CN-122020301-A - Real-time diagnosis method and system for corrosion state of fishing light complementary power station grounding system based on double-layer information fusion and intelligent prediction

CN122020301ACN 122020301 ACN122020301 ACN 122020301ACN-122020301-A

Abstract

The invention discloses a real-time diagnosis method for corrosion state of a fishing light complementary power station grounding system based on double-layer information fusion and intelligent prediction, which belongs to the field of power equipment state monitoring and predictive maintenance and comprises the steps of synchronously collecting environment data of an environment where a grounding system is located, real-time corrosion rate data of a grounding body and grounding resistance data obtained by adopting a pulse synchronous measurement technology; based on the collected data, calculating to obtain the real-time health degree SOH of the grounding system through a first layer weighted fusion model, predicting the future health trend of the grounding system through a long-short-term memory network LSTM model based on the historical SOH sequence, and calculating the residual life RUL of the grounding system. The invention realizes real-time, multidimensional, accurate diagnosis and prospective prediction of the corrosion state of the grounding system of the fishing light complementary power station through anti-interference measurement and double-layer fusion analysis.

Inventors

Jiang Sushu
GONG TIEYU
WEI XING
CHEN TIEJUN
ZANG JUNLIN
LIU PENG

Assignees

上海尤汶新能源有限公司

Dates

Publication Date: 20260512
Application Date: 20260128

Claims (10)

1. A real-time diagnosis method for corrosion state of a fishing light complementary power station grounding system based on double-layer information fusion and intelligent prediction is characterized by comprising the following steps: S1, synchronously acquiring environment data of an environment where a grounding system is located, real-time corrosion rate data of a grounding body and grounding resistance data, wherein the grounding resistance data is obtained by adopting a pulse synchronous measurement technology; S2, calculating the real-time health degree of the grounding system through a first layer weighted fusion model based on the environmental data, the real-time corrosion rate data and the grounding resistance data acquired in the step S1; S3, predicting a future health trend sequence of the grounding system through a long-short-term memory network LSTM model based on the historical real-time health degree sequence, and calculating the residual life of the grounding system.
2. The method of claim 1, wherein the environmental data collected in step S1 includes salt spray concentration, ambient temperature and relative humidity, and wherein the leakage current data of the grounding system is also collected.
3. The method of claim 1, wherein the pulse synchronous measurement technique is specifically to inject a current pulse signal with a specific frequency into a ground network and synchronously detect a corresponding voltage response to calculate a ground resistance.
4. The method according to claim 2, wherein the calculation of the first layer weighted fusion model in step S2 comprises: s21, calculating an environmental corrosiveness index based on the environmental data; s22, calculating an electrical performance index EPI based on the ground resistance data and the leakage current data; And S23, carrying out weighted fusion according to the environment corrosiveness index, the real-time corrosion rate and the electrical performance index, and calculating to obtain the real-time health degree.
5. The method according to claim 4, wherein in step S23, the weighted fusion model is: SOH real−time =α*(1−ECI)+β*(1−V corr )+γ*EPI; Wherein, α, β, γ are weight coefficients, and α+β+γ=1, v corr is the normalized real-time corrosion rate.
6. The method according to claim 1, wherein in step S3, calculating the remaining life of the grounding system is specifically determining a time point when the future health trend curve predicted by the long-short-term memory network LSTM model is lower than a preset failure threshold for the first time, and a time difference between the time point and a current time point is the remaining life.
7. The method of claim 1, further comprising step S4 of generating hierarchical early warning information or predictive maintenance work orders based on the real-time wellness and the remaining life.
8. A real-time diagnosis system for corrosion state of a fishing light complementary power station grounding system based on double-layer information fusion and intelligent prediction, which is used for realizing the method as set forth in any one of claims 1-7, and is characterized by comprising: the on-site sensing layer comprises an environment sensor for collecting environment data, a corrosion rate sensor for collecting real-time corrosion rate data and a ground resistance on-line monitor adopting a pulse synchronous measurement technology; the data transmission layer is used for transmitting the data of the field sensing layer to the cloud intelligent analysis platform through wireless communication; And the cloud intelligent analysis platform is configured to execute the first-layer weighted fusion model calculation and the long-short-term memory network LSTM model prediction.
9. The system of claim 8, wherein the ground resistance online monitor comprises a host, a current pole and a voltage pole, the current pole and the voltage pole being disposed at a periphery of a ground grid, the host being connected to the current pole and the voltage pole by a measurement cable and configured to perform the pulsed synchronous measurement technique.
10. The system of claim 8, wherein the long-term memory network LSTM model is input as a time series data window comprising a historical real-time health, a historical environmental corrosion index, and a historical real-time corrosion rate, and output as a health prediction sequence over a future period of time.

