Real-time detection and fault early warning system for live equipment

Abstract

The invention relates to the technical field of power equipment monitoring, in particular to a live equipment real-time detection and fault early warning system which comprises a corona detection module, an environment monitoring module, a data integration and transmission module, a data analysis module, a fault prediction module and an early warning module. Wherein the corona detection module is used for capturing a corona discharge phenomenon on the surface of equipment; the data integration and transmission module is used for synchronizing and integrating output data of the two modules; the data analysis module is used for quantifying the discharge intensity and range and performing correction based on the environmental data; the fault prediction module is used for generating a discharge trend prediction result; the early warning module is used for triggering an early warning signal and automatically generating an early warning information report; according to the invention, through real-time monitoring, environment correction, trend prediction and multi-channel early warning, the problems of real-time performance, accuracy and foresight of corona discharge detection are solved, and the operation safety and fault prevention capability of live equipment are significantly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power equipment monitoring, and in particular to a real-time detection and fault early warning system for live equipment. Background Art

[0002] With the widespread application of power systems and high-voltage transmission and distribution equipment, the operational safety of live equipment has become a key factor in ensuring the stable operation of power systems. Corona discharge is one of the common phenomena in live equipment, which usually occurs at high-voltage insulators, busbar connections and conductor joints. When the corona discharge intensity is too high or the range continues to expand, it will not only cause equipment energy loss, but may also cause the surface insulation performance of the equipment to deteriorate, and even cause serious electrical failures. Traditional corona discharge detection mainly relies on manual inspections or regular offline testing. This method has problems such as insufficient real-time performance, obvious monitoring blind spots, and difficulty in timely warning of sudden discharges. It cannot meet the high requirements of modern power systems for equipment operating status.

[0003] Although the corona discharge detection system in the existing technology has introduced ultraviolet imaging technology and intelligent data processing methods, it still has many limitations. On the one hand, the existing system lacks an effective correction mechanism for changes in environmental conditions, which limits the accuracy of the detection results; on the other hand, fault prediction and risk assessment are often based only on the current state, and fail to fully combine historical data for trend prediction, lacking foresight. In addition, the existing system also has deficiencies in the triggering of warning signals and information transmission, making it difficult to effectively and timely transmit key warning information to maintenance personnel, affecting the efficiency of fault response. Therefore, the development of a real-time detection and fault warning system for live equipment has important technical significance and urgent practical needs. Summary of the invention

[0004] Based on the above objectives, the present invention provides a real-time detection and fault warning system for live equipment.

[0005] The real-time detection and fault warning system for live equipment includes a corona detection module, an environmental monitoring module, a data integration and transmission module, a data analysis module, a fault prediction module and a warning module; among which:

[0006] Corona detection module: deployed at designated locations of live equipment, using ultraviolet imaging technology to capture the corona discharge phenomenon on the equipment surface in real time, generating two-dimensional images and video data of corona discharge;

[0007] Environmental monitoring module: used to collect real-time environmental data around live equipment, including temperature, humidity and air pressure;

[0008] Data integration and transmission module: connects the corona detection module and the environmental monitoring module to synchronize and integrate the output data of the two modules, generate a standardized data set, and transmit it to the data analysis module;

[0009] Data analysis module: receives the standardized data set sent by the data integration and transmission module, and uses the extraction algorithm of the discharge edge detection to analyze the two-dimensional image and video data of the corona discharge to quantify the discharge intensity and range, and corrects it based on the environmental data to output the current corona discharge risk level;

[0010] Fault prediction module: Based on the current corona discharge risk level output by the data analysis module, combined with historical data and deep learning models, predict the future risk level and generate discharge trend prediction results;

[0011] Early warning module: When the discharge trend prediction results show that the future risk level is about to exceed the preset threshold, the early warning signal is triggered, an early warning information report is automatically generated, and maintenance personnel are notified through multiple channels.

[0012] Optionally, the corona detection module includes a mounting unit, an ultraviolet sensing unit, an image processing unit and a data output unit; wherein:

[0013] Installation unit: It is configured at the high-voltage contacts, insulator surfaces and wire joints of live equipment to fix the equipment;

[0014] UV sensing unit: used to capture the UV light signal generated on the surface of the charged equipment during the corona discharge process, and convert the UV light signal in the corona discharge area into a digital signal using a UV imaging detector with a wavelength range of 240 to 280 nanometers;

[0015] Image processing unit: connected to the ultraviolet sensor unit, used to receive the digital signal generated by the ultraviolet sensor unit, extract the discharge intensity and spatial position information through signal enhancement, noise filtering and target area recognition algorithm, and generate two-dimensional images and video data reflecting the corona discharge phenomenon in real time;

[0016] Data output unit: connected to the image processing unit, used to transmit the generated corona discharge two-dimensional image and video data to the data integration and transmission module.

