Online monitoring system for insulation state of power equipment

Abstract

The invention relates to the technical field of power equipment monitoring, and discloses a power equipment insulation state online monitoring system. The system comprises a multi-source data fusion module which collects operation parameters, environment sensing and historical insulation degradation record data streams in real time; and the space insulation field intensity distribution module performs field intensity dynamic mapping on the operation parameter data flow in combination with an equipment structure topological graph and an insulation degradation feature library to generate space insulation field intensity distribution feature information. The reference state topology module constructs an insulation state reference topology network through dynamic weight distribution according to element service duration distribution and operation modes; the insulation degradation dynamic detection module analyzes the field intensity deviation degree of the two regions and generates a primary insulation degradation index. When the primary index exceeds the threshold value, the delay response verification module executes continuous time sequence degradation trajectory tracking verification; the insulation state trend prediction module performs multi-step state evolution simulation to obtain predicted field intensity distribution characteristics, and the monitoring compensation optimization module corrects the primary indexes according to the predicted field intensity distribution characteristics and generates optimized insulation degradation indexes.

Description

Technical Field

[0001] This invention relates to the field of power equipment monitoring technology, specifically to an online monitoring system for the insulation status of power equipment. Background Technology

[0002] During the operation of a power system, the insulation condition of electrical equipment directly affects the stability and safety of the entire system. If insulation performance deteriorates and is not detected in time, it can easily lead to equipment failure or even large-scale power outages. Currently, monitoring methods for the insulation condition of power equipment are mainly divided into two categories: offline monitoring and online monitoring. Offline monitoring requires shutting down the equipment from the operating system, which not only affects the normal power supply but also has limitations such as long monitoring cycles and the inability to capture changes in insulation condition in real time. The applicability of offline monitoring is significantly reduced, especially for critical equipment that undertakes important power supply tasks and cannot be frequently shut down.

[0003] While online monitoring technology can collect insulation parameters of equipment during operation, existing online monitoring systems generally suffer from a lack of data collection dimensions. They mostly focus only on the equipment's own operating parameters (such as partial discharge and dielectric loss), neglecting the impact of environmental factors (such as temperature, humidity, and pollution levels) on the insulation condition. This results in the collected data failing to comprehensively reflect the true state of the equipment's insulation. Furthermore, existing systems lack effective spatial field distribution modeling capabilities when analyzing and processing the collected data. They struggle to construct insulation condition assessment models that fit the actual operating scenario based on the equipment's structural topology, often only able to qualitatively assess insulation degradation without accurately locating the deteriorated areas and their degree.

[0004] Existing online monitoring systems often use fixed threshold modes for baseline state setting, failing to consider the impact of differences in equipment component service life and operational mode changes on insulation baseline state. As equipment operating time increases, component insulation performance naturally degrades, and the heating and electric field distribution of equipment differ significantly under different operating modes (such as full load, light load, start-up, and shutdown). Fixed threshold modes easily lead to misjudgments or missed detections. Regarding insulation degradation trend prediction, existing systems mostly rely on simple linear predictions based on historical data, failing to simulate the dynamic evolution of insulation state under the combined effects of multiple factors, resulting in insufficient accuracy and foresight in the prediction results. Furthermore, when the system detects suspected insulation degradation signals, the lack of an effective delayed response verification mechanism makes it difficult to distinguish between genuine degradation signals and interference signals, further reducing the reliability of monitoring results. These problems make existing power equipment insulation condition monitoring technologies insufficient to meet the demands of power systems developing towards high reliability and high intelligence. A new online insulation condition monitoring system capable of multi-source data fusion, accurate modeling, dynamic baseline setting, reliable detection, and accurate prediction is needed. Summary of the Invention

[0005] The purpose of this invention is to provide an online monitoring system for the insulation status of power equipment to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides an online monitoring system for the insulation status of power equipment, the system comprising: The multi-source data fusion module is used to collect real-time data streams of operating parameters of power equipment, environmental sensor data streams, and historical insulation degradation records. The spatial insulation field strength distribution module is used to dynamically map the field strength of the operating parameter data stream based on the structural topology map of the power equipment and in combination with a preset insulation degradation feature library, and generate spatial insulation field strength distribution feature information. The baseline state topology module is used to construct an insulation state baseline topology network based on the component service duration distribution and equipment operation mode of the power equipment through a dynamic weight allocation mechanism. The insulation degradation dynamic detection module is used to perform regional field strength deviation analysis based on the spatial insulation field strength distribution characteristics and the output of the insulation state reference topology network, and generate primary insulation degradation index. The delayed response verification module is used to execute the continuous time-series degradation trajectory tracking verification rule when the primary insulation degradation index exceeds the preset deviation threshold. The insulation state trend prediction module is used to perform multi-step state evolution simulation on the spatial insulation field strength distribution characteristic information and environmental sensing data stream to generate predicted insulation field strength distribution characteristic information. The monitoring, compensation, and optimization module is used to perform field strength compensation correction on the primary insulation degradation index based on the predicted insulation field strength distribution characteristics information, and generate an optimized insulation degradation index.

[0007] Preferably, the multi-source data fusion module is used to collect real-time operating parameter data streams of power equipment, environmental sensor data streams, and historical insulation degradation record data streams, including: Obtain the equipment model identifier and installation location coordinates of the power equipment; Load the insulation parameter acquisition protocol set corresponding to the device model according to the device model identifier; Activate the environmental sensor array acquisition node based on the installation location coordinates; Simultaneously retrieve the historical insulation degradation feature database index associated with the equipment model identifier; The multi-channel concurrent acquisition task generates the operating parameter data stream, environmental sensor data stream, and historical insulation degradation record data stream.

[0008] Preferably, the reference state topology module is used to construct an insulation state reference topology network based on the component service duration distribution and equipment operating modes of the power equipment through a dynamic weight allocation mechanism, including: Analyze the service duration distribution of the components to generate a discrete feature vector of service duration; Extract the modal feature codes corresponding to the operating modes of the device; Establish an initial topology graph with power equipment components as vertices and insulation connections between components as edges; Calculate the initial weight factor for each vertex based on the discrete feature vector of service duration; The modal feature encoding is fused to dynamically calibrate the edge connection strength; Output the insulation state reference topology network with weighted attributes.

