Electrical equipment insulation state evaluation and maintenance system and method thereof

Abstract

The invention relates to the technical field of power systems, in particular to an electrical equipment insulation state evaluation and maintenance system and method. The system comprises an ultrasonic input module, a data processing module, a data analysis and evaluation module, a server and a display module. The ultrasonic input module transmits and collects an insulating layer ultrasonic signal; the data processing module extracts characteristic parameters; the data analysis and evaluation module calculates an insulation state evaluation score and generates a maintenance suggestion; the server stores historical data and optimizes maintenance suggestions; and the display module displays the evaluation score and the maintenance suggestion. Through technical innovation and intelligent integration, the reliability, safety and economical efficiency of a power system are improved, the existing technical problems are solved, a new road is opened up for intelligent and informatization development of the power industry, and the system has great practical value and far-reaching social significance.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, in particular to an electrical equipment insulation status evaluation and maintenance system and its method. Background Art

[0002] The safe and stable operation of the power system is an important foundation of modern society, and the evaluation and maintenance of the insulation status of electrical equipment are the key links to ensure the reliability of the power system. With the continuous growth of power demand and the continuous expansion of the power grid scale, the evaluation and maintenance of the insulation status of electrical equipment are facing increasing challenges.

[0003] Traditional insulation status evaluation methods mainly rely on regular off-line detection and empirical judgment. This method has problems such as long detection cycle, great influence of human factors, and inability to reflect the equipment status in real time. In recent years, with the development of sensing technology and data analysis technology, some on-line monitoring systems have begun to be applied to the evaluation of the insulation status of electrical equipment. However, these systems often have the following deficiencies:

[0004] Firstly, existing on-line monitoring systems mostly adopt single detection means, such as partial discharge detection or infrared thermal imaging detection, etc. Although these methods can reflect some aspects of the insulation status, it is difficult to comprehensively and accurately evaluate the overall condition of the insulation material. Especially for complex insulation structures, single detection means often cannot accurately identify and locate potential problems.

[0005] Secondly, the existing systems have limited capabilities in data processing and analysis. Most systems can only perform simple threshold judgment or trend analysis, and it is difficult to capture the complex non-linear relationships in the insulation deterioration process. This results in low prediction accuracy of the system and it is difficult to provide precise decision support for equipment maintenance.

[0006] Furthermore, existing evaluation and maintenance systems are often fragmented, lacking effective information integration and intelligent decision-making mechanisms. This makes there be a gap between the evaluation results and specific maintenance actions, and it is difficult to achieve true intelligent maintenance.

[0007] Finally, existing systems generally lack the ability of self-learning and optimization. As time goes by and data accumulates, the performance of the system often does not increase but decrease, and it is difficult to adapt to the dynamic changes of equipment status and operating environment. Summary of the Invention

[0008] In view of the above problems, the present invention proposes an innovative electrical equipment insulation status evaluation and maintenance system and its method. The system realizes the comprehensive, accurate and real-time evaluation of the insulation status of electrical equipment through integrating advanced ultrasonic detection technology, intelligent data processing algorithms, deep learning analysis models and intelligent decision-making mechanisms, and can provide personalized maintenance suggestions.

[0009] The present invention proposes an insulation status evaluation and maintenance system for electrical equipment, including:

[0010] An ultrasonic input module, configured to:

[0011] Transmit ultrasonic signals to the insulation layer of the electrical equipment to be measured;

[0012] Collect ultrasonic reflection signals of the insulation layer;

[0013] A data processing module, communicatively connected to the ultrasonic input module, configured to:

[0014] Receive the ultrasonic reflection signals sent by the ultrasonic input module;

[0015] Extract characteristic parameters based on the ultrasonic reflection signals;

[0016] A data analysis and evaluation module, communicatively connected to the data processing module, configured to:

[0017] Calculate an insulation status evaluation score based on the characteristic parameters;

[0018] Generate maintenance suggestions according to the insulation status evaluation score;

[0019] A server, communicatively connected to the data analysis and evaluation module, configured to:

[0020] Store historical insulation data, historical maintenance data, and historical maintenance plans;

[0021] Optimize the maintenance suggestions based on the historical data;

[0022] A display module, communicatively connected to the data analysis and evaluation module and the server, configured to:

[0023] Display the insulation status evaluation score and the maintenance suggestions.

