Electrical equipment state monitoring method and device, terminal equipment and storage medium

Abstract

The invention discloses an electrical equipment state monitoring method and device, terminal equipment and a storage medium. The method comprises the following steps: acquiring an electromagnetic signal of to-be-monitored equipment; inputting characteristic parameters extracted according to the electromagnetic signal into a preset fuzzy logic control model for fuzzification conversion to obtain a first fault prediction value for representing the fault risk level of the to-be-monitored equipment; inputting the characteristic parameters into a preset support vector machine model for fault identification, and obtaining a second fault prediction value used for representing the fault probability of the to-be-monitored equipment; according to the first fault prediction value, the second fault prediction value and a preset weight, determining whether the to-be-monitored device has a partial discharge fault; if yes, generating a detection measure according to the generated detection command; if not, performing evaluation according to the characteristic parameters and a preset parameter threshold value to obtain a state evaluation result of the to-be-monitored equipment; electromagnetic signals are collected through a quantum sensor. According to the invention, the monitoring accuracy and efficiency of the electrical equipment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical equipment condition monitoring, and in particular to an electrical equipment condition monitoring method, device, terminal device and storage medium. Background Art

[0002] With the increasing complexity of the usage environment and operating load of electrical equipment, the fault detection and condition monitoring technologies of equipment are also constantly evolving. Traditional electrical equipment condition monitoring mostly relies on traditional sensor technologies, such as the monitoring of physical quantities such as current, voltage, and temperature. In the face of complex environmental changes and minor equipment faults, there are often problems such as slow response speed, low accuracy, and insufficient fault warning capabilities. In recent years, quantum sensor technology has become a new direction for electrical equipment monitoring due to its high sensitivity and high-precision measurement capabilities. Especially in the field of partial discharge detection, quantum sensors can capture weak electromagnetic signals in electrical equipment, providing the possibility for early fault prediction of equipment.

[0003] However, at present, the electrical equipment condition monitoring methods based on quantum sensors mostly focus on the acquisition of electromagnetic signals and simple partial discharge detection, lacking effective signal deconstruction and fault detection. Most of the existing partial discharge signal processing relies on simple frequency analysis, ignoring the time-varying and amplitude attenuation characteristics of the signals, making it difficult to detect many fault characteristics of electrical equipment in a timely manner, thus affecting the operation and maintenance efficiency of electrical equipment. Summary of the Invention

[0004] Embodiments of the present invention provide an electrical equipment condition monitoring method, device, terminal device and storage medium, which can effectively solve the problem that the existing technology cannot discover the fault characteristics of electrical equipment, resulting in low operation and maintenance efficiency of electrical equipment.

[0005] An embodiment of the present invention provides an electrical equipment condition monitoring method, including:

[0006] Obtain the electromagnetic signal of the device to be monitored;

[0007] Extract the corresponding characteristic parameters according to the electromagnetic signal;

[0008] Input the characteristic parameters into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value for characterizing the fault risk level of the device to be monitored;

[0009] Input the characteristic parameters into a preset support vector machine model for fault identification to obtain a second fault prediction value for characterizing the fault probability of the device to be monitored;

[0010] Judge whether the device to be monitored has a partial discharge fault according to the first fault prediction value, the second fault prediction value and a preset weight;

[0011] If so, generate a detection command to generate corresponding detection measures according to the detection command;

[0012] If not, evaluate the device to be monitored according to the characteristic parameters and the preset parameter threshold to obtain a status evaluation result of the device to be monitored;

[0013] Among them, the electromagnetic signal is collected by a quantum sensor; the characteristic parameters include: amplitude, frequency, and attenuation coefficient.

[0014] Further, input the characteristic parameters into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value for characterizing the fault risk level of the device to be monitored, including:

[0015] Input the characteristic parameters into a preset fuzzy logic control model to determine the value range of each characteristic parameter;

[0016] According to the characteristic parameters and the value range, create a membership function for each characteristic parameter;

[0017] Perform fuzzy conversion on the characteristic parameters according to the membership function to obtain a fuzzy set for each characteristic parameter;

[0018] Convert the fuzzy set through a preset defuzzification method to obtain a first fault prediction value for characterizing the fault risk level of the device to be monitored.

[0019] Further, the training of the support vector machine model includes:

[0020] Obtain the historical characteristic parameters, historical fault probabilities, and initial model parameters of the electrical equipment;

[0021] Input the historical characteristic parameters and the current model parameters into the support vector machine model to be trained for training to obtain a predicted fault probability;

[0022] Calculate the loss value of a preset loss function according to the historical fault probability and the predicted fault probability; and determine whether the loss value converges;

[0023] When the loss value converges, obtain the trained support vector machine model;

[0024] When the loss value does not converge, update the current model parameters according to the loss value, and use the updated current model parameters as the current model parameters for the next training; among them, the current model parameters for the first training are the initial model parameters.

[0025] Further, judging whether a partial discharge fault occurs in the device to be monitored according to the first fault prediction value, the second fault prediction value and a preset weight includes:

[0026] Calculate the judgment value of the device to be monitored according to the first fault prediction value, the second fault prediction value and the preset weight;

[0027] Comparing the judgment value with a preset threshold;

[0028] When the judgment value is greater than or equal to a preset threshold, it is determined that a partial discharge fault occurs in the equipment to be monitored;

[0029] When the judgment value is less than the preset threshold, it is determined that no partial discharge fault occurs in the device to be monitored.