[0009] Preferably, the insulation degradation dynamic detection module is used to perform regional field strength deviation analysis based on the spatial insulation field strength distribution characteristic information and the output of the insulation state reference topology network, and generate primary insulation degradation indices, including: The spatial insulation field strength distribution characteristic information is divided into field strength partition datasets according to the equipment area; Match the reference field strength interval of each region in the insulation state reference topology network; Calculate the multidimensional deviation of the real-time field strength value of each zone from the reference field strength interval; Aggregate the multidimensional deviations of each partition to generate a comprehensive regional degradation coefficient; Based on the preset degradation level mapping table, the regional comprehensive degradation coefficient is converted into the primary insulation degradation index.

[0010] Preferably, the delay response verification module is used to execute continuous time-series degradation trajectory tracking verification rules when the primary insulation degradation index exceeds a preset deviation threshold, including: The time when the primary insulation degradation index first exceeds the limit is recorded as the starting verification time stamp; Construct a sliding time window based on the initial verification time scale; The trajectory of changes in insulation degradation indicators is continuously captured within the sliding time window; When the degradation trajectory meets the duration threshold and degradation monotonicity condition, an insulation anomaly confirmation signal is triggered. Otherwise, reset the sliding time window and continue monitoring.

[0011] Preferably, the insulation state trend prediction module is used to perform multi-step state evolution simulation on the spatial insulation field strength distribution characteristic information and environmental sensing data stream to generate predicted insulation field strength distribution characteristic information, including: Extracting the temporal variation patterns of historical insulation field strength distribution characteristics; Temperature and humidity change gradients in the associated environmental sensor data stream; Construct a field strength-environment coupled evolution model; Simulate the future multi-step insulation field strength distribution using a rolling time-domain prediction algorithm; Output the predicted insulation field strength distribution characteristics with confidence intervals.

[0012] Preferably, the monitoring compensation optimization module is used to perform field strength compensation correction on the primary insulation degradation index based on the predicted insulation field strength distribution characteristic information, and generate an optimized insulation degradation index, including: Compare the deviation between the real-time spatial insulation field strength distribution characteristics and the predicted insulation field strength distribution characteristics; When the deviation exceeds the tolerance threshold, calculate the field strength distribution compensation coefficient; The primary insulation degradation index is linearly corrected based on the field strength distribution compensation coefficient. The optimized insulation degradation index is output after compensation calibration.

[0013] Preferably, the system further includes: a baseline topology adaptive module, used to dynamically reallocate node weights in the insulation state baseline topology network according to the optimized insulation degradation index, including: The optimized insulation degradation index is analyzed to obtain the severity level of degradation in each region; Calculate the weight decay factor of the corresponding topology node based on the severity of degradation. The vertex weights of the insulation state baseline topology are updated based on the weight decay factor. Reconstruct the mapping relationship between edge connection strength and vertex weight.

[0014] Preferably, the system further includes: an insulation field strength reconstruction module, used to reconstruct the equipment insulation field strength map based on the compensated spatial insulation field strength distribution characteristic information when the optimized insulation degradation index triggers the reconstruction condition, including: When the optimized insulation degradation index continues to exceed the limit and the compensation correction is ineffective, activate the field strength spectrum reconstruction command. Obtain the compensated and corrected spatial insulation field strength distribution characteristics at the current moment; Spatial interpolation calculation of insulation field strength is performed using the integrated equipment structure topology map; Generate a new device insulation field strength map with three-dimensional coordinate attributes.

[0015] Preferably, the system further includes: an anomaly handling decision module, used to generate an insulation status maintenance strategy based on the reconstructed equipment insulation field strength map, including: Identify the coordinates of weak areas in the insulation field strength spectrum of newly built equipment; Match the emergency response plan template with the associated operation and maintenance procedure library of the power equipment; Generate a priority sequence for handling based on the real-time load rate of the equipment; The output includes the insulation status maintenance strategy, which includes the coordinates of the disposal location, the disposal method, and the execution sequence.

[0016] Compared with the prior art, the beneficial effects of the present invention are: Through a multi-source data fusion module, the system collects and integrates real-time data streams of power equipment operating parameters, environmental sensor data, and historical insulation degradation records. This breaks through the limitations of existing monitoring systems that rely on a single data collection dimension. It comprehensively captures various factors affecting the insulation status of equipment, ensuring that the collected data more closely reflects the actual operating scenario of the equipment insulation. This provides richer and more comprehensive basic data support for subsequent insulation status analysis. Compared to traditional systems that only focus on the equipment's own operating parameters, this system considers the influence of environmental factors and historical degradation records. This allows the assessment of insulation status to no longer be limited to a single dimension, but to reflect the equipment's insulation condition from multiple dimensions and angles, effectively avoiding assessment bias caused by data limitations.

[0017] The spatial insulation field strength distribution module, based on the equipment's structural topology map and combined with a pre-set insulation degradation feature library, dynamically maps the field strength of the operating parameter data stream to generate spatial insulation field strength distribution feature information, thus solving the problem of existing systems lacking spatial field strength modeling capabilities. By constructing a spatial field strength model that matches the actual equipment structure, it can accurately present the electric field distribution in different regions inside the equipment. This not only enables qualitative judgment of insulation degradation but also accurately locates the degraded areas, providing a clear direction for subsequent maintenance and repair, avoiding the drawbacks of traditional systems that cannot locate degraded areas and suffer from significant blindness in maintenance work. Furthermore, the introduction of the pre-set insulation degradation feature library allows the dynamic field strength mapping process to incorporate known degradation patterns, further improving the accuracy and relevance of the spatial insulation field strength distribution feature information.

[0018] The baseline topology module constructs an insulation state baseline topology network based on the service life distribution of equipment components and the equipment's operating modes, using a dynamic weight allocation mechanism. This changes the existing system's approach of setting baseline states with fixed thresholds. This module fully considers the natural degradation of insulation performance due to differences in the service life of equipment components, as well as the impact of changes in equipment operating conditions under different operating modes on the insulation baseline state. Through dynamic weight allocation, the baseline state can adaptively adjust with changes in equipment operating time and operating modes, ensuring that the baseline state always closely matches the actual operating conditions of the equipment. This dynamic baseline setting method effectively avoids the misjudgments and omissions caused by equipment aging or changes in operating modes under the fixed threshold mode, improving the accuracy of insulation state assessment.