[0024] Preferably, the ultrasonic input module includes:

[0025] An ultrasonic sensor pan-tilt, configured to:

[0026] Realize three-dimensional spatial positioning of the ultrasonic sensor;

[0027] Control the ultrasonic sensor to perform omnidirectional scanning;

[0028] A plurality of ultrasonic sensors, installed on the ultrasonic sensor pan-tilt, configured to:

[0029] Transmit ultrasonic signals;

[0030] Receive ultrasonic reflection signals;

[0031] Among them, the ultrasonic sensor includes a transmitting sensor and a receiving sensor. The transmitting sensors are arranged at intervals in the clockwise direction along the inner upper surface of the insulating layer to be measured, and the receiving sensors are arranged at intervals in the counterclockwise direction along the outer lower surface of the insulating layer to be measured.

[0032] Preferably, the ultrasonic sensor cloud platform is implemented based on a three-degree-of-freedom mechanical platform and includes:

[0033] An X-axis motor, horizontally arranged, for driving the Z-axis motor to rotate around the X-axis on a horizontal plane;

[0034] A Y-axis motor, vertically arranged with respect to the X-axis motor, for driving the entire mechanical platform to move left and right;

[0035] A Z-axis motor, vertically arranged, for:

[0036] driving the ultrasonic sensor to move vertically up and down;

[0037] driving the ultrasonic sensor to move left and right on a vertical line;

[0038] Among them, the X-axis motor, Y-axis motor, and Z-axis motor are all equipped with encoders for feedback control of their respective motion parameters.

[0039] Preferably, the data processing module includes:

[0040] A data preprocessing unit for performing noise reduction and frequency reduction processing on the ultrasonic reflection signal;

[0041] A feature extraction unit, connected to the data preprocessing unit, for extracting feature quantities from the preprocessed signal;

[0042] A feature data analysis unit, connected to the feature extraction unit, for performing comparative analysis on the feature quantities and a preset database.

[0043] Preferably, the data analysis and evaluation module is used for:

[0044] calculating the insulation aging degree I, and its calculation formula is:

[0045]

[0046] Among them, a is an insulation material constant, b is an inherent viscosity coefficient, ρ is the deviation of the medium density, d is the thickness of the electrical equipment, and n is the number of measurements; based on the insulation aging degree I, calculating the real-time insulation state evaluation score S, and its calculation formula is:

[0047] S = S0 + ΔI + f(T, M, G, V, t);

[0048] Among them, S0 is the initial score, ΔI is the change value of the insulation aging degree, T is the temperature, M is the material type, G is the insulation grade, V is the working voltage grade, and t is the historical usage time.

[0049] Preferably, it further includes:

[0050] A remote terminal, communicatively connected to the server, for:

[0051] Collecting on-site operation data of the electrical equipment;

[0052] Sending the on-site operation data to the server;

[0053] A mobile terminal, communicatively connected to the server, for:

[0054] Receiving equipment information input by the operator;

[0055] Displaying the predicted result of the remaining insulation life sent by the server.

[0056] Preferably, the remote terminal further includes:

[0057] An electrical operator input module, for:

[0058] Collecting the identity information of the operator;

[0059] Verifying the operation authority of the operator;

[0060] Among them, the identity information includes at least face information, palmprint information, fingerprint information, and voiceprint information.

[0061] Preferably, the server is further used for:

[0062] Analyzing the aging law of the insulation material based on the deep learning algorithm;

[0063] Predicting the insulation aging life data;

[0064] Generating a personalized maintenance plan according to the insulation aging life data.

[0065] Preferably, the ultrasonic signal frequency range of the ultrasonic input module is 10 MHz - 150 MHz.

[0066] A method for evaluating and maintaining the insulation state of electrical equipment by an electrical equipment insulation state evaluation and maintenance system includes the following steps:

[0067] S1. Transmitting an ultrasonic signal to the insulation layer of the electrical equipment to be measured through the ultrasonic input module, and collecting the ultrasonic reflection signal;

[0068] S2. The ultrasonic reflection signal is preprocessed by the data processing module to extract characteristic parameters;

[0069] S3. The insulation status evaluation score is calculated based on the characteristic parameters by the data analysis and evaluation module, and maintenance suggestions are generated;

[0070] S4. The insulation status evaluation score and the maintenance suggestions are transmitted to the server;

[0071] S5. The server optimizes the maintenance suggestions based on historical data;

[0072] S6. The optimized insulation status evaluation score and maintenance suggestions are displayed through the display module;

[0073] S7. Maintenance operations are performed according to the maintenance suggestions;

[0074] S8. The maintenance results are recorded and fed back to the server for further optimization of the maintenance strategy.