[0030] Furthermore, the device to be monitored is evaluated according to the characteristic parameters and the preset parameter thresholds to obtain a status evaluation result of the device to be monitored, including:

[0031] Calculating the health value of the device to be monitored according to the characteristic parameters, the preset characteristic weights and the preset health decay parameters;

[0032] Making a judgment based on the health value and a preset parameter threshold;

[0033] When the health value is less than or equal to a preset parameter threshold, determining that a state assessment result of the device to be monitored is that a non-partial discharge fault exists;

[0034] When the health value is greater than a preset parameter threshold, it is determined that the status evaluation result of the device to be monitored is normal.

[0035] Furthermore, it also includes:

[0036] After determining that a partial discharge fault occurs in the equipment to be monitored, determining whether insulation material aging exists in the equipment to be monitored according to the characteristic parameters;

[0037] When the attenuation coefficient is less than the preset attenuation coefficient threshold and the amplitude of the frequency is greater than the preset amplitude, it is determined that the insulation material of the monitored equipment is aged.

[0038] Furthermore, it also includes:

[0039] Determine the layout position of the quantum sensor according to the type of the device to be monitored;

[0040] When the device to be monitored is a transformer, determining that the quantum sensor is arranged in a bushing of the transformer;

[0041] When the device to be monitored is a switching device, determine that the deployment position of the quantum sensor is at the switching contact;

[0042] When the device to be monitored is a power line joint, determine that the deployment positions of the quantum sensor are at the power line joint, the insulating material, and the device interface;

[0043] Generate a quantum sensor deployment plan according to the deployment positions.

[0044] As an improvement to the above solution, another embodiment of the present invention correspondingly provides an electrical equipment status monitoring device, including:

[0045] An electromagnetic signal acquisition module, configured to acquire the electromagnetic signal of the device to be monitored;

[0046] A signal feature extraction module, configured to extract the corresponding characteristic parameters according to the electromagnetic signal;

[0047] A first prediction module, configured to input the characteristic parameters into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value for characterizing the fault risk level of the device to be monitored;

[0048] A second prediction module, configured to input the characteristic parameters into a preset support vector machine model for fault identification to obtain a second fault prediction value for characterizing the fault probability of the device to be monitored;

[0049] A first judgment module, configured to judge whether the device to be monitored has a partial discharge fault according to the first fault prediction value, the second fault prediction value, and a preset weight;

[0050] A measure generation module, configured to generate a detection command in the case of judging that the device to be monitored has a partial discharge fault, so as to generate corresponding detection measures according to the detection command;

[0051] A second judgment module, configured to evaluate the device to be monitored according to the characteristic parameters and a preset parameter threshold in the case of judging that the device to be monitored has no partial discharge fault, to obtain a status evaluation result of the device to be monitored;

[0052] Wherein, the electromagnetic signal is collected by a quantum sensor; the characteristic parameters include: amplitude, frequency, and attenuation coefficient.

[0053] Another embodiment of the present invention provides a terminal device, including 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 an electrical equipment status monitoring method as described in the above embodiment.

[0054] Another embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the electrical equipment status monitoring method described in the above embodiment.

[0055] By implementing the present invention, at least the following beneficial effects are achieved:

[0056] The present invention provides an electrical equipment status monitoring method, device, terminal device, and storage medium. The method can acquire the electromagnetic signal of the device to be monitored; extract the corresponding characteristic parameters according to the electromagnetic signal; input the characteristic parameters into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value representing the fault risk level of the device to be monitored; input the characteristic parameters into a preset support vector machine model for fault identification to obtain a second fault prediction value representing the fault probability of the device to be monitored; judge whether the device to be monitored has a partial discharge fault according to the first fault prediction value, the second fault prediction value, and a preset weight; if so, generate a detection command so as to generate corresponding detection measures according to the detection command; if not, evaluate the device to be monitored according to the characteristic parameters and a preset parameter threshold to obtain a status evaluation result of the device to be monitored; wherein, the electromagnetic signal is collected by a quantum sensor; the characteristic parameters include: amplitude, frequency, and attenuation coefficient. By collecting the electromagnetic signal of the device to be monitored through a quantum sensor and obtaining characteristic parameters, it is possible to achieve fine monitoring of the electromagnetic signal of the device to be monitored, such as subtle changes in amplitude, frequency, and attenuation coefficient, and timely detect faults in electrical equipment; at the same time, the fuzzy logic control model is used to handle uncertain problems, such as the boundary state with high amplitude but low frequency. The support vector machine model is used to identify the characteristic parameters and output a second fault prediction value representing the fault probability of the device to be monitored. Multiple models are combined to judge whether a partial discharge fault occurs, and then the device to be monitored is further evaluated according to the characteristic parameters, thereby improving the accuracy of identifying partial discharge faults in the device to be monitored. Thus, when the electromagnetic signal of the device to be monitored changes due to a fault, the state change of the device to be monitored can be detected in a timely manner, providing a data basis for equipment operation and maintenance and improving the efficiency of operation and maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a flowchart of an electrical equipment status monitoring method provided by an embodiment of the present invention;

[0058] Figure 2 is a structural diagram of an electrical equipment status monitoring device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0060] See Figure 1 , which is a schematic flow chart of a method for monitoring the state of an electrical device provided by an embodiment of the present invention, including:

[0061] S1. Obtain the electromagnetic signal of the device to be monitored;

[0062] Specifically, the electromagnetic signal is collected by a quantum sensor.