[0019] The dynamic insulation degradation detection module analyzes regional field strength deviations based on spatial insulation field strength distribution characteristics and the output of the insulation state reference topology network, generating primary insulation degradation indices and enabling quantitative analysis of insulation degradation. Through regional field strength deviation analysis, the difference between the insulation state and the reference state in each region can be accurately calculated, clarifying the severity of degradation. Compared to the qualitative judgment of traditional systems, this approach is more scientific and practical. The delay response verification module executes continuous time-series degradation trajectory tracking verification rules when the primary insulation degradation index exceeds a preset deviation threshold, effectively distinguishing between real degradation signals and interference signals. Continuous time-series tracking observes the changing trends of degradation indices, avoiding false alarms caused by instantaneous interference signals, improving the reliability of monitoring results, and reducing unnecessary maintenance costs and workload.

[0020] The insulation condition trend prediction module performs multi-step state evolution simulations based on spatial insulation field strength distribution characteristics and environmental sensor data streams to generate predicted insulation field strength distribution characteristics, overcoming the limitations of simple linear predictions in traditional systems. This multi-step state evolution simulation comprehensively considers the dynamic changes in equipment operating parameters and environmental factors, simulating the development trend of insulation condition under different conditions. This makes the prediction results more forward-looking and accurate, helping maintenance personnel to grasp the possible direction and extent of insulation degradation in advance, providing a valuable reference for developing preventative maintenance strategies. The monitoring compensation and optimization module performs field strength compensation correction on the primary insulation degradation index based on the predicted insulation field strength distribution characteristics, generating optimized insulation degradation indices and further improving the accuracy of the degradation indices. By introducing predicted information for compensation correction, errors caused by the limitations of current monitoring data or environmental fluctuations can be eliminated, making the final output optimized insulation degradation index more accurately reflect the actual degradation status of the equipment insulation, providing a more reliable basis for subsequent maintenance decisions. Attached Figure Description

[0021] Figure 1 This is a timing diagram of the online monitoring system for the insulation status of power equipment according to the present invention; Figure 2 A flowchart illustrating the operation of the baseline state topology module; Figure 3 A flowchart illustrating the operation of the dynamic insulation degradation detection module. Figure 4 A flowchart illustrating the operation of the insulation state trend prediction module; Figure 5 A flowchart illustrating the operation of the baseline topology adaptive module. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Please see Figure 1 This invention provides an online monitoring system for the insulation status of power equipment, the system comprising: The multi-source data fusion module collects real-time data streams of operating parameters, environmental sensor data, and historical insulation degradation records from power equipment. The spatial insulation field strength distribution module, based on the structural topology map of the power equipment and combined with a pre-set insulation degradation feature library, dynamically maps the field strength of the operating parameter data streams to generate spatial insulation field strength distribution characteristic information. The baseline state topology module constructs an insulation state baseline topology network based on the component service duration distribution and equipment operating modes of the power equipment through a dynamic weight allocation mechanism. The insulation degradation dynamic detection module performs regional field strength deviation analysis based on the spatial insulation field strength distribution characteristic information and the output of the insulation state baseline topology network, generating primary insulation degradation indicators. The delay response verification module executes continuous time-series degradation trajectory tracking verification rules when the primary insulation degradation indicator exceeds a pre-set deviation threshold. The insulation state trend prediction module performs multi-step state evolution simulation on the spatial insulation field strength distribution characteristic information and environmental sensor data streams to generate predicted insulation field strength distribution characteristic information. The monitoring compensation optimization module performs field strength compensation correction on the primary insulation degradation indicators based on the predicted insulation field strength distribution characteristic information, generating optimized insulation degradation indicators.

[0024] Example 1: See Figure 2 The multi-source data fusion module begins with the identification of the target power equipment. The system first obtains the unique equipment model identifier, typically provided by the equipment nameplate or system registration database, containing key information such as the manufacturer, specifications, and rated parameters. Next, it acquires the equipment's installation location coordinates, using a Geographic Information System (GIS) standard format with meter-level accuracy, to pinpoint the equipment's physical location. Based on the equipment model identifier, the system loads the corresponding insulation parameter acquisition protocol set from a pre-configured protocol library. This protocol set details the various data acquisition parameters: for operational parameters, the acquisition frequency is set to 10 times per second, with a measurement accuracy of 0.5%, and the data format uses the IEEE floating-point standard; for environmental parameters, the acquisition interval is set to once every 30 seconds, with a temperature measurement range of -40℃ to +85℃ and a humidity measurement accuracy of ±3%RH. The protocol set also includes data verification rules, such as the generation and verification methods for CRC cyclic redundancy check codes.

[0025] The installation location coordinates are used to activate the environmental sensor array acquisition nodes deployed around the equipment. These nodes are distributed according to a pre-planned topology, including temperature sensors mounted on the transformer tank surface, humidity sensors located near the radiators, and air quality monitoring points set up around the equipment. Each acquisition node has an independent network address, and the system establishes communication connections with each node via the Modbus TCP protocol. Simultaneously with initiating real-time data acquisition, the system retrieves the historical insulation degradation feature database index associated with the equipment model identifier. This index is organized using a B+ tree data structure, supporting fast retrieval. The index content contains all insulation-related events recorded during the past operation of this equipment model, such as historical partial discharge data, dielectric loss factor change records, and insulating oil chromatographic analysis results. This historical data is sorted by timestamp and stored in a large-capacity time-series database.

[0026] When performing multi-channel concurrent data acquisition tasks, the system adopts a multi-threaded architecture. The operational parameter acquisition thread obtains real-time operational data through the device's monitoring IED (Intelligent Electronic Device), including the effective values ​​and phase angles of three-phase voltage and current, instantaneous values ​​of active and reactive power, and readings from the winding thermometer. The environmental sensor data acquisition thread simultaneously acquires environmental parameters such as temperature, humidity, and air pressure from multiple sensor nodes. The historical data retrieval thread extracts insulation degradation-related records from the historical database in parallel over the past three years. All data streams are timestamped with a uniform time synchronization accuracy down to the millisecond level, ensuring consistency of data from different sources across time. The acquired operational parameter data stream contains electrical quantity measurements from the device, which undergo digital filtering to eliminate noise interference. In addition to basic temperature and humidity readings, the environmental sensor data stream includes sensor status information such as battery level and communication signal strength. The historical insulation degradation record data stream provides trend data on changes in insulation performance during long-term device operation. These three types of data streams are cached in different memory buffers, awaiting calls from subsequent processing modules.