[0075] While solving the problems of the prior art, the system of the present invention also brings significant benefits in many aspects. From a macroscopic perspective, the system greatly improves the reliability and safety of the power system. Through real-time and accurate insulation status evaluation, the system can effectively prevent potential equipment failures and reduce accidental power outages, thus ensuring the stability of power supply. This not only reduces the operation risks and economic losses of power enterprises, but also provides a reliable energy guarantee for the stable development of the whole society.

[0076] From a technical perspective, the present invention has achieved several key breakthroughs. First, the high-frequency ultrasonic detection technology adopted by the system, combined with the innovative sensor layout scheme, greatly improves the accuracy and comprehensiveness of insulation status detection. This enables the system to detect tiny insulation defects in a timely manner, providing the possibility for preventive maintenance. Second, the data analysis model based on deep learning breaks through the limitations of traditional analysis methods and can accurately capture the complex non-linear relationships in the insulation degradation process, significantly improving the accuracy of prediction. Third, the intelligent decision-making mechanism of the system realizes the seamless connection between the evaluation results and maintenance actions, greatly improving the pertinence and efficiency of maintenance.

[0077] From the perspective of economic benefits, the system of the present invention significantly reduces the maintenance cost of equipment by optimizing the maintenance strategy. Through precise status evaluation and prediction, the system can perform maintenance at the best time, avoiding the waste of resources caused by premature maintenance and the high losses brought by late maintenance. At the same time, the self-learning and optimization ability of the system ensures long-term performance improvement, bringing continuous economic benefits to users.

[0078] From the perspective of environmental protection, the system of the present invention indirectly reduces the equipment replacement frequency and raw material consumption by extending the equipment life and reducing the failure rate, which is in line with the concept of sustainable development. In addition, the remote monitoring and diagnosis functions of the system reduce the need for on-site inspections and lower the carbon emissions caused by personnel travel.

[0079] Generally speaking, the electrical equipment insulation state evaluation and maintenance system of the present invention comprehensively improves the reliability, safety and economy of the power system through technological innovation and intelligent integration. It not only solves many problems faced by the existing technologies, but also opens up a new path for the intelligent and informatized development of the power industry, with great practical value and profound social significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 is the top-level block diagram of the system of the present invention;

[0081] Figure 2 is the logic block diagram of the ultrasonic input module of the present invention;

[0082] Figure 3 is the working flow chart of the ultrasonic input module of the present invention;

[0083] Figure 4 is the internal structure diagram of the data processing module of the present invention;

[0084] Figure 5 is the structure diagram of the remote terminal of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0085] Please refer to the attached Figures 1-5 , the present invention provides an electrical equipment insulation state evaluation and maintenance system and its method. The system includes an ultrasonic input module, a data processing module, a data analysis and evaluation module, a server and a display module, and these modules work together to achieve a comprehensive evaluation and intelligent maintenance of the insulation state of electrical equipment.

[0086] Specifically, the electrical equipment insulation state evaluation and maintenance system of the present invention includes the following modules:

[0087] The ultrasonic input module 1 is used to emit ultrasonic signals to the insulation layer of the electrical equipment to be tested and collect the reflected signals. By emitting high-frequency ultrasonic waves and receiving the reflected waves, this module can effectively detect the microscopic structural changes inside the insulation layer. Preferably, the frequency of the ultrasonic signal can be set in the range of 10 MHz - 150 MHz, and this frequency range can take into account both the detection accuracy and the signal penetration ability.

[0088] The data processing module 2 is communicatively connected to the ultrasonic input module 1 and is used to receive ultrasonic reflection signals and extract characteristic parameters. In an embodiment of the present invention, the data processing module 2 adopts advanced signal processing algorithms, such as wavelet transform and Hilbert-Huang transform, etc., to improve the accuracy of feature extraction.