[0063] Preferably, according to the type of the device to be monitored, determine the layout position of the quantum sensor;

[0064] When the device to be monitored is a transformer, determine that the layout position of the quantum sensor is inside the transformer bushing;

[0065] When the device to be monitored is a switchgear, determine that the layout position of the quantum sensor is at the switch contact;

[0066] When the device to be monitored is a power line joint, determine that the layout position of the quantum sensor is at the joint, insulating material, and device interface of the power line;

[0067] Generate a quantum sensor layout plan according to the layout position.

[0068] In a preferred embodiment of the present invention, a superconducting quantum interference device can be selected as a quantum sensor to detect the magnetic field fluctuations of the device to be monitored, and an electromagnetic signal is obtained. The sampling rate should be set above 100 MHz. The electromagnetic signals collected by the quantum sensor include the electromagnetic waves caused by partial discharge, as well as the frequency, amplitude, and duration of the discharge. When the device to be monitored is a transformer, it is installed in the bushing to monitor the partial discharge phenomenon in the transformer; when the device to be monitored is a switchgear, a quantum sensor is installed inside the high-voltage switchgear or at the switch contact to monitor the partial discharge during switch operation in real time; when the device to be monitored is a power line joint, quantum sensors are arranged at the joints of the power line, insulating materials, and equipment interfaces. To enhance the reliability and accuracy of the data, the quantum sensor adopts a multi-channel synchronous acquisition technology to collect electric field and magnetic field signals (electromagnetic signals) respectively, ensuring real-time monitoring of each key point of the device to be monitored. The data transmission of the quantum sensor includes two methods: optical fiber communication: in a high electromagnetic interference environment, optical fiber communication has the characteristics of strong anti-interference ability and high transmission rate, and is suitable for data transmission from the quantum sensor to the central processing unit. Wireless transmission: for devices to be monitored that are difficult to wire, 5G or Wi-Fi wireless communication technology is used for data transmission. The entire system has an adaptive frequency modulation function, which can dynamically adjust the transmission frequency according to environmental changes to ensure the stability of data transmission.

[0069] S2. Extract the corresponding characteristic parameters according to the electromagnetic signal;

[0070] Specifically, the characteristic parameters include: amplitude, frequency, and attenuation coefficient.

[0071] In a preferred embodiment of the present invention, the characteristic parameters corresponding to the electromagnetic signal are extracted based on the electromagnetic wave propagation model. The electromagnetic wave propagation model performs mathematical modeling on the electromagnetic signal through physical formulas to obtain the characteristic parameters of the electromagnetic signal, such as amplitude, frequency, attenuation constant, etc. The characteristic parameters are quantitative descriptions of the partial discharge signal and can be expressed as:

[0072]

[0073] Among them, A is the amplitude of the electromagnetic signal, f(t) is the time-varying frequency, φ(t) is the time-varying phase, τ is the attenuation coefficient of the electromagnetic signal, f is the actual frequency of the electromagnetic signal, f0 is the center frequency of the electromagnetic signal, σ f is the width of frequency filtering, α is the parameter controlling the noise attenuation rate, and t is the time.

[0074] S3. Input the characteristic parameters into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value for characterizing the fault risk level of the device to be monitored;

[0075] Specifically, inputting the characteristic parameters into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value for characterizing the fault risk level of the device to be monitored, including:

[0076] Inputting the characteristic parameters into a preset fuzzy logic control model to determine the value range of each of the characteristic parameters;

[0077] Creating a membership function for each of the characteristic parameters according to the characteristic parameters and the value range;

[0078] Performing fuzzy conversion on the characteristic parameters according to the membership function to obtain a fuzzy set for each of the characteristic parameters;

[0079] Converting the fuzzy set through a preset defuzzification method to obtain a first fault prediction value for characterizing the fault risk level of the device to be monitored.

[0080] In a preferred embodiment of the present invention, the first fault prediction value is used to represent the fault risk level of the device to be monitored, such as high, medium, or low. The process of fuzzification is actually to convert quantitative features (feature parameters) into fuzzy sets, so as to facilitate reasoning and analysis through a fuzzy logic control model (FLC). That is to say, the output of the electromagnetic wave propagation model (such as amplitude, frequency, attenuation constant) is the input of the fuzzy logic control model. Therefore, fuzzification is closely related to the electromagnetic wave propagation model. The purpose of fuzzification is to convert the quantitative features output by the electromagnetic wave model into fuzzy languages ("high", "medium", "low"), so that the fuzzy logic control model can process uncertain and fuzzy information and obtain the fault risk level, and obtain the first fault prediction value. First, input the feature parameters into a preset fuzzy logic control model, determine the value range of each feature parameter, and determine the value range of the feature parameters (such as amplitude, frequency, attenuation constant) of the electromagnetic signal according to historical data or theoretical analysis. For example, the range of amplitude may be from 0 to a certain maximum value (such as 0 to 10V), and the frequency range may be from 0Hz to 100kHz. Then, according to the feature parameters and the value range, create a membership function for each feature parameter, establish a membership function for each feature parameter, and describe the relationship between the input value and the fuzzy set. The membership function is usually a triangular or Gaussian function. For example, the amplitude may have a "high" membership function, whose value is close to 1 when it is greater than a certain threshold and close to 0 when it is lower than the threshold; similarly, membership functions for "high", "medium", and "low" also need to be created for frequency and attenuation constant. Next, perform a fuzzification conversion on the feature parameters according to the membership function to obtain a fuzzy set for each feature parameter. Finally, convert the fuzzy set through a preset defuzzification method to obtain the first fault prediction value used to characterize the fault risk level of the device to be monitored. For example, if the frequency of a certain signal is 50kHz, according to the membership function, the membership value of "medium" frequency may be 0.8, the membership value of "low" frequency may be 0.2, and the membership value of "high" frequency may be 0. Specifically, "high" means that the feature parameters (such as amplitude, frequency, attenuation constant) of the electromagnetic signal reach a certain threshold, indicating that the state of the device to be monitored is abnormal and the fault risk is relatively high, which may be a sign of problems such as device damage and aging. "Medium" means that the feature parameters of the electromagnetic signal are within the normal range but close to the threshold, indicating that the device has a certain risk and needs further monitoring. "Low" means that the feature parameters of the electromagnetic signal are within the normal working range of the device, the device state is good, and the risk is relatively low.

