Oil level detection method and detection device for oil-filled electric equipment

Abstract

The invention discloses an oil level detection method and a detection device for oil-filled electric equipment, and relates to the technical field of oil level detection of oil-filled electric equipment, comprising the following steps: inputting the obtained interference-free radar signal data and electromagnetic data into a preset radar reflection path evaluation model to obtain a path evaluation coefficient of each reflection path; screening the path evaluation coefficient of each reflection path based on a plurality of preset radar sensing threshold values ​​to obtain a plurality of interference-free reflection path sets; performing a plurality of random electromagnetic evaluation superpositions to obtain a plurality of interference reflection path sets and performing path matching with the plurality of interference-free reflection path sets to obtain a plurality of matching path sets; calculating the path change values ​​in the plurality of matching path sets to obtain an optimal radar sensing threshold value, thereby solving the problem that the radar signal threshold value of a radar oil level gauge set under electromagnetic interference cannot eliminate the interference path to the maximum extent, thereby reducing the oil level measurement accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of oil level, and more specifically, to a method and a device for detecting the oil level of oil-filled electric power equipment. Background Art

[0002] Oil-filled power equipment (such as transformers, oil circuit breakers, etc.) plays a vital role in the power system, and its operating status is directly related to the stability and safety of the power grid. The oil level in the equipment is one of the important parameters of its health status. The change of oil level reflects the temperature and pressure changes inside the equipment or the failures such as oil leakage and oil seepage. Timely and accurate monitoring of oil level is a key means to ensure the safe operation of equipment and prevent failures.

[0003] For example, the invention patent with announcement number: CN115326172B announces a transformer oil level detection method, system, storage medium and intelligent terminal, which obtains oil level detection information, ambient temperature detection information and a QR code on the oil tank; if the temperature value corresponding to the ambient temperature detection information falls into the oil temperature range corresponding to the reference temperature information, then the expansion correction value corresponding to the ambient temperature detection information is calculated according to the correction database; the insulation coefficient corresponding to the QR code is determined according to the insulation database; the oil level correction value corresponding to the insulation coefficient and the expansion correction value is determined according to the oil level correction database; the oil level value corresponding to the oil level detection information is corrected according to the oil level correction value to generate an oil level height value.

[0004] The above disclosed technical solutions have at least the following technical problems:

[0005] When detecting the oil level of oil-filled electrical equipment, the conventional radar oil level gauge is susceptible to electromagnetic interference from the surrounding environment, and the interference path cannot be eliminated to the maximum extent when setting the radar signal threshold value of the radar oil level gauge, which reduces the oil level measurement accuracy. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an oil level detection method and a detection device for oil-filled power equipment. By providing an oil level detection method and a detection device for oil-filled power equipment, the problem that the radar signal threshold value of the radar oil level gauge set under electromagnetic interference cannot eliminate the interference path to the maximum extent, thereby reducing the oil level measurement accuracy is solved.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for detecting oil level of oil-filled electric equipment comprises the following steps: obtaining non-interference radar signal data and electromagnetic data of a radar oil level gauge under different working conditions and performing preprocessing; inputting the non-interference radar signal data and electromagnetic data into a preset radar reflection path evaluation model to obtain a path evaluation coefficient of each reflection path; screening the path evaluation coefficient of each reflection path based on a plurality of preset radar sensing threshold values ​​to obtain a plurality of non-interference reflection path sets; using an electromagnetic evaluation generator to perform a plurality of random electromagnetic evaluation superpositions on a working area of ​​the radar oil level gauge to obtain a plurality of interference reflection path sets, and performing path matching on the plurality of non-interference reflection path sets and the plurality of interference reflection path sets based on a preset path matching principle to obtain a plurality of matching path sets; calculating the path change values ​​of the matching paths in the plurality of matching path sets, drawing a path change value curve, and obtaining an optimal radar sensing threshold value based on the path change value curve.