[0027] The module first analyzes the service life distribution of power equipment components, information sourced from the equipment ledger management system. For equipment like transformers, components include high-voltage windings, low-voltage windings, cores, insulating oil, bushings, tap changers, etc. The commissioning time of each component is precisely recorded, and the system calculates the cumulative operating hours since commissioning, generating a discrete feature vector of service life. This vector is normalized, mapping the actual operating time to the interval [0,1], where 0 represents a brand-new state and 1 represents reaching the design life. Equipment operating mode identification is based on real-time operating parameter data streams. The system continuously monitors parameters such as equipment load rate, operating temperature, and cooling system status. According to preset mode classification rules, the equipment operating state is divided into multiple modes: light-load operation mode, rated operation mode, overload operation mode, and no-load operation mode. Each mode corresponds to a unique mode feature code, represented using one-hot encoding for easy computer processing.

[0028] When establishing the initial topology, each component of the power equipment is used as a vertex, and each vertex contains the component's identifier, type attribute, and physical location information. Insulation relationships between components are represented as edges, indicating mutual influence on insulation. The initial connection strength of the edges is set based on the physical distance between components and the characteristics of the insulation material. A negative correlation algorithm is used in the calculation, meaning that components with longer service lives have smaller weight factors, indicating that the component's influence on the overall insulation state should be reduced. The specific values ​​of the weight factors are calculated using a linear decay function, while also considering the importance coefficient of the component type.

[0029] When the equipment is in overload operation mode, the connection strength of insulation-related edges associated with heat generation increases; when the ambient humidity is high, the connection strength of edges associated with surface insulation adjusts. The calibration process uses a weighted average algorithm, with the weight coefficients for real-time operation modes being higher than historical statistical values. The final output insulation state benchmark topology network is a directed graph structure with weighted attributes. Each vertex has a corresponding weight factor, and each edge has a dynamically adjusted connection strength value. This network serves as a benchmark model for insulation state assessment, realistically reflecting the expected insulation performance level of the equipment under current operating conditions and service duration. This benchmark topology network is continuously updated as the equipment operates over time and its operating state changes, ensuring that the benchmark model remains synchronized with the actual condition of the equipment.

[0030] Example 2: See Figure 3The processing of the insulation degradation dynamic detection module begins with the regional segmentation of the spatial insulation field strength distribution characteristics. The high-voltage winding region is further subdivided into upper, middle, and lower winding zones, while the low-voltage winding region uses the same subdivision method. The core and clamping parts region is divided into three-phase zones (A, B, and C) according to phase, and the insulating oil region is divided into multiple flow zones based on the oil channel structure. This fine regional segmentation ensures the accuracy of the field strength analysis. During the generation of the field strength zone dataset, each zone contains a set of field strength sampling points. For the winding region, the sampling points are evenly distributed along the winding height; for the insulating oil region, the sampling points are set at the main oil channel locations; for the bushing region, the sampling points are distributed along the bushing axial and radial directions. Each sampling point records its three-dimensional coordinate information and real-time field strength measurement value, which comes from the field strength sensor array installed inside the equipment. The raw data collected by the sensors undergoes signal conditioning and digital processing to convert it into standardized field strength values.

[0031] When matching the reference field strength intervals for each region in the insulation state reference topology network, the system calls a pre-established reference model. This model is derived from the insulation performance data at the initial stage of equipment commissioning and historical normal operation data. Each region corresponds to a reference field strength interval, which is defined by an upper limit and a lower limit. When calculating the multidimensional deviation of the real-time field strength value of each region relative to the reference field strength interval, the system uses multiple deviation indices. Absolute value deviation calculates the absolute deviation between the real-time field strength and the median of the reference interval; gradient deviation analyzes the rate of change of field strength, obtained by comparing the difference between the current field strength value and historical data; distribution pattern deviation assesses the uniformity of field strength distribution within the region, using statistical methods to calculate the dispersion of field strength values. Each deviation is normalized and converted into a dimensionless value within the range of [0,1].

[0032] The process of aggregating the multidimensional deviations of each zone to generate the regional comprehensive degradation coefficient employs a weighted fusion algorithm. The weights of different zones are allocated according to their importance in the insulation system, with winding zones having a higher weight than other zones, and high-voltage zones having a higher weight than low-voltage zones. The weight of each deviation index is also adjusted according to its sensitivity, with gradient deviation typically having a higher weight than other indices. The aggregation calculation uses a linear weighted summation method to produce a comprehensive degradation coefficient value that reflects the regional insulation state. When converting the regional comprehensive degradation coefficient into a primary insulation degradation index based on a preset degradation level mapping table, the system calls predefined mapping rules. This mapping table discretizes continuous degradation coefficient values ​​into several insulation state levels; for example, 0-0.2 corresponds to a normal state, 0.2-0.4 to a warning state, 0.4-0.6 to an abnormal state, and above 0.6 to a dangerous state. Each state level is accompanied by a detailed state description and handling suggestions. The converted primary insulation degradation index includes a zone identifier, state level, and timestamp information.

[0033] A sliding time window is constructed based on the initial verification timescale, and the window size is dynamically adjusted according to the equipment type and operating status. For transformer equipment, the default window is 30 minutes, but the window length can be fixed or variable. The variable window adaptively adjusts according to changes in degradation indicators, shrinking when indicators change drastically and expanding when they are stable. Within the window, the system continuously captures the trajectory of insulation degradation indicators, including the degradation indicator value, rate of change, and acceleration. The acquisition frequency is adjusted according to the degree of degradation, initially once per second, increasing as the indicator rises. The trajectory data is stored in a circular buffer for easy access and analysis. When the degradation trajectory meets the duration threshold and degradation monotonicity condition, the system triggers an insulation anomaly confirmation signal. The duration threshold requires the indicator to exceed the limit for a certain period within the window, such as exceeding the warning state for 5 consecutive minutes. Degradation monotonicity requires the indicator to generally increase, allowing for small fluctuations, which are determined by calculating the linear regression slope through trend analysis. If the conditions are not met, the system resets the window and continues monitoring. Resetting includes clearing the buffer, resetting the timescale, and adjusting parameters. During monitoring, the system records changes in the environment and equipment operating status for subsequent analysis and diagnosis.