[0081] In a preferred embodiment of the present invention, the signal amplitude represents the intensity of the electromagnetic signal, which is usually directly related to the health status of the device to be monitored. A higher amplitude usually means a greater risk of failure of the device to be monitored. The frequency can indicate certain specific types of failures of the device to be monitored (such as insulation problems or contact wear). A change in frequency may represent a change in the operating state of the device to be monitored at a certain moment. The decay constant can reflect the aging status of the device to be monitored. A faster decay may mean that the device is aging or damaged. It is collected by a quantum sensor and converted into a first fault prediction value through fuzzy processing (such as the "high", "medium", "low" fuzzy sets mentioned above) for inferring the status of the device to be monitored. The fuzzy rules for fuzzy processing are based on previous research experiences. For example: if the amplitude is high and the frequency is low, the status of the device to be monitored is abnormal (possibly a contact fault); if the amplitude is medium and the decay constant is fast, there may be an aging problem with the status of the device to be monitored; if the amplitude is low and the frequency is medium, the status of the device to be monitored is good. The fuzzy set is converted into a crisp value, for example, the result of fuzzy processing is converted into a specific first fault prediction value through the weighted average method or the maximum membership method. The first fault prediction value is usually a numerical value representing the fault risk of the device to be monitored. For example, 0.2 may represent a low risk, and 0.8 may represent a high risk. The relationship with the fault risk level is mapped through a set threshold (such as "low risk", "medium risk", "high risk"). The advantage of the FLC model is its strong real-time performance, which can quickly respond to the changing device status and make a preliminary fault judgment.

[0082] S4. Input the characteristic parameters into a preset support vector machine model for fault identification to obtain a second fault prediction value for characterizing the fault probability of the device to be monitored;

[0083] Specifically, the training of the support vector machine model includes:

[0084] Obtain the historical characteristic parameters, historical fault probabilities, and initial model parameters of the electrical equipment;

[0085] Input the historical characteristic parameters and the current model parameters into the support vector machine model to be trained for training to obtain a predicted fault probability;

[0086] Calculate the loss value of a preset loss function according to the historical fault probability and the predicted fault probability; and determine whether the loss value converges;

[0087] In the case where the loss value converges, obtain the trained support vector machine model;

[0088] In the case where the loss value does not converge, update the current model parameters according to the loss value, and use the updated current model parameters as the current model parameters for the next training; wherein, the current model parameters for the first training are the initial model parameters.

[0089] Preferably, the support vector machine model (SVM) is a supervised learning model based on machine learning and is commonly used in classification and regression problems. In fault diagnosis, the SVM trains a classifier by learning historical feature parameters in historical electromagnetic signals, such as signal amplitude, frequency, etc., to determine whether a device has a fault. Through the trained classification model (support vector machine model), the real-time collected electromagnetic signals are compared with historical data to predict whether the device to be monitored has a fault, and a second fault prediction value is output. The larger the value, the higher the probability of the device having a fault.

[0090] S5. Determine whether the device to be monitored has a partial discharge fault according to the first fault prediction value, the second fault prediction value, and a preset weight;

[0091] Specifically, determining whether the device to be monitored has a partial discharge fault according to the first fault prediction value, the second fault prediction value, and a preset weight includes:

[0092] Calculate a judgment value for the device to be monitored according to the first fault prediction value, the second fault prediction value, and a preset weight;

[0093] Compare the judgment value with a preset threshold;

[0094] In the case where the judgment value is greater than or equal to the preset threshold, determine that the device to be monitored has a partial discharge fault;

[0095] In the case where the judgment value is less than the preset threshold, determine that the device to be monitored has not had a partial discharge fault.

[0096] In a preferred embodiment of the present invention, if the output of the FLC shows "high fault risk" and the fault prediction value of the SVM is greater than a certain threshold (such as 0.7), it can be determined that the equipment has a partial discharge fault. If the output of the FLC is "medium or low fault risk", even if the prediction value of the SVM is relatively high (such as greater than 0.5), it can still be determined that the equipment has not failed, but is only in a potential risk state. When a power transformer is monitoring its electromagnetic signals, the amplitude, frequency, and attenuation constant of the electromagnetic signals of the transformer are collected through a quantum sensor, and these data are input into the FLC model and the SVM model. The results obtained by analyzing the FLC model show that the amplitude of the equipment is high, the frequency is low, and the attenuation constant is relatively fast. Therefore, the FLC model gives a high fault risk level, and the first fault prediction value is 0.8. The second fault prediction value obtained by training the SVM model with historical data is 0.9, indicating a high probability of partial discharge fault in the equipment. In the comprehensive judgment stage, assuming that the weight of the FLC model is 0.4 and the weight of the SVM model is 0.6, the judgment value of the final comprehensive judgment result is:

[0097]

[0098] Since the final fault prediction value is greater than 0.8, the equipment is determined to have a partial discharge fault and further maintenance or detection measures need to be taken.