[0009] In a preferred embodiment, the preset radar reflection path evaluation model is specifically:

[0010] The historical interference-free radar signal data and electromagnetic data are preprocessed to construct a radar reflection data set; the radar reflection data set is divided into a training set and a test set according to a preset ratio, the training set is input into a neural network model to train the model, and the model parameters are initialized to obtain the trained neural network model; the trained neural network model is verified using a test set to obtain a radar reflection path evaluation model.

[0011] In a preferred embodiment, the path evaluation coefficient of each reflection path is screened based on several preset radar sensing threshold values ​​to obtain several interference-free reflection path sets, specifically: calculating the mean and variance of the path evaluation coefficients of all reflection paths; performing data analysis on the mean and variance of the path evaluation coefficients of all reflection paths, and setting several radar sensing threshold values ​​using the quantile method; comparing the path evaluation coefficient of each reflection path with the several radar sensing threshold values; and taking the radar signal reflection path whose path evaluation coefficient is less than the radar sensing threshold value as an interference-free reflection path.

[0012] In a preferred embodiment, the electromagnetic evaluation generator is used to perform several random electromagnetic evaluation superpositions on the working area of ​​the radar oil level gauge to obtain several interference reflection path sets, specifically: selecting the interference source type, using the electromagnetic evaluation generator to generate a random interference signal, and injecting it into the radar oil level gauge; obtaining the radar signal power, the interference signal power and the frequency offset value, and obtaining the interference coefficient based on the interference coefficient calculation formula; classifying the interference scenarios of the radar oil level gauge based on the interference coefficient to obtain interference scenarios with different interference coefficients; performing electromagnetic evaluation superpositions of different intensities based on the interference scenarios with different interference coefficients to obtain several interference reflection path sets.

[0013] In a preferred embodiment, the plurality of non-interference reflection path sets and the plurality of interference reflection path sets are path matched based on a preset path matching principle to obtain a plurality of matching path sets, specifically: creating an empty matching path set; based on each path in the non-interference reflection path set, traversing all paths in the interference reflection path set in turn for matching; if the path meets the preset matching conditions, adding the successfully matched path pair to the empty matching path set; if the path does not meet the preset matching conditions, skipping it; repeating the matching steps and calculating the matching coverage to perform matching quality evaluation to obtain a plurality of matching path sets.

[0014] In a preferred embodiment, the path change values ​​of the matching paths in the plurality of matching paths are calculated, a path change value curve is drawn, and an optimal radar sensing threshold value is obtained based on the path change value curve, specifically: the path change value is obtained based on the path change calculation formula for the matching paths in the plurality of matching paths; a two-dimensional path change value curve is drawn with the radar sensing threshold value as the horizontal coordinate and the path change value as the vertical coordinate; the second-order derivative of the two-dimensional path change value curve is calculated to obtain an inflection point, and the inflection point is used as the optimal radar sensing threshold value.

[0015] The detection device of any one of the oil level detection methods for oil-filled electric equipment is characterized by comprising a data acquisition module, a path evaluation module, a path screening module, a path matching module, and a path change module, and there are connections between the modules:

[0016] A data acquisition module, used to acquire non-interference radar signal data and electromagnetic data of the radar oil level gauge under different working conditions and perform pre-processing;

[0017] Path evaluation module: used for inputting the interference-free radar signal data and electromagnetic data into a preset radar reflection path evaluation model to obtain a path evaluation coefficient of each reflection path;

[0018] Path screening module: used to screen the path evaluation coefficient of each reflection path based on several preset radar sensing threshold values ​​to obtain several interference-free reflection path sets;

[0019] Path matching module: used to use an electromagnetic evaluation generator to perform several random electromagnetic evaluation superpositions on the working area of ​​the radar oil level gauge to obtain several interference reflection path sets, and perform path matching on the several non-interference reflection path sets and the several interference reflection path sets based on a preset path matching principle to obtain several matching path sets;

[0020] Path change module: used to calculate the path change value of the matching path in the plurality of matching path sets, draw a path change value curve, and obtain the optimal radar sensing threshold value based on the path change value curve.