[0017] Optionally, the image processing unit specifically includes:

[0018] Signal enhancement: Use contrast stretching technology to process the received digital signal to enhance the brightness and details of the image;

[0019] Noise filtering: Gaussian filtering technology is used to remove high-frequency noise in the enhanced image, retaining the relevant features of the discharge area while eliminating background noise;

[0020] Target area recognition: Using threshold segmentation technology, the area with brightness exceeding the preset threshold is identified from the enhanced image, and the target area of ​​corona discharge is extracted; the part with high brightness is marked as the discharge area, and the rest is marked as the non-discharge area, forming a preliminary binary image of the target area;

[0021] Feature extraction: Calculate the sum of the brightness of all pixels in the target area and quantify it as the discharge intensity; and calculate the coordinates of the center point of the target area based on the pixel distribution of the target area to determine the spatial position of the discharge phenomenon;

[0022] Two-dimensional image generation: Using the target area and discharge intensity data, the boundary information and brightness distribution of the discharge area are integrated to generate a two-dimensional image containing the discharge area and intensity distribution;

[0023] Video data generation: The continuously generated two-dimensional images are encoded in time series to form video data that can dynamically reflect the changes in the corona discharge phenomenon.

[0024] Optionally, the environment monitoring module includes a temperature acquisition unit, a humidity acquisition unit and an air pressure acquisition unit; wherein:

[0025] Temperature acquisition unit: installed around live equipment to monitor ambient temperature in real time. It measures the ambient air temperature through a built-in thermistor and converts the measured value into a digital signal output.

[0026] Humidity collection unit: The relative humidity in the air is detected in real time through the configured capacitive humidity sensor. The sensor calculates the humidity by sensing the change of capacitance value and converts the calculation result into a standardized digital signal;

[0027] Air pressure acquisition unit: A micro air pressure sensor is used to sense the atmospheric pressure in the environment in real time, and the detected air pressure changes are converted into standardized digital signal output through the internal piezoresistive measurement component.

[0028] Optionally, the data integration and transmission module includes a data receiving unit, a synchronization processing unit, a data integration unit and a formatting unit; wherein:

[0029] Data receiving unit: connected to the corona detection module and the environment monitoring module respectively, used for receiving in real time the corona discharge image and video data output by the corona detection module, and the environmental data of temperature, humidity and air pressure output by the environment monitoring module;

[0030] Synchronous processing unit: time-stamps the multi-source data collected by the data receiving unit, and performs synchronous correction on the data according to a unified clock signal to ensure the consistency of the corona detection data and the environmental monitoring data in the time dimension;

[0031] Data integration unit: Integrate the synchronized data according to the association rules, associate and store the corona discharge data with the environmental data of the corresponding time period to form a unified multi-dimensional data set;

[0032] Formatting unit: Converts the multidimensional data set generated by the integration unit into a standardized data set, including fields and parameter values ​​in a fixed format.

[0033] Optionally, the data analysis module includes a discharge feature extraction unit, an environment correction unit and a risk assessment unit; wherein:

[0034] Discharge feature extraction unit: Based on the two-dimensional image and video data in the standardized data set, the edge detection algorithm is used to extract the boundary information of the corona discharge area, and the discharge intensity is calculated by accumulating pixel values. At the same time, the regional statistical method is used to determine the coverage of the discharge; the discharge intensity represents the total energy of the discharge light signal, and the discharge range represents the physical size of the discharge area;

[0035] Environmental correction unit: based on the corresponding environmental data in the standardized data set, the discharge intensity and range are corrected; the temperature data is used to correct the influence of the change in sensor sensitivity, and the humidity and air pressure data are used to compensate for the fluctuation of the air dielectric constant;

[0036] Risk assessment unit: Based on the preset risk level judgment rules, the comprehensive risk score of the current corona discharge is calculated in combination with the corrected discharge intensity and range, and the corresponding risk level is matched; the risk score is generated by multi-dimensional parameter weighting, and the risk level is divided into three levels: low, medium and high.

[0037] Optionally, the discharge feature extraction unit specifically includes:

[0038] Boundary information extraction: Gradient calculation is performed on the two-dimensional image in the standardized data set, and the brightness change rate of each pixel is calculated using the Sobel operator. The formula is:

[0039] Where M(p, q) is the gradient amplitude at the pixel position (p, q); M x (p, q) and M y (p, q) are the gradient components in the horizontal and vertical directions respectively; based on the calculated gradient amplitude map, the non-maximum suppression algorithm is applied to remove the response of the non-boundary part, and the boundary pixels of the corona discharge area are extracted by the double threshold method to generate boundary information;

[0040] Discharge intensity calculation: Using boundary information, all pixels in the corona discharge area are extracted from the original image; and the brightness values ​​of all pixels are accumulated to quantify the total brightness of the discharge area. The formula is: F = ∑ (p ,q)∈Z L(p, q), where F represents the total discharge intensity value, Z is the pixel set in the discharge area, and L(p, Q) is the brightness value of the pixel point (p, q);

[0041] Discharge coverage calculation: Count the total number of pixels in the discharge area and calculate the actual physical area of ​​the discharge area in combination with the image resolution. The formula is: Where A is the actual area covered by the discharge; N is the total number of pixels in the discharge area; R s The physical size corresponding to a single pixel.

