Breakpoint sniffing method and system for wind power grid load storage architecture

Abstract

The present invention is applicable to the field of wind power management and provides a breakpoint sniffing method and system for a wind power grid-load-storage architecture. The method comprises the following steps: obtaining gas detection data from wind power cables and analyzing it to obtain gas composition distribution data; establishing a steady-state thermal circuit model based on cable structural parameters, and inferring the thermal conductivity distribution of the insulation layer using load current and surface temperature data from a SCADA system; establishing a multi-dimensional mapping relationship between gas distribution, electrical parameters, and thermal conductivity distribution of the insulation layer using electrical parameters from the SCADA system; and generating a comprehensive risk score based on gas concentration thresholds and electrical parameter mutation thresholds. This invention eliminates the need for traditional contact sensors, reducing deployment cost and complexity while enabling real-time monitoring in the live state, significantly improving the early identification of cable faults and operational efficiency.

Description

Technical Field

[0001] The present invention relates to the field of wind power management, and in particular to a breakpoint sniffing method and system for a wind power source grid-load-storage architecture. Background Art

[0002] The source-grid-load-storage architecture is a new energy system architecture that integrates four key components: power sources, grids, loads, and energy storage. This architecture aims to improve the flexibility, stability, and efficiency of power systems, which is particularly important given the intermittent generation of renewable energy.

[0003] As the penetration of renewable energy sources like wind power continues to increase in integrated power generation systems, the inherent volatility and intermittency of their generation process present unprecedented challenges to the stable operation of power systems. Wind power generation is significantly influenced by natural meteorological conditions, exhibiting significant uncertainty and uncontrollability. This complicates grid frequency regulation and leads to frequent power flow fluctuations, potentially leading to voltage instability, increased harmonic distortion, and localized overloads.

[0004] In this context, the safe operation of power cables, as the key link between power sources, loads, and energy storage devices, is paramount. However, traditional methods for detecting hidden cable breakages rely on hardware sensors (such as ultrasonic sensors, vibration sensors, and partial discharge detectors) or physical contact testing methods (such as impedance spectroscopy and insulation resistance measurement). These methods present challenges such as high cost, complex deployment, and limited live monitoring. Therefore, a breakage detection method and system for wind power grid-load-storage architectures have been proposed to address these issues. Summary of the Invention

[0005] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a breakpoint sniffing method and system for a wind power grid-load-storage architecture to solve the problems existing in the above-mentioned background technology.

[0006] The present invention is implemented as follows: a breakpoint sniffing method for a wind power grid-load-storage architecture, the method comprising the following steps:

[0007] Obtaining gas detection data of wind power cables and analyzing it to obtain gas composition distribution data, wherein the gas detection data is obtained by a gas sampling device deployed in the cable trench;

[0008] Establishing a steady-state thermal circuit model based on cable structural parameters, including conductor diameter, insulation layer thickness, and material thermal resistance, and inferring insulation layer thermal conductivity distribution in combination with load current and surface temperature data from a SCADA system;

[0009] Combined with the electrical parameters of the SCADA system, a multi-dimensional mapping relationship between gas distribution, electrical parameters, and insulation layer thermal conductivity distribution is established to identify the aging characteristics of different wind power cables;

[0010] A comprehensive risk score is generated based on the gas concentration threshold and electrical parameter mutation threshold, and a hierarchical early warning mechanism is used to assign maintenance priorities to cable abnormal points.

[0011] As a further solution of the present invention: the step of obtaining gas detection data of the wind power cable and analyzing to obtain gas composition distribution data specifically includes:

[0012] Obtain gas detection data at key nodes of wind power cables through a gas sampling device with a configured sampling frequency;

[0013] Use the operating parameters retrieved from the SCADA system to align the data timestamps of gas detection data;

[0014] Identify the concentration of characteristic gases representing cable aging through an online mass spectrometer and record their changing trends over time;

[0015] Combined with the ambient temperature and humidity sensor data, an algorithm is used to eliminate the interference of non-cable related background gases, and the gas composition distribution map and concentration gradient curve distributed along each key node of the cable are obtained.

[0016] As a further solution of the present invention, the step of establishing a steady-state thermal circuit model based on cable structural parameters and calculating the thermal conductivity distribution of the insulation layer in combination with the load current and surface temperature data of the SCADA system specifically includes:

[0017] Obtain cable structural parameters and thermal resistance and heat capacity data from the material database, and use a deep learning model to automatically extract key features to generate standardized parameter inputs for a steady-state thermal circuit model.

[0018] A steady-state thermal circuit model based on thermoelectric analogy is established based on the cable layered structure, and the heat flux distribution is simulated using the Monte Carlo method to output the temperature field prediction value of the cable surface;

[0019] Extract real-time load current, ambient temperature, and cable surface temperature data from the SCADA system, calculate the calorific value using the Joule heat formula, and eliminate noise through Kalman filtering to obtain the SCADA temperature measured value.

[0020] By comparing the temperature field prediction value with the SCADA temperature measured value, and combining the least squares method to infer the equivalent thermal conductivity of the insulation layer;

[0021] The steady-state heat circuit model is calibrated by using historical fault gas data to constrain parameters of the steady-state heat circuit model.

