Corrosion state early warning method and system for transformer substation grounding grid

Abstract

The invention discloses a corrosion state early warning method and system for a transformer substation grounding grid, and the method comprises the steps: carrying out the preprocessing and feature engineering of collected multi-source data, and forming a feature data set; inputting the feature data set into a pre-trained corrosion state evaluation model, and outputting a current corrosion state quantitative index of the grounding grid conductor; the corrosion state evaluation model is a machine learning model trained based on historical grounding grid corrosion sample data; predicting the corrosion development trend and the remaining service life of the grounding grid based on the corrosion state evaluation result of the time sequence; and the evaluation and prediction result is visually displayed, and an early warning is given out. Large-scale excavation inspection is avoided, a large amount of manpower, material resources and time cost are saved, and on-line or off-line intelligent diagnosis and early warning of the state of the grounding grid are achieved.

Description

Technical Field

[0001] This invention belongs to the field of power facility condition monitoring technology, and particularly relates to a method and system for early warning of corrosion status of substation grounding grid. Background Technology

[0002] Substation grounding grids are critical facilities for ensuring the safe and stable operation of power systems and protecting equipment and personnel. Buried underground for extended periods, they are subject to various factors such as soil chemical corrosion, electrochemical corrosion, and stray current corrosion. These factors can lead to a reduction in conductor cross-section and deterioration of connection points, resulting in increased grounding resistance and decreased current discharge capacity, seriously threatening power grid safety.

[0003] Currently, the assessment of corrosion status of grounding grids mainly relies on periodic excavation and spot checks. This method has drawbacks such as being destructive, costly, and lacking sufficient sample representativeness. In recent years, some trenchless detection methods have emerged, such as electrochemical measurement methods (e.g., polarization resistance method), electromagnetic detection methods, and infrared thermography. However, these methods often have the following limitations:

[0004] Singleness: Most rely on a single parameter (such as grounding resistance) or a single physical field (such as electromagnetic field) for judgment, which makes it difficult to fully reflect the complex corrosion situation, has poor anti-interference ability, and limited accuracy.

[0005] High dependence on experience: Traditional analysis methods are mostly based on threshold comparisons or simple empirical formulas, which cannot deeply explore the complex nonlinear relationship between monitoring data and corrosion status.

[0006] Insufficient predictability: It is difficult to accurately predict the remaining life of the grounding grid and its future corrosion trend, and it cannot provide forward-looking decision support for condition-based maintenance.

[0007] Therefore, there is an urgent need for a trenchless analysis method that can integrate multi-dimensional information, has a high degree of intelligence, and can accurately assess and predict the corrosion status of grounding grids. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method and system for early warning of corrosion status of substation grounding grids that is highly accurate, predictive, and requires no excavation. The first aspect of this invention provides...

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] A method for early warning of corrosion status in substation grounding grids includes the following steps:

[0011] S1: Multi-source data acquisition: Acquire data on various characteristic parameters that affect the corrosion status of the grounding grid, including soil physicochemical parameters, grounding grid electrical parameters, operating environment parameters, and historical maintenance data;

[0012] S2: Data preprocessing and feature engineering: Clean, align, and normalize the multi-source data collected in step S1, and construct derived features that are strongly correlated with erosion to form a high-quality feature dataset.

[0013] S3: Intelligent Corrosion Status Assessment: Input the feature dataset obtained in step S2 into the pre-trained corrosion status assessment model, and output the current average corrosion depth and / or remaining cross-sectional percentage of the grounding grid conductor; the corrosion status assessment model is a machine learning model trained based on historical grounding grid corrosion sample data;

[0014] S4: Corrosion Trend Prediction and Lifetime Assessment: Based on the corrosion status assessment results of the time series, the corrosion development trend in a specific future time period is predicted using a time series prediction algorithm, and the remaining service life of the grounding grid is estimated.

[0015] S5: Visualization and Early Warning: Visualizes the assessment results, predicted trends and lifespan information, and issues early warning information when the corrosion status exceeds the safety threshold or the lifespan is lower than the preset value.

[0016] Preferably, in step S1, the soil physicochemical parameters include soil resistivity, pH value, water content, redox potential, Cl- ion concentration, and SO42-. 2 - Ion concentration; The grounding grid electrical parameters include conductivity resistance, grounding impedance and conductor potential; The operating environment parameters include grounding grid current, leakage current of adjacent cables and historical records of short-circuit faults in the station; The historical maintenance data includes corrosion depth and corrosion location information from excavation inspection records.

[0017] Preferably, in step S3, the machine learning model is a Gradient Boosting Decision Tree (GBDT), Random Forest, or Support Vector Machine (SVM) model.

[0018] Preferably, in step S4, the time series prediction algorithm is an autoregressive integral moving average model (ARIMA) or a long short-term memory network (LSTM).

[0019] The beneficial effects of this invention are as follows:

[0020] Multi-source information fusion: It integrates information from multiple dimensions such as soil, electrical, and environment, comprehensively reflects the causes of corrosion, and significantly improves the accuracy and reliability of the assessment.

[0021] Artificial intelligence driven: Employing machine learning algorithms, it can automatically learn complex nonlinear relationships, overcoming the limitations of traditional empirical formulas, and making the evaluation results more scientific and objective.

[0022] Condition prediction capability: It can not only assess the current condition, but also predict future corrosion trends and remaining lifespan, providing strong data support for the full life cycle management and predictive maintenance of equipment.