[0089] The data analysis and evaluation module 3 is communicatively connected to the data processing module 2. Its core function is to calculate the insulation status evaluation score based on the extracted characteristic parameters and generate maintenance suggestions. This module adopts an innovative evaluation algorithm, combining the advantages of machine learning and expert systems, and can accurately evaluate the insulation status.

[0090] The server 4 is communicatively connected to the data analysis and evaluation module 3 and is used to store historical data and optimize maintenance suggestions. The server 4 of the present invention adopts a distributed storage architecture, ensuring the efficient storage and rapid retrieval of a large amount of historical data.

[0091] The display module 5 is communicatively connected to the data analysis and evaluation module 3 and the server 4 and is used to intuitively display the insulation status evaluation results and maintenance suggestions. The display module 5 of the present invention adopts a user-friendly interface design, facilitating operators to quickly understand and make decisions.

[0092] Furthermore, the ultrasonic input module 1 of the present invention includes an ultrasonic sensor pan-tilt 11 and a plurality of ultrasonic sensors. The ultrasonic sensor pan-tilt 11 realizes the three-dimensional spatial positioning of the ultrasonic sensors, enabling the system to perform all-round and multi-angle scanning detection on electrical equipment. This design greatly improves the comprehensiveness and accuracy of detection.

[0093] Preferably, the ultrasonic sensors include transmitting sensors and receiving sensors. The transmitting sensors are arranged at intervals along the inner upper surface of the insulation layer to be measured in the clockwise direction, while the receiving sensors are arranged at intervals along the outer lower surface of the insulation layer to be measured in the counterclockwise direction. This unique arrangement ensures that the ultrasonic signals can fully cover the insulation layer to be measured, greatly improving the reliability of detection.

[0094] A prominent feature of the present invention is that the ultrasonic sensor pan-tilt 11 is realized based on a three-degree-of-freedom mechanical platform. This platform includes an X-axis motor, a Y-axis motor, and a Z-axis motor, which are respectively responsible for motion control in different directions. The X-axis motor is horizontally arranged and is used to drive the Z-axis motor to rotate around the X-axis on the horizontal plane. The Y-axis motor is vertically arranged with the X-axis motor and is used to drive the entire mechanical platform to move left and right. The Z-axis motor is vertically arranged and can not only drive the ultrasonic sensor to move vertically up and down but also move left and right on the vertical line.

[0095] To achieve precise motion control, encoders are installed on the X-axis motor, Y-axis motor, and Z-axis motor respectively, which are used to feedback and control their respective motion parameters. This design enables the system to achieve precise positioning and flexible movement of the ultrasonic sensor, thus ensuring the comprehensiveness and accuracy of detection.

[0096] In practical applications, the system of the present invention can flexibly adjust the position and scanning path of the ultrasonic sensor according to the characteristics of different electrical equipment. For example, for large transformers, a larger scanning range and denser scanning points can be set; while for linear equipment such as cables, a continuous scanning mode along the length direction can be adopted. This flexibility enables the present invention to meet the insulation status evaluation requirements of various types of electrical equipment.

[0097] Through the above innovative design, the present invention realizes a comprehensive, accurate, and efficient evaluation of the insulation status of electrical equipment, providing strong technical support for the safe and stable operation of the power system.

[0098] The data processing module 2 of the present invention includes a data preprocessing unit 21, a feature extraction unit 22, and a feature data analysis unit 23. These three units work together to realize the conversion process from the original signal to effective features.

[0099] The data preprocessing unit 21 first performs noise reduction and frequency down-conversion processing on the ultrasonic reflection signal. In a preferred embodiment of the present invention, the wavelet threshold denoising method is adopted, which can effectively remove the high-frequency noise in the signal while retaining the important features of the signal. The frequency down-conversion processing adopts digital down-conversion technology to convert the high-frequency ultrasonic signal to a lower frequency for subsequent processing.

[0100] The feature extraction unit 22 is connected to the data preprocessing unit 21 and is used to extract feature quantities from the preprocessed signal. The present invention adopts a variety of advanced feature extraction algorithms, including but not limited to time-domain analysis, frequency-domain analysis, and time-frequency joint analysis. Preferably, the Hilbert-Huang transform (HHT) is used for time-frequency analysis. This method is particularly suitable for processing non-linear and non-stationary signals and can effectively capture the signal features caused by the microscopic structure changes of insulating materials.