[0021] The technical effects and advantages of the oil level detection method and detection device of the oil-filled electric equipment of the present invention are as follows:

[0022] 1. The present invention traverses each radar sensing threshold value, compares the path evaluation coefficients in descending order with the threshold value one by one, and generates an interference-free reflection path set based on the evaluation results. This method has significant advantages. First, it realizes the accurate screening of reflection paths, can effectively filter out interference paths with low evaluation coefficients, and only retains effective reflection paths with high credibility. Secondly, the multi-threshold value screening strategy provides flexibility and comprehensiveness, and the screening result sets corresponding to different threshold values ​​can be used for comparative analysis to help optimize the setting of threshold values. Thirdly, the method processes complex path data in a hierarchical manner, which significantly improves the anti-interference ability and measurement reliability of the radar oil level gauge in complex environments. At the same time, through the generation of the screening result set, high-quality basic data is provided for subsequent steps such as path change analysis and sensing threshold optimization, thereby improving the measurement efficiency and accuracy of the overall system.

[0023] 2. The present invention introduces an electromagnetic evaluation generator to simulate the interference environment under real working conditions, systematically generates an interference reflection path set, and combines the interference-free reflection path set to obtain a highly correlated matching path set through the path matching principle. This method has multiple advantages: enhancing anti-interference ability, identifying the impact of interference on radar signals by comparing the change values ​​of interference-free and interference paths; improving the scientific nature of threshold value setting, using the path change value curve to accurately locate the optimal radar sensing threshold value, and realizing effective screening of reliable paths; adapting to complex environments, using multiple evaluation superpositions and dynamic matching mechanisms to ensure the adaptability of the method to different interference intensities and types; improving measurement accuracy, and significantly improving the radar oil level gauge's accurate detection of liquid levels and signal stability through optimal threshold value optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1The present invention is a schematic structural diagram of an oil level detection method for oil-filled electrical equipment.

[0025] Figure 2 The present invention is a schematic diagram of the structure of a detection device for an oil level detection method for oil-filled electrical equipment. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] Embodiment 1, Figure 1 The present invention provides a method for detecting the oil level of oil-filled electric equipment, comprising the following steps:

[0028] S1, obtaining interference-free radar signal data and electromagnetic data of the radar oil level gauge under different working conditions and performing preprocessing;

[0029] In this example, the interference-free radar signal data and electromagnetic data are specifically:

[0030] Non-interference radar signal data refers to the original signal data collected by the radar oil level gauge without external electromagnetic interference, inner wall reflection interference or other interference factors, including the following data:

[0031] Original radar echo data: records the change of the original echo intensity of the radar signal over time, including reflection delay, echo amplitude and echo shape;

[0032] Signal characteristic data: key characteristic data extracted from the original radar echo;

[0033] Radar configuration parameters: mainly include transmission frequency, bandwidth, transmission power and pulse repetition frequency.

[0034] It should be noted that obtaining interference-free radar signal data and electromagnetic data of the radar oil level gauge under different working conditions can fully reflect the actual measurement performance of the equipment under diverse conditions. By accurately collecting liquid level reflection signals, inner wall reflection signals and other path information, the effective path and the interference path can be effectively distinguished, providing a reliable basis for subsequent data analysis. Combined with interference-free background electromagnetic data, the baseline characteristics of the equipment working area can be accurately established to ensure that abnormal changes in the signal when it is interfered with are easy to identify. This process covers a variety of working conditions, such as different liquid levels, medium characteristics, temperature, humidity and environmental pressure, which enhances the comprehensiveness and robustness of the data and provides high-quality sample data for the training and verification of the path evaluation model. This method significantly improves the reliability and anti-interference ability of the radar oil level gauge in complex environments.