[0023] Trenchless inspection: It avoids large-scale excavation inspection, saving a lot of manpower, material resources and time costs, and realizes online or offline intelligent diagnosis and early warning of the grounding grid status. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart illustrating a corrosion state early warning method for a substation grounding grid according to an embodiment of the present invention;

[0026] Figure 2 This is a structural block diagram of a corrosion status early warning system for a substation grounding grid, provided in an embodiment of the present invention.

[0027] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0029] Example 1:

[0030] Reference Figure 1 The specific steps of the early warning method for corrosion status of substation grounding grid in this embodiment are as follows:

[0031] S101: Multi-source data acquisition. Soil resistivity, pH, and moisture content data are acquired at different locations within the station using soil resistivity meters and multi-parameter soil analyzers; electrical parameters are acquired using grounding grid continuity testers and grounding impedance testers; operating environment parameters and historical maintenance data are obtained from the SCADA system or historical records. All data is timestamped and tagged with location.

[0032] S102: Data Preprocessing and Feature Engineering. Missing data is imputed using interpolation, and outliers are identified and removed using box plots. Data of different dimensions (such as resistivity and ion concentration) are Z-score normalized. Derivative features are constructed, such as: "Soil Corrosivity Index" (combining pH, resistivity, and ion concentration), "Annual Average Fault Current Density," etc.

[0033] S103: Intelligent Corrosion Status Assessment. This embodiment selects the GBDT model as the corrosion status assessment model. This model is trained using a large number of historical excavation samples (feature parameters as input, measured corrosion depth as output label). The real-time feature data processed in step S102 is input into the trained model, and the model can then output a predicted value for the average corrosion depth of the grounding grid conductor.

[0034] S104: Corrosion Trend Prediction and Lifetime Assessment. Following step S103, a monthly average corrosion depth time series over a period (e.g., the past 3 years) is obtained. An LSTM network is used to learn from this time series to predict corrosion depth changes over the next 12 months. Assuming the original conductor radius is R and the critical corrosion depth is R_critical (e.g., 80%*R), the remaining lifespan can be estimated based on the predicted corrosion rate.

[0035] S105: Visualization and Early Warning. The system interface displays historical and predicted corrosion depth changes using a line graph, and shows the current remaining cross-section percentage and remaining lifespan in a dashboard format. A threshold can be set (e.g., remaining lifespan < 2 years). When the threshold is triggered, the system sends an alert to maintenance personnel via interface flashing, SMS, or other means.

[0036] Reference Figure 2 This embodiment of a substation grounding grid corrosion early warning system includes:

[0037] Database: Stores all relevant data.

[0038] Data acquisition module: Interfaces with field testing equipment and SCADA system to acquire data.

[0039] Data preprocessing module and feature engineering module: perform data cleaning and feature construction.

[0040] Analysis and computation engine: This is the core computing unit, with a built-in model management module (storing GBDT and LSTM models) that performs model calls and calculations.

[0041] Visualization and Early Warning Module: Provides a web interface for users to view results and receive early warnings.

[0042] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

[0043] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for early warning of corrosion status in substation grounding grids, characterized in that, include: S1: Collect data on various characteristic parameters that affect the corrosion status of the grounding grid, including soil physicochemical parameters, grounding grid electrical parameters, operating environment parameters, and historical maintenance data; S2: Preprocess and feature-engineer the multi-source data collected in step S1 to form a feature dataset; S3: Input the feature dataset into the pre-trained corrosion state assessment model and output a quantitative index of the current corrosion state of the grounding grid conductor; the corrosion state assessment model is a machine learning model trained based on historical grounding grid corrosion sample data; S4: Based on the corrosion status assessment results of the time series, predict the corrosion development trend and remaining service life of the grounding grid; S5: Visualize the assessment and prediction results and issue early warnings.

2. The corrosion status early warning method for substation grounding grids according to claim 1, characterized in that, In step S1, the soil physicochemical parameters include multiple parameters such as soil resistivity, pH value, water content, redox potential, Cl- ion concentration, and SO42- ion concentration; the grounding grid electrical parameters include multiple parameters such as conductivity resistance, grounding impedance, and conductor potential; and the operating environment parameters include grounding grid current and historical records of short-circuit faults within the station.

3. A method for early warning of corrosion status of a substation grounding grid according to claim 1 or 2, characterized in that, In step S2, the preprocessing includes data cleaning, alignment, and normalization; the feature engineering includes constructing derived features that are strongly correlated with the corrosion rate.

4. The corrosion status early warning method for substation grounding grids according to claim 1, characterized in that, In step S3, the corrosion state quantification index is the average corrosion depth and / or the percentage of remaining cross section; the machine learning model is a gradient boosting decision tree, random forest, or support vector machine model.

5. A method for early warning of corrosion status in a substation grounding grid according to claim 1, characterized in that, In step S4, a time series prediction algorithm is used for prediction, wherein the algorithm is an ARIMA model or an LSTM network.

6. A corrosion status early warning system for substation grounding grids, used to implement the analysis method described in any one of claims 1-5, characterized in that, include: The data acquisition module is used to acquire multi-source feature parameter data; The data preprocessing module is used to clean, align, and normalize the collected data. The feature engineering module is used to construct derived features from preprocessed data; The model management module is used to store, update, and recall the corrosion state assessment model and trend prediction model; An analysis and calculation engine is used to invoke the model and perform corrosion status assessment and trend prediction calculations. The visualization and early warning module is used to display analysis results and provide early warnings.

7. A corrosion condition early warning system for substation grounding grids according to claim 6, characterized in that, It also includes a database for storing the collected raw data, preprocessed data, feature datasets, model parameters, and historical analysis results. The system according to claim 6 is characterized in that the data acquisition module can interface with field testing equipment and power grid monitoring system to acquire data automatically or manually.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method as described in any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1 to 5.

Images