[0042] Optionally, the environmental correction unit specifically includes:

[0043] Temperature correction: First, obtain the current temperature value in the standardized data set and set it as T; then calculate the sensitivity correction coefficient K based on the temperature drift characteristics of the sensor sensitivity T , the formula is: K T =1+α·(TT 0 ), where α is the temperature sensitivity coefficient of the sensor, T 0 is the reference operating temperature of the sensor, T is the current temperature value; finally, the correction coefficient K is used T Correct the total discharge intensity value F, the formula is: Among them, F T is the corrected discharge intensity, and F is the uncorrected discharge intensity;

[0044] Humidity and air pressure correction: Get the current humidity value H and air pressure value P in the standardized data set, and calculate the correction coefficient of the current air dielectric constant ε based on the atmospheric characteristics. The formula is: ε=ε 0 ·(1+β H H+β P ·(PP 0 ), where ε 0 is the dielectric constant under standard atmospheric conditions, β H is the humidity influence coefficient, β P is the air pressure influence coefficient, P 0 is the standard gas pressure value; then the discharge range A is corrected using the corrected dielectric constant ε, the formula is: Among them, A C is the corrected discharge coverage, and A is the uncorrected discharge coverage.

[0045] Optionally, the fault prediction module includes a historical data storage unit, a data processing unit, a deep learning prediction unit and a trend generation unit; wherein:

[0046] Historical data storage unit: used to store historical corona discharge data during the operation of live equipment, including historical risk level, discharge intensity, coverage and environmental condition data; arrange historical corona discharge data in chronological order;

[0047] Data processing unit: receives the current corona discharge risk level output by the data analysis module, formats it, and generates input data containing timestamp, risk level, discharge intensity and coverage; at the same time, aligns the current data with the historical data in the historical data storage unit to generate time series data, ensuring the time series of the input data is complete;

[0048] Deep learning prediction unit: Based on the time series data generated by the data processing unit, it uses the long short-term memory network model to make predictions and output the future risk level of multiple time steps;

[0049] Trend generation unit: The future risk levels of multiple time steps output by the deep learning prediction unit are combined into a discharge trend prediction result, and expressed in the form of a time series curve to show the changing trend of the corona discharge risk level in the future period of time.

[0050] Optionally, the warning module includes a threshold judgment unit, a warning generation unit and an information transmission unit; wherein:

[0051] Threshold judgment unit: receives the discharge trend prediction result of the trend generation unit in the fault prediction module, and compares the predicted risk level of each time step with the preset threshold. When the predicted risk level of any time step is greater than the preset threshold, an early warning signal will be triggered;

[0052] Warning generation unit: used to generate a warning information report according to the discharge trend prediction result after the threshold judgment unit triggers the warning signal. The warning information report includes equipment information, risk information, time information and recommended measures;

[0053] Information transmission unit: used to send the generated early warning report to the maintenance personnel through various channels, including SMS, email and system notification.

[0054] Beneficial effects of the present invention:

[0055] The present invention combines ultraviolet imaging technology with environmental parameter correction, and the system can effectively eliminate interference caused by environmental conditions such as temperature, humidity and air pressure, ensuring that the quantitative results of corona discharge intensity and range are more accurate. The fault prediction module is based on a deep learning model and combines current and historical data to perform trend prediction and generate discharge trend prediction results, which significantly improves the system's forward-looking judgment ability on potential risks.

[0056] The present invention can trigger a warning signal in real time through the warning module, and notify maintenance personnel of a detailed warning report through multiple channels, thereby achieving a fast and reliable fault response, effectively reducing equipment operation risks, and improving the safety and stability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 A schematic diagram of a real-time detection and fault warning system according to an embodiment of the present invention;

[0059] Figure 2 Schematic diagram of a data analysis module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0061] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0062] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0063] like Figure 1-Figure 2As shown, the real-time detection and fault warning system for live equipment includes a corona detection module, an environmental monitoring module, a data integration and transmission module, a data analysis module, a fault prediction module and a warning module; wherein:

[0064] Corona detection module: deployed at designated locations of live equipment, using ultraviolet imaging technology to capture the corona discharge phenomenon on the equipment surface in real time, generating two-dimensional images and video data of corona discharge;

[0065] Environmental monitoring module: used to collect real-time environmental data around live equipment, including temperature, humidity and air pressure, to provide environmental background information required for corona discharge analysis;

[0066] Data integration and transmission module: connects the corona detection module and the environmental monitoring module to synchronize and integrate the output data of the two modules, generate a standardized data set, and transmit it to the data analysis module;

[0067] Data analysis module: receives the standardized data set sent by the data integration and transmission module, and uses the extraction algorithm of the discharge edge detection to analyze the two-dimensional image and video data of the corona discharge to quantify the discharge intensity and range, and corrects it based on the environmental data to output the current corona discharge risk level;

[0068] Fault prediction module: Based on the current corona discharge risk level output by the data analysis module, combined with historical data and deep learning models, predict the future risk level and generate discharge trend prediction results;

[0069] Early warning module: When the discharge trend prediction results show that the future risk level is about to exceed the preset threshold, the early warning signal is triggered, an early warning information report is automatically generated, and maintenance personnel are notified through multiple channels.