[0101] The feature data analysis unit 23 is connected to the feature extraction unit 22, and its main function is to compare and analyze the extracted feature quantities with a preset database. In the system of the present invention, the preset database contains a large number of feature templates of different types and different degrees of insulation deterioration. By adopting an improved dynamic time warping (DTW) algorithm, the system can accurately identify which mode in the database is closest to the current insulation status, thereby realizing an accurate judgment of the insulation status.

[0102] The data analysis and evaluation module 3 is the core component of the system of the present invention, and its innovation is mainly reflected in two aspects: the calculation of the degree of insulation aging and the generation of the real-time insulation status evaluation score.

[0103] First, the present invention proposes a novel formula for calculating the degree of insulation aging:

[0104] I = a·e (b·ρ / d·n) ,

[0105] Here, I represents the degree of insulation aging, a is the insulation material constant (the value range is approximately between 100 and 400), b is the inherent viscosity coefficient (usually between 0.01 and 0.1), ρ represents the deviation of the dielectric density, d is the thickness of the electrical equipment, and n is the number of measurements. This formula comprehensively considers material characteristics, equipment parameters, and measurement factors, and can more accurately reflect the actual situation of insulation aging.

[0106] In practical applications, the value range of the insulation material constant a is usually between 100 and 400, and the specific value needs to be calibrated according to different types of insulation materials. For example, for epoxy resin insulation materials, the value of a may be close to 400, while for polyethylene insulation materials, the value of a may be close to 100. The inherent viscosity coefficient b reflects the internal structural characteristics of the material, and its value range is generally between 0.01 and 0.1.

[0107] Based on the calculated degree of insulation aging I, the system of the present invention further calculates the real-time insulation status evaluation score S. Its calculation formula is as follows:

[0108] S = S0 + ΔI + f(T, M, L, V, H, R),

[0109] Among them, S0 is the initial score, ΔI is the change value of the degree of insulation aging, and f(T, M, L, V, H, R) considers the influence of factors such as temperature T, material type M, insulation grade G, working voltage grade V, and historical usage time H. R represents the correlation degree between variables, and the specific implementation is:

[0110]

[0111] The value range of the elements of the R matrix is [-1, 1], which is used to quantify the correlation between variables. When |R i,j | > 0.7, the system will automatically adjust the weight calculation method of the corresponding variables.

[0112] The present invention classifies variables into two categories:

[0113] The set of variables causing aging C = {T, V} (temperature, voltage, etc.);

[0114] The set of aging variables E = {I, H} (degree of insulation aging, historical usage time, etc.);

[0115] The weight distribution formula is:

[0116] f(T, M, L, V, H) = ∑ i∈C α i ·X i +∑ j∈E β j ·g(X j ),

[0117] where α i is the weight coefficient (negative impact) of the aging variable, β j is the weight coefficient reflecting the aging variable, and g(X j ) is the non - linear conversion function reflecting the aging variable.

[0118] To avoid "assigning weights with the wrong direction to risk variables", the present invention introduces a weight direction calibration mechanism:

[0119] 1. Set up a prior knowledge base K, which includes the theoretical direction of the influence of each variable on the insulation state;

[0120] 2. After training the model, check whether the actual weight direction sign(α i ) is consistent with the theoretical direction in the prior knowledge K;

[0121] 3. If they are inconsistent, perform constraint optimization: s.t.sign(α i ) = K i , this mechanism ensures that the weight direction learned by the model is consistent with the theory and avoids misinterpretation.

[0122] The system of the present invention further includes a remote terminal 6 and a mobile terminal 7, which are communicatively connected to the server 4, greatly improving the flexibility and practicality of the system.

[0123] The remote terminal 6 is mainly used to collect on - site operation data of electrical equipment and transmit this data to the server 4 in real - time. In an embodiment of the present invention, the remote terminal 6 is also equipped with an infrared thermal imager, which can monitor the temperature distribution of the equipment in real - time and provide additional reference information for insulation state assessment.