[0022] As a further solution of the present invention, the step of establishing a steady-state thermal circuit model based on a thermoelectric analogy of the cable layered structure and simulating the heat flow distribution by the Monte Carlo method specifically includes:

[0023] Based on the physical characteristics of wind power cables, a steady-state thermal circuit model with thermoelectric analogy layers is established, and the thermal resistance units and material properties of each layer are divided. The cable layers are composed of conductor, insulation layer, sheath and environment.

[0024] Equivalently equate the conductor, insulation layer, sheath and environment to series thermal resistance, and quantify the material thermal resistance value and contact surface thermal resistance distribution;

[0025] By randomly sampling, different load currents, ambient temperatures, and cable surface temperatures are input into the steady-state thermal circuit model to simulate the distribution of heat flow in the steady-state thermal circuit model.

[0026] The predicted value of the cable surface temperature field is generated based on the calculation results of the steady-state thermal circuit model.

[0027] As a further solution of the present invention, the steps of generating a comprehensive risk score based on the gas concentration threshold and the electrical parameter mutation threshold, and assigning maintenance priorities to cable abnormal points using a graded early warning mechanism, specifically include:

[0028] Based on gas composition, electrical parameters and steady-state thermal circuit model output, the algorithm quantifies the weight of each feature and dynamically adjusts the weight distribution;

[0029] A weighted summation method is used to generate risk scores, and classification thresholds are set based on historical fault data. The thresholds can be adjusted adaptively.

[0030] Overlay the risk score onto the cable GIS map to generate a spatial heat map, and mark outliers with a color gradient;

[0031] Three levels of warning are divided according to the scores, and the work order system is linked to push maintenance work orders to operation and maintenance personnel in a graded manner.

[0032] As a further embodiment of the present invention, the method further comprises:

[0033] Extract spatial coordinates from anomaly points marked on the GIS map and generate drone inspection target points based on the cable wiring diagram;

[0034] Calculate the optimal inspection path for drones based on the priority of outliers and the algorithm;

[0035] Obtain image data, thermal imaging data, and gas concentration information during drone inspections;

[0036] The spatial heat map is dynamically adjusted based on the inspection results.

[0037] Another object of the present invention is to provide a breakpoint sniffing system for a wind power grid-load-storage architecture, the system comprising:

[0038] A sampling and analysis module is used to obtain gas detection data of wind power cables and analyze it to obtain gas composition distribution data. The gas detection data is obtained by a gas sampling device deployed in the cable trench;

[0039] A thermodynamic modeling module is used to establish a steady-state thermal circuit model based on cable structural parameters, including conductor diameter, insulation thickness, and material thermal resistance, and to infer the thermal conductivity distribution of the insulation layer using load current and surface temperature data from the SCADA system;

[0040] A multi-dimensional analysis module is used to combine the electrical parameters of the SCADA system to establish a multi-dimensional mapping relationship between gas distribution, electrical parameters, and insulation layer thermal conductivity distribution, which is used to identify the aging characteristics of different wind power cables;

[0041] The dynamic early warning assessment module is used to generate a comprehensive risk score based on the gas concentration threshold and the electrical parameter mutation threshold, and use the hierarchical early warning mechanism to assign maintenance priorities to cable abnormal points.

[0042] As a further solution of the present invention: the sampling and analysis module includes:

[0043] A gas sampling unit is used to obtain gas detection data at key nodes of wind power cables through a gas sampling device with a configured sampling frequency;

[0044] A data alignment unit is used to align the data timestamps of the gas detection data using the operating parameters retrieved from the SCADA system;

[0045] Gas analysis unit, used to identify the concentration of characteristic gases representing cable aging through an online mass spectrometer and record their changing trends over time;

[0046] The interference elimination unit is used to combine the ambient temperature and humidity sensor data, use the algorithm to eliminate the interference of non-cable related background gases, and obtain the gas composition distribution map and concentration gradient curve distributed along each key node of the cable.

[0047] As a further solution of the present invention: the thermodynamic modeling module includes:

[0048] A feature extraction unit is used to obtain cable structural parameters and thermal resistance and thermal capacity data from the material database. It uses a deep learning model to automatically extract key features and generate standardized parameter inputs for the steady-state thermal circuit model.

[0049] A model building unit is used to establish a steady-state thermal circuit model based on the thermoelectric analogy of the cable layered structure, simulate the heat flow distribution through the Monte Carlo method, and output the temperature field prediction value of the cable surface;

[0050] The data fusion unit is used to extract real-time load current, ambient temperature and cable surface temperature data from the SCADA system, calculate the calorific value by combining the Joule heat formula, and eliminate noise through Kalman filtering to obtain the SCADA temperature measured value;

[0051] The data inversion unit is used to compare the temperature field prediction value with the SCADA temperature measured value and use the least squares method to infer the equivalent thermal conductivity of the insulation layer;

[0052] The model calibration unit is used to constrain the steady-state heat circuit model parameters by using historical fault gas data to calibrate the steady-state heat circuit model.