[0034] The output of the insulation degradation dynamic detection module not only includes simple status indicators but also provides detailed analytical data: deviation components for each region, aggregate weight allocation, and historical records of state transitions. This additional information provides rich data support for subsequent trend prediction and maintenance decisions. The verification results of the delay response verification module also include a detailed trajectory analysis report, including time window parameters, trajectory feature descriptions, and verification decision basis. The system's parameter settings support online adjustment, allowing users to modify various thresholds and parameters based on equipment characteristics and operational experience.

[0035] Example 3: See Figure 4 The insulation condition trend prediction module's processing begins with extracting time-series variation patterns from historical insulation field strength distribution characteristics. This module accesses the system's historical database to obtain field strength distribution data for various areas of the equipment over a past period. This data is organized in a time series, containing the field strength measurement value and its spatial distribution information at each sampling time point. Pattern extraction employs time series analysis methods to identify periodic variations, trend variations, and random fluctuations in the field strength data. For equipment like transformers, field strength variations often exhibit diurnal periodic characteristics related to load changes, as well as long-term periodic characteristics related to seasonal variations. The system uses a sliding window technique to analyze recent data, with the window size adaptively adjusted based on equipment type and operating history. During pattern recognition, the system pays particular attention to abrupt changes in field strength, which may correspond to significant changes in equipment operating status or sudden deterioration of insulation conditions.

[0036] The temperature and humidity gradients in the associated environmental sensor data stream are a key element in trend prediction. Environmental data comes from a multi-sensor network deployed around the device, including temperature, humidity, and barometric pressure sensors. The temperature gradient is obtained by calculating the difference between temperature values ​​at adjacent time points, and the humidity gradient is calculated using a similar method. These gradient values ​​reflect the dynamic characteristics of environmental conditions and are closely related to changes in the performance of insulating materials. For example, increased temperature accelerates the aging process of insulating materials, and increased humidity may reduce surface insulation strength. The system establishes a correlation model between environmental parameters and insulation field strength. This model, based on statistical analysis of historical data, quantifies the typical variation patterns of field strength under different environmental conditions.

[0037] Constructing a coupled field strength-environment evolution model is the core step in trend forecasting. This model links changes in field strength with changes in environmental factors, forming a multivariate forecasting framework. The model considers the temporal correlation of field strength itself, as well as the lagged effects of environmental factors on field strength. The model parameters are trained from historical data using machine learning methods and are updated periodically to maintain forecast accuracy. The mathematical expression of the model is as follows:

[0038] in: Indicates the future Predicted field strength at time [time]. This represents the current and historical field strength observations. and This indicates the current temperature and humidity measurements. and This indicates the current temperature and humidity gradient. This represents the set of model parameters. This nonlinear functional relationship is implemented through a neural network structure, which can capture the complex interaction between field strength and environmental factors.

[0039] A rolling time-domain prediction algorithm simulates the future multi-step insulation field strength distribution. This algorithm employs an iterative prediction approach, first predicting the field strength distribution at the next moment, then using the predicted value as a new observation to continue predicting the state at more distant moments. This rolling prediction method continuously incorporates the latest observation data, reducing the accumulation of prediction errors. During the multi-step prediction process, the system simultaneously predicts the changing trends of environmental parameters as input conditions for field strength prediction. The prediction step size is set according to application requirements, typically including short-term predictions (minutes to hours) and medium-term predictions (hours to days). Each prediction step generates complete spatial distribution information of the field strength, including predicted values ​​for all monitored areas.

[0040] The system outputs predicted insulation field strength distribution characteristics with confidence intervals. These confidence intervals are derived by analyzing historical prediction error statistics, reflecting the range of uncertainty in the prediction results. The system provides upper and lower limit estimates for each predicted value to form a prediction interval. The confidence level can be set as needed, with 95% being commonly used. Prediction results are output in a standardized format, including fields such as timestamp, predicted value, and confidence interval boundaries, and are stored in a prediction database for subsequent module analysis. The monitoring compensation optimization module is implemented based on a comparative analysis of the predicted insulation field strength distribution characteristics and real-time monitoring data. The module first compares the deviation between the real-time and predicted insulation field strength distribution characteristics, using a point-by-point comparison method to calculate the absolute and relative differences between the real-time measured value and the predicted value at each monitoring point. It records the spatial distribution and temporal variation characteristics of the deviation, analyzing deviation patterns and anomalies. When the deviation exceeds the tolerance threshold, the system calculates the field strength distribution compensation coefficient. The tolerance threshold is dynamically adjusted based on equipment type, operating status, and historical deviation statistics, and is typically set to the percentile value of the historical deviation distribution. The calculation of the compensation coefficient takes into account the magnitude of the deviation, the duration, and spatial consistency. Local deviations are compensated using local compensation strategies, while global deviations are adjusted using system-level compensation. The compensation coefficient is a multi-dimensional vector corresponding to the compensation amount for different regions and time points.

[0041] The primary insulation degradation index is linearly corrected based on the field strength distribution compensation coefficient. The correction process maps the compensation coefficient to the calculation formula of the degradation index, adjusting the index value to reflect the predicted information. The correction weights are dynamically allocated based on the reliability of the prediction results; higher weights are assigned when the prediction confidence interval is narrow, and lower weights are assigned when the confidence interval is wide. The corrected degradation index more accurately reflects the actual changing trend of the insulation condition. The optimized insulation degradation index after compensation calibration is output. This index includes basic condition level information and detailed compensation adjustment records. The system records metadata such as the time, compensation amount, and correction reason for each compensation operation, forming a complete compensation log. The optimized index value is used for subsequent alarm judgment and maintenance decisions, and is also fed back to the prediction model to improve future prediction accuracy.