[0124] The mobile terminal 7 is mainly for operators, used to receive equipment information input by the operators and display the predicted result of the remaining insulation life sent by the server 4. Preferably, the mobile terminal 7 adopts an intuitive graphical interface, enabling the operators to quickly understand the insulation state and predicted life of the equipment.

[0125] To further improve the security of the system, the remote terminal 6 of the present invention further includes an electrical operator input module 61. This module is used to collect the identity information of the operator and verify their operation authority. A prominent feature of the present invention is the adoption of a variety of biometric technologies, including face information, palmprint information, fingerprint information, and voiceprint information. The system can flexibly select single or multiple biometric features for identity verification according to different security level requirements.

[0126] For example, for ordinary daily inspection operations, fingerprint verification may be sufficient; while for situations involving important equipment or critical operations, the system may require face recognition and voiceprint verification simultaneously to ensure the absolute security of the operation. This multi-level identity authentication mechanism greatly improves the security of the system and effectively prevents illegal operations by unauthorized personnel.

[0127] Through the above design, the electrical equipment insulation state evaluation and maintenance system of the present invention realizes the full-process intelligent management from data collection, processing, analysis to remote operation, providing strong technical support for the safe and stable operation of the power system.

[0128] The server 4 of the present invention is not just a simple data storage unit. It also undertakes the important functions of in-depth data analysis and intelligent decision-making. In the preferred embodiment of the present invention, the server 4 adopts an algorithm based on deep learning to analyze the aging law of insulating materials. This method can better capture complex non-linear relationships compared with traditional statistical analysis methods, thereby improving the accuracy of prediction.

[0129] Specifically, the present invention adopts an improved long short-term memory network (LSTM) model to predict the insulation aging life data. The input of this model includes multi-dimensional time series data such as historical insulation state evaluation scores, ambient temperature, and load conditions. The structure of the model is as follows:

[0130] h t =σ*(W x x t +W h h t-1 +b h ),

[0131]

[0132] y t =W y (o t ⊙tanh(c t ))+b y ,

[0133] where h t is the hidden state, c t is the cell state, xt is the input vector, y t is the output vector, W x , W h , W y is the weight matrix, b h , b y is the bias vector, σ is the sigmoid activation function, and ⊙ represents Hadamard.

[0134] One innovation of this model is the introduction of an attention mechanism, which enables the model to automatically identify and focus on the factors that have the greatest impact on insulation aging. For example, for certain types of insulation materials, temperature may be the most critical influencing factor; while for other types, mechanical stress may be more important. Through the attention mechanism, the model can adaptively adjust the weights of various factors according to different situations.

[0135] Based on the predicted insulation aging life data, the system of the present invention can generate personalized maintenance plans. The "personalization" here is reflected in the following aspects:

[0136] 1. Equipment characteristic adaptation: Formulate differentiated maintenance strategies according to the characteristics of different electrical equipment, such as type, importance, operating environment, etc.

[0137] 2. Dynamic adjustment: As new data accumulates continuously, the system will automatically adjust the maintenance plan to achieve continuous optimization of the maintenance strategy.

[0138] 3. Cost-benefit balance: On the premise of ensuring safety, comprehensively consider the maintenance cost and the benefit of extending the equipment life to find the optimal maintenance time point.

[0139] Preferably, the system of the present invention also introduces a decision optimization model based on reinforcement learning. This model models the maintenance decision problem as a Markov decision process (MDP), where the state space includes information such as the current insulation state and the predicted remaining life, and the action space includes different maintenance operations (such as continuing to operate, partial repair, complete replacement, etc.). Through continuous interaction and learning with the environment, the system can gradually optimize its maintenance decision-making strategy to maximize the long-term benefit.

[0140] An important feature of the present invention is that the ultrasonic signal frequency range of the ultrasonic input module 1 is 10 MHz - 150 MHz. The selection of this frequency range is the optimal result obtained through in-depth research and a large number of experiments. Within this frequency range, ultrasonic waves can effectively detect the internal structure of the insulation layer while maintaining a high spatial resolution.

[0141] Specifically, ultrasonic waves with lower frequencies (such as 10 MHz - 50 MHz) have stronger penetration ability and are suitable for detecting thicker or denser insulating materials. For example, for the oil-paper insulation system of large transformers, ultrasonic waves around 50 MHz can effectively penetrate the insulation layer and detect internal structural changes.

