Electrical equipment monitoring method, system and equipment based on deep learning and medium

Abstract

The invention discloses an electrical equipment monitoring method and system based on deep learning, equipment and a medium, and the method comprises the following steps: S1, obtaining the historical data of the electrical equipment, and carrying out the preprocessing of the historical data, and obtaining the preprocessed data; s2, constructing a deep learning model based on the preprocessed data; s3, acquiring first operation data of the electrical equipment at the first time and second operation data of the electrical equipment at the second time; s4, inputting the first operation data into the deep learning model to obtain a continuous first comparison data graph; s5, obtaining a comparison point corresponding to the second time according to the first comparison data graph; s6, obtaining a comparison interval of the second operation data according to the comparison point; and S7, comparing the second operation data with the comparison interval, and if the second operation data is not located in the comparison interval, outputting an alarm signal, thereby realizing accurate prevention of the fault of the electrical equipment.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of electrical equipment management, and particularly relates to an electrical equipment monitoring method, system, device and medium based on deep learning. BACKGROUND

[0002] Stable operation of electrical equipment is the core guarantee of industrial production, but its operation data is affected by multiple factors such as environment, service life and load, and the traditional monitoring method has obvious limitations: only relying on existing data threshold judgment, it cannot early warn potential faults, and is easy to lead to production line paralysis; the maintenance cost is high, if the early warning is excessive, the maintenance workload will be increased, if the early warning is insufficient, safety accidents may be caused; the device aging law in the historical data is not fully mined, and it is difficult to adapt to the dynamic changes of the whole life cycle of the equipment; the same batch of equipment may have common defects, but the traditional method cannot be monitored in linkage to avoid systemic risks.

[0003] Therefore, it is urgent to design an electrical equipment monitoring method, system, device and medium based on deep learning to solve the above problems. SUMMARY

[0004] The purpose of the application is to provide an electrical equipment monitoring method, system, device and medium based on deep learning, which realizes accurate prevention of faults of electrical equipment.

[0005] To achieve the above purpose, the technical scheme of the electrical equipment monitoring method, system, device and medium based on deep learning is as follows: An electrical equipment monitoring method based on deep learning, comprising the following steps: S1, obtaining historical data of electrical equipment, and preprocessing the historical data to obtain preprocessed data; S2, constructing a deep learning model based on the preprocessed data; S3, obtaining first running data of the electrical equipment at a first time and second running data of the electrical equipment at a second time; S4, inputting the first running data into the deep learning model to obtain continuous first comparison data graph; S5, obtaining a comparison point corresponding to the second time according to the first comparison data graph; S6, obtaining a comparison interval of the second running data according to the comparison point; S7, comparing the second running data with the comparison interval, and outputting an alarm signal if the second running data is not located in the comparison interval.

[0006] Further, preprocessing the historical data to obtain preprocessed data comprises the following steps: Filtering the historical data by using one or a combination of wavelet transform, smoothing filtering, low-pass filtering, high-pass filtering and band-pass filtering to obtain filtered data; Performing denoising and normalization processing on the filtered data by using a time series correlation and density clustering algorithm to obtain preprocessed data.

[0007] Further, the comparison interval of the second running data is obtained according to the comparison point, including the following steps: Setting the upper distance and the lower distance corresponding to the ordinate of the comparison point, obtaining the historical data of the third time to obtain the third running data; Judging whether the second running data is the same as the historical data corresponding to the second time, if not, increasing the upper distance and the lower distance by a first proportion, if the same, judging whether the difference between the second running data and the third running data is within the data fluctuation range, if exceeding the data fluctuation range, reducing the upper distance and the lower distance by a first proportion; The comparison interval is obtained by subtracting the adjusted lower distance from the ordinate of the comparison point to the ordinate of the comparison point plus the adjusted upper distance.

[0008] Further, the comparison interval of the second running data is obtained according to the comparison point, including the following steps: Setting the upper distance and the lower distance corresponding to the ordinate of the comparison point, obtaining the historical data of the third time to obtain the third running data; Judging whether the first running data is the same as the historical data corresponding to the first time, if not, increasing the upper distance and the lower distance by a first proportion, if the same, judging whether the difference between the first running data and the third running data is within the data fluctuation range, if exceeding the data fluctuation range, reducing the upper distance and the lower distance by a first proportion; The comparison interval is obtained by subtracting the adjusted lower distance from the ordinate of the comparison point to the ordinate of the comparison point plus the adjusted upper distance.

[0009] Further, the electrical equipment monitoring method based on deep learning further includes the following steps: S8, after outputting the alarm signal, obtaining the electrical equipment outputting the alarm signal, obtaining the electrical equipment of the same batch as the electrical equipment outputting the alarm signal, and detecting the electrical equipment of the same batch.