[0042] The data storage and management adopt a layered architecture, with raw monitoring data, forecast data, and compensation data stored in different database layers. The data access interface provides a unified data retrieval service, supporting the data needs of various analytical applications. The system performance monitoring module continuously tracks the processing latency and resource consumption of forecast and compensation calculations, ensuring the system meets the real-time requirements of online monitoring. The collaborative work of the trend forecasting module and the compensation optimization module forms a closed-loop optimization system. Forecast results guide compensation adjustments, and the compensated monitoring data is used to improve the forecasting model. This iterative optimization mechanism enables the system to continuously adapt to changes in equipment status, improving the accuracy and reliability of monitoring results. The system also provides a manual intervention interface, allowing experienced engineers to review and adjust the automatic forecast and compensation results, organically combining artificial intelligence analysis with human expert experience.

[0043] Example 4: See Figure 5 The baseline topology adaptive module's processing is based on the analytical results of the optimized insulation degradation index. This module receives the optimized insulation degradation index data stream from the monitoring compensation optimization module. This data is organized by equipment region and includes an identifier for each region, the degradation index value, a timestamp, and confidence level information. The analytical process first verifies the validity of the index data, checking data integrity and time consistency, and then processes it by region. The system obtains the degradation severity level of each region through predefined level classification rules, which map continuous degradation index values ​​to discrete severity levels. For example, index values ​​between 0.0 and 0.2 correspond to level 0 (normal state), 0.2 to 0.4 correspond to level 1 (slight degradation), 0.4 to 0.6 correspond to level 2 (moderate degradation), 0.6 to 0.8 correspond to level 3 (severe degradation), and values ​​above 0.8 correspond to level 4 (dangerous state). Each level has a corresponding color code and descriptive text for easy visualization and subsequent processing.

[0044] When calculating the weight decay factor for the corresponding topology node based on the severity of degradation, the system employs a level mapping algorithm. This algorithm assigns a decay coefficient to each degradation level; the higher the level, the larger the decay coefficient, indicating that the node's weight in the baseline topology network should be reduced accordingly. The calculation considers the node's historical state changes. If the node's state continues to deteriorate, a larger decay step size is used; if the state improves, the decay degree is appropriately reduced. The calculation of the decay factor also references the node's initial weight and the states of neighboring nodes to maintain the overall consistency of the network structure.

[0045] When updating the vertex weights of the insulation state baseline topology network based on the weight decay factor, the system adopts a gradual adjustment strategy. The update process does not completely reset the weights, but rather applies a decay factor to the original weights, ensuring a smooth transition. Each vertex's weight update record includes a timestamp, original weight value, new weight value, and reason for adjustment. The weight update algorithm ensures that the total weight of the network remains unchanged; it only redistributes the weight ratios of each node. The system also records the historical trajectory of weight adjustments to analyze the evolution trend of the network structure. Edge connection strength reflects the degree of insulation association between nodes. When a vertex weight changes, the strength of the edges connected to it also needs to be adjusted accordingly. The system establishes a functional relationship between edge strength and vertex weight. When a vertex weight decays, the connection strength of the output edge weakens accordingly, while the connection strength of the input edge is adjusted comprehensively based on the state of adjacent nodes. The reconstruction of the mapping relationship uses an iterative optimization algorithm to ensure the stability and consistency of the network structure. The adjusted insulation state baseline topology network will serve as the benchmark for the next round of insulation state assessment, achieving dynamic adaptation and continuous optimization of the assessment standard.

[0046] When the optimized insulation degradation index continues to exceed limits and compensation correction is ineffective, the system activates a field strength spectrum reconstruction command. The criterion for continuous exceeding limits is that the degradation index exceeds the threshold for multiple consecutive monitoring cycles, and the compensation operation fails to bring the index back to the normal range. The determination of ineffective compensation correction is based on the trend of the index change after compensation; if the index value continues to deteriorate or shows no signs of improvement, the compensation correction is considered ineffective. The reconstruction command includes parameters such as reconstruction range, reconstruction accuracy, and reconstruction time, which are dynamically set according to the severity of degradation and the importance of the equipment. When acquiring the spatial insulation field strength distribution characteristics information after compensation correction at the current moment, the system reads the latest field strength data from the data cache. This data has already been processed by the monitoring compensation optimization module and includes compensation adjustments based on predictive information. The data format includes the three-dimensional coordinates of each monitoring point, the measured field strength value, the compensation amount, and the timestamp. The system performs a data quality check, removing outliers and unreliable data, and then performs data alignment and format conversion to prepare for subsequent reconstruction calculations.

[0047] The core step in the reconstruction process is to perform spatial interpolation calculations of insulation field strength using the integrated equipment structural topology map. The equipment structural topology map provides information on the equipment's geometry and material distribution, including structural features such as winding locations, insulation barriers, and cooling channels. The spatial interpolation calculation employs a physics-based interpolation algorithm that considers the dielectric properties of the insulating material, the electric field distribution, and the equipment's structural characteristics. The interpolation process first discretizes the equipment space into a three-dimensional mesh. Then, based on known monitoring point data, the field strength values ​​at the mesh nodes are calculated by solving the electric field distribution equation. The interpolation algorithm uses an adaptive mesh refinement technique, employing a finer mesh in regions with drastic field changes and a coarser mesh in regions with uniform field strength, balancing computational accuracy and efficiency.

[0048] When generating a new equipment insulation field strength map with three-dimensional coordinate attributes, the system combines the interpolation calculation results with the equipment coordinate system. Each grid node contains three-dimensional coordinate information and the calculated field strength value, forming a complete spatial distribution model of the field strength. The map data is organized in a hierarchical structure, including a basic geometry layer, a field strength data layer, a material attribute layer, and a labeling information layer. The system adds metadata information to the map, including generation time, data version, calculation parameters, and accuracy indicators. Version management of new and old maps adopts an incremental update method, recording the changes and impact scope of each reconstruction. Refer to Table 1 to show the correspondence between optimized insulation degradation indicators and weighted attenuation factors.