[0099] S6. If so, generate a detection command so that corresponding detection measures are generated according to the detection command;

[0100] Specifically, the equipment to be monitored is determined to have a partial discharge fault, and further maintenance or detection measures need to be taken.

[0101] Schematically, after determining that the equipment to be monitored has a partial discharge fault, it is judged whether the equipment to be monitored has aging of insulating materials according to the characteristic parameters;

[0102] In the case where the attenuation coefficient is less than a preset attenuation coefficient threshold and the amplitude of the frequency is greater than a preset amplitude, it is judged that the equipment to be monitored has aging of insulating materials. When the frequency and attenuation constant of the partial discharge signal are abnormally identified as aging of insulating materials, an aging maintenance warning is sent to the operation and maintenance personnel.

[0103] In a preferred embodiment of the present invention, after it is determined that a partial discharge fault has occurred in the device to be monitored, a fluctuation image is constructed based on the characteristic parameters for visual display. The frequency of the electromagnetic signal reflects the time period during which the discharge phenomenon occurs. Under the normal operating condition of the electrical device, the discharge frequency should be within a fixed range. If the frequency exceeds this normal range, it is generally considered that there is an abnormality. For example: when the frequency is higher than the normal range, it may indicate that a relatively intense partial discharge has occurred at the electrical contact point of the electrical device, which may be due to poor contact or device damage. When the frequency is lower than the normal range, it may indicate that the insulating material is aging or the fault occurs in some relatively slow discharge processes. The attenuation constant reflects the attenuation rate of the electromagnetic signal over time. The attenuation of the electromagnetic signal in a normal device should follow a stable law, and if the attenuation rate is too fast or too slow, it may indicate that there is a problem with the electrical device. An overly fast attenuation constant may indicate the aging phenomenon of the device, especially the aging of the insulating material, resulting in a more rapid attenuation of the discharge process. An overly slow attenuation constant may indicate that some problems in the device have not been fully resolved, or the attenuation of the signal is insufficient to reflect the normal electrical discharge phenomenon. As one of the characteristic parameters of the electrical device, the electromagnetic signal attenuation coefficient can help determine whether there is a partial discharge phenomenon in the device. For example, a relatively fast attenuation constant (or attenuation coefficient) may indicate that a partial discharge has occurred in the device, which is usually an initial sign of an electrical fault. After determining whether a partial discharge has occurred in the device, it is possible to further analyze whether the insulating material is aging based on characteristic parameters such as the attenuation constant, the amplitude, and the frequency of the signal. If the attenuation speed is fast (the attenuation constant is small) and the frequency of the discharge signal shows a significant abnormality, it can generally be inferred that the insulating material is aging. Under the condition that the partial discharge signal is normal, the abnormal changes in the attenuation coefficient and the frequency can be used as the basis for further diagnosis, indicating that the device may be aging or damaged. The aging of the insulating material is usually accompanied by the offset of the electromagnetic signal frequency and the change of the attenuation characteristics. Traditional monitoring methods are difficult to accurately identify this process. However, in this embodiment, due to the high sensitivity of the quantum sensor, these weak signal changes can be accurately captured to ensure that the aging phenomenon is detected in a timely manner.

[0104] S7. If not, evaluate the device to be monitored according to the characteristic parameters and the preset parameter thresholds to obtain the status evaluation result of the device to be monitored;

[0105] Specifically, evaluating the device to be monitored according to the characteristic parameters and the preset parameter thresholds to obtain the status evaluation result of the device to be monitored includes:

[0106] Calculate the health value of the device to be monitored according to the characteristic parameters, the preset characteristic weights, and the preset health decline parameters;

[0107] Make a judgment according to the health value and the preset parameter thresholds;

[0108] When the health value is less than or equal to the preset parameter threshold, it is determined that the status evaluation result of the device to be monitored is a non - partial discharge fault;

[0109] When the health value is greater than the preset parameter threshold, it is determined that the status evaluation result of the device to be monitored is normal.

[0110] Preferably, when it is recognized that there is no partial discharge in the device to be monitored, the health status of the device to be monitored is evaluated. According to the characteristic parameters, preset characteristic weights, and preset health decay parameters, the health value of the device to be monitored is calculated, expressed as:

[0111]