[0053] As a further solution of the present invention: the dynamic early warning assessment module includes:

[0054] The scoring weight allocation unit is used to quantify the weight of each feature through an algorithm based on gas composition, electrical parameters and steady-state thermal circuit model output, and dynamically adjust the weight allocation;

[0055] The risk score calculation unit is used to generate risk scores using a weighted summation method and set classification thresholds based on historical fault data. The thresholds can be adjusted adaptively.

[0056] A visualization unit is used to overlay the risk score onto the cable GIS map to generate a spatial heat map and mark abnormal points with a color gradient;

[0057] The hierarchical processing unit is used to divide warnings into three levels according to the scores, and to link with the work order system to push maintenance work orders to operation and maintenance personnel in a hierarchical manner.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] The present invention achieves efficient detection and risk warning of hidden breakpoints in wind power cables through multi-dimensional data fusion. Gas sampling devices deployed in the cable trench collect gas composition data in real time, and an online mass spectrometer is used to analyze the characteristic gas distribution corresponding to different aging mechanisms (such as oxidation, thermal decomposition, and partial discharge), providing a chemical basis for cable condition assessment. A steady-state thermal circuit model is established based on cable structural parameters, and the thermal conductivity distribution of the insulation layer is inferred by combining the load current and surface temperature data from the SCADA system. The degree of material aging is reflected through thermodynamic properties. Furthermore, the gas composition data is correlated with electrical parameters and modeled to construct a multi-dimensional mapping relationship to accurately identify different aging types. Finally, a comprehensive risk score is generated by setting gas concentration thresholds and electrical parameter mutation thresholds, and a hierarchical warning mechanism is used to dynamically assign maintenance priorities. This invention eliminates the need for traditional contact sensors, reducing deployment cost and complexity. It also enables real-time monitoring in the energized state, significantly improving the early identification of cable faults and operational efficiency, and providing reliable protection for the stable operation of the source-grid-load-storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 The flowchart of the breakpoint sniffing method for the wind power grid load storage architecture is shown.

[0061] Figure 2 The flowchart of obtaining gas composition distribution data in the breakpoint sniffing method of wind power grid load storage architecture.

[0062] Figure 3 This is a flow chart for estimating the thermal conductivity distribution of the insulation layer in the breakpoint sniffing method of the wind power grid load storage architecture.

[0063] Figure 4 Flowchart for simulating heat flow distribution in the breakpoint sniffing method for wind power grid-load-storage architecture.

[0064] Figure 5 A flow chart for generating a comprehensive risk score based on gas concentration thresholds and electrical parameter mutation thresholds in the breakpoint sniffing method for wind power grid-load-storage architecture.

[0065] Figure 6 This is a structural diagram of the breakpoint sniffing system for the wind power grid load storage architecture.

[0066] Figure 7 This is a structural diagram of the sampling and analysis module in the breakpoint sniffing system of the wind power grid load storage architecture.

[0067] Figure 8 This is a structural diagram of the thermodynamic modeling module in the breakpoint sniffing system of the wind power grid load storage architecture.

[0068] Figure 9This is a structural diagram of the dynamic early warning assessment module in the breakpoint sniffing system of the wind power grid load storage architecture. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0070] The specific implementation of the present invention is described in detail below with reference to specific embodiments.

[0071] like Figure 1 As shown, an embodiment of the present invention provides a breakpoint sniffing method for a wind power grid load storage architecture, the method comprising the following steps:

[0072] S100, obtaining gas detection data of a wind power cable and analyzing it to obtain gas composition distribution data, wherein the gas detection data is obtained by a gas sampling device deployed in a cable trench;

[0073] S200, establishing a steady-state thermal circuit model based on cable structural parameters, and inferring insulation layer thermal conductivity distribution in combination with load current and surface temperature data from a SCADA system, wherein the cable structural parameters include conductor diameter, insulation layer thickness, and material thermal resistance;

[0074] S300, combined with the electrical parameters of the SCADA system, establishes a multi-dimensional mapping relationship between gas distribution, electrical parameters, and insulation layer thermal conductivity distribution to identify the aging characteristics of different wind power cables;

[0075] S400 generates a comprehensive risk score based on the gas concentration threshold and electrical parameter mutation threshold, and uses a graded early warning mechanism to assign maintenance priorities to cable anomaly points.

[0076] It should be noted that traditional methods for detecting hidden cable breakpoints mostly rely on hardware sensors (such as ultrasonic sensors, vibration sensors, partial discharge detectors) or physical contact testing methods (such as impedance spectroscopy and insulation resistance measurement), which have problems such as high cost and complex deployment.

[0077] In one embodiment of the present invention, multi-dimensional data fusion is used to efficiently detect hidden breakpoints in wind turbine cables and provide risk warnings. Gas sampling devices deployed in the cable trench collect gas composition data in real time. This data is then combined with an online mass spectrometer to analyze the characteristic gas distribution corresponding to different aging mechanisms (such as oxidation, thermal decomposition, and partial discharge), providing a chemical basis for cable condition assessment. A steady-state thermal circuit model is established based on cable structural parameters, and the insulation layer thermal conductivity distribution is inferred through load current and surface temperature data from the SCADA system. The degree of material aging is then reflected through thermodynamic properties. Furthermore, the gas composition data is correlated with electrical parameters to create a multi-dimensional mapping relationship for accurate identification of different aging types. Finally, a comprehensive risk score is generated by setting gas concentration thresholds and electrical parameter mutation thresholds, and a hierarchical warning mechanism is used to dynamically assign maintenance priorities. This invention eliminates the need for traditional contact sensors, reducing deployment cost and complexity. It also enables real-time monitoring in the live state, significantly improving the early identification of cable faults and operational efficiency, and providing reliable assurance for the stable operation of the source-grid-load-storage system.