[0035] S2, inputting the interference-free radar signal data and electromagnetic data into a preset radar reflection path evaluation model to obtain a path evaluation coefficient of each reflection path;

[0036] In this example, the preset radar reflection path evaluation model is as follows:

[0037] Preprocess the historical interference-free radar signal data and electromagnetic data to construct a radar reflection data set;

[0038] Dividing the radar reflection data set into a training set and a test set according to a preset ratio, inputting the training set into a neural network model to train the model, and initializing the model parameters to obtain a trained neural network model;

[0039] The trained neural network model is verified using a test set to obtain a radar reflection path evaluation model.

[0040] In this example, the interference-free radar signal data and electromagnetic data are input into a preset radar reflection path evaluation model to obtain a path evaluation coefficient for each reflection path. The specific formula is as follows:

[0041]

[0042] in is the path evaluation coefficient of the i-th reflection path, is the radar transmitting power, is the reflection coefficient of the reflecting surface, is the wavelength of non-interfering radar signals, is the total path length, is the propagation loss of the path, and are the gains of the transmitting antenna and the receiving antenna respectively, is the propagation delay of the i-th reflection path, is the radar signal frequency, is the index.

[0043] It should be noted that inputting the interference-free radar signal data and electromagnetic data into the preset radar reflection path evaluation model can accurately calculate the path evaluation coefficient of each reflection path, which has multiple advantages. First, the path evaluation coefficient reflects the importance of each path. The evaluation results can be used to sort the paths, give priority to retaining valid paths with high evaluation coefficients, reduce the amount of path data, and lay the foundation for subsequent path screening. Secondly, the path evaluation model comprehensively considers multi-dimensional characteristics such as delay, intensity, and direction to ensure the comprehensiveness and reliability of the evaluation results. In addition, by inputting interference-free data, the model can construct an accurate basic evaluation baseline, thereby providing a comparative reference for subsequent interference path identification. This process simplifies the screening process of complex paths and reduces the misjudgment rate. At the same time, it lays a solid foundation for path change analysis and induction threshold optimization under dynamic conditions, and significantly improves the measurement stability and anti-interference ability of the radar oil level gauge.

[0044] S3, screening the path evaluation coefficient of each reflection path based on a plurality of preset radar sensing threshold values, and obtaining a plurality of interference-free reflection path sets;

[0045] In this example, the path evaluation coefficient of each reflection path is screened based on several preset radar sensing threshold values ​​to obtain several interference-free reflection path sets, specifically:

[0046] Calculate the mean and variance of the path evaluation coefficients for all reflection paths;

[0047] Performing data analysis on the mean values ​​and variances of the path evaluation coefficients of all the reflection paths, and setting a number of radar sensing threshold values ​​using a quantile method;

[0048] Comparing the path evaluation coefficient of each reflection path with the plurality of radar sensing threshold values ​​respectively;

[0049] The radar signal reflection path whose path evaluation coefficient is less than the radar sensing threshold value is regarded as the interference-free reflection path.

[0050] In this example, the radar sensing threshold is specifically:

[0051] The radar sensing threshold value is a key parameter used by radar oil level gauges to screen effective reflection paths in signal processing, and determines the response sensitivity of the device to signals of different strengths. The threshold value sets a lower limit for signal strength or path evaluation coefficient to eliminate invalid signals (such as inner wall reflection interference, noise, etc.) below the threshold, and only retains qualified liquid level reflection signals. Reasonable setting of the radar sensing threshold value can balance the sensitivity and anti-interference ability of signal processing: too low a threshold may introduce noise or interference signals, and too high a threshold may miss effective signals. Therefore, by analyzing path changes and optimizing threshold values, the measurement accuracy and stability of radar oil level gauges in complex environments can be significantly improved.