[0070] The corona detection module includes a mounting unit, an ultraviolet sensing unit, an image processing unit and a data output unit; wherein:

[0071] Installation unit: It is configured at the high-voltage contacts, insulator surfaces and conductor joints of live equipment to fix the equipment and ensure that the monitoring field of view covers all potential discharge areas;

[0072] UV sensing unit: used to capture the UV light signal generated on the surface of the charged equipment during the corona discharge process, and convert the UV light signal in the corona discharge area into a digital signal using a UV imaging detector with a wavelength range of 240 to 280 nanometers;

[0073] Image processing unit: connected to the ultraviolet sensor unit, used to receive the digital signal generated by the ultraviolet sensor unit, extract the discharge intensity and spatial position information through signal enhancement, noise filtering and target area recognition algorithm, and generate two-dimensional images and video data reflecting the corona discharge phenomenon in real time;

[0074] Data output unit: connected to the image processing unit, used to transmit the generated corona discharge two-dimensional image and video data to the data integration and transmission module for subsequent analysis; the above steps arrange ultraviolet sensing units at high-voltage insulators, busbar connection parts and conductor joints, and use ultraviolet imaging detectors with a wavelength range of 240 to 280 nanometers to accurately capture the initial signal of corona discharge, and combine the real-time enhancement and recognition capabilities of the image processing unit to generate high-quality image and video data, thereby ensuring an accurate description of the corona discharge phenomenon and providing a high-precision input source for subsequent data analysis and fault prediction.

[0075] The image processing unit specifically includes:

[0076] Signal enhancement: The received digital signal is processed using contrast stretching technology to enhance the brightness and details of the image, making the corona discharge area clearer and more prominent, while compressing the background information of the non-discharge area; the formula for enhancing the image is: Where I(i, j) is the pixel intensity at position (i, j) in the input image; I min and I max are the minimum and maximum pixel intensities in the image, respectively; L min and L max is the target intensity range of the enhanced image;

[0077] Noise filtering: Gaussian filtering technology is used to remove high-frequency noise in the enhanced image, retain the relevant features of the discharge area, and eliminate background noise that may interfere with the extraction of discharge features. The specific formula of Gaussian filtering is: Among them, G(i, j) is the Gaussian kernel function; σ is the standard deviation of the Gaussian filter, which is used to control the smoothness of the filter;

[0078] Target area recognition: Threshold segmentation technology is used to identify areas with brightness exceeding the preset threshold from the enhanced image, and the target area of ​​corona discharge is extracted; the part with high brightness is marked as the discharge area, and the rest is marked as the non-discharge area, forming a preliminary binary image of the target area; the formula for identifying the target area is: Where R(i, j) is the value of position (i, j) in the binary image, 1 indicates a discharge area, and 0 indicates a non-discharge area; T is a preset intensity threshold;

[0079] Feature extraction: Calculate the sum of the brightness of all pixels in the target area and quantify it as the discharge intensity; and calculate the coordinates of the center point of the target area based on the pixel distribution of the target area to determine the spatial position of the discharge phenomenon; the calculation formula for the discharge intensity is: S = ∑ (i , j)∈RI′(i, j), where S is the discharge intensity and I′(i, j) is the enhanced pixel intensity; the formula for calculating the centroid position of the target area is: Among them, (x c ,y c ) is the centroid coordinate of the discharge area; i, j are the row and column indexes of the image;

[0080] Two-dimensional image generation: Using the target area and discharge intensity data, the boundary information and brightness distribution of the discharge area are integrated to generate a two-dimensional image containing the discharge area and intensity distribution;

[0081] Video data generation: The continuously generated two-dimensional images are encoded in time series to form video data that can dynamically reflect the changes in the corona discharge phenomenon. The above steps are carried out through feature extraction and two-dimensional image generation. The image processing unit generates a two-dimensional image that accurately reflects the discharge area and intensity distribution while clarifying the corona discharge intensity and spatial position, and generates dynamic video data in combination with the time dimension. This process can provide high-quality visual information and provide reliable input data for the subsequent fault prediction module.

[0082] The environmental monitoring module includes a temperature acquisition unit, a humidity acquisition unit and an air pressure acquisition unit; wherein:

[0083] Temperature acquisition unit: installed around live equipment to monitor ambient temperature in real time. It measures the ambient air temperature through a built-in thermistor and converts the measured value into a digital signal output.