[0078] like Figure 2 As shown, as a preferred embodiment of the present invention, the step of obtaining gas detection data of the wind power cable and analyzing to obtain gas composition distribution data specifically includes:

[0079] S101, obtaining gas detection data of key nodes of wind power cables through a gas sampling device configured with a sampling frequency;

[0080] S102, aligning the data timestamps of the gas detection data using the operating parameters retrieved from the SCADA system;

[0081] S103, identifying the concentration of characteristic gases representing cable aging through an online mass spectrometer and recording their changing trend over time;

[0082] S104, combining the ambient temperature and humidity sensor data, using an algorithm to eliminate interference from non-cable related background gases, and obtaining a gas composition distribution map and concentration gradient curve distributed along each key node of the cable.

[0083] In an embodiment of the present invention, during the gas detection process for wind turbine cables, gas samples are first collected at key cable nodes, such as connectors and branch boxes, using a gas sampling device with a pre-set sampling frequency. Then, relevant operating parameters (such as load current and surface temperature) are obtained from the SCADA system, ensuring that these data are precisely aligned with the timestamps of the gas samples to accurately correlate changes in electrical status with gas composition. Next, an online mass spectrometer analyzes the collected gas samples to identify the concentrations of characteristic gases indicative of cable aging (such as carbon monoxide (CO) and carbon dioxide (CO2)), and records the trends of these concentrations over time, which helps track the progression of the aging process. Finally, combined with data from ambient temperature and humidity sensors, a specific algorithm is used to eliminate non-cable-related background gas interference, such as ambient emissions or soil releases, to generate gas composition distribution maps and concentration gradient curves along each key cable node. This specific algorithm can employ ICA (with a reference vector) or a machine learning classifier (such as a random forest). For example, if the CO concentration in the air around a cable segment is significantly higher than in other areas and is increasing, while the ambient temperature and humidity data for that area indicate no external pollution sources, it can be determined that the cable segment may be experiencing oxidation and aging issues and requires further inspection and maintenance. This approach not only improves fault location accuracy but also reduces the possibility of false alarms.

[0084] like Figure 3 As shown, as a preferred embodiment of the present invention, the step of establishing a steady-state thermal circuit model based on cable structural parameters and calculating the thermal conductivity distribution of the insulation layer in combination with the load current and surface temperature data of the SCADA system specifically includes:

[0085] S201, obtaining cable structural parameters and thermal resistance and thermal capacity data from a material database, and automatically extracting key features using a deep learning model to generate standardized parameter inputs for a steady-state thermal circuit model;

[0086] S202, establishing a steady-state thermal circuit model based on thermoelectric analogy based on the layered structure of the cable, and simulating heat flux distribution using a Monte Carlo method to output a predicted value of the temperature field on the cable surface;

[0087] S203, extracting real-time load current, ambient temperature, and cable surface temperature data from the SCADA system, calculating the calorific value using the Joule heat formula, and eliminating noise through Kalman filtering to obtain the SCADA temperature measured value;

[0088] S204, by comparing the temperature field prediction value with the SCADA temperature measured value, and combining the least square method to infer the equivalent thermal conductivity of the insulation layer;

[0089] S205 , using historical fault gas data to constrain steady-state heat circuit model parameters, and calibrating the steady-state heat circuit model.

[0090] In an embodiment of the present invention, the geometric parameters of the cable (such as conductor diameter, insulation layer thickness) are first obtained through laser scanning or image recognition technology, and combined with the thermal resistance (Rth) and heat capacity (Cth) data in the material database, a deep learning model (such as U-Net) is used to automatically extract key features (such as non-uniform areas of the insulation layer) to generate a standardized parameter input file. For example, if the insulation layer of a certain section of cable is locally thinned or there is an air gap, the model will automatically mark the area as a thermal resistance anomaly point, providing accurate input for subsequent modeling. Subsequently, based on the conductor-insulation layer-sheath-environment layered structure of the cable, a steady-state thermal circuit model of thermoelectric analogy is established (each layer is equivalent to a series thermal resistance), and the Monte Carlo method is used to simulate the heat flux distribution under different loads and ambient temperatures, and the predicted value of the cable surface temperature field is output. For example, under high temperature conditions in summer, the model can simulate local temperature hotspots caused by the decrease in thermal conductivity of the insulation layer. Next, the real-time load current, ambient temperature and cable surface temperature data are extracted from the SCADA system, and combined with the Joule heat formula (I is the conductor current, R is the conductor resistance) to calculate the conductor's heat generation. Kalman filtering is then used to eliminate sensor noise, resulting in more accurate measured temperature values. For example, if the load fluctuates significantly on a particular day, the filtering algorithm can eliminate transient anomalies and preserve the true temperature trend. By comparing the model's predicted temperature field with the actual SCADA measurements and inferring the insulation's equivalent thermal conductivity (λ_insulation) using the least squares method, the degree of aging can be quantitatively reflected. For example, if the predicted temperature of a cable section is consistently higher than the measured value and the thermal conductivity drops by 15%, this may indicate thermal decomposition or partial discharge within the insulation. Finally, gas detection data from historical fault cases (such as sudden CO2 concentration increases) is incorporated as a constraint to calibrate the thermal resistance and thermal conductivity parameters of the thermal circuit model, improving the model's adaptability in real-world operations. For example, if an abnormal increase in CO2 concentration prior to a fault aligns with the model's predicted downward trend in thermal conductivity, the validity of the model parameter calibration can be verified, providing a reliable basis for cable condition assessment.

