layer; f sub This is the operation of taking the average value.

6. A circuit breaker hydraulic mechanical failure detection device, characterized in that: include: Current hydraulic oil status data acquisition module and circuit breaker hydraulic mechanical fault detection module; The current hydraulic oil state data acquisition module is used to acquire the current hydraulic oil state data of the circuit breaker; wherein the current hydraulic oil state data includes: the water content, viscosity and contamination degree of the hydraulic oil; The circuit breaker hydraulic machinery fault detection module is used to input the current hydraulic oil state data into a preset circuit breaker hydraulic machinery fault detection model, so that the circuit breaker hydraulic machinery fault detection model extracts a number of local features from the current hydraulic oil state data, and fuses the local features to obtain corresponding global features, and then identifies the circuit breaker hydraulic machinery fault according to the global features, and outputs the corresponding circuit breaker hydraulic machinery fault type; wherein the circuit breaker hydraulic machinery fault detection model is obtained by training a preset neural network model with the historical hydraulic oil state data of the circuit breaker as input and the circuit breaker hydraulic machinery fault type corresponding to the historical hydraulic oil state data as output.

7. The circuit breaker hydraulic mechanical failure detection device according to claim 6, characterized in that: The generation of the circuit breaker hydraulic mechanical fault detection model includes: Acquire several historical hydraulic oil status data of the circuit breaker and the hydraulic mechanical fault type of the circuit breaker corresponding to each historical hydraulic oil status data; The historical hydraulic oil status data and the corresponding circuit breaker hydraulic mechanical fault types are used as sample training sets. A preset neural network model is trained according to the sample training set and a preset loss function. If the neural network model converges during the training process, the current model parameters are saved, and then the corresponding circuit breaker hydraulic mechanical fault detection model is obtained according to the model parameters.

8. The circuit breaker hydraulic mechanical failure detection device according to claim 6, characterized in that: The circuit breaker hydraulic mechanical fault detection model includes: a convolution layer, a pooling layer and a fully connected output layer; The extracting of a number of local features from the current hydraulic oil state data and fusing the local features to obtain corresponding global features, and then identifying the circuit breaker hydraulic mechanical fault according to the global features and outputting the corresponding circuit breaker hydraulic mechanical fault type, including: The convolution layer extracts a number of local features from the hydraulic oil state data, and generates a corresponding feature matrix according to the extracted local features; The pooling layer converts the feature matrix into a corresponding one-dimensional representation; The fully connected output layer fuses the one-dimensional representations to generate corresponding global features, identifies the circuit breaker hydraulic mechanical fault according to the global features, and outputs the corresponding circuit breaker hydraulic mechanical fault type.

9. An electronic device, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the method for detecting hydraulic mechanical failure of a circuit breaker according to any one of claims 1 to 5 when executing the computer program.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is executed, the device where the storage medium is located is controlled to execute the method for detecting hydraulic mechanical failure of a circuit breaker according to any one of claims 1 to 5.

Images