[0142] On the other hand, ultrasonic waves with higher frequencies (such as 100 MHz - 150 MHz) have higher spatial resolution and are suitable for detecting thin layers or minute defects. For example, for the polyethylene insulation layer of cables, ultrasonic waves around 120 MHz can accurately locate minute air bubbles or impurities.

[0143] The system of the present invention can automatically select the most suitable ultrasonic frequency according to the characteristics of the device to be tested, or use multiple frequencies for scanning in one detection, so as to obtain comprehensive and detailed insulation status information.

[0144] Finally, the present invention also provides a method for evaluating and maintaining the insulation status of electrical equipment by applying the above-mentioned electrical equipment insulation status evaluation and maintenance system. The method includes the following steps:

[0145] S1. Transmit ultrasonic signals to the insulation layer of the electrical equipment to be tested through the ultrasonic input module 1, and collect ultrasonic reflection signals;

[0146] S2. Preprocess the ultrasonic reflection signals through the data processing module 2 to extract characteristic parameters;

[0147] S3. Calculate the insulation status evaluation score based on the characteristic parameters through the data analysis and evaluation module 3, and generate maintenance suggestions;

[0148] S4. Transmit the insulation status evaluation score and maintenance suggestions to the server 4;

[0149] S5. The server 4 optimizes the maintenance suggestions based on historical data;

[0150] S6. Display the optimized insulation status evaluation score and maintenance suggestions through the display module 5;

[0151] S7. Perform maintenance operations according to the maintenance suggestions;

[0152] S8. Record the maintenance results and feedback them to the server 4 for further optimizing the maintenance strategy.

[0153] In practical applications, to address the issue of "pseudo closed-loop", the present invention improves the design of the self-optimization system: a device age stratified learning mechanism. To solve problems such as the incorrect association of "high temperature = high score", the present system introduces a device age stratified learning mechanism:

[0154] M a={M1, M2,..., M k},

[0155] where M a represents k model sets divided according to the service life a of the device. The system selects the corresponding model for evaluation according to the device age to avoid the problem of data confusion between new and old devices.

[0156] Secondly, by integrating the physical aging model and the data-driven model, the present invention takes aging as an independent deduction item and adopts a dual-model integration method:

[0157] S = S base - I phy - I data ,

[0158] where: S base is the basic score, I phy is the aging deduction item based on the physical mechanism, and the formula is l phy = a · e (b -T·t) / d , I data is the aging deduction item based on historical data. This dual-model integration method ensures that the system has both a theoretical basis and can adapt to actual data changes.

[0159] To ensure the effectiveness of the closed-loop system, the present invention introduces a multi-level verification mechanism:

[0160] Internal model verification: Use cross-validation to evaluate the generalization ability of the model on different device types;

[0161] Actual effect verification: Take the improvement degree of the device performance after maintenance as the model evaluation criterion;

[0162] Expert knowledge calibration: Regularly introduce expert evaluation to calibrate the model parameters;

[0163] Through the above improvements, the electrical equipment insulation state evaluation system of the present invention solves potential problems such as variable dependence, wrong weight direction, and "pseudo closed-loop", and realizes more accurate and reliable insulation state evaluation and maintenance.

[0164] It should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. Electrical equipment insulation state evaluation and maintenance system, characterized in that, Comprising: An ultrasonic input module, configured to: Transmit an ultrasonic signal to the insulating layer of the electrical device to be measured; Collect the ultrasonic reflection signal of the insulating layer; A data processing module, communicatively connected to the ultrasonic input module, configured to: Receive the ultrasonic reflection signal sent by the ultrasonic input module; Extract characteristic parameters based on the ultrasonic reflection signal; A data analysis and evaluation module, communicatively connected to the data processing module, configured to: Calculate an insulation status evaluation score based on the characteristic parameters; Generate a maintenance suggestion according to the insulation status evaluation score; A server, communicatively connected to the data analysis and evaluation module, configured to: Store historical insulation data, historical maintenance data, and historical maintenance plans; Optimize the maintenance suggestion based on the historical data; A display module, communicatively connected to the data analysis and evaluation module and the server, configured to: Display the insulation status evaluation score and the maintenance suggestion.