[0052] In this example, the path evaluation coefficient of each reflection path is compared with the plurality of radar sensing threshold values, specifically:

[0053] Traversing the path evaluation coefficient of each reflection path, and comparing the path evaluation coefficient of each reflection path with each radar sensing threshold value;

[0054] When the path evaluation coefficient is less than the radar sensing threshold value, the reflection path corresponding to the path evaluation coefficient is added to the screening result set of the current radar sensing threshold value, otherwise it is removed;

[0055] For each radar sensing threshold value, a corresponding set of interference-free reflection paths is generated.

[0056] It should be noted that the path evaluation coefficient of each reflection path is traversed one by one and compared with multiple preset radar sensing threshold values. Specifically, the path evaluation coefficients of all reflection paths are traversed first, and each path evaluation coefficient is compared with each radar sensing threshold value one by one. If a path evaluation coefficient is less than the corresponding threshold value, the path is considered to be a qualified "interference-free reflection path" and is added to the screening result set corresponding to the threshold value; otherwise, the path is eliminated. In this way, multiple different sets of interference-free reflection paths can be generated, each set corresponding to a different radar sensing threshold value. This screening method can hierarchically screen the reflection paths according to different threshold values, thereby effectively filtering out reflection paths with better quality under different interference environments, reducing unnecessary interference, and improving the stability and accuracy of the radar system.

[0057] S4, using an electromagnetic evaluation generator to perform several random electromagnetic evaluation superpositions on the working area of ​​the radar oil level gauge to obtain several interference reflection path sets, and performing path matching on the several non-interference reflection path sets and the several interference reflection path sets based on a preset path matching principle to obtain several matching path sets;

[0058] In this example, an electromagnetic evaluation generator is used to perform several random electromagnetic evaluation superpositions on the working area of ​​the radar oil level gauge to obtain several interference reflection path sets, specifically:

[0059] Select the interference source type, use an electromagnetic evaluation generator to generate a random interference signal, and inject it into the radar oil level gauge;

[0060] Obtain radar signal power, interference signal power and frequency offset value, and obtain interference coefficient based on interference coefficient calculation formula;

[0061] Based on the interference coefficient, the interference scenarios of the radar oil level gauge are classified to obtain interference scenarios with different interference coefficients;

[0062] Electromagnetic evaluation superposition with different intensities is performed based on the interference scenarios with different interference coefficients to obtain a plurality of interference reflection path sets.

[0063] In this example, the interference scenarios of the radar oil level gauge are classified based on the interference coefficient to obtain interference scenarios with different interference coefficients, specifically:

[0064] The interference scenarios of the radar oil level gauge are classified based on the interference coefficient, and the scenarios are divided into low interference scenarios, medium interference scenarios and high interference scenarios;

[0065] Record the path change characteristics under different levels of interference scenarios to provide a reference for optimizing the radar sensing threshold.

[0066] In this example, the interference coefficient calculation formula is as follows:

[0067]

[0068] in, is the interference coefficient, is the interference signal power, is the radar signal power, is the offset between the center frequency of the interference signal and the radar operating frequency, is the bandwidth of the radar operating frequency band, is the cosine value of the angle between the interference signal incident angle and the main beam direction of the radar antenna, is the distance between the interference source and the radar antenna, , and are weight coefficients respectively.

[0069] It should be noted that this method can systematically evaluate the performance of radar oil level gauges under different interference conditions by generating and injecting random interference signals, combined with interference coefficient calculation and scenario classification, and has significant advantages. First, by selecting the type of interference source and calculating the interference coefficient, the intensity, frequency offset and other characteristics of the interference signal can be accurately quantified, making the interference analysis more scientific and detailed. Secondly, by classifying the interference scenarios according to the interference coefficient, low, medium and high interference conditions can be clearly divided, which is helpful to design optimization solutions for different scenarios. At the same time, by superimposing electromagnetic evaluations of different intensities, the interference effects in complex environments are simulated, and the potential impacts of actual working conditions are fully covered. This method provides an efficient strategy for the screening and elimination of interference paths, improves the anti-interference ability and measurement reliability of radar oil level gauges, and provides strong support for system optimization and practical application.