[0084] Humidity collection unit: The relative humidity in the air is detected in real time through the configured capacitive humidity sensor. The sensor calculates the humidity by sensing the change of capacitance value and converts the calculation result into a standardized digital signal;

[0085] Air pressure acquisition unit: It uses a micro air pressure sensor to sense the atmospheric pressure in the environment in real time, and converts the detected air pressure changes into standardized digital signal output through the internal piezoresistive measurement component; through the collaborative work of the above units, it can generate comprehensive environmental parameter data including temperature, humidity and air pressure in real time, and use the internal data integrator to uniformly format multiple environmental parameters for processing by the data integration and transmission module.

[0086] The data integration and transmission module includes a data receiving unit, a synchronization processing unit, a data integration unit and a formatting unit; wherein:

[0087] Data receiving unit: connected to the corona detection module and the environment monitoring module respectively, used for receiving in real time the corona discharge image and video data output by the corona detection module, and the environmental data of temperature, humidity and air pressure output by the environment monitoring module;

[0088] Synchronous processing unit: time-stamps the multi-source data collected by the data receiving unit, and performs synchronous correction on the data according to a unified clock signal to ensure the consistency of the corona detection data and the environmental monitoring data in the time dimension;

[0089] Data integration unit: Integrate the synchronized data according to the association rules, associate and store the corona discharge data with the environmental data of the corresponding time period, and form a unified multi-dimensional data set to fully express the relationship between the corona discharge phenomenon and environmental conditions;

[0090] Formatting unit: converts the multidimensional data set generated by the integration unit into a standardized data set, including fields and parameter values ​​in a fixed format; the standardized data set adopts preset data format specifications to adapt to the analysis and processing requirements of subsequent modules; through the collaborative work of the above units, the data integration and transmission module can effectively synchronize and integrate the multi-source data output by the corona detection module and the environmental monitoring module, generate a data set that meets the standardized specifications, and transmit it to the data analysis module, thereby improving the data processing efficiency and accuracy of the corona discharge monitoring system.

[0091] The data analysis module includes a discharge feature extraction unit, an environment correction unit and a risk assessment unit; wherein:

[0092] Discharge feature extraction unit: Based on the two-dimensional image and video data in the standardized data set, the edge detection algorithm is used to extract the boundary information of the corona discharge area, and the discharge intensity is calculated by accumulating pixel values. At the same time, the regional statistical method is used to determine the coverage of the discharge; the discharge intensity represents the total energy of the discharge light signal, and the discharge range represents the physical size of the discharge area;

[0093] Environmental correction unit: Based on the corresponding environmental data in the standardized data set, the discharge intensity and range are corrected; the temperature data is used to correct the influence of the change in sensor sensitivity, and the humidity and air pressure data are used to compensate for the fluctuation of the air dielectric constant. The correction formula is dynamically adjusted through the set parameter association table to generate a correction result consistent with the actual environmental conditions;

[0094] Risk assessment unit: Combined with the corrected discharge intensity and range, based on the preset risk level judgment rules, the comprehensive risk score of the current corona discharge is calculated and matched with the corresponding risk level; the risk score is generated by multi-dimensional parameter weighting, and the risk level is divided into three levels: low, medium and high; through the collaborative work of the discharge feature extraction unit, the environmental correction unit and the risk assessment unit, the discharge features can be accurately extracted from the standardized data set, corrected in combination with environmental conditions, and the risk level representing the severity of the current corona discharge is output, providing accurate input data for the fault prediction module.

[0095] The discharge feature extraction unit specifically includes:

[0096] Boundary information extraction: Gradient calculation is performed on the two-dimensional image in the standardized data set, and the brightness change rate of each pixel is calculated using the Sobel operator. The formula is:

[0097] Where M(p, q) is the gradient amplitude at the pixel position (p, q); M x (p, q) and M y (p, q) are the gradient components in the horizontal and vertical directions respectively; based on the calculated gradient amplitude map, the non-maximum suppression algorithm is applied to remove the response of the non-boundary part, and the boundary pixels of the corona discharge area are extracted by the double threshold method to generate boundary information;

[0098] Discharge intensity calculation: Using boundary information, all pixels in the corona discharge area are extracted from the original image; and the brightness values ​​of all pixels are accumulated to quantify the total brightness of the discharge area. The formula is: F = ∑ (p , q)∈Z L(p, q), where F represents the total discharge intensity value, Z is the pixel set in the discharge area, and L(p, q) is the brightness value of the pixel point (p, q);

[0099] Discharge coverage calculation: Count the total number of pixels in the discharge area and calculate the actual physical area of ​​the discharge area in combination with the image resolution. The formula is: Where A is the actual area covered by the discharge; N is the total number of pixels in the discharge area; R s is the physical size corresponding to a single pixel; through the above steps, the discharge feature extraction unit can accurately identify the boundary information of the corona discharge area, calculate the discharge intensity and coverage, and provide a reliable data basis for subsequent correction and risk assessment.