[0010] Further, the detection of the electrical equipment of the same batch includes the following steps: S81, set the first preset number of times, and after the upper distance and the lower distance of the same batch of electrical equipment are reduced by the second ratio, the same batch of electrical equipment jumps to step S1, and steps S1-S7 are run, and if the alarm signal is output again, steps S81 are repeated until the number of jumps exceeds the first preset number of times, and the same batch of electrical equipment all output alarm signals.

[0011] Further, when the number of alarms of the electrical equipment in a unit time exceeds the third preset number of times, the fourth running data corresponding to the second time in the first comparison data graph of the electrical equipment outputting the alarm signal is configured as the fifth running data, the average value of the fourth running data and the second running data is configured as the fifth running data, the fourth running data in the first comparison data graph of the electrical equipment outputting the alarm signal is replaced by the fifth running data, and the second comparison data graph is generated.

[0012] An electrical equipment monitoring system based on deep learning adopts the above-mentioned electrical equipment monitoring method based on deep learning, and comprises a preprocessing module, a deep learning module, an acquisition module, an input module and a judgment module. The preprocessing module is used for preprocessing historical data to obtain preprocessed data. The deep learning module is used for constructing a deep learning model according to the preprocessed data. The acquisition module is used for acquiring first running data of the electrical equipment at a first time and second running data of the electrical equipment at a second time. The input module is used for inputting the first running data into the deep learning model to obtain a continuous first comparison data graph. The judgment module is used for obtaining a comparison point corresponding to the second time and a comparison interval of the second running data according to the first comparison data graph, comparing the second running data with the comparison interval, and outputting an alarm signal if the second running data is not located in the comparison interval.

[0013] An apparatus comprising at least one processor and at least one memory; The at least one memory is used for storing a computer program. The at least one processor is used for executing the computer program to realize the above-mentioned method.

[0014] A computer readable storage medium for storing executable instructions, the executable instructions being executed by a processor to cause the processor to realize the above-mentioned method.

[0015] The electrical equipment monitoring method, system, device and medium based on deep learning have the following advantages: 1. Strong prediction: predict the data trend of electrical equipment through deep learning model, and discover potential faults of electrical equipment in advance combined with dynamic interval; 2. Closed-loop prevention: form a complete closed loop from alarm to maintenance and effect verification to reduce fault recurrence rate; 3. Batch risk control: linkage monitoring and prevention of devices in the same batch to avoid systemic risks; 4. Maintenance cost optimization: dynamically adjust the maintenance cycle of electrical equipment based on data to reduce over-maintenance and downtime loss of electrical equipment; 5. High universality: suitable for multiple types of electrical equipment, and customized models are used to adapt to the fault characteristics of different equipment. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a flowchart of the electrical equipment monitoring method based on deep learning of the present application; Figure 2 is a confusion matrix heat map of predicted values and true values under different working conditions of the present application; Figure 3 is an accuracy diagnosis diagram of the learning curve of the electrical equipment monitoring method based on deep learning of the present application. DETAILED DESCRIPTION

[0017] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0018] Those skilled in the art can understand that although some embodiments herein include certain features rather than other features included in other embodiments, the combination of features of different embodiments means to be within the scope of the present application and form different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0019] The following Figures 1 to 3 describes a deep learning-based electrical equipment monitoring method, system, device and medium.

[0020] A deep learning-based electrical equipment monitoring method, as shown in Figure 1 includes the following steps: S1, obtaining historical data of electrical equipment, and pre-processing the historical data to obtain pre-processed data; S2, constructing a deep learning model based on the preprocessed data; S3, obtaining first running data of the electrical equipment at a first time and second running data of the electrical equipment at a second time; S4, inputting the first running data into the deep learning model to obtain a continuous first comparison data graph; S5, obtaining a comparison point corresponding to the second time according to the first comparison data graph; S6, obtaining a comparison interval of the second running data according to the comparison point; S7, comparing the second running data with the comparison interval, and outputting an alarm signal if the second running data is not located in the comparison interval.