[0112] where w a ,w b ,w c are the weighting coefficients of the amplitude, frequency, and attenuation coefficient of the characteristic parameters respectively, δ is the preset health decay parameter, the larger the HI, the greater the health value of the device to be monitored. When HI is lower than the preset parameter threshold, it is recognized that there is a non - partial discharge fault. When the health value is greater than the preset parameter threshold, it is determined that the status evaluation result of the device to be monitored is normal. Non - partial discharge faults include contact faults and electrical short - circuits. When the signal amplitude and frequency are abnormal, it is recognized as a contact fault, and a fault detection is sent to the operation and maintenance personnel; when the signal amplitude is greater than the preset peak, the frequency is lower than the preset value, and the attenuation speed is fast, it is recognized that there is an electrical short - circuit and a short - circuit warning is sent to the operation and maintenance personnel. Contact faults often occur at the electrical contact points of switchgear. When the contacts are worn or poorly contacted, the electrical performance of the device will be seriously affected, and the abnormal change of the signal is an important signal of the contact fault. In traditional methods, when detecting contact faults, they often rely only on the external physical manifestations of the device, such as temperature or mechanical damage, which results in a lagging and inaccurate diagnosis. In this embodiment, by monitoring the signal combination with a signal amplitude greater than the preset peak, a frequency lower than the preset value, and a fast attenuation speed, the electrical short - circuit problem can be accurately identified. Electrical short - circuits are usually accompanied by sudden changes in current and voltage, which will generate intense electromagnetic wave changes in the device, and these changes can be accurately captured by quantum sensors.

[0113] In a preferred embodiment of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments. First, to verify the effectiveness of the electrical equipment status monitoring method based on quantum sensors, a high-voltage transformer is specifically selected as the test object to conduct experiments on partial discharge detection, fault diagnosis, and equipment health status assessment. In the experiment, a superconducting quantum interference device (SQUID) is selected as the quantum sensor to detect the magnetic field fluctuation changes inside the electrical equipment. The experimental equipment configuration is as follows: Test object: High-voltage transformer; Sensor: Superconducting quantum interference device (SQUID); Data acquisition: The system sampling rate is set to 150 MHz, and the quantum sensor collects the electromagnetic signals inside the transformer in real time, including the electromagnetic waves caused by partial discharge, the frequency, amplitude, and duration of the discharge. Decomposition of partial discharge signals: Based on the electromagnetic wave propagation model, the partial discharge signals are deconstructed to extract characteristic parameters such as time-varying frequency, amplitude, and attenuation coefficient. Detection of partial discharge faults: A diagnostic method combining fuzzy logic control (FLC) and support vector machine (SVM) is used to detect whether there are partial discharge faults. Equipment health assessment: According to the characteristics of the partial discharge signals, the health value (HI) of the equipment is calculated. When the HI value is lower than the threshold, it is determined that there is a non-partial discharge fault. Construction of fluctuation images: According to the changes in the partial discharge signals, fluctuation images are constructed for visual display to issue early warnings of insulation material aging, contact faults, or electrical short circuits to the operation and maintenance personnel. In this embodiment, the partial discharge signals of the transformer are collected and analyzed under different operating states, and data such as the frequency, amplitude, and attenuation speed of the signals are recorded. For each set of experimental data, the corresponding health value (HI) is calculated and compared with the traditional electrical equipment monitoring method to evaluate the advantages of the technology of the present invention.

[0114] Table 1 Comparison table of experimental data

[0115]

[0116] Referring to Table 1, for the normal operating state of the transformer, the health value (HI) fluctuates between 0.85 and 0.78, indicating the relative health of the equipment. As faults occur in the equipment, especially in the faulty transformer, the frequency and amplitude increase significantly (e.g., the frequency increases from 4700 Hz to 6000 Hz, and the amplitude increases from 100 mV to 200 mV), while the health value (HI) rises to 0.92. This shows that the occurrence of partial discharge faults leads to fluctuations in equipment performance, and the change in the health index is highly consistent with the fault characteristics. From the data comparison, the advantages of this implementation in partial discharge fault detection and equipment health assessment can be seen. Compared with traditional monitoring methods, by collecting electromagnetic signals through quantum sensors and combining advanced signal deconstruction and fault diagnosis algorithms, the present invention can accurately capture tiny equipment fault characteristics. For example, during the aging process of the transformer, changes in frequency and attenuation coefficient can be identified at an early stage, while these changes may be overlooked in traditional methods.

[0117] By comparing the experimental data, the key advantage of this embodiment lies in its real-time performance and accuracy during equipment operation. Traditional methods mostly rely on the monitoring of physical quantities such as temperature and pressure, and often require a long time of accumulation to detect faults. In contrast, this embodiment can immediately detect tiny changes in the equipment state through the high sensitivity of quantum sensors, thus greatly improving the timeliness and accuracy of fault diagnosis. In addition, the quantitative evaluation of the health value (HI) also provides a scientific basis for equipment management, making the evaluation of the equipment operating state more intuitive and reliable.

[0118] By implementing this embodiment, the electromagnetic signal of the device to be monitored is acquired; the corresponding characteristic parameters are extracted according to the electromagnetic signal; the characteristic parameters are input into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value for characterizing the fault risk level of the device to be monitored; the characteristic parameters are input into a preset support vector machine model for fault identification to obtain a second fault prediction value for characterizing the fault probability of the device to be monitored; it is judged whether the device to be monitored has a partial discharge fault according to the first fault prediction value, the second fault prediction value and a preset weight; if so, a detection command is generated so that corresponding detection measures are generated according to the detection command; if not, the device to be monitored is evaluated according to the characteristic parameters and a preset parameter threshold to obtain a status evaluation result of the device to be monitored; wherein, the electromagnetic signal is acquired by a quantum sensor; the characteristic parameters include: amplitude, frequency and attenuation coefficient. By acquiring the electromagnetic signal of the device to be monitored through a quantum sensor and obtaining characteristic parameters, the subtle monitoring of the electromagnetic signal of the device to be monitored can be realized, such as the subtle changes in amplitude, frequency and attenuation coefficient, and the faults of electrical equipment can be discovered in time; at the same time, the fuzzy logic control model is used to process uncertain problems, such as the boundary state with high amplitude but low frequency, the support vector machine model is used to identify the characteristic parameters, and the second fault prediction value for characterizing the fault probability of the device to be monitored is output. Multiple models are combined to judge whether a partial discharge fault occurs, and then the device to be monitored is further evaluated according to the characteristic parameters, thereby improving the accuracy of identifying the partial discharge fault of the device to be monitored. Thus, when the electromagnetic signal of the device to be monitored changes due to a fault, the state change of the device to be monitored can be discovered in time, providing a data basis for equipment operation and maintenance and improving the efficiency of operation and maintenance work.