[0091] like Figure 4 As shown, as a preferred embodiment of the present invention, the steps of establishing a steady-state thermal circuit model based on a thermoelectric analogy based on the cable layered structure and simulating the heat flow distribution by the Monte Carlo method specifically include:

[0092] S212: Based on the physical characteristics of wind power cables, a steady-state thermal circuit model based on thermoelectric analogy layers is established, and the thermal resistance units and material properties of each layer are divided. The cable layers are composed of conductor, insulation layer, sheath, and environment.

[0093] S222: Equivalently equate the conductor, insulation layer, sheath, and environment to series thermal resistances, and quantify the material thermal resistance and contact surface thermal resistance distribution.

[0094] S232, inputting parameters of different load currents, ambient temperatures, and cable surface temperatures into the steady-state thermal circuit model by random sampling to simulate the distribution of heat flow in the steady-state thermal circuit model;

[0095] S242: Generate a predicted value of the cable surface temperature field based on the calculation results of the steady-state thermal circuit model.

[0096] In the embodiment of the present invention, a steady-state thermal circuit model of thermoelectric analogy layers is first established based on the physical properties of the wind power cable (such as the conductor, insulation layer, sheath, and environment), and the multi-layer structure of the cable is equivalent to a series thermal resistance unit (similar to a series circuit of resistors). For example, the thermal resistance of the conductor ( ) is determined by the thermal conductivity and geometric dimensions of its material, and the thermal resistance of the insulation layer ( ) The thermal performance changes caused by aging need to be considered, and the thermal resistance of the sheath ( ) and the thermal resistance of the environment (such as soil or air) ( ) together form a complete heat conduction path. Subsequently, the thermal resistance of each layer is parameterized by quantifying the material thermal resistance value and the contact surface thermal resistance distribution. For example, there may be a small air gap at the contact surface between the conductor and the insulation layer, resulting in the contact thermal resistance ( ) increases significantly and needs to be calibrated using experimental data or literature values. In this case, the total thermal resistance can be expressed as:

[0097]

[0098] During the Monte Carlo simulation phase, random sampling is used to input various parameters, such as load current, ambient temperature, and cable surface temperature, into the steady-state thermal circuit model. For example, the load current may vary between 50 A and 200 A due to wind power fluctuations, and the ambient temperature may randomly distribute between -20°C and 40°C. These parameters all follow a specific probability distribution (such as a normal distribution or a uniform distribution). By generating a large number of random samples (e.g., 10,000 simulations), the heat flux distribution pattern for each sampling period is calculated, and the predicted cable surface temperature field is recorded. For example, for a wind turbine cable operating in a high-temperature summer environment, a Monte Carlo simulation shows that when the load current increases from 100 A to 150 A, the predicted surface temperature increases from 45°C to 65°C. Furthermore, 30% of the simulations show a sudden increase in the insulation thermal resistance. This indicates possible localized aging of the cable insulation, requiring further verification with gas detection data (such as abnormal CO2 concentrations).

[0099] like Figure 5 As shown, as a preferred embodiment of the present invention, the steps of generating a comprehensive risk score based on the gas concentration threshold and the electrical parameter mutation threshold, and assigning maintenance priorities to cable abnormal points using a graded early warning mechanism, specifically include:

[0100] S401, based on gas composition, electrical parameters and steady-state thermal circuit model output, quantifies the weight of each feature through an algorithm and dynamically adjusts the weight distribution;

[0101] S402: Generate risk scores using a weighted summation method, set grading thresholds based on historical fault data, and enable adaptive adjustment of the thresholds.

[0102] S403, overlaying the risk score onto the cable GIS map to generate a spatial heat map, and marking abnormal points using a color gradient;

[0103] S404: Classify warnings into three levels based on the scores, and link them to the work order system to push maintenance work orders to operation and maintenance personnel in different levels;

[0104] S405, extracting spatial coordinates from the abnormal points marked in the GIS map and generating drone inspection target points in combination with the cable wiring diagram;

[0105] S406, calculating the optimal inspection path of the UAV based on the priority of the abnormal points and the algorithm;

[0106] S407, obtaining image data, thermal imaging data, and gas concentration information during the drone inspection;

[0107] S408: Dynamically adjust the spatial heat map based on the inspection results.