Description

Real-time diagnosis method and system for corrosion state of fishing light complementary power station grounding system based on double-layer information fusion and intelligent prediction Technical Field The invention belongs to the technical field of power equipment state monitoring and predictive maintenance, and particularly relates to a real-time diagnosis method and system for corrosion state of a fishing light complementary power station grounding system based on double-layer information fusion and intelligent prediction. Background The grounding system is a key component for ensuring the safe and stable operation of electric power facilities, and is particularly important for a fishing light complementary power station built in a water area environment. The grounding body is in severe environments such as high humidity, high salt and alkali for a long time, electrochemical corrosion is easy to occur, the electrical performance is deteriorated, and the safety of a power station is threatened. At present, the monitoring and maintenance of the state of the grounding system mainly depend on the following prior art that firstly, the grounding system is periodically and manually inspected by inspection and excavation, and the grounding system is measured and visually inspected by tools such as a grounding resistance megger, and the method can find existing problems, but has long period and low efficiency, and belongs to post-treatment. Secondly, the on-line monitoring device with single function can realize automatic collection and uploading of specific parameters such as the grounding resistance, and avoid inconvenience of manual measurement, but the monitoring dimension is single, and only the electrical parameters are usually concerned. Thirdly, in the corrosion monitoring field, there are electrochemical sensors based on principles such as a hanging method (average corrosion rate is calculated by weighing) or a linear polarization resistance method, which can be used for acquiring corrosion information, but the former has serious data lag, and the latter is mostly used for laboratory or isolated point monitoring. However, the above prior art has significant drawbacks when applied to state management of a fishing light complementary power station grounding system. Firstly, the monitoring means is passive and lagged, so that real-time and continuous perception of the corrosion state cannot be realized, and the operation and maintenance decision lacks in-time data support. Secondly, each monitoring technology bar is segmented to present a data island state, and the complete fault chain of 'environment-driven corrosion process and electric performance degradation caused by corrosion process' is difficult to reveal only by means of isolated grounding resistance, environment or corrosion data, so that the dimension diagnosis is on one side, and accurate root cause analysis cannot be performed. Again, the existing methods are based entirely on current or historical conditions, lack predictive capabilities for corrosion trends and system remaining life, so that the operation and maintenance plan can only rely on fixed cycles or post-remediation, and cannot achieve prospective predictive maintenance. Finally, under the complex electromagnetic environment (particularly the existence of direct current components and stray currents) of the fishing light complementary power station, the traditional grounding resistance measurement method is easy to interfere, so that key basic data are distorted, and the reliability of any upper layer state evaluation is affected. Therefore, how to realize real-time, multidimensional, accurate diagnosis and prospective prediction of corrosion state of the grounding system of the fishing light complementary power station is a technical problem to be solved in the field. Disclosure of Invention In order to solve the technical problems, the invention provides a real-time diagnosis method and a real-time diagnosis system for the corrosion state of a grounding system of a fishing light complementary power station based on double-layer information fusion and intelligent prediction, so as to solve the problems in the prior art. In order to achieve the above purpose, in a first aspect, the present invention provides a real-time diagnosis method for corrosion state of a grounding system of a fishing light complementary power station based on double-layer information fusion and intelligent prediction, comprising: S1, synchronously acquiring environment data of an environment where a grounding system is located, real-time corrosion rate data of a grounding body and grounding resistance data, wherein the grounding resistance data is obtained by adopting a pulse synchronous measurement technology; S2, calculating the real-time health degree of the grounding system through a first layer weighted fusion model based on the environmental data, the real-time corrosion rate data and the grounding res