[0021] Specifically, first running data of the electrical equipment at a first time (t1) (such as the temperature of the high-voltage winding of the dry-type transformer at t1, which is 88.3°C) and second running data of the electrical equipment at a second time (t2, t2>t1) (such as the actual temperature at t2, which is 91.4°C) are obtained. Historical data of the electrical equipment in the whole life cycle (including parameter records and maintenance logs in normal operation, before failure, and failure state) are taken as a training set to construct a deep learning model (such as an improved LSTM network). The deep learning model predicts the trend of the running data of the electrical equipment in the interval from t1 to t2 by learning the time sequence characteristics of the data. The first running data is input into the trained model to generate a continuous first comparison data graph (the horizontal axis is time, and the vertical axis is the predicted value of the running data). The vertical coordinate point corresponding to the second time t2 is defined as a “comparison point” (such as the model predicting that the temperature at t2 is 90°C). The comparison interval is generated according to the comparison point, and an alarm signal is output if the second running data exceeds the interval range. Through trend analysis of the first comparison data graph, if it is found that the second running data deviates from the comparison point (such as the predicted temperature value being 90°C, and the actual value rising from 88°C to 91°C) although it is within the interval, the maintenance personnel can be prompted to check the heat dissipation system of the equipment in advance to avoid the expansion of the failure.

[0022] Specifically, the deep learning model and the historical data are used to know the data change rule of the corresponding electrical equipment, and the first comparison data graph is generated in combination with the first running data. Comparing the relatively continuous first comparison data graph with the second running data can know whether the electrical equipment should alarm through the difference between the predicted data and the actual data, so that the staff can maintain multiple electrical equipment according to the existing data.

[0023] Specifically, the application is based on historical data, integrates electrical equipment parameter information, inspection data and other information materials, realizes the integrated supervision of data integration transmission, centralized display, monitoring and early warning, fault diagnosis of electrical equipment, and comprehensively masters the state information of key electrical equipment.

[0024] It should be noted that due to various complex factors such as environment and quantity, although the trend of operation data of each device changes over time is roughly constant and can be simulated by a deep learning model, it needs to be calibrated with the first time and first operation data. That is, when the deep learning model outputs the first comparison data graph, it is through the calibration and fine-tuning of the first time and first operation data that the first comparison data graph is generated. Relatively speaking, electrical equipment involves service life and calendar life, and a large number of historical data samples are needed to generate a deep learning model as a prediction model. This deep learning model can be an existing technology, such as a language model (CHATGPT) based on artificial intelligence technology or other deep learning algorithms, which will not be described in detail.

[0025] Specifically, the line or broken line of the first comparison data graph of the deep learning model can be derived from any one of artificial intelligence, neural networks, prediction algorithms according to the first operation data of the first time, and the existing technology will not be described in detail. That is, through the first time and the first operation data, a corresponding first comparison data graph and a line composed of time and operation data can be output by the deep learning model. This line should be a continuous line with an arc, or a relatively dense broken line graph.

[0026] Preferably, the first time is earlier than the second time, and the first time can be the time of the last week, and the second time can be the time of the present time, and the alarm signal reflects whether the electrical equipment has a fault at the present time.

[0027] Specifically, the electrical equipment includes switch cabinets, transformers, uninterruptible power supplies (UPS), and operation data includes temperature, voltage, current, insulation resistance, vibration frequency and other key parameters.

[0028] Further, the pre-processing of the historical data to obtain the pre-processed data includes the following steps: Wavelet transform, smoothing filtering, low-pass filtering, high-pass filtering and band-pass filtering are used to filter the historical data to obtain filtered data; The filtered data is denoised and normalized by using a time series correlation and density clustering algorithm to obtain pre-processed data.

[0029] Specifically, to meet the needs of monitoring alarm, fault diagnosis, state evaluation, data mining and other advanced applications, the collected data should be screened and preprocessed for usefulness and effectiveness, and data filtering is performed to filter out obviously erroneous data. For different types of data, wavelet transform, smoothing filter, low-pass filter, high-pass filter, band-pass filter and other digital filtering methods are selected to filter the data, and interpolation algorithm is used to supplement the missing data. A data cleaning method based on time series correlation and density clustering algorithm is used to preprocess the historical data in the database, identify abnormal data in the historical data, solve the missing part of the abnormal data and fill it in, and complete the cleaning of the abnormal data.

[0030] Optionally, a plurality of historical data of the same electrical equipment are used to draw a comparison data reference graph, the horizontal coordinate of the comparison data reference graph is time, the vertical coordinate is operation data, and the comparison data reference graph is not a continuous line; and the above comparison data reference graph is drawn into a relatively continuous line by using a deep learning model, and the discontinuous points in the line drawing process are deleted to generate a first comparison data graph.

[0031] Optionally, historical data of the electrical equipment are obtained, the historical data including operation parameters, fault records and maintenance records of the electrical equipment; the historical data are preprocessed, including data cleaning, data normalization and data division; a deep learning model is constructed, the deep learning model including an input layer, a hidden layer and an output layer, wherein the input layer is used to receive the preprocessed data, the hidden layer is used to extract data features, and the output layer is used to output prediction results; the deep learning model is trained using the preprocessed data to obtain a trained deep learning model; and the trained deep learning model is used to predict the operation state of the electrical equipment to construct a first comparison data graph.