[0100] The environmental correction unit specifically includes:

[0101] Temperature correction: First, obtain the current temperature value in the standardized data set and set it as T; then calculate the sensitivity correction coefficient K based on the temperature drift characteristics of the sensor sensitivity T , the formula is: K T =1+α·(TT 0 ), where α is the temperature sensitivity coefficient of the sensor, T 0 is the reference operating temperature of the sensor, T is the current temperature value; finally, the correction coefficient K is used T Correct the total discharge intensity value F, the formula is: Among them, F T is the corrected discharge intensity, and F is the uncorrected discharge intensity;

[0102] Humidity and air pressure correction: Get the current humidity value H and air pressure value P in the standardized data set, and calculate the correction coefficient of the current air dielectric constant ε based on the atmospheric characteristics. The formula is: ε=ε 0 ·(1+β H H+β p ·(PP 0 ), where ε 0 is the dielectric constant under standard atmospheric conditions, β H is the humidity influence coefficient, β P is the air pressure influence coefficient, P 0 is the standard gas pressure value; then the discharge range A is corrected using the corrected dielectric constant ε, the formula is: Among them, A C is the corrected discharge coverage, and A is the uncorrected discharge coverage. Through the above steps, the environmental correction unit can effectively compensate for the sensor sensitivity drift and air dielectric constant fluctuation caused by changes in temperature, humidity and air pressure, and output the corrected discharge intensity and discharge coverage data to provide accurate input for the risk assessment unit.

[0103] The fault prediction module includes a historical data storage unit, a data processing unit, a deep learning prediction unit and a trend generation unit; wherein:

[0104] Historical data storage unit: used to store historical corona discharge data during the operation of live equipment, including historical risk level, discharge intensity, coverage and environmental condition data; historical corona discharge data are arranged in chronological order and updated regularly to ensure the timeliness of the prediction model;

[0105] Data processing unit: receives the current corona discharge risk level output by the data analysis module, formats it, and generates input data containing timestamp, risk level, discharge intensity and coverage; at the same time, aligns the current data with the historical data in the historical data storage unit to generate time series data, ensuring the time series of the input data is complete;

[0106] Deep learning prediction unit: Based on the time series data generated by the data processing unit, the long short-term memory network (LSTM) model is used for prediction, and the future risk level of multiple time steps is output;

[0107] The calculation process of the model is as follows:

[0108] Assume that the input time series data is X t , where each time step contains the current risk level R t , discharge intensity F t 、Coverage A t and environmental condition parameters;

[0109] The hidden state h at time step t is calculated by the hidden state update formula of LSTM t , the update formula is: t =f(W h ·h t-1 +W x ·X t +b h ), where W h and W x are the weight matrices of hidden state and input, respectively, and b h is the bias term, f is the activation function;

[0110] The output layer is used to predict the future risk level. The formula is: in, is the predicted risk level for the next time step, W o is the output layer weight matrix, b o is the output bias term, and g is the output layer activation function.

[0111] Trend generation unit: combines the future risk levels of multiple time steps output by the deep learning prediction unit into a discharge trend prediction result, and expresses it in the form of a time series curve, showing the changing trend of the corona discharge risk level in the future period of time; through the collaborative work of the above units, the fault prediction module can use the deep learning model to achieve high-precision prediction of the future risk level based on the current corona discharge risk level and historical data, and generate a discharge trend prediction result to provide accurate input for the early warning module.

[0112] The early warning module includes a threshold judgment unit, an early warning generation unit and an information transmission unit; wherein:

[0113] Threshold judgment unit: receives the discharge trend prediction result of the trend generation unit in the fault prediction module, and compares the predicted risk level of each time step with the preset threshold; the preset threshold is set according to the equipment operation specification and risk management requirements. When the predicted risk level of any time step is greater than the preset threshold, an early warning signal will be triggered;

[0114] Warning generation unit: used to generate a warning information report based on the discharge trend prediction result after the threshold judgment unit triggers the warning signal. The warning information report includes equipment information, risk information, time information and recommended measures; the equipment information is used to indicate the equipment where corona discharge occurs and the specific monitoring location; the risk information is used to list the risk level and change trend predicted in the future; the time information is used to indicate the specific time when the risk level exceeds the preset threshold in the prediction; the recommended measures are used to provide maintenance suggestions for the current risk situation, including detection, maintenance or operation adjustment plans;

[0115] Information transmission unit: used to send the generated early warning report to the maintenance personnel through various channels, including SMS, email and system notification; SMS is used to send the summary information of the early warning report to the maintenance personnel's mobile terminal via SMS; email is used to send the complete early warning report as an attachment to the designated maintenance team mailbox; system notification is used to pop up an alarm window on the interface of the system monitoring platform, display the content of the early warning report, and provide operational options for risk handling suggestions.

