Grounding grid corrosion rate prediction method and system based on machine learning model
By using a machine learning model that integrates multi-source data, the corrosion rate of the grounding grid can be predicted in real time and an early warning can be generated. This solves the problem of difficulty in monitoring the corrosion status of the grounding grid in existing technologies, and improves the safety and operation and maintenance efficiency of the power grid.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 国网江西省电力有限公司九江供电分公司
- Filing Date
- 2025-11-27
- Publication Date
- 2026-04-10
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing technologies cannot achieve long-term, continuous, and accurate monitoring of the corrosion status of grounding grids, making it difficult to prevent potential safety hazards in power systems.
A machine learning model that integrates multi-source data fusion is used to construct an ensemble learning prediction model by collecting soil, electrical, and meteorological parameters. This model can predict the corrosion rate of the grounding grid in real time and generate early warning information.
It enables high-precision, real-time prediction of grounding grid corrosion rates, improves operation and maintenance efficiency, and can issue early warnings of corrosion risks to ensure power grid safety.
Smart Images
Figure CN121834149A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power system equipment condition monitoring and corrosion protection technology, and particularly relates to a method and system for predicting the corrosion rate of grounding grids based on machine learning models. Background Technology
[0002] The grounding grid is a critical facility for ensuring the safe and stable operation of the power system. It provides a discharge path for fault currents, protecting personnel and equipment. However, because the grounding grid is buried deep underground for extended periods, it is exposed to a complex soil chemical and electrochemical environment, making it highly susceptible to corrosion. Corrosion leads to a reduction in the cross-sectional area of the grounding grid conductors and deterioration of connection points, resulting in a decline in its electrical performance, such as increased grounding resistance and insufficient thermal stability. In severe cases, this can lead to major safety accidents.
[0003] Currently, monitoring the corrosion status of grounding grids mainly relies on periodic excavation and spot checks. This method has drawbacks such as high randomness, large workload, and inability to achieve continuous monitoring. Some electrochemical detection methods (such as linear polarization method and electrochemical impedance spectroscopy) have also been tried for on-site measurement, but these methods usually only provide point-like and instantaneous corrosion information, making it difficult to accurately reflect the long-term and macroscopic corrosion trend of the entire grounding grid.
[0004] In recent years, some studies have attempted to predict corrosion rates by measuring soil physicochemical properties (such as pH, water content, resistivity, etc.) and combining them with empirical models. However, the soil environment is complex and variable, and there is a high degree of nonlinearity and uncertainty between a single or a few parameters and the corrosion rate, resulting in generally low prediction accuracy.
[0005] Therefore, there is an urgent need for an intelligent method and system that can accurately predict the overall corrosion rate of the grounding grid without relying on excavation, over a long period of time, online. Summary of the Invention
[0006] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a high-precision, real-time-efficient, and excavation-free method and system for predicting grounding grid corrosion rates based on machine learning models. This method achieves accurate prediction of corrosion rates in future time periods by fusing multi-source heterogeneous data and utilizing machine learning models to capture the complex nonlinear relationship between data and corrosion rates.
[0007] In a first aspect, the present invention provides a method for predicting the corrosion rate of a grounding grid based on a machine learning model, comprising:
[0008] Step S1: Multi-source data acquisition and preprocessing: Collect multi-source historical data of the environment where the target grounding grid is located. The multi-source data includes soil environmental parameters, electrical parameters and meteorological parameters. Clean, align and normalize the collected data to form a regular historical dataset.
[0009] Step S2: Feature Engineering and Dataset Construction: Extract time-domain and frequency-domain features related to corrosion rate from the preprocessed data, and associate them with the actual corrosion rate label data of the corresponding time period obtained through offline detection to construct a feature-label dataset for model training;
[0010] Step S3: Machine learning model construction and training: Build an ensemble learning prediction model, use the dataset built in step S2 to train the model and optimize hyperparameters to obtain a trained corrosion rate prediction model.
[0011] Step S4: Real-time prediction: Collect soil environmental parameters, electrical parameters and meteorological parameters in real time, and after the same preprocessing and feature extraction as in steps S1 and S2, input them into the prediction model trained in step S3, and output the predicted value of corrosion rate for a future time period.
[0012] Step S5: Prediction Result Output and Early Warning: Visualize the prediction results and compare them with the preset corrosion rate threshold. If the predicted value exceeds the threshold, generate an early warning message and send it to the operation and maintenance personnel.
[0013] Secondly, the present invention provides a grounding grid corrosion rate prediction system based on a machine learning model, comprising:
[0014] The acquisition module is configured for multi-source data acquisition and preprocessing: it acquires multi-source historical data of the environment where the target grounding grid is located, including soil environmental parameters, electrical parameters, and meteorological parameters; it cleans, aligns, and normalizes the acquired data to form a regular historical dataset.