[0032] Further, a comparison interval of the second operation data is obtained according to the comparison point, including the following steps: The upper distance and the lower distance corresponding to the vertical coordinate of the comparison point are set, historical data of a third time are obtained, and third operation data are obtained; It is judged whether the second operation data and the historical data corresponding to the second time are the same, if not, the upper distance and the lower distance are increased by a first proportion, if the same, it is judged whether the difference between the second operation data and the third operation data is within a data fluctuation range, if the difference exceeds the data fluctuation range, the upper distance and the lower distance are decreased by the first proportion; The vertical coordinate of the comparison point minus the adjusted lower distance to the vertical coordinate of the comparison point plus the adjusted upper distance is obtained as the comparison interval.

[0033] Specifically, by judging the same and different of the second time historical data and the second running data, when the second time historical data is different from the second running data, it can be known that the point on the first comparison data graph of the second time is the point on the historical data as the reference or the point predicted by the deep learning model. If it is the predicted point, the upper distance and the lower distance are increased to reduce the influence of the error of the predicted data. When the second time historical data is the same as the second running data, it is necessary to judge whether the third time historical data of the third running data is similar to the second running data, that is, although the horizontal coordinates are the same, the vertical coordinates exceed the data fluctuation range; it may be that the deep learning algorithm generates the first comparison data graph by excluding the third time historical data in order to generate a more continuous first comparison data graph, so as to produce the above-mentioned dissimilarity problem. Then, this problem point should be paid special attention to, and the upper distance and the lower distance of the second time are reduced to increase the alarm probability of the second time through a smaller comparison interval, so as to make the operator pay more attention to this point.

[0034] Specifically, if the number of alarm signals of the second time exceeds the third preset number of times, the historical data of the second time can be excluded to avoid the repeatedly influence of the excluded historical data on the data source of the deep learning model, and ensure the accuracy of the deep learning model. Preferably, the third preset number of times is 5 times.

[0035] Optionally, the first proportion is 1% to 30%, and preferably, the first proportion is 10%.

[0036] Optionally, the data fluctuation range is 5% to 30% of the second running data, and preferably, 10% of the second running data.

[0037] Specifically, the database pre-stores the upper distance 5℃ and the lower distance 5℃ corresponding to the comparison point (90℃). If there is no t2 time record in the historical data, the upper and lower distances are increased by the first proportion (10%), and the interval is expanded to 84.5℃ to 95.5℃; if there is t2 time historical data (such as 88℃), and the second running data (91.4℃) is different from the third time historical data by more than 10% (i.e. dissimilar), the upper and lower distances are reduced by 10%, and the interval is reduced to 85.5℃ to 94.5℃.

[0038] Specifically, for the time nodes (such as the afternoon of the high temperature season) prone to faults in the historical data, the monitoring sensitivity is improved by narrowing the comparison interval to capture abnormalities (such as abnormal rise of winding temperature) in advance, and trigger preventive inspection (such as cleaning the heat sink and tightening the terminal).

[0039] Further, the comparison interval of the second running data is obtained according to the comparison point, comprising the following steps: Set the upper distance and the lower distance corresponding to the ordinate of the contrast point, obtain the historical data of the third time, and obtain the third running data; Determine whether the first running data and the historical data corresponding to the first time are the same, if not, increase the upper distance and the lower distance by a first proportion, if the same, determine whether the difference between the first running data and the third running data is within the data fluctuation range, if it exceeds the data fluctuation range, reduce the upper distance and the lower distance by a first proportion; The ordinate of the contrast point minus the adjusted lower distance to the ordinate of the contrast point plus the adjusted upper distance, to obtain the contrast interval.

[0040] Specifically, by judging the same and not the same of the historical data of the first time and the first running data, when the historical data of the first time and the first running data are not the same, it can be known that the point on the first contrast data graph of the first time is the point on the historical data as the reference, or the point predicted by the deep learning model. If it is a predicted point, increase the upper distance and the lower distance to expand the inclusive degree of the point to the prediction algorithm, so as to reduce the influence caused by the error of the prediction data. When the historical data of the first time and the first running data are the same, it is necessary to determine whether the third running data of the historical data of the third time is similar to the first running data, that is, although the abscissa is the same, the ordinate exceeds the data fluctuation range; It may be that the deep learning algorithm has excluded this historical data of the third time to generate the first contrast data graph, so as to generate the above problem of not similar. Then, this problem point should be paid special attention to, and the upper distance and the lower distance of this first time are reduced, so as to increase the alarm probability of this first time through a smaller contrast interval, so that the operator pays more attention to this point.