[0119] See Figure 2 , which is a schematic structural diagram of an electrical equipment status monitoring device provided by an embodiment of the present invention, includes:

[0120] An electromagnetic signal acquisition module, configured to acquire the electromagnetic signal of the device to be monitored;

[0121] A signal feature extraction module, configured to extract the corresponding characteristic parameters according to the electromagnetic signal;

[0122] A first prediction module, configured to input the characteristic parameters into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value for characterizing the fault risk level of the device to be monitored;

[0123] A second prediction module, configured to input the characteristic parameters into a preset support vector machine model for fault identification to obtain a second fault prediction value for characterizing the fault probability of the device to be monitored;

[0124] The first judgment module is used to judge whether the device to be monitored has a partial discharge fault according to the first fault prediction value, the second fault prediction value and a preset weight;

[0125] The measure generation module is used to generate a detection command in the case of judging that the device to be monitored has a partial discharge fault, so as to generate a corresponding detection measure according to the detection command;

[0126] The second judgment module is used to evaluate the device to be monitored according to the characteristic parameters and a preset parameter threshold in the case of judging that the device to be monitored does not have a partial discharge fault, so as to obtain a state evaluation result of the device to be monitored;

[0127] Wherein, the electromagnetic signal is collected by a quantum sensor; the characteristic parameters include: amplitude, frequency and attenuation coefficient.

[0128] The present invention provides an electrical equipment status monitoring device, which acquires the electromagnetic signal of the device to be monitored through an electromagnetic signal acquisition module; in a signal feature extraction module, extracts the corresponding feature parameters according to the electromagnetic signal; through a first prediction module, inputs the feature parameters into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value for characterizing the fault risk level of the device to be monitored; through a second prediction module, inputs the feature parameters into a preset support vector machine model for fault identification to obtain a second fault prediction value for characterizing the fault probability of the device to be monitored; in a first judgment module, judges whether the device to be monitored has a partial discharge fault according to the first fault prediction value, the second fault prediction value and a preset weight; according to a measure generation module, generates a detection command in the case of judging that the device to be monitored has a partial discharge fault, so as to generate a corresponding detection measure according to the detection command; finally, through a second judgment module, in the case of judging that the device to be monitored does not have a partial discharge fault, evaluates the device to be monitored according to the feature parameters and a preset parameter threshold to obtain a status evaluation result of the device to be monitored; wherein, the electromagnetic signal is collected by a quantum sensor; the feature parameters include: amplitude, frequency and attenuation coefficient. By collecting the electromagnetic signal of the device to be monitored through a quantum sensor and obtaining feature parameters, it is possible to achieve fine monitoring of the electromagnetic signal of the device to be monitored, such as fine changes in amplitude, frequency and attenuation coefficient, and timely detect faults of electrical equipment; at the same time, the fuzzy logic control model is used to process uncertain problems, such as the boundary state with high amplitude but low frequency, the support vector machine model is used to identify the feature parameters, and the second fault prediction value representing the fault probability of the device to be monitored is output. Multiple models are combined to judge whether a partial discharge fault occurs, and then the device to be monitored is further evaluated according to the feature parameters, thereby improving the accuracy of identifying partial discharge faults of the device to be monitored. Thus, when the electromagnetic signal of the device to be monitored changes due to a fault, the status change of the device to be monitored can be timely detected, providing a data basis for equipment operation and maintenance and improving the efficiency of operation and maintenance work.

[0129] 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 attached drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that they have a communication connection, 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.

[0130] Those skilled in the art can clearly understand that for the sake of convenience and conciseness, 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 herein.

[0131] Another embodiment of the present invention further provides a terminal device, including 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 an electrical device state monitoring method as described in the above embodiments. The terminal device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The terminal device may include, but is not limited to, a processor and a memory.

[0132] The 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 terminal device, and connects various parts of the entire terminal device through various interfaces and lines.

[0133] The memory can be used to store the computer program. The processor realizes various functions of the terminal 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, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0134] Another embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the electrical equipment status monitoring method described in the above embodiment.

[0135] 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 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, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0136] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, 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 monitoring the state of electrical equipment, characterized in that: include: Obtain electromagnetic signals from the equipment to be monitored; Extracting corresponding characteristic parameters according to the electromagnetic signal; Inputting the characteristic parameter into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value for characterizing the fault risk level of the equipment to be monitored; Inputting the characteristic parameters into a preset support vector machine model for fault identification to obtain a second fault prediction value for characterizing the fault probability of the device to be monitored; Determining whether a partial discharge fault occurs in the device to be monitored according to the first fault prediction value, the second fault prediction value and a preset weight; If yes, a detection command is generated, so that corresponding detection measures are generated according to the detection command; If not, the device to be monitored is evaluated according to the characteristic parameters and the preset parameter threshold to obtain a status evaluation result of the device to be monitored; Wherein, the electromagnetic signal is collected by a quantum sensor; the characteristic parameters include: amplitude, frequency and attenuation coefficient.

