CN-122023564-A - Map drawing method and system for contribution of environmental factors to metal corrosion

CN122023564ACN 122023564 ACN122023564 ACN 122023564ACN-122023564-A

Abstract

A map drawing method and system for contribution of environmental factors to metal corrosion comprises the steps of disposing a plurality of sensors in a region to be predicted, monitoring the environmental factors and metal corrosion rates, carrying out correspondence between the environmental factors and environmental type colors, building and training a corrosion prediction model of each sensor according to the monitored environmental factors and the metal corrosion rates, obtaining corrosion contribution rates corresponding to the environmental factors of places where the sensors are disposed according to the trained corrosion prediction model and the monitored environmental factors and the metal corrosion rates, and drawing a corrosion contribution rate map according to the environmental type colors corresponding to the environmental factors with the highest corrosion contribution rates of the places. According to the invention, the corrosion contribution rate is constructed by applying the prediction model, and the corrosion contribution rate map is drawn, so that the real-time corrosion condition and future corrosion trend can be analyzed, the contribution effect of each environmental factor on corrosion is displayed, and a decision maker is helped to select corrosion prevention measures and strategies according to the environments of different areas.

Inventors

HE CHENG
YANG BINGKUN
WANG XIAOFANG
WANG XINXIN
LIU YANG
BUALLAY ADAK
LIU JIYAN
HUANG LUYAO
WANG ZONGJIANG
YOU YI
ZHANG QIANG
LU YILIANG
CHEN YUN
LIU HUI
HAN YU

Assignees

国网新疆电力有限公司电力科学研究院
国网智能电网研究院有限公司
中国电力科学研究院有限公司
国家电网有限公司

Dates

Publication Date: 20260512
Application Date: 20241106

Claims (10)

1. A method of mapping the contribution of environmental factors to metal corrosion, comprising: Disposing a plurality of sensors in a region to be predicted, monitoring environmental factors and metal corrosion rates, and performing correspondence between the environmental factors and environmental category colors; establishing and training a corrosion prediction model of each sensor according to the monitored environmental factors and the metal corrosion rate; aiming at each sensor, according to the trained corrosion prediction model, the monitored environmental factors and the metal corrosion rate, obtaining corrosion contribution rates corresponding to the environmental factors of each sensor deployment site; and drawing a corrosion contribution rate map according to the environment category colors corresponding to the environment factors with the highest corrosion contribution rate at each location.
2. The method of claim 1, wherein said building and training a corrosion prediction model for each sensor based on the monitored environmental factors and the metal corrosion rate comprises: integrating the monitored environmental factors and the metal corrosion rate for each sensor; the integrated environmental factors and the metal corrosion rate are taken as input, the corrosion contribution rate of each environmental factor is taken as output, and a random forest algorithm is adopted to establish a corrosion prediction model; and repeatedly testing, verifying, adjusting and optimizing the established corrosion prediction model.
3. The method of claim 2, wherein integrating the monitored environmental factors and the metal corrosion rate for each sensor comprises at least one or more of: And cleaning, removing abnormal data and normalizing the monitored environmental factors and the monitored metal corrosion rate.
4. The method of claim 2, wherein the environmental factors include at least one or more of: atmospheric temperature, atmospheric humidity, hour rainfall, wind speed, altitude, or air primary contaminant concentration.
5. The method of claim 1, wherein the deriving, for each sensor, a corrosion contribution rate corresponding to each environmental factor at a location where each sensor is deployed based on the trained corrosion prediction model and the monitored environmental factors and the metal corrosion rate comprises: Inputting the current environmental factors into a model to obtain the metal corrosion rate for each sensor, and calculating a correlation coefficient between the environmental factors and the metal corrosion rate; And accumulating correlation coefficients between the environmental factors and the metal corrosion rate in a preset time interval, and carrying out normalization processing to obtain the corrosion contribution rate corresponding to the environmental factors of each sensor deployment place in the current time period.
6. The method of claim 5, wherein the corrosion contribution rate for each environmental factor for each sensor deployment site within the current time period is determined as follows: wherein P a is the corrosion contribution rate corresponding to the environmental factor a of the place where each sensor is deployed in the current time period, c a is the accumulated value corresponding to w a in the time period from t o to t 1 , w a is the correlation coefficient in the model corresponding to the environmental factor a, c i is the accumulated value corresponding to w i in the time period from t o to t 1 , w i is the correlation coefficient in the model corresponding to the environmental factor i, and n is the number of the environmental factors; the accumulated value c i corresponding to w i in the time period t o to t 1 is calculated as follows: the accumulated value c a corresponding to w a in the time period t o to t 1 is calculated as follows:
7. The method of claim 1, wherein the mapping corrosion contribution rates according to the environmental category colors corresponding to the environmental factors having the highest corrosion contribution rates for each site comprises: calculating the corrosion contribution rate of each point according to the corrosion contribution rate corresponding to each environmental factor of each sensor deployment site; And drawing a corrosion contribution rate map by utilizing GIS software according to the color corresponding to the environmental factor with the highest corrosion contribution rate of each point.
8. The method of claim 7, wherein said calculating the corrosion contribution rate for each point based on the corrosion contribution rate for each environmental factor for each sensor deployment site comprises: And calculating the corrosion contribution rate of each point by using an inverse distance weight method and a Kriging interpolation method according to the corrosion contribution rate corresponding to each environmental factor of the place where each sensor is deployed.
9. The method of claim 1, wherein after the mapping the corrosion contribution rate according to the environmental category color corresponding to the environmental factor with the highest corrosion contribution rate for each location, further comprising: Text of the metal corrosion rate monitored by the corresponding sensor is marked in a map.
10. A mapping system for the contribution of environmental factors to metal corrosion, comprising: The monitoring module is used for deploying a plurality of sensors in the area to be predicted, monitoring environmental factors and metal corrosion rates, and performing the correspondence between the environmental factors and environmental category colors; the model building and training module is used for building and training a corrosion prediction model of each sensor according to the monitored environmental factors and the metal corrosion rate; the corrosion contribution rate acquisition module is used for acquiring the corrosion contribution rate corresponding to each environmental factor of each sensor deployment site according to the trained corrosion prediction model, the monitored environmental factors and the metal corrosion rate; And the drawing module is used for drawing a corrosion contribution rate map according to the environment category colors corresponding to the environment factors with the highest corrosion contribution rate at each location.