[0041] Further, the electrical equipment monitoring method based on deep learning further comprises the following steps: S8, after outputting the alarm signal, obtaining the electrical equipment outputting the alarm signal, obtaining the electrical equipment of the same batch as the electrical equipment outputting the alarm signal, and detecting the electrical equipment of the same batch.

[0042] Further, detecting the electrical equipment of the same batch comprises the following steps: S81, set a first preset number, reduce the upper distance and the lower distance of the electrical equipment of the same batch by a second proportion, and the electrical equipment of the same batch jumps to step S1, runs steps S1-S7, if the alarm signal is output again, repeat step S81 until the jump number exceeds the first preset number, and the electrical equipment of the same batch outputs the alarm signal.

[0043] Further, a second preset number is set, when the number of alarms of the electrical equipment in a unit time exceeds the second preset number, the running data corresponding to the second time in the first comparison data graph of the electrical equipment outputting the alarm signal is configured as fourth running data, the average value of the fourth running data and the second running data is configured as fifth running data, the fourth running data in the first comparison data graph of the electrical equipment outputting the alarm signal is replaced by the fifth running data, and a second comparison data graph is generated.

[0044] When the dry-type transformer winding temperature (such as 91.4℃) exceeds the interval (85.5℃~94.5℃), an alarm signal is output, and the fault module (such as the winding) is located. At this time, preventive measures are started: Immediately stop and check the insulation state of the module (such as partial discharge detection); Based on the maintenance records of the same type of faults in the historical data, an automatic repair scheme (such as replacing the insulation material) is pushed.

[0045] The up / down distance (such as adjusted to 4.75℃) of the same batch of transformers is reduced by a second proportion (such as 5%), and the model construction and monitoring are re-executed (jump to the step of claim 1); if the number of jumps exceeds a first preset number (such as 3), the same batch of devices are all alarmed, and batch preventive maintenance (such as replacing all the aged components) is started.

[0046] If the number of alarms in 1 hour exceeds a second preset number (such as 5), the average value (90.7℃) of the predicted value (90℃) and the actual value (91.4℃) of the second time in the first comparison data graph is replaced with the original historical data, the model is retrained, and a second comparison data graph is generated. Optimize the adaptability of the model to the aging trend of the equipment and prevent the recurrence of the same type of faults in the long term.

[0047] Optionally, the second proportion is 1%~50%, and preferably, the second proportion is 5%. The second proportion is less than the first proportion, which improves the screening probability of the same batch of electrical equipment.

[0048] An electrical equipment monitoring system based on deep learning adopts the above-mentioned electrical equipment monitoring method based on deep learning, comprising: a preprocessing module, a deep learning module, an acquisition module, an input module and a judgment module; The preprocessing module is used for preprocessing the historical data to obtain preprocessed data; The deep learning module is used for constructing a deep learning model according to the preprocessed data; The acquisition module is used for acquiring first running data of a first time of the electrical equipment and second running data of a second time of the electrical equipment; The input module is used for inputting the first running data into the deep learning model to obtain a continuous first comparison data graph; The judgment module is configured to obtain a comparison point corresponding to the second time according to the first comparison data graph and a comparison interval of the second operation data, compare the second operation data with the comparison interval, and output an alarm signal if the second operation data is not located in the comparison interval.

[0049] Further, the electrical equipment monitoring system based on deep learning further comprises a prevention module configured to automatically generate a maintenance work order based on the alarm signal, sort the health degrees of the electrical equipment of the same batch, preferentially maintain the electrical equipment with higher risks, and record the effect of the preventive measures.

[0050] An apparatus comprising at least one processor and at least one memory; The at least one memory is configured to store a computer program; The at least one processor is configured to execute the computer program to implement the method described above.

[0051] The apparatus according to the embodiments of the present application includes a processor which can perform various appropriate actions and processes according to a program stored in a read only memory (ROM) or a program loaded from a storage section into a random access memory (RAM). The processor may, for example, include a general purpose microprocessor (such as a CPU), an instruction set processor, and / or a related chipset, and / or a special purpose microprocessor (such as an application specific integrated circuit (ASIC)), and so on. The processor can also include an on-board memory for cache use. The processor can include a single processing unit or a plurality of processing units for performing different actions of the method processes according to the embodiments of the present application.

[0052] In the RAM, various programs and data required for operation of the electronic device are stored. The processor, the ROM, and the RAM are connected to each other through a bus. The processor performs various operations of the method processes according to the embodiments of the present application by executing the programs in the ROM and / or the RAM. It is noted that the programs can also be stored in one or more memories other than the ROM and the RAM. The processor can also perform various operations of the method processes according to the embodiments of the present application by executing the programs stored in the one or more memories.