[0049] Table 1: Correspondence between Deterioration Level and Weighted Decay Factor

[0050] The system performance monitoring module tracks the computational load of baseline topology adaptation and field strength reconstruction in real time, dynamically adjusting the allocation of computational resources. When the device state changes drastically, the system automatically increases the computational priority, accelerating the update speed of the network and graph; when the state is stable, the computation frequency is appropriately reduced to conserve system resources. Resource management strategies ensure that the system can complete baseline adjustment and graph reconstruction tasks in a timely manner under various operating conditions. Error handling and recovery mechanisms ensure the reliability of implementation. When an anomaly occurs during computation, the system can automatically save intermediate results, roll back to the previous stable state, and continue execution after the problem is resolved. All abnormal events are recorded in the system log, including the anomaly type, occurrence time, processing procedure, and final result. Log analysis tools help operations and maintenance personnel identify system problems and improve implementation processes.

[0051] The visualization module provides an intuitive graphical interface for the implementation results. Adjustments to the baseline topology network are displayed as a topology graph, with vertex size representing weight values, edge thickness representing connection strength, and color indicating state level. The field strength graph is displayed using 3D rendering technology, employing color gradients to represent field strength, and supports interactive operations such as rotation, zooming, and profile viewing. Visualization tools help users understand the implementation effects and decision-making basis. All data exchanges during implementation use standardized data formats and interface protocols. The baseline topology network data uses a graph structure description language, and the field strength graph data uses a scientific data format, ensuring data portability and interoperability. The system provides data export functionality, supporting the output of implementation results to other analysis tools or management systems for further processing.

[0052] Example 5: Matching emergency response plan templates with the associated power equipment operation and maintenance procedure library is a crucial step in decision generation. The operation and maintenance procedure library stores standard maintenance procedures provided by equipment manufacturers, industry standard requirements, and experience-based response plans accumulated by operating units. The system retrieves matching emergency response plans from the library based on multi-dimensional characteristics such as equipment type, failure mode, and degree of degradation. The matching process employs a rule-based reasoning algorithm. First, it determines the applicable subset of procedures based on the equipment model; then, it filters possible response methods based on the degradation type; and finally, it adjusts the response intensity based on the degree of degradation. Each matched emergency response plan template includes detailed information such as response operation steps, required tools and materials, safety precautions, estimated working hours, and acceptance criteria. The system also records historical execution effect data for each plan as a reference for this matching.

[0053] When generating a priority sequence based on real-time equipment load rates, the system comprehensively considers technical factors and operational requirements. Real-time load rate data comes from the equipment monitoring system, reflecting the current operational importance and the scope of power outage impact. Power outages for maintenance are generally avoided during periods of high load rates, while low load rates are more suitable for scheduling maintenance work. The priority calculation algorithm combines technical urgency with operational constraints. Technical urgency is based on the rate of degradation and risk consequence assessment, while operational constraints include load levels, weather conditions, and the status of backup equipment. The system calculates a comprehensive priority score for each item to be handled, with items having higher scores placed at the top of the execution sequence. The priority sequence also considers dependencies between items; some operations require other operations to be completed before they can be performed. These logical relationships are modeled as constraints and incorporated into the sorting algorithm.

[0054] The output insulation condition maintenance strategy is the final outcome of the entire implementation process. The strategy document is organized in a structured format and includes three main parts: the coordinates of the treatment location, the treatment method, and the execution sequence. The treatment location coordinates are described using both the equipment coordinate system and the geographic coordinate system, encompassing both the relative position within the equipment structure and the actual geographic location information, facilitating on-site personnel location. The treatment method describes detailed operational steps and technical requirements, including specific specifications for different treatment methods such as cleaning, drying, touch-up painting, and replacement. The execution sequence provides suggested implementation time windows, estimated durations, and schedules, taking into account factors such as equipment operation plans, weather forecasts, and resource availability. The strategy output also includes a risk assessment report, explaining the potential consequences of not taking action and the possible risks during the treatment process. The system provides multiple output formats, including machine-readable XML, human-readable PDF documents, and API interfaces for integration with other management systems.

[0055] The implementation of the anomaly handling decision-making module fully embodies the combination of professional knowledge and data-driven approaches. The system not only relies on mathematical models and algorithms for analysis but also deeply integrates domain knowledge and operational experience in equipment insulation management. The decision-making process considers the balance between technical feasibility and actual operating conditions, ensuring that the generated maintenance strategies are both technically sound and practically operable on-site. The implementation results provide a scientific basis and technical support for equipment maintenance management, helping operating units optimize the allocation of maintenance resources and improve the accuracy and effectiveness of equipment maintenance. Data security and access control were fully considered during implementation. Sensitive equipment information contained in the maintenance strategies is protected by access control; users at different levels can only view content within their corresponding access permissions. Encryption measures are used during strategy transmission and storage to prevent information leakage. All access and modification operations to the strategies are recorded in detail, forming a complete audit trail.

[0056] The system also provides strategy simulation and impact analysis functions, allowing users to simulate the effects of strategy execution before implementation, assess the impact of different time schedules, and compare the advantages and disadvantages of various handling options. Simulation results help users make more informed decisions and optimize the overall effectiveness of maintenance plans. Impact analysis can assess the comprehensive impact of maintenance work on equipment reliability, operating costs, and system risks, providing a comprehensive reference for decision-making. The implementation of the anomaly handling decision-making module completes a closed loop from condition monitoring to maintenance action. The system can not only detect equipment insulation problems but also provide specific handling solutions and implementation suggestions, transforming monitoring data into actual productivity. This complete solution embodies the ultimate value of a condition monitoring system, providing reliable technical support for the safe operation and intelligent maintenance of power equipment. The decision-making data and experience accumulated during implementation will continuously enrich the system's knowledge base, driving the continuous evolution and improvement of decision-making capabilities.

[0057] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0058] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended related limiting contents and their equivalents.

Claims

1. An online monitoring system for the insulation status of power equipment, characterized in that, The system includes: The multi-source data fusion module is used to collect real-time data streams of operating parameters of power equipment, environmental sensor data streams, and historical insulation degradation records. The spatial insulation field strength distribution module is used to dynamically map the field strength of the operating parameter data stream based on the structural topology map of the power equipment and in combination with a preset insulation degradation feature library, and generate spatial insulation field strength distribution feature information. The baseline state topology module is used to construct an insulation state baseline topology network based on the component service duration distribution and equipment operation mode of the power equipment through a dynamic weight allocation mechanism. The insulation degradation dynamic detection module is used to perform regional field strength deviation analysis based on the spatial insulation field strength distribution characteristics and the output of the insulation state reference topology network, and generate primary insulation degradation index. The delayed response verification module is used to execute the continuous time-series degradation trajectory tracking verification rule when the primary insulation degradation index exceeds the preset deviation threshold. The insulation state trend prediction module is used to perform multi-step state evolution simulation on the spatial insulation field strength distribution characteristic information and environmental sensing data stream to generate predicted insulation field strength distribution characteristic information. The monitoring, compensation, and optimization module is used to perform field strength compensation correction on the primary insulation degradation index based on the predicted insulation field strength distribution characteristics information, and generate an optimized insulation degradation index.