[0070] In this example, the plurality of non-interference reflection path sets and the plurality of interference reflection path sets are path matched based on a preset path matching principle to obtain a plurality of matching path sets, specifically:

[0071] Create an empty set of matching paths;

[0072] Based on each path in the non-interference reflection path set, all paths in the interference reflection path set are traversed in turn for matching;

[0073] If the path meets the preset matching condition, the successfully matched path pair is added to the empty matching path set;

[0074] If the path does not meet the preset matching conditions, it will be skipped;

[0075] Repeat the matching steps and calculate the matching coverage to evaluate the matching quality, and obtain several matching path sets.

[0076] It should be noted that the process of establishing an empty set, matching paths one by one, and dynamically adding matching results has multiple advantages. First, this method ensures the reliability and accuracy of matching by setting precise path matching conditions, effectively filters irrelevant or severely interfered paths, and retains high-quality reflection path pairs. Secondly, the interference-free path set and the interference path set are traversed one by one to fully cover potential matching pairs and avoid missing important path information. The matching quality is evaluated by calculating the matching coverage rate, which provides a scientific basis for optimizing matching conditions and radar sensing threshold values. In addition, the method is highly flexible and can adapt to the path change requirements under different working conditions, and generate a variety of matching path sets for multi-scenario analysis and system optimization. The overall anti-interference ability and measurement accuracy of the radar oil level gauge are improved, and at the same time, a reliable guarantee is provided for path identification and optimization in complex environments.

[0077] S5, calculating the path change values ​​of the matching paths in the plurality of matching path sets, drawing a path change value curve, and obtaining an optimal radar sensing threshold value based on the path change value curve;

[0078] In this example, the path change values ​​of the matching paths in the plurality of matching path sets are calculated, a path change value curve is drawn, and an optimal radar sensing threshold value is obtained based on the path change value curve, specifically:

[0079] Concentrating the matching paths among the plurality of matching paths based on a path change calculation formula to obtain a path change value;

[0080] Draw a two-dimensional path change value curve using the radar sensing threshold value as the horizontal coordinate and the path change value as the vertical coordinate;

[0081] The second-order derivative of the two-dimensional path change value curve is calculated to obtain an inflection point, and the inflection point is used as an optimal radar sensing threshold value.

[0082] In this example, the path change calculation formula is as follows:

[0083]

[0084] in, is the path change value, is the path signal strength change value, is the reflected signal power of the path under interference-free conditions, is the propagation delay of the path under interference-free conditions, is the delay variation value, is the direction change value, , and is the weight factor.

[0085] It should be noted that the two-dimensional curve analysis of the path change value and the radar sensing threshold value can accurately determine the optimal radar sensing threshold value, which has significant advantages. First, the change characteristics of the matching path are quantified by the path change calculation formula, combined with different threshold value screening, to fully reflect the sensitivity of the path change to the threshold adjustment. Secondly, the second-order derivative is used to calculate the inflection point of the curve, which can scientifically and automatically locate the key turning point of the threshold value, avoid the uncertainty of human selection, and ensure the objectivity and accuracy of the results. In addition, this method intuitively displays the trend of path changes, makes the path evaluation process transparent, and facilitates analysis and optimization. The optimal threshold value determined by the final inflection point can effectively retain the target path while maximally eliminating the interference path, significantly improving the measurement accuracy and anti-interference ability of the radar oil level gauge. The overall process is scientific and rigorous, providing a reliable basis for system optimization under complex working conditions.

[0086] Embodiment 2, as Figure 2 The figure shows a functional module diagram of a detection device for an oil-filled electric equipment oil level detection method. According to the functions to be implemented, the detection device for an oil-filled electric equipment oil level detection method may include a data acquisition module, a path evaluation module, a path screening module, a path matching module, and a path change module. The module described in the invention may also be referred to as a unit, which refers to a series of computer program segments that can be executed by a processor of a personnel performance data processing device and can complete fixed functions, and is stored in a memory of the personnel performance data processing device.