[0053] According to embodiments of the present invention, the electronic device may further include an input / output (I / O) interface, which is also connected to a bus. The system may also include one or more of the following components connected to the I / O interface: an input section including a keyboard, mouse, etc.; an output section including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section including a hard disk, etc.; and a communication section including a network interface card such as a LAN card, modem, etc. The communication section performs communication processing via a network such as the Internet. A drive is also connected to the I / O interface as needed. Removable media, such as magnetic disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed so that computer programs read from them can be installed into the storage section as needed.

[0054] According to embodiments of the present invention, the method flow according to embodiments of the present invention can be implemented as a computer software program. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the method shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a processor, it performs the functions defined above in the system of the embodiments of the present invention. According to embodiments of the present invention, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0055] The present invention also provides a computer-readable storage medium for storing executable instructions that, when executed by a processor, cause the processor to implement the above-described method.

[0056] The computer-readable storage medium may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more executable instructions, which, when executed, implement the method according to the embodiments of the present invention.

[0057] According to embodiments of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium. Examples include, but are not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0058] Example 1 Taking dry-type transformer fault diagnosis as an example, the first step is to distinguish the data differences between normal operation and various faults: (1) Normal operation Normal operation refers to the condition where the temperature of all components of a dry-type transformer is within the design range and the temperature changes smoothly without drastic fluctuations or temperature rises exceeding the limits. This state meets the requirements of rated capacity, rated voltage, and rated frequency, with stable electrical parameters and insulation resistance meeting the specifications.

[0059] Based on the finite element method, the temperature distribution of various parts inside the dry-type transformer during normal operation is shown in the figure, including the winding temperature, core temperature, and air temperature. By comparing the temperature values ​​of different parts, it can be seen that the heat distribution of the dry-type transformer is relatively uniform. The internal temperature changes relatively gradually over a certain period of time. Meanwhile, the temperature of the high-voltage winding is higher than that of the low-voltage winding, and the core also has relatively high heat due to the dense magnetic circuit.

[0060] When a dry-type transformer is in normal operation, its electrical parameters, including rated capacity, rated voltage, and rated power, are stable. At the same time, the insulation condition can be indirectly monitored and evaluated through test methods such as resistance testing. The insulation resistance of a dry-type transformer should meet the requirements during normal operation.

[0061] In the simulation method for normal operation, five load voltage scenarios were set: 1V, 0.95V, 0.98V, 1.02V, and 1.05V. The heat source values ​​were set to the values ​​of core loss and winding loss, and the temperature field under these scenarios was simulated. 36 temperature data points were obtained for each scenario, for a total of 180 temperature data points.

[0062] (2) Overheating of the iron core When the core of a dry-type transformer overheats, it will emit noticeable noise. The most obvious characteristic is that the internal temperature of the transformer rises due to core overheating. Normally, the operating temperature of a transformer should be within the specified range. If the core temperature exceeds the normal range, it will lead to a decline in transformer performance. At the same time, core overheating will generate a large amount of heat and gas. If this heat is not dissipated in time, it will increase the risk of safety accidents such as transformer fires and explosions. When overheating occurs inside the core, it will cause localized temperature increases. In this case, the temperature in a localized area of ​​the core will be significantly higher than the temperature of the surrounding area, and the temperature distribution will be uneven.

[0063] Overheating of the iron core will also cause the winding temperature to rise. The degree of temperature rise depends on the extent and location of the iron core overheating. When the overheating range of the iron core is large, the temperature rise of the winding will also be more significant. Iron core overheating will cause rapid temperature rise and fluctuations because the temperature rises very quickly when the iron core is overheated, resulting in large temperature fluctuations. When the iron core temperature exceeds the design value, the dry-type transformer will be in a dangerous operating state.

[0064] The method of applying a heat source to the core surface is used to simulate the overheating operation of the dry-type transformer core. Simulations were performed on the core sections of the three-phase transformer, obtaining 180 temperature data points.

[0065] (3) Partial discharge Partial discharge is a discharge phenomenon occurring in specific areas of an insulating medium, usually caused by defects in the insulating material, excessively high electric field strength, or other reasons. Partial discharge generates noise of specific frequencies and amplitudes, significantly increasing the noise level of a transformer. It also generates a large amount of heat, causing a rise in the local temperature of the transformer. Furthermore, partial discharge can form discharge channels inside the transformer, affecting its normal operation and increasing maintenance difficulty and costs. Additionally, the number and intensity of partial discharges tend to gradually increase over time.

[0066] During partial discharge, the discharge area experiences a localized temperature rise due to energy release. This temperature rise can be monitored and measured in real time using equipment such as infrared thermal imagers. Simultaneously, the transformer's temperature sensor can also detect temperature changes and transmit the data to a monitoring system for recording and analysis.