[0015] The module is configured for feature engineering and dataset construction: extract time-domain and frequency-domain features related to corrosion rate from the preprocessed data, and associate them with the actual corrosion rate label data of the corresponding time period obtained through offline detection to construct a feature-label dataset for model training;
[0016] The training module is configured for machine learning model building and training: it builds an ensemble learning prediction model, uses the dataset built in step S2 to train the model and optimize its hyperparameters, and obtains a trained corrosion rate prediction model.
[0017] The prediction module is configured for real-time prediction: it collects soil environmental parameters, electrical parameters and meteorological parameters at the current moment in real time, and after the same preprocessing and feature extraction as in steps S1 and S2, it is input into the prediction model trained in step S3 and outputs the predicted value of corrosion rate for a future time period.
[0018] The output module is configured to output prediction results and issue warnings: it visualizes the prediction results and compares them with a preset corrosion rate threshold. If the predicted value exceeds the threshold, it generates a warning message and sends it to the maintenance personnel.
[0019] Thirdly, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the grounding grid corrosion rate prediction method based on a machine learning model according to any embodiment of the present invention.
[0020] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the steps of the grounding grid corrosion rate prediction method based on a machine learning model according to any embodiment of the present invention.
[0021] The grounding grid corrosion rate prediction method and system based on machine learning models in this application have the following specific advantages:
[0022] Multi-source data fusion: It comprehensively utilizes multi-dimensional parameters such as soil chemistry, electrical engineering and meteorology, providing comprehensive information and overcoming the limitations of single-parameter prediction.
[0023] Artificial intelligence driven: Employing advanced machine learning algorithms (such as LightGBM), it can automatically learn and capture the complex nonlinear mapping relationship between multi-source data and corrosion rate, with prediction accuracy far exceeding that of traditional empirical models.
[0024] Online real-time prediction: It enables online and continuous corrosion status monitoring and prediction without relying on excavation, which greatly improves operation and maintenance efficiency and status awareness.
[0025] Proactive early warning: It can issue early warnings of corrosion risks based on the prediction results, guide operation and maintenance personnel to carry out proactive protection and precise maintenance, transform "passive response" into "proactive defense", and effectively ensure the safety of the power grid. Attached Figure Description
[0026] 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.
[0027] Figure 1 A flowchart illustrating a grounding grid corrosion rate prediction method based on a machine learning model, as provided in an embodiment of the present invention;
[0028] Figure 2 A structural block diagram of a grounding grid corrosion rate prediction system based on a machine learning model is provided in an embodiment of the present invention.
[0029] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation
[0030] 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.
[0031] Please see Figure 1 The diagram shows a flowchart of a grounding grid corrosion rate prediction method based on a machine learning model, as described in this application.
[0032] like Figure 1 As shown, the grounding grid corrosion rate prediction method based on machine learning models specifically includes the following steps:
[0033] Step S1: Multi-source data acquisition and preprocessing: Collect multi-source historical data of the environment where the target grounding grid is located. The multi-source data includes soil environmental parameters, electrical parameters and meteorological parameters. Clean, align and normalize the collected data to form a regular historical dataset.
[0034] In step S1, the soil environmental parameters include soil resistivity, pH value, redox potential, water content, chloride ion concentration and sulfate ion concentration; the electrical parameters include leakage current density and / or ground potential gradient of the grounding grid conductor; and the meteorological parameters include ambient temperature, humidity, precipitation and evaporation.
[0035] Step S2: Feature Engineering and Dataset Construction: Extract time-domain and frequency-domain features related to corrosion rate from the preprocessed data, and associate them with the actual corrosion rate label data for the corresponding time period obtained through offline detection to construct a feature-label dataset for model training.
[0036] In step S2, the actual corrosion rate label data is obtained by weightlessness or based on measurements from a pre-embedded corrosion sensor array.
[0037] Step S3: Machine learning model construction and training: Build an ensemble learning prediction model, use the dataset built in step S2 to train the model and optimize its hyperparameters to obtain a trained corrosion rate prediction model.
[0038] In step S3, the ensemble learning prediction model employs the LightGBM algorithm, and its objective function is:
[0039]
[0040] In the formula, Obj (t) The loss function y i To represent the actual corrosion rate, For the predicted value, Ω(f) t ) is a regularization term used to control model complexity and prevent overfitting.
[0041] Step S4: Real-time prediction: Collect soil environmental parameters, electrical parameters and meteorological parameters in real time, and after the same preprocessing and feature extraction as in steps S1 and S2, input them into the prediction model trained in step S3, and output the predicted value of corrosion rate for the next time period.