[0087] In this embodiment, the functions of each module are as follows:

[0088] A data acquisition module, used to acquire non-interference radar signal data and electromagnetic data of the radar oil level gauge under different working conditions and perform pre-processing;

[0089] Path evaluation module: used for inputting the interference-free radar signal data and electromagnetic data into a preset radar reflection path evaluation model to obtain a path evaluation coefficient of each reflection path;

[0090] Path screening module: based on several preset radar sensing threshold values, the path evaluation coefficient of each reflection path is screened to obtain several sets of interference-free reflection paths;

[0091] Path matching module: used to use an electromagnetic evaluation generator to perform several random electromagnetic evaluation superpositions on the working area of ​​the radar oil level gauge to obtain several interference reflection path sets, and perform path matching on the several non-interference reflection path sets and the several interference reflection path sets based on a preset path matching principle to obtain several matching path sets;

[0092] Path change module: used to calculate the path change value of the matching path in the plurality of matching path sets, draw a path change value curve, and obtain the optimal radar sensing threshold value based on the path change value curve.

[0093] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0094] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0095] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0096] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0097] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0098] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting the oil level of oil-filled electrical equipment, characterized in that: The following steps are involved: Obtain interference-free radar signal data and electromagnetic data of the radar oil level gauge under different working conditions and perform preprocessing; The interference-free radar signal data and electromagnetic data are input into a preset radar reflection path evaluation model to obtain a path evaluation coefficient for each reflection path. The specific formula is as follows: in is the path evaluation coefficient of the i-th reflection path, is the radar transmitting power, is the reflection coefficient of the reflecting surface, is the wavelength of non-interfering radar signals, is the total path length, is the propagation loss of the path, and are the gains of the transmitting antenna and the receiving antenna respectively, is the propagation delay of the i-th reflection path, is the radar signal frequency, is the index; Based on a plurality of preset radar sensing threshold values, the path evaluation coefficient of each reflection path is screened respectively to obtain a plurality of interference-free reflection path sets; An electromagnetic evaluation generator is used to perform several random electromagnetic evaluation superpositions on the working area of ​​the radar oil level gauge to obtain several interference reflection path sets, and the several non-interference reflection path sets and the several interference reflection path sets are path matched based on a preset path matching principle to obtain several matching path sets; Calculate the path change values ​​of the matching paths in the plurality of matching path sets, draw a path change value curve, and obtain an optimal radar sensing threshold value based on the path change value curve.

2. The oil level detection method for oil-filled electrical equipment according to claim 1, characterized in that: The preset radar reflection path evaluation model is specifically: Preprocess the historical interference-free radar signal data and electromagnetic data to construct a radar reflection data set; Dividing the radar reflection data set into a training set and a test set according to a preset ratio, inputting the training set into a neural network model to train the model, and initializing the model parameters to obtain a trained neural network model; The trained neural network model is verified using a test set to obtain a radar reflection path evaluation model.

3. The oil level detection method for oil-filled electrical equipment according to claim 2, characterized in that: Based on several preset radar sensing threshold values, the path evaluation coefficient of each reflection path is screened respectively to obtain several interference-free reflection path sets, specifically: Calculate the mean and variance of the path evaluation coefficients for all reflection paths; Performing data analysis on the mean values ​​and variances of the path evaluation coefficients of all the reflection paths, and setting a number of radar sensing threshold values ​​using a quantile method; Comparing the path evaluation coefficient of each reflection path with the plurality of radar sensing threshold values ​​respectively; The radar signal reflection path whose path evaluation coefficient is less than the radar sensing threshold value is regarded as the interference-free reflection path.