[0067] By analyzing the temperature data during partial discharge, the location and intensity of the discharge can be determined, and the insulation condition of the transformer can be assessed. At this time, due to the thermal resistance limitation of the insulation material, the temperature rises rapidly, and local temperature anomalies are significant.

[0068] In the simulation method for partial discharge, a segmented approach was used, dividing the high-voltage winding into four equal parts on a plane with a cutting surface, and applying an additional heat source at one-quarter of the winding. 180 data points were obtained by simulating the heat released during partial discharge in a dry-type transformer.

[0069] (4) Winding short circuit A winding short circuit refers to the phenomenon where a short circuit occurs between windings due to factors such as aging or damage to the insulation material, leading to a decrease in insulation performance and strength. A winding short circuit can cause excessive local current, generating a large amount of heat and causing a rapid increase in the transformer's internal temperature. It can also cause instantaneous changes in the transformer's output voltage and current, severely impacting electrical performance and potentially leading to serious malfunctions. Furthermore, it can cause drastic changes in current and magnetic field, generating noise and vibration, which is usually one of the most noticeable symptoms during operation. Under winding short-circuit conditions, the temperature data characterization and analysis inside a dry-type transformer exhibits the following characteristics: 1) Winding temperature rise: Due to the excessive current caused by a short circuit in the winding, the winding temperature will rise accordingly. The magnitude of the temperature rise is related to the degree of short circuit and the duration of the short circuit.

[0070] 2) Localized temperature anomalies: At the location of a short circuit, the winding temperature will be higher, leading to localized temperature anomalies. This may cause aging of the insulation material and a decrease in insulation performance.

[0071] 3) Uneven temperature distribution along the current direction: At the location of a short circuit, the temperature rises where the current flows, but the temperature distribution along the current direction is uneven. At the two ends of the short circuit, the temperature is lower.

[0072] For the case of winding short circuits, this invention mainly considers short circuits occurring in the high-voltage windings of dry-type transformers, and then performs temperature field distribution simulation. A cut surface is made inside the high- and low-voltage windings of the transformer. It is assumed that an inter-turn short circuit occurs in the winding region of the three-phase windings within the cut surface, and a heat source is applied to the short-circuited winding portion. Winding short circuits are simulated separately on the three-phase high-voltage windings.

[0073] Secondly, the data is preprocessed and analyzed: The feature set consists of three-phase temperature data, and the last column is a label set used to represent the fault type. For example... Figure 2 As shown, the label set uses label encoding: 1 represents normal operation, 2 represents core overheating, 3 represents partial discharge, and 4 represents winding short circuit, to facilitate data processing. 20 sets of data are randomly selected for each operating state as test data, for a total of 80 sets of data as the test set. The remaining 70 sets of data for each operating state, for a total of 280 sets of data, are used as the training set.

[0074] See Figure 2As shown, by analyzing the confusion matrix heatmaps of predicted and true values ​​under different operating conditions, the correlation between each feature variable can be analyzed. The heatmap is used for the confusion matrix, where the color of each cell represents the magnitude of the value at that position, and the color gradient shows the intensity of the data values. The closer the value is to 1, the stronger the positive correlation between the two predicted and true values, proving the accuracy of the prediction.

[0075] In dry-type transformer fault identification based on machine learning multi-classification algorithms, the data normalization method employed is Min-Max Normalization. This normalization method is widely used in data preprocessing in machine learning. For input data, the value range of each feature is first mapped to the interval between 0 and 1. Specifically, each feature value is subtracted from its minimum value, and then divided by the feature's value range (i.e., maximum value minus minimum value). The purpose of normalization is to unify the value ranges of different features to the same interval, thereby avoiding excessive influence of certain features on model training.

[0076] The historical data is preprocessed using max-min normalization, as shown in the following formula: in, For the preprocessed data, For historical data to be preprocessed, The maximum value of the historical data to be preprocessed. This represents the minimum value of the historical data to be preprocessed.

[0077] Mapping the value range of each feature to between 0 and 1 makes the value range of different features uniform, which is more conducive to model training.

[0078] Finally, a deep learning model is developed, and its accuracy is verified: Cross-validation is used to optimize generalization ability, and pipeline techniques are employed to preprocess the data and train the model. For example... Figure 3 As shown in the learning rate curve, the accuracy of both the training and validation sets is above 95%, proving the accuracy of the model.