[0116] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0117] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. Real-time detection and fault warning system for live equipment, characterized in that: It includes corona detection module, environmental monitoring module, data integration and transmission module, data analysis module, fault prediction module and early warning module; among which: Corona detection module: deployed at designated locations of live equipment, using ultraviolet imaging technology to capture the corona discharge phenomenon on the equipment surface in real time, generating two-dimensional images and video data of corona discharge; Environmental monitoring module: used to collect real-time environmental data around live equipment, including temperature, humidity and air pressure; Data integration and transmission module: connects the corona detection module and the environmental monitoring module to synchronize and integrate the output data of the two modules, generate a standardized data set, and transmit it to the data analysis module; Data analysis module: receives the standardized data set sent by the data integration and transmission module, and uses the extraction algorithm of the discharge edge detection to analyze the two-dimensional image and video data of the corona discharge to quantify the discharge intensity and range, and corrects it based on the environmental data to output the current corona discharge risk level; Fault prediction module: Based on the current corona discharge risk level output by the data analysis module, combined with historical data and deep learning models, predict the future risk level and generate discharge trend prediction results; Early warning module: When the discharge trend prediction results show that the future risk level is about to exceed the preset threshold, the early warning signal is triggered, an early warning information report is automatically generated, and maintenance personnel are notified through multiple channels.

2. The real-time detection and fault warning system for live equipment according to claim 1 is characterized in that: The corona detection module includes a mounting unit, an ultraviolet sensing unit, an image processing unit and a data output unit; wherein: Installation unit: It is configured at the high-voltage contacts, insulator surfaces and wire joints of live equipment to fix the equipment; UV sensing unit: used to capture the UV light signal generated on the surface of the charged equipment during the corona discharge process, and convert the UV light signal in the corona discharge area into a digital signal using a UV imaging detector with a wavelength range of 240 to 280 nanometers; Image processing unit: connected to the ultraviolet sensor unit, used to receive the digital signal generated by the ultraviolet sensor unit, extract the discharge intensity and spatial position information through signal enhancement, noise filtering and target area recognition algorithm, and generate two-dimensional images and video data reflecting the corona discharge phenomenon in real time; Data output unit: connected to the image processing unit, used to transmit the generated corona discharge two-dimensional image and video data to the data integration and transmission module.

3. The real-time detection and fault warning system for live equipment according to claim 2 is characterized in that: The image processing unit specifically includes: Signal enhancement: Use contrast stretching technology to process the received digital signal to enhance the brightness and details of the image; Noise filtering: Use Gaussian filtering technology to remove high-frequency noise in the enhanced image, retain the relevant features of the discharge area, and eliminate background noise; Target area recognition: Using threshold segmentation technology, the area with brightness exceeding the preset threshold is identified from the enhanced image, and the target area of ​​corona discharge is extracted; the part with high brightness is marked as the discharge area, and the rest is marked as the non-discharge area, forming a preliminary binary image of the target area; Feature extraction: Calculate the sum of the brightness of all pixels in the target area and quantify it as the discharge intensity; and calculate the coordinates of the center point of the target area based on the pixel distribution of the target area to determine the spatial position of the discharge phenomenon; Two-dimensional image generation: Using the target area and discharge intensity data, the boundary information and brightness distribution of the discharge area are integrated to generate a two-dimensional image containing the discharge area and intensity distribution; Video data generation: The continuously generated two-dimensional images are encoded in time series to form video data that can dynamically reflect the changes in the corona discharge phenomenon.

4. The real-time detection and fault warning system for live equipment according to claim 1 is characterized in that: The environmental monitoring module includes a temperature acquisition unit, a humidity acquisition unit and an air pressure acquisition unit; wherein: Temperature acquisition unit: installed around live equipment to monitor ambient temperature in real time. It measures the ambient air temperature through a built-in thermistor and converts the measured value into a digital signal output. Humidity collection unit: The relative humidity in the air is detected in real time through the configured capacitive humidity sensor. The sensor calculates the humidity by sensing the change of capacitance value and converts the calculation result into a standardized digital signal; Air pressure acquisition unit: A micro air pressure sensor is used to sense the atmospheric pressure in the environment in real time, and the detected air pressure changes are converted into standardized digital signal output through the internal piezoresistive measurement component.

5. The real-time detection and fault warning system for live equipment according to claim 1 is characterized in that: The data integration and transmission module includes a data receiving unit, a synchronization processing unit, a data integration unit and a formatting unit; wherein: Data receiving unit: connected to the corona detection module and the environment monitoring module respectively, used for receiving in real time the corona discharge image and video data output by the corona detection module, and the environmental data of temperature, humidity and air pressure output by the environment monitoring module; Synchronous processing unit: time-stamps the multi-source data collected by the data receiving unit, and performs synchronous correction on the data according to a unified clock signal to ensure the consistency of the corona detection data and the environmental monitoring data in the time dimension; Data integration unit: Integrate the synchronized data according to the association rules, associate and store the corona discharge data with the environmental data of the corresponding time period to form a unified multi-dimensional data set; Formatting unit: Converts the multidimensional data set generated by the integration unit into a standardized data set, including fields and parameter values ​​in a fixed format.