[0108] In this embodiment of the present invention, cable anomaly detection begins with an algorithm dynamically adjusting the weights of various features based on gas composition (such as CO and CH4 concentrations), electrical parameters (such as sudden changes in load current), and temperature field data output by a steady-state thermal circuit model. For example, if historical fault data for a region indicates a strong correlation between a sudden increase in CO concentration and insulation aging, the algorithm automatically increases the weight of CO and decreases the weight of sudden changes in current. A weighted summation method is then used to generate a comprehensive risk score. Three levels of warning (e.g., low risk, medium risk, and high risk) are then set based on historical fault thresholds. A machine learning model is then used to adaptively adjust the thresholds. After the risk score is overlaid on a GIS map, anomalies are marked with a gradient of red (high risk), orange (medium risk), and yellow (low risk). For example, a cable joint with a high risk score due to localized overheating appears as a red hotspot on the map. When the tiered warning system is linked to the work order system, high-risk points are prioritized for emergency maintenance tasks, medium-risk points are scheduled for next-day inspections, and low-risk points are only recorded for monitoring. During drone inspection route planning, an algorithm calculates the optimal route based on outlier priority and cable routing diagrams. For example, high-risk points are prioritized for flight, thermal imaging is collected, and gas concentration data is verified. If a red hotspot is detected during an inspection and is actually due to external contamination rather than a cable fault, the system dynamically adjusts the thermal map weight (for example, reducing the weight of CO) and optimizes subsequent warning strategies, creating a closed-loop management system.

[0109] like Figure 6 As shown, an embodiment of the present invention further provides a breakpoint sniffing system for a wind power grid load storage architecture, the system comprising:

[0110] The sampling and analysis module 100 is used to obtain gas detection data of the wind power cable and analyze it to obtain gas composition distribution data. The gas detection data is obtained by a gas sampling device deployed in the cable trench;

[0111] Thermodynamic modeling module 200 is used to establish a steady-state thermal circuit model based on cable structural parameters, including conductor diameter, insulation layer thickness, and material thermal resistance, and to infer insulation layer thermal conductivity distribution in combination with load current and surface temperature data from a SCADA system;

[0112] The multi-dimensional analysis module 300 is used to combine the electrical parameters of the SCADA system to establish a multi-dimensional mapping relationship between gas distribution, electrical parameters, and insulation layer thermal conductivity distribution, so as to identify the aging characteristics of different wind power cables;

[0113] The dynamic early warning assessment module 400 is used to generate a comprehensive risk score based on the gas concentration threshold and the electrical parameter mutation threshold, and to assign maintenance priorities to cable abnormal points using a hierarchical early warning mechanism.

[0114] In an embodiment of the present invention, the breakpoint sniffing system of the wind power source grid load storage architecture realizes full-link monitoring and intelligent early warning of wind power cable anomalies by integrating four modules: sampling analysis, thermodynamic modeling, multi-dimensional analysis, and dynamic early warning evaluation. The system first acquires real-time gas detection data (such as carbon monoxide, ammonia, and hydrogen concentrations) using gas sampling devices (such as infrared spectrometers or electrochemical sensors) deployed in the cable trench. The system then quantifies the gas composition distribution based on the GB / T 2951 standard. For example, a sudden increase in CO concentration indicates thermal decomposition of the insulation layer, while abnormal H2 concentrations are associated with partial discharge. A steady-state thermal circuit model is then established based on structural parameters such as the cable conductor diameter, insulation thickness, and material thermal resistance. The thermal conductivity distribution of the insulation layer is then estimated using load current and surface temperature data from the SCADA system and the Joule heating formula. Furthermore, a multi-dimensional analysis module integrates gas distribution, electrical parameters (such as current mutation ΔI), and thermal circuit model output to establish a multi-feature mapping relationship, identifying cable aging characteristics such as oxidation, thermal decomposition, and partial discharge. Finally, the dynamic early warning assessment module generates a comprehensive risk score using a weighted summation method. Adaptive classification thresholds are set based on historical fault data, and the module, linked to a GIS map, generates a spatial heat map, annotates outliers with a red, orange, and yellow color gradient, and transmits the risk score to the work order system.

[0115] like Figure 7 As shown, as a preferred embodiment of the present invention, the sampling and analysis module 100 includes:

[0116] The gas sampling unit 101 is used to obtain gas detection data of key nodes of the wind power cable through a gas sampling device with a configured sampling frequency;

[0117] The data alignment unit 102 is used to align the data timestamps of the gas detection data using the operating parameters retrieved from the SCADA system;

[0118] The gas analysis unit 103 is used to identify the concentration of characteristic gases representing cable aging through an online mass spectrometer and record their changing trend over time;

[0119] The interference elimination unit 104 is used to combine the ambient temperature and humidity sensor data, use an algorithm to eliminate the interference of non-cable related background gases, and obtain the gas composition distribution map and concentration gradient curve distributed along each key node of the cable.