[0042] Step S5: Prediction Result Output and Early Warning: Visualize the prediction results and compare them with the preset corrosion rate threshold. If the predicted value exceeds the threshold, generate an early warning message and send it to the operation and maintenance personnel.
[0043] Please see Figure 2 The diagram shows a structural block diagram of a grounding grid corrosion rate prediction system based on a machine learning model, according to this application.
[0044] like Figure 2 As shown, the grounding grid corrosion rate prediction system 200 based on a machine learning model includes an acquisition module 210, a construction module 220, a training module 230, a prediction module 240, and an output module 250.
[0045] The acquisition module 210 is configured for multi-source data acquisition and preprocessing: acquiring multi-source historical data of the environment where the target grounding grid is located, including soil environmental parameters, electrical parameters and meteorological parameters; cleaning, aligning and normalizing the acquired data to form a regular historical dataset;
[0046] Module 220 is configured for feature engineering and dataset construction: extract time-domain and frequency-domain features related to corrosion rate from the preprocessed data, and associate them with the actual corrosion rate label data of the corresponding time period obtained by offline detection to construct a feature-label dataset for model training.
[0047] Training module 230 is configured for machine learning model building and training: build an ensemble learning prediction model, use the dataset built in step S2 to train the model and optimize hyperparameters to obtain a trained corrosion rate prediction model.
[0048] The prediction module 240 is configured for real-time prediction: it collects soil environmental parameters, electrical parameters and meteorological parameters at the current moment in real time, and after the same preprocessing and feature extraction as in steps S1 and S2, it is input into the prediction model trained in step S3 and outputs the predicted value of corrosion rate for a future time period.
[0049] Output module 250 is configured to output prediction results and issue warnings: it visualizes the prediction results and compares them with a preset corrosion rate threshold. If the predicted value exceeds the threshold, it generates a warning message and sends it to the maintenance personnel.
[0050] It should be understood that Figure 2 The modules and references described in the document Figure 1 The steps described in the text correspond to those in the method described above. Therefore, the operations, features, and corresponding technical effects described above also apply to the method described in the text. Figure 2 The various modules in the document will not be described in detail here.
[0051] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the grounding grid corrosion rate prediction method based on a machine learning model in any of the above method embodiments.
[0052] In one embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, which are configured as follows:
[0053] Multi-source data acquisition and preprocessing: Collect multi-source historical data of the environment where the target grounding grid is located. The multi-source data includes soil environmental parameters, electrical parameters and meteorological parameters. Clean, align and normalize the collected data to form a regular historical dataset.
[0054] Feature engineering and dataset construction: Extract time-domain and frequency-domain features related to corrosion rate from the preprocessed data, and associate them with the actual corrosion rate label data of the corresponding time period obtained by offline detection to construct a feature-label dataset for model training;
[0055] Machine learning model construction and training: Construct an ensemble learning prediction model, use the dataset constructed in step S2 to train the model and optimize its hyperparameters to obtain a trained corrosion rate prediction model;
[0056] Real-time prediction: The soil environmental parameters, electrical parameters and meteorological parameters at the current moment are collected in real time. After the same preprocessing and feature extraction as in steps S1 and S2, they are input into the prediction model trained in step S3 and output the predicted value of corrosion rate for a future time period.
[0057] Prediction results output and early warning: The prediction results are visualized and compared with the preset corrosion rate threshold. If the predicted value exceeds the threshold, an early warning message is generated and sent to the operation and maintenance personnel.
[0058] Computer-readable storage media may include a stored program area and a stored data area, wherein the stored program area may store an operating system and an application program required for at least one function; the stored data area may store data created based on the use of the grounding grid corrosion rate prediction system based on a machine learning model. Furthermore, the computer-readable storage medium may include high-speed random access memory, and may also include memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely configured relative to a processor, which can be connected to the grounding grid corrosion rate prediction system based on a machine learning model via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
[0059] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 3 As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 3 Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby implementing the grounding grid corrosion rate prediction method based on a machine learning model as described in the above method embodiment. The input device 330 can receive input digital or character information and generate key signal inputs related to user settings and function control of the grounding grid corrosion rate prediction system based on the machine learning model. The output device 340 may include a display device such as a screen.
[0060] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.
[0061] In one implementation, the above-described electronic device is applied to a grounding grid corrosion rate prediction system based on a machine learning model, and is used as a client, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:
[0062] Multi-source data acquisition and preprocessing: Collect multi-source historical data of the environment where the target grounding grid is located. The multi-source data includes soil environmental parameters, electrical parameters and meteorological parameters. Clean, align and normalize the collected data to form a regular historical dataset.