2. The online monitoring system for insulation status of power equipment according to claim 1, characterized in that, The multi-source data fusion module is used to collect real-time data streams of operating parameters of power equipment, environmental sensor data streams, and historical insulation degradation records, including: Obtain the equipment model identifier and installation location coordinates of the power equipment; Load the insulation parameter acquisition protocol set corresponding to the device model according to the device model identifier; Activate the environmental sensor array acquisition node based on the installation location coordinates; Simultaneously retrieve the historical insulation degradation feature database index associated with the equipment model identifier; The multi-channel concurrent acquisition task generates the operating parameter data stream, environmental sensor data stream, and historical insulation degradation record data stream.

3. The online monitoring system for insulation status of power equipment according to claim 1, characterized in that, The baseline state topology module is used to construct an insulation state baseline topology network based on the component service life distribution and equipment operating modes of the power equipment through a dynamic weight allocation mechanism, including: Analyze the service duration distribution of the components to generate a discrete feature vector of service duration; Extract the modal feature codes corresponding to the operating modes of the device; Establish an initial topology graph with power equipment components as vertices and insulation connections between components as edges; Calculate the initial weight factor for each vertex based on the discrete feature vector of service duration; The modal feature encoding is fused to dynamically calibrate the edge connection strength; Output the insulation state reference topology network with weighted attributes.

4. The online monitoring system for insulation status of power equipment according to claim 1, characterized in that, The insulation degradation dynamic detection module is used to perform regional field strength deviation analysis based on the spatial insulation field strength distribution characteristics and the output of the insulation state reference topology network, and to generate primary insulation degradation indices, including: The spatial insulation field strength distribution characteristic information is divided into field strength partition datasets according to the equipment area; Match the reference field strength interval of each region in the insulation state reference topology network; Calculate the multidimensional deviation of the real-time field strength value of each zone from the reference field strength interval; Aggregate the multidimensional deviations of each partition to generate a comprehensive regional degradation coefficient; Based on the preset degradation level mapping table, the regional comprehensive degradation coefficient is converted into the primary insulation degradation index.

5. The online monitoring system for insulation status of power equipment according to claim 1, characterized in that, The delay response verification module is used to execute continuous time-series degradation trajectory tracking verification rules when the primary insulation degradation index exceeds a preset deviation threshold, including: The time when the primary insulation degradation index first exceeds the limit is recorded as the starting verification time stamp; Construct a sliding time window based on the initial verification time scale; The trajectory of changes in insulation degradation indicators is continuously captured within the sliding time window; When the degradation trajectory meets the duration threshold and degradation monotonicity condition, an insulation anomaly confirmation signal is triggered. Otherwise, reset the sliding time window and continue monitoring.

6. The online monitoring system for insulation status of power equipment according to claim 1, characterized in that, The insulation state trend prediction module is used to perform multi-step state evolution simulation on the spatial insulation field strength distribution characteristic information and environmental sensing data stream to generate predicted insulation field strength distribution characteristic information, including: Extracting the temporal variation patterns of historical insulation field strength distribution characteristics; Temperature and humidity change gradients in the associated environmental sensor data stream; Construct a field strength-environment coupled evolution model; Simulate the future multi-step insulation field strength distribution using a rolling time-domain prediction algorithm; Output the predicted insulation field strength distribution characteristics with confidence intervals.

7. The online monitoring system for insulation status of power equipment according to claim 1, characterized in that, The monitoring compensation optimization module is used to perform field strength compensation correction on the primary insulation degradation index based on the predicted insulation field strength distribution characteristic information, and generate an optimized insulation degradation index, including: Compare the deviation between the real-time spatial insulation field strength distribution characteristics and the predicted insulation field strength distribution characteristics; When the deviation exceeds the tolerance threshold, calculate the field strength distribution compensation coefficient; The primary insulation degradation index is linearly corrected based on the field strength distribution compensation coefficient. The optimized insulation degradation index is output after compensation calibration.

8. The online monitoring system for insulation status of power equipment according to claim 1, characterized in that, The system further includes: a baseline topology adaptive module, used to dynamically redistribute node weights in the insulation state baseline topology network according to the optimized insulation degradation index, including: The optimized insulation degradation index is analyzed to obtain the severity level of degradation in each region; Calculate the weight decay factor of the corresponding topology node based on the severity of degradation. The vertex weights of the insulation state baseline topology are updated based on the weight decay factor. Reconstruct the mapping relationship between edge connection strength and vertex weight.

9. The online monitoring system for insulation status of power equipment according to claim 1, characterized in that, The system further includes: an insulation field strength reconstruction module, used to reconstruct the equipment insulation field strength map based on the compensated spatial insulation field strength distribution characteristic information when the optimized insulation degradation index triggers the reconstruction condition, including: When the optimized insulation degradation index continues to exceed the limit and the compensation correction is ineffective, activate the field strength spectrum reconstruction command. Obtain the compensated and corrected spatial insulation field strength distribution characteristics at the current moment; Spatial interpolation calculation of insulation field strength is performed using the integrated equipment structure topology map; Generate a new device insulation field strength map with three-dimensional coordinate attributes.

10. The online monitoring system for insulation status of power equipment according to claim 1, characterized in that, The system further includes: an anomaly handling decision module, used to generate an insulation status maintenance strategy based on the reconstructed equipment insulation field strength map, including: Identify the coordinates of weak areas in the insulation field strength spectrum of newly built equipment; Match the emergency response plan template with the associated operation and maintenance procedure library of the power equipment; Generate a priority sequence for handling based on the real-time load rate of the equipment; The output includes the insulation status maintenance strategy, which includes the coordinates of the disposal location, the disposal method, and the execution sequence.

Images