[0120] like Figure 8 As shown, as a preferred embodiment of the present invention, the thermodynamic modeling module 200 includes:

[0121] A feature extraction unit 201 is used to obtain cable structural parameters and thermal resistance and thermal capacity data from a material database, and automatically extract key features using a deep learning model to generate standardized parameter inputs for a steady-state thermal circuit model;

[0122] A model building unit 202 is used to establish a steady-state thermal circuit model of thermoelectric analogy based on the cable layered structure, and simulate the heat flow distribution through the Monte Carlo method to output a temperature field prediction value on the cable surface;

[0123] The data fusion unit 203 is used to extract the real-time load current, ambient temperature and cable surface temperature data from the SCADA system, calculate the calorific value by combining the Joule heat formula, and eliminate the noise by Kalman filtering to obtain the SCADA temperature measured value;

[0124] The data inverse deduction unit 204 is used to compare the temperature field prediction value with the SCADA temperature measured value and inversely deduce the equivalent thermal conductivity of the insulation layer by combining the least square method;

[0125] The model calibration unit 205 is configured to constrain the steady-state heat circuit model parameters using historical fault gas data to calibrate the steady-state heat circuit model.

[0126] like Figure 9 As shown, as a preferred embodiment of the present invention, the dynamic early warning assessment module 400 includes:

[0127] The scoring weight allocation unit 401 is used to quantify the weight of each feature through an algorithm based on gas composition, electrical parameters and steady-state thermal circuit model output, and dynamically adjust the weight allocation;

[0128] The risk score calculation unit 402 is used to generate a risk score using a weighted summation method and set a classification threshold based on historical fault data, and the threshold can be adaptively adjusted;

[0129] The visualization unit 403 is used to overlay the risk score onto the cable GIS map to generate a spatial heat map and mark abnormal points using a color gradient;

[0130] The grading processing unit 404 is used to divide the warning into three levels according to the scores, and link the work order system to push maintenance work orders to the operation and maintenance personnel in a graded manner.

[0131] The above is only a detailed description of the preferred embodiments of the present invention, which is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0132] It should be understood that, although the various steps in the flow chart of each embodiment of the present invention are shown in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in order, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0133] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0134] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the disclosure in the specification and examples. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered merely as exemplary, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A breakpoint sniffing method for a wind power grid load storage architecture, characterized in that: The method comprises the following steps: Obtaining gas detection data of wind power cables and analyzing it to obtain gas composition distribution data, wherein the gas detection data is obtained by a gas sampling device deployed in the cable trench; Establishing a steady-state thermal circuit model based on cable structural parameters, including conductor diameter, insulation layer thickness, and material thermal resistance, and inferring insulation layer thermal conductivity distribution in combination with load current and surface temperature data from a SCADA system; Combined with the electrical parameters of the SCADA system, a multi-dimensional mapping relationship between gas distribution, electrical parameters, and insulation layer thermal conductivity distribution is established to identify the aging characteristics of different wind power cables; Generate a comprehensive risk score based on gas concentration thresholds and electrical parameter mutation thresholds, and use a graded early warning mechanism to assign maintenance priorities to cable anomalies; The step of establishing a steady-state thermal circuit model based on cable structural parameters and calculating the thermal conductivity distribution of the insulation layer in combination with the load current and surface temperature data of the SCADA system specifically includes: Obtain cable structural parameters and thermal resistance and heat capacity data from the material database, and use a deep learning model to automatically extract key features to generate standardized parameter inputs for a steady-state thermal circuit model. A steady-state thermal circuit model based on thermoelectric analogy is established based on the cable layered structure, and the heat flux distribution is simulated using the Monte Carlo method to output the temperature field prediction value of the cable surface; Extract real-time load current, ambient temperature, and cable surface temperature data from the SCADA system, calculate the calorific value using the Joule heat formula, and eliminate noise through Kalman filtering to obtain the SCADA temperature measured value. By comparing the temperature field prediction value with the SCADA temperature measured value, and combining the least squares method to infer the equivalent thermal conductivity of the insulation layer; The steady-state heat circuit model is calibrated by using historical fault gas data to constrain parameters of the steady-state heat circuit model.

2. The breakpoint sniffing method of the wind power grid load storage architecture according to claim 1 is characterized in that: The steps of obtaining gas detection data of the wind power cable and analyzing to obtain gas composition distribution data specifically include: Obtain gas detection data at key nodes of wind power cables through a gas sampling device with a configured sampling frequency; Use the operating parameters retrieved from the SCADA system to align the data timestamps of gas detection data; Identify the concentration of characteristic gases representing cable aging through an online mass spectrometer and record their changing trends over time; Combined with the ambient temperature and humidity sensor data, an algorithm is used to eliminate the interference of non-cable related background gases, and the gas composition distribution map and concentration gradient curve distributed along each key node of the cable are obtained.

3. The breakpoint sniffing method of the wind power grid load storage architecture according to claim 1 is characterized in that: The steps of establishing a steady-state thermal circuit model based on a thermoelectric analogy based on the cable layered structure and simulating heat flow distribution by using the Monte Carlo method specifically include: Based on the physical characteristics of wind power cables, a steady-state thermal circuit model with thermoelectric analogy layers is established, and the thermal resistance units and material properties of each layer are divided. The cable layers are composed of conductor, insulation layer, sheath and environment. Equivalently equate the conductor, insulation layer, sheath and environment to series thermal resistance, and quantify the material thermal resistance value and contact surface thermal resistance distribution; By randomly sampling, different load currents, ambient temperatures, and cable surface temperatures are input into the steady-state thermal circuit model to simulate the distribution of heat flow in the steady-state thermal circuit model. The predicted value of the cable surface temperature field is generated based on the calculation results of the steady-state thermal circuit model.