Description

Map drawing method and system for contribution of environmental factors to metal corrosion Technical Field The invention relates to the technical field of corrosion detection, in particular to a map drawing method and system for metal corrosion contribution by environmental factors. Background Metal corrosion refers to the act of metal destruction by the chemical action of the surrounding medium. The statistics of the global related research institutions show that the loss caused by corrosion accounts for 3.4% of the global economic total yield value, and the use environment of most metals is related to the atmospheric medium, so that the amount of the lost metals accounts for more than half of the total value. Therefore, the atmospheric corrosion is monitored in real time and predicted, and the method is a foundation for equipment material selection, service life evaluation and maintenance method formulation. The existing method for evaluating the atmospheric corrosiveness is a corrosion rate map. The map drawing method is to sample, rate and draw the limited points in the research area. This method has the following disadvantages: (1) Insufficient data range Due to the limitations of manpower, cost, etc., it is impossible to sample and measure all unknown areas, and only relevant corrosion grade data for a limited number of sampling points in the investigation region is often available. This means that the data obtained cannot cover all atmospheric corrosion features in the area, affecting the final analysis accuracy and practicality, causing errors in the knowledge of the monitoring personnel's assessment of the corrosion of critical components in the area. (2) Difficulty in data update Existing data acquisition methods are static and one-time sampling may not reflect time-varying dynamic corrosion changes within the region. For obvious fluctuation (such as weather change, human intervention and the like) of atmospheric environmental conditions, the prior art cannot update data in real time or periodically, and mislead the safety evaluation of monitoring personnel. (3) High cost and low efficiency The traditional corrosion map rating system needs a large amount of manpower and material resources to sample regularly, and cannot analyze data acquired in severe environment areas in real time, so that the cost for comprehensively acquiring accurate data is very high. (4) Poor practicality At present, the atmospheric corrosiveness evaluation is only output of technical values or text display, and cannot display the evaluation of atmospheric corrosiveness, and the continuous change process in the atmospheric environment corrosiveness time course in the region to be detected is observed in real time, so that the method is not beneficial to grasping the corrosion rule of the material in the atmospheric environment, and the practicability is poor. And only atmospheric corrosion registration is output, so that the influence of other environmental factors on corrosion behavior cannot be measured. Disclosure of Invention In order to solve the defects in the prior art, the invention provides a map drawing method and a map drawing system for contribution of environmental factors to metal corrosion. The technical scheme provided by the invention is as follows: a method of mapping a contribution of environmental factors to metal corrosion, the method comprising: Disposing a plurality of sensors in a region to be predicted, monitoring environmental factors and metal corrosion rates, and performing correspondence between the environmental factors and environmental category colors; establishing and training a corrosion prediction model of each sensor according to the monitored environmental factors and the metal corrosion rate; aiming at each sensor, according to the trained corrosion prediction model, the monitored environmental factors and the metal corrosion rate, obtaining corrosion contribution rates corresponding to the environmental factors of each sensor deployment site; and drawing a corrosion contribution rate map according to the environment category colors corresponding to the environment factors with the highest corrosion contribution rate at each location. Preferably, the building and training of the corrosion prediction model of each sensor based on the monitored environmental factors and the metal corrosion rate includes: integrating the monitored environmental factors and the metal corrosion rate for each sensor; the integrated environmental factors and the metal corrosion rate are taken as input, the corrosion contribution rate of each environmental factor is taken as output, and a random forest algorithm is adopted to establish a corrosion prediction model; and repeatedly testing, verifying, adjusting and optimizing the established corrosion prediction model. Preferably, said integrating said environmental factor and said metal corrosion rate monitored for each sensor comprises at least one or more of: A