2. The system according to claim 1, wherein The ultrasonic input module includes: An ultrasonic sensor pan-tilt, configured to: Realize the three-dimensional spatial positioning of the ultrasonic sensor; Control the ultrasonic sensor to perform an omnidirectional scan; A plurality of ultrasonic sensors, mounted on the ultrasonic sensor pan-tilt, configured to: Transmit ultrasonic signals; Receive ultrasonic reflection signals; Wherein, the ultrasonic sensor includes a transmitting sensor and a receiving sensor, the transmitting sensors are arranged at intervals along the inner upper surface of the insulating layer to be measured in a clockwise direction, and the receiving sensors are arranged at intervals along the outer lower surface of the insulating layer to be measured in a counterclockwise direction.

3. The system according to claim 2, wherein The ultrasonic sensor pan-tilt is implemented based on a three-degree-of-freedom mechanical platform and includes: An X-axis motor, horizontally arranged, configured to drive the Z-axis motor to rotate around the X-axis on a horizontal plane; A Y-axis motor, vertically arranged perpendicular to the X-axis motor, configured to drive the entire mechanical platform to move left and right; A Z-axis motor, vertically arranged, configured to: Drive the ultrasonic sensor to move vertically up and down; Drive the ultrasonic sensor to move left and right on a vertical line; Wherein, the X-axis motor, Y-axis motor, and Z-axis motor are all provided with encoders for feedback control of their respective motion parameters.

4. The system according to claim 1, wherein The data processing module includes: A data preprocessing unit, configured to perform noise reduction and frequency reduction processing on the ultrasonic reflection signal; A feature extraction unit, connected to the data preprocessing unit, configured to extract feature quantities from the preprocessed signal; A feature data analysis unit, connected to the feature extraction unit, configured to perform comparative analysis on the feature quantities and a preset database.

5. The system according to claim 1, characterized in that, The data analysis and evaluation module is configured to: Calculate the insulation aging degree I, and its calculation formula is: Wherein, a is the insulation material constant, b is the inherent viscosity coefficient, ρ is the deviation of the medium density, d is the thickness of the electrical device, and n is the number of measurements; Based on the insulation aging degree I, calculate the real-time insulation status evaluation score S, and its calculation formula is: S = S0 + ΔI + f(T, M, G, V, t), Wherein, S0 is the initial score, ΔI is the change value of the insulation aging degree, T is the temperature, M is the material type, G is the insulation grade, V is the working voltage grade, and t is the historical usage time.

6. The system according to claim 1, wherein It further includes: A remote terminal, communicatively connected to the server, configured to: Collect the on-site operation data of the electrical device; Send the on-site operation data to the server; A mobile terminal, communicatively connected to the server, is configured to: Receive device information input by an operator; Display the insulation remaining life prediction result sent by the server.

7. The system according to claim 6, characterized in that, The remote terminal further includes: An electrical operator input module, configured to: Collect the identity information of the operator; Verify the operation authority of the operator; Wherein, the identity information includes at least face information, palmprint information, fingerprint information, and voiceprint information.

8. The system according to claim 1, wherein The server is further configured to: Analyze the aging law of the insulating material based on a deep learning algorithm; Predict the insulation aging life data; Generate a personalized maintenance plan according to the insulation aging life data.

9. The system according to claim 1, wherein The ultrasonic signal frequency range of the ultrasonic input module is 10 MHz - 150 MHz.

10. A method for evaluating and maintaining the insulation status of an electrical equipment by using the electrical equipment insulation status evaluation and maintenance system according to any one of claims 1-9, characterized in that, It includes the following steps: S1. Transmit an ultrasonic signal to the insulation layer of the electrical equipment to be measured through the ultrasonic input module, and collect the ultrasonic reflection signal; S2. Preprocess the ultrasonic reflection signal through the data processing module to extract characteristic parameters; S3. Calculate the insulation status evaluation score based on the characteristic parameters through the data analysis and evaluation module, and generate a maintenance suggestion; S4. Transmit the insulation status evaluation score and the maintenance suggestion to the server; S5. The server optimizes the maintenance suggestion based on historical data; S6. Display the optimized insulation status evaluation score and maintenance suggestion through the display module; S7. Perform maintenance operations according to the maintenance suggestion; S8. Record the maintenance result and feedback it to the server for further optimizing the maintenance strategy.

Images