4. The breakpoint sniffing method of the wind power grid load storage architecture according to claim 1 is characterized in that: The steps of generating a comprehensive risk score based on the gas concentration threshold and the electrical parameter mutation threshold, and assigning maintenance priorities to cable abnormal points using a graded early warning mechanism, specifically include: Based on gas composition, electrical parameters and steady-state thermal circuit model output, the algorithm quantifies the weight of each feature and dynamically adjusts the weight distribution; A weighted summation method is used to generate risk scores, and classification thresholds are set based on historical fault data. The thresholds can be adjusted adaptively. Overlay the risk score onto the cable GIS map to generate a spatial heat map, and mark outliers with a color gradient; Three levels of warning are divided according to the scores, and the work order system is linked to push maintenance work orders to operation and maintenance personnel in a graded manner.

5. The breakpoint sniffing method of the wind power grid load storage architecture according to claim 4 is characterized in that: The method further comprises: Extract spatial coordinates from anomaly points marked on the GIS map and generate drone inspection target points based on the cable wiring diagram; Calculate the optimal inspection path for drones based on the priority of outliers and the algorithm; Obtain image data, thermal imaging data, and gas concentration information during drone inspections; The spatial heat map is dynamically adjusted based on the inspection results.

6. The breakpoint sniffing system of wind power grid load storage architecture is characterized by: The system comprises: A sampling and analysis module is used to obtain gas detection data of wind power cables and analyze it to obtain gas composition distribution data. The gas detection data is obtained by a gas sampling device deployed in the cable trench; A thermodynamic modeling module is used to establish a steady-state thermal circuit model based on cable structural parameters, including conductor diameter, insulation thickness, and material thermal resistance, and to infer the thermal conductivity distribution of the insulation layer using load current and surface temperature data from the SCADA system; A multi-dimensional analysis module is used to combine the electrical parameters of the SCADA system to establish a multi-dimensional mapping relationship between gas distribution, electrical parameters, and insulation layer thermal conductivity distribution, which is used to identify the aging characteristics of different wind power cables; Dynamic early warning assessment module, used to generate comprehensive risk scores based on gas concentration thresholds and electrical parameter mutation thresholds, and use a hierarchical early warning mechanism to assign maintenance priorities to cable anomalies; The thermodynamic modeling module includes: A feature extraction unit is used to obtain cable structural parameters and thermal resistance and thermal capacity data from the material database. It uses a deep learning model to automatically extract key features and generate standardized parameter inputs for the steady-state thermal circuit model. A model building unit is used to establish a steady-state thermal circuit model based on the thermoelectric analogy of the cable layered structure, simulate the heat flow distribution through the Monte Carlo method, and output the temperature field prediction value of the cable surface; The data fusion unit is used to extract real-time load current, ambient temperature and cable surface temperature data from the SCADA system, calculate the calorific value by combining the Joule heat formula, and eliminate noise through Kalman filtering to obtain the SCADA temperature measured value; The data inversion unit is used to compare the temperature field prediction value with the SCADA temperature measured value and use the least squares method to infer the equivalent thermal conductivity of the insulation layer; The model calibration unit is used to constrain the steady-state heat circuit model parameters by using historical fault gas data to calibrate the steady-state heat circuit model.

7. The breakpoint sniffing system of the wind power grid load storage architecture according to claim 6 is characterized in that: The sampling and analysis module includes: A gas sampling unit is used to obtain gas detection data at key nodes of wind power cables through a gas sampling device with a configured sampling frequency; A data alignment unit is used to align the data timestamps of the gas detection data using the operating parameters retrieved from the SCADA system; Gas analysis unit, used to identify the concentration of characteristic gases representing cable aging through an online mass spectrometer and record their changing trends over time; The interference elimination unit is used to combine the ambient temperature and humidity sensor data, use the algorithm to eliminate the interference of non-cable related background gases, and obtain the gas composition distribution map and concentration gradient curve distributed along each key node of the cable.

8. The breakpoint sniffing system of the wind power grid load storage architecture according to claim 6 is characterized in that: The dynamic early warning assessment module includes: The scoring weight allocation unit is used to quantify the weight of each feature through an algorithm based on gas composition, electrical parameters and steady-state thermal circuit model output, and dynamically adjust the weight allocation; The risk score calculation unit is used to generate risk scores using a weighted summation method and set classification thresholds based on historical fault data. The thresholds can be adjusted adaptively. A visualization unit is used to overlay the risk score onto the cable GIS map to generate a spatial heat map and mark abnormal points with a color gradient; The hierarchical processing unit is used to divide warnings into three levels according to the scores, and to link with the work order system to push maintenance work orders to operation and maintenance personnel in a hierarchical manner.

Images