[0063] Feature engineering and dataset construction: Extract time-domain and frequency-domain features related to corrosion rate from the preprocessed data, and associate them with the actual corrosion rate label data of the corresponding time period obtained by offline detection to construct a feature-label dataset for model training;
[0064] Machine learning model construction and training: Construct an ensemble learning prediction model, use the dataset constructed in step S2 to train the model and optimize its hyperparameters to obtain a trained corrosion rate prediction model;
[0065] Real-time prediction: The soil environmental parameters, electrical parameters and meteorological parameters at the current moment are collected in real time. After the same preprocessing and feature extraction as in steps S1 and S2, they are input into the prediction model trained in step S3 and output the predicted value of corrosion rate for a future time period.
[0066] Prediction results output and early warning: The prediction results are visualized and compared with the preset corrosion rate threshold. If the predicted value exceeds the threshold, an early warning message is generated and sent to the operation and maintenance personnel.
[0067] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.
[0068] 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 predicting the corrosion rate of grounding grids based on a machine learning model, characterized in that, include: Step S1: Multi-source data acquisition and preprocessing: Acquire multi-source historical data of the environment where the target grounding grid is located. The multi-source data includes soil environmental parameters, electrical parameters and meteorological parameters. The collected data is cleaned, aligned, and normalized to form a well-organized historical dataset. Step S2: Feature Engineering and Dataset Construction: Extract time-domain and frequency-domain features related to corrosion rate from the preprocessed data, and associate them with the actual corrosion rate label data of the corresponding time period obtained through offline detection to construct a feature-label dataset for model training; Step S3: Machine learning model construction and training: Build an ensemble learning prediction model, use the dataset built in step S2 to train the model and optimize hyperparameters to obtain a trained corrosion rate prediction model. Step S4: Real-time prediction: Collect soil environmental parameters, electrical parameters and meteorological parameters in real time, and after the same preprocessing and feature extraction as in steps S1 and S2, input them into the prediction model trained in step S3, and output the predicted value of corrosion rate for a future time period. Step S5: Prediction Result Output and Early Warning: Visualize the prediction results and compare them with the preset corrosion rate threshold. If the predicted value exceeds the threshold, generate an early warning message and send it to the operation and maintenance personnel.
2. The grounding grid corrosion rate prediction method based on a machine learning model according to claim 1, characterized in that, In step S1, the soil environmental parameters include soil resistivity, pH value, redox potential, water content, chloride ion concentration and sulfate ion concentration; the electrical parameters include leakage current density and / or ground potential gradient of the grounding grid conductor; and the meteorological parameters include ambient temperature, humidity, precipitation and evaporation.
3. The grounding grid corrosion rate prediction method based on a machine learning model according to claim 1, characterized in that, In step S2, the actual corrosion rate label data is obtained by the weightlessness method or by measurements based on a pre-embedded corrosion sensor array.
4. The grounding grid corrosion rate prediction method based on a machine learning model according to claim 1, characterized in that, In step S3, the ensemble learning prediction model uses the LightGBM algorithm, and its objective function is: In the formula, Obj (t) The loss function y i To represent the actual corrosion rate, For the predicted value, Ω(f) t ) is a regularization term used to control model complexity and prevent overfitting.
5. A grounding grid corrosion rate prediction system based on a machine learning model, characterized in that, include: The acquisition module is configured for multi-source data acquisition and preprocessing: it acquires multi-source historical data of the environment where the target grounding grid is located, including soil environmental parameters, electrical parameters and meteorological parameters; The collected data is cleaned, aligned, and normalized to form a well-organized historical dataset. The module is configured for feature engineering and dataset construction: extract time-domain and frequency-domain features related to corrosion rate from the preprocessed data, and associate them with the actual corrosion rate label data of the corresponding time period obtained through offline detection to construct a feature-label dataset for model training; The training module is configured for machine learning model building and training: it builds an ensemble learning prediction model, uses the dataset built in step S2 to train the model and optimize its hyperparameters, and obtains a trained corrosion rate prediction model. The prediction module is configured for real-time prediction: it collects soil environmental parameters, electrical parameters and meteorological parameters at the current moment in real time, and after the same preprocessing and feature extraction as in steps S1 and S2, it is input into the prediction model trained in step S3 and outputs the predicted value of corrosion rate for a future time period. The output module is configured to output prediction results and issue warnings: it visualizes the prediction results and compares them with a preset corrosion rate threshold. If the predicted value exceeds the threshold, it generates a warning message and sends it to the maintenance personnel.
6. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 4.
7. 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 method described in any one of claims 1 to 4.