6. The real-time detection and fault warning system for live equipment according to claim 1 is characterized in that: The data analysis module includes a discharge feature extraction unit, an environment correction unit and a risk assessment unit; wherein: Discharge feature extraction unit: Based on the two-dimensional image and video data in the standardized data set, the edge detection algorithm is used to extract the boundary information of the corona discharge area, and the discharge intensity is calculated by accumulating pixel values. At the same time, the regional statistical method is used to determine the coverage of the discharge; the discharge intensity represents the total energy of the discharge light signal, and the discharge range represents the physical size of the discharge area; Environmental correction unit: based on the corresponding environmental data in the standardized data set, the discharge intensity and range are corrected; the temperature data is used to correct the influence of the change in sensor sensitivity, and the humidity and air pressure data are used to compensate for the fluctuation of the air dielectric constant; Risk assessment unit: Based on the preset risk level judgment rules, the comprehensive risk score of the current corona discharge is calculated in combination with the corrected discharge intensity and range, and the corresponding risk level is matched; the risk score is generated by multi-dimensional parameter weighting, and the risk level is divided into three levels: low, medium and high.

7. The real-time detection and fault warning system for live equipment according to claim 6 is characterized in that: The discharge feature extraction unit specifically includes: Boundary information extraction: Gradient calculation is performed on the two-dimensional image in the standardized data set, and the brightness change rate of each pixel is calculated using the Sobel operator. The formula is: Where M(p, q) is the gradient amplitude at the pixel position (p, q); M x (p, q) and M y (p, q) are the gradient components in the horizontal and vertical directions respectively; based on the calculated gradient amplitude map, the non-maximum suppression algorithm is applied to remove the response of the non-boundary part, and the boundary pixels of the corona discharge area are extracted by the double threshold method to generate boundary information; Discharge intensity calculation: Using boundary information, all pixels in the corona discharge area are extracted from the original image; and the brightness values ​​of all pixels are accumulated to quantify the total brightness of the discharge area. The formula is: F = ∑ (p , q)∈Z l(p, Q), where F represents the total discharge intensity value, Z is the pixel set in the discharge area, and L(p, q) is the brightness value of the pixel point (p, q); Discharge coverage calculation: Count the total number of pixels in the discharge area and calculate the actual physical area of ​​the discharge area in combination with the image resolution. The formula is: Where A is the actual area covered by the discharge; N is the total number of pixels in the discharge area; R s The physical size corresponding to a single pixel.

8. The real-time detection and fault warning system for live equipment according to claim 7 is characterized in that: The environmental correction unit specifically comprises: Temperature correction: First, obtain the current temperature value in the standardized data set and set it as T; then calculate the sensitivity correction coefficient K based on the temperature drift characteristics of the sensor sensitivity T , the formula is: K T =1+α·(T-T0), where α is the temperature sensitivity coefficient of the sensor, T0 is the reference operating temperature of the sensor, and T is the current temperature value; finally, the correction coefficient K is used T Correct the total discharge intensity value F, the formula is: Among them, F T is the corrected discharge intensity, and F is the uncorrected discharge intensity; Humidity and air pressure correction: Get the current humidity value H and air pressure value P in the standardized data set, and calculate the correction coefficient of the current air dielectric constant ε based on the atmospheric characteristics. The formula is: ε=ε0·(1+β H H+β P ·(P-P0)), where ε0 is the dielectric constant under standard atmospheric conditions, β H is the humidity influence coefficient, β P is the air pressure influence coefficient, P0 is the standard air pressure value; then the discharge range A is corrected using the corrected dielectric constant ε, the formula is: Among them, A C is the corrected discharge coverage, and A is the uncorrected discharge coverage.

9. The real-time detection and fault warning system for live equipment according to claim 6, characterized in that: The fault prediction module includes a historical data storage unit, a data processing unit, a deep learning prediction unit and a trend generation unit; wherein: Historical data storage unit: used to store historical corona discharge data during the operation of live equipment, including historical risk level, discharge intensity, coverage and environmental condition data; arrange historical corona discharge data in chronological order; Data processing unit: receives the current corona discharge risk level output by the data analysis module, formats it, and generates input data containing timestamp, risk level, discharge intensity and coverage; at the same time, aligns the current data with the historical data in the historical data storage unit to generate time series data, ensuring the time series of the input data is complete; Deep learning prediction unit: Based on the time series data generated by the data processing unit, it uses the long short-term memory network model to make predictions and output the future risk level of multiple time steps; Trend generation unit: The future risk levels of multiple time steps output by the deep learning prediction unit are combined into a discharge trend prediction result, and expressed in the form of a time series curve to show the changing trend of the corona discharge risk level in the future period of time.

10. The real-time detection and fault warning system for live equipment according to claim 1, characterized in that: The early warning module includes a threshold judgment unit, an early warning generation unit and an information transmission unit; wherein: Threshold judgment unit: receives the discharge trend prediction result of the trend generation unit in the fault prediction module, and compares the predicted risk level of each time step with the preset threshold. When the predicted risk level of any time step is greater than the preset threshold, an early warning signal will be triggered; Warning generation unit: used to generate a warning information report according to the discharge trend prediction result after the threshold judgment unit triggers the warning signal. The warning information report includes equipment information, risk information, time information and recommended measures; Information transmission unit: used to send the generated early warning report to the maintenance personnel through various channels, including SMS, email and system notification.

Images