4. The oil level detection method for oil-filled electrical equipment according to claim 3, characterized in that: The electromagnetic evaluation generator is used to perform several random electromagnetic evaluation superpositions on the working area of ​​the radar oil level gauge to obtain several interference reflection path sets, specifically: Selecting the type of interference source, using an electromagnetic evaluation generator to generate a random interference signal to inject into the radar oil level gauge; Obtain radar signal power, interference signal power and frequency offset value, and obtain interference coefficient based on interference coefficient calculation formula; Based on the interference coefficient, the interference scenarios of the radar oil level gauge are classified to obtain interference scenarios with different interference coefficients; Electromagnetic evaluation superposition with different intensities is performed based on the interference scenarios with different interference coefficients to obtain a plurality of interference reflection path sets.

5. The oil level detection method for oil-filled electrical equipment according to claim 4, characterized in that: The plurality of non-interference reflection path sets and the plurality of interference reflection path sets are path matched based on a preset path matching principle to obtain a plurality of matching path sets, specifically: Create an empty set of matching paths; Based on each path in the non-interference reflection path set, all paths in the interference reflection path set are traversed in turn for matching; If the path meets the preset matching condition, the successfully matched path pair is added to the empty matching path set; If the path does not meet the preset matching conditions, it will be skipped; Repeat the matching steps and calculate the matching coverage to evaluate the matching quality, and obtain several matching path sets.

6. The oil level detection method for oil-filled electrical equipment according to claim 5, characterized in that: Calculate the path change values ​​of the matching paths in the plurality of matching path sets, draw a path change value curve, and obtain the optimal radar sensing threshold value based on the path change value curve, specifically: Concentrating the matching paths among the plurality of matching paths based on a path change calculation formula to obtain a path change value; Draw a two-dimensional path change value curve by taking the radar sensing threshold value as the horizontal coordinate and the path change value as the vertical coordinate; The second-order derivative of the two-dimensional path change value curve is calculated to obtain an inflection point, and the inflection point is used as an optimal radar sensing threshold value.

7. The oil level detection method for oil-filled electrical equipment according to claim 6, characterized in that: The interference coefficient calculation formula is specifically: in, is the interference coefficient, is the interference signal power, is the radar signal power, is the offset between the center frequency of the interference signal and the radar operating frequency, is the bandwidth of the radar operating frequency band, is the cosine value of the angle between the interference signal incident angle and the main beam direction of the radar antenna, is the distance between the interference source and the radar antenna, , and are weight coefficients respectively.

8. The oil level detection method for oil-filled electrical equipment according to claim 7, characterized in that: The path change calculation formula is specifically: in, is the path change value, is the path signal strength change value, is the reflected signal power of the path under interference-free conditions, is the propagation delay of the path under interference-free conditions, is the delay variation value, is the direction change value, , and is the weight factor.

9. A detection device using the oil level detection method for oil-filled electrical equipment as claimed in any one of claims 1 to 8, characterized in that: It includes data acquisition module, path evaluation module, path screening module, path matching module, path change module, and there are connections between modules: A data acquisition module, used to acquire non-interference radar signal data and electromagnetic data of the radar oil level gauge under different working conditions and perform pre-processing; Path evaluation module: used for inputting the interference-free radar signal data and electromagnetic data into a preset radar reflection path evaluation model to obtain a path evaluation coefficient of each reflection path; Path screening module: used to screen the path evaluation coefficient of each reflection path based on several preset radar sensing threshold values ​​to obtain several interference-free reflection path sets; Path matching module: used to use an electromagnetic evaluation generator to perform several random electromagnetic evaluation superpositions on the working area of ​​the radar oil level gauge to obtain several interference reflection path sets, and perform path matching on the several non-interference reflection path sets and the several interference reflection path sets based on a preset path matching principle to obtain several matching path sets; Path change module: used to calculate the path change value of the matching path in the plurality of matching path sets, draw a path change value curve, and obtain the optimal radar sensing threshold value based on the path change value curve.

Images