[0079] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A deep learning-based electrical equipment monitoring method, characterized by, The method comprises the following steps: S1, obtaining historical data of an electrical equipment, and preprocessing the historical data to obtain preprocessed data; S2, constructing a deep learning model based on the preprocessed data; S3, obtaining first running data of the electrical equipment at a first time and second running data of the electrical equipment at a second time; S4, inputting the first running data into the deep learning model to obtain a continuous first comparison data graph; S5, obtaining a comparison point corresponding to the second time according to the first comparison data graph; S6, obtaining a comparison interval of the second running data according to the comparison point; S7, comparing the second running data with the comparison interval, and outputting an alarm signal if the second running data is not located in the comparison interval.

2. The deep learning-based electrical equipment monitoring method according to claim 1, characterized by, The preprocessing of the historical data to obtain preprocessed data comprises the following steps: filtering the historical data by using one or a combination of wavelet transform, smoothing filtering, low-pass filtering, high-pass filtering and band-pass filtering to obtain filtered data; performing denoising and normalization processing on the filtered data by using a time series correlation and density clustering algorithm to obtain preprocessed data. 3.The deep learning-based electrical equipment monitoring method of claim 1, wherein, The comparison interval of the second running data according to the comparison point comprises the following steps: setting an upper distance and a lower distance corresponding to the ordinate of the comparison point, obtaining historical data at a third time to obtain third running data; judging whether the second running data is the same as the historical data corresponding to the second time, if not, increasing the upper distance and the lower distance by a first proportion, if the same, judging whether the difference between the second running data and the third running data is within a data fluctuation range, if exceeding the data fluctuation range, decreasing the upper distance and the lower distance by the first proportion; obtaining the comparison interval from the ordinate of the comparison point minus the adjusted lower distance to the ordinate of the comparison point plus the adjusted upper distance. 4.The deep learning-based electrical equipment monitoring method of claim 1, wherein, The comparison interval of the second running data according to the comparison point comprises the following steps: setting an upper distance and a lower distance corresponding to the ordinate of the comparison point, obtaining historical data at a third time to obtain third running data; judging whether the first running data is the same as the historical data corresponding to the first time, if not, increasing the upper distance and the lower distance by a first proportion, if the same, judging whether the difference between the first running data and the third running data is within a data fluctuation range, if exceeding the data fluctuation range, decreasing the upper distance and the lower distance by the first proportion; obtaining the comparison interval from the ordinate of the comparison point minus the adjusted lower distance to the ordinate of the comparison point plus the adjusted upper distance.

5. The deep learning-based electrical equipment monitoring method according to claim 3 or 4, characterized by, The method further comprises the following steps: S8, after outputting the alarm signal, obtaining the electrical equipment outputting the alarm signal, obtaining electrical equipment of the same batch as the electrical equipment outputting the alarm signal, and detecting the electrical equipment of the same batch.

6. The deep learning-based electrical equipment monitoring method according to claim 5, characterized by, The detection of the electrical equipment of the same batch comprises the following steps: S81, set the first preset number of times, and the same batch of electrical equipment jumps to step S1 after reducing the upper distance and the lower distance of the same batch of electrical equipment by the second ratio, runs steps S1-S7, and if the alarm signal is output again, steps S81 are repeated until the number of jumps exceeds the first preset number of times, and the same batch of electrical equipment all output alarm signals.

7. The deep learning-based electrical equipment monitoring method according to claim 6, characterized by, A second preset number of times is set, when the number of alarms of the electrical equipment in a unit time exceeds a third preset number of times, the running data corresponding to the second time in the first comparison data graph of the electrical equipment outputting the alarm signal is configured as fourth running data, the average value of the fourth running data and the second running data is configured as fifth running data, the fourth running data in the first comparison data graph of the electrical equipment outputting the alarm signal is replaced by the fifth running data, and a second comparison data graph is generated.

8. A deep learning-based electrical equipment monitoring system employing the deep learning-based electrical equipment monitoring method according to any one of claims 1 to 7. Comprise: A preprocessing module, a deep learning module, an acquisition module, an input module and a judgment module; The preprocessing module is used for preprocessing historical data to obtain preprocessed data; The deep learning module is used for constructing a deep learning model according to the preprocessed data; The acquisition module is used for acquiring first running data of the electrical equipment at a first time and second running data of the electrical equipment at a second time; The input module is used for inputting the first running data into the deep learning model to obtain a continuous first comparison data graph; The judgment module is used for obtaining a comparison point corresponding to the second time and a comparison interval of the second running data according to the first comparison data graph, comparing the second running data with the comparison interval, and outputting an alarm signal if the second running data is not located in the comparison interval.

9. An apparatus, comprising: Comprise at least one processor and at least one memory; The at least one memory is used for storing a computer program; The at least one processor is used for executing the computer program to realize the method of claims 1-7.

10. A computer-readable storage medium, characterized in that, For storing executable instructions, the executable instructions are executed by the processor to make the processor realize the method of any one of claims 1-7.

Images