2. The method for monitoring the state of electrical equipment according to claim 1, characterized in that: The characteristic parameters are input into a preset fuzzy logic control model for fuzzy conversion to obtain a first fault prediction value for characterizing the fault risk level of the equipment to be monitored, including: Inputting the characteristic parameters into a preset fuzzy logic control model to determine the value range of each characteristic parameter; Creating a membership function for each of the characteristic parameters according to the characteristic parameters and the value range; Fuzzy transform the characteristic parameters according to the membership function to obtain a fuzzy set of each characteristic parameter; The fuzzy set is transformed by a preset clarification method to obtain a first fault prediction value for characterizing the fault risk level of the equipment to be monitored.

3. The method for monitoring the state of electrical equipment according to claim 1, characterized in that: The training of the support vector machine model includes: Obtain historical characteristic parameters, historical failure probabilities and initial model parameters of electrical equipment; Inputting the historical characteristic parameters and the current model parameters into the support vector machine model to be trained to obtain the predicted failure probability; Calculate the loss value of a preset loss function according to the historical failure probability and the predicted failure probability; and determine whether the loss value shown converges; When the loss value converges, a trained support vector machine model is obtained; When the loss value has not converged, the current model parameters are updated according to the loss value, and the updated current model parameters are used as the current model parameters for the next training; wherein, the current model parameters for the first training are the initial model parameters.

4. The method for monitoring the state of electrical equipment according to claim 1, characterized in that: Judging whether a partial discharge fault occurs in the device to be monitored according to the first fault prediction value, the second fault prediction value and a preset weight includes: Calculate the judgment value of the device to be monitored according to the first fault prediction value, the second fault prediction value and the preset weight; Comparing the judgment value with a preset threshold; When the judgment value is greater than or equal to a preset threshold, it is determined that a partial discharge fault occurs in the equipment to be monitored; When the judgment value is less than the preset threshold, it is determined that no partial discharge fault occurs in the device to be monitored.

5. The method for monitoring the state of electrical equipment according to claim 1, characterized in that: The device to be monitored is evaluated according to the characteristic parameters and the preset parameter thresholds to obtain a status evaluation result of the device to be monitored, including: Calculating the health value of the device to be monitored according to the characteristic parameters, the preset characteristic weights and the preset health decay parameters; Making a judgment based on the health value and a preset parameter threshold; When the health value is less than or equal to a preset parameter threshold, determining that a state assessment result of the device to be monitored is that a non-partial discharge fault exists; When the health value is greater than a preset parameter threshold, it is determined that the status evaluation result of the device to be monitored is normal.

6. The method for monitoring the state of electrical equipment according to claim 1, characterized in that: Also includes: After determining that a partial discharge fault occurs in the equipment to be monitored, determining whether insulation material aging exists in the equipment to be monitored according to the characteristic parameters; When the attenuation coefficient is less than the preset attenuation coefficient threshold and the amplitude of the frequency is greater than the preset amplitude, it is determined that the insulation material of the monitored equipment is aged.

7. The method for monitoring the state of electrical equipment according to claim 1, characterized in that: Also includes: Determine the layout position of the quantum sensor according to the type of the device to be monitored; When the device to be monitored is a transformer, determining that the quantum sensor is arranged in a bushing of the transformer; When the device to be monitored is a switch device, determining that the quantum sensor is arranged at a switch contact; When the device to be monitored is a power line joint, determining the layout position of the quantum sensor at the power line joint, insulating material and device interface; A quantum sensor deployment plan is generated according to the deployment positions.

8. An electrical equipment status monitoring device, characterized in that: include: An electromagnetic signal acquisition module, used to acquire electromagnetic signals of the equipment to be monitored; A signal feature extraction module, used to extract corresponding feature parameters according to the electromagnetic signal; A first prediction module, used for inputting the characteristic parameter into a preset fuzzy logic control model for fuzzy conversion, and obtaining a first fault prediction value for characterizing the fault risk level of the equipment to be monitored; A second prediction module is used to input the characteristic parameters into a preset support vector machine model for fault identification, and obtain a second fault prediction value for characterizing the failure probability of the device to be monitored; A first judgment module, used for judging whether a partial discharge fault occurs in the device to be monitored according to the first fault prediction value, the second fault prediction value and a preset weight; A measure generation module is used to generate a detection command when it is determined that a partial discharge fault occurs in the device to be monitored, so as to generate corresponding detection measures according to the detection command; A second judgment module is used to evaluate the device to be monitored according to the characteristic parameters and the preset parameter threshold value to obtain a status evaluation result of the device to be monitored when it is determined that no partial discharge fault occurs in the device to be monitored; Wherein, the electromagnetic signal is collected by a quantum sensor; the characteristic parameters include: amplitude, frequency and attenuation coefficient.

9. A terminal 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 when the processor executes the computer program, an electrical equipment status monitoring method as claimed in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute an electrical equipment status monitoring method as claimed in any one of claims 1 to 7.

Images