Method and device for evaluating corrosion risk of steel structure building, electronic equipment and medium
By constructing a corrosion database and selecting a prediction model with a high R² value, the problem of lack of prediction methods in corrosion monitoring of steel structures was solved, enabling accurate assessment and efficient prediction of corrosion risks.
Patent Information
- Application Number
- CN202511054145.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-30
- Publication Date
- 2026-01-16
AI Technical Summary
Existing methods for monitoring corrosion in steel structures lack effective predictive means, cannot detect corrosion problems in a timely manner, rely on manual monitoring, are costly, and their accuracy depends on the experience of the monitoring personnel, making it difficult to support preventive maintenance of the structure.
By acquiring corrosion data and meteorological data of steel structure buildings, a corrosion database is constructed. After preprocessing, a prediction model for corrosion increment and corrosion rate is built. The optimal model is selected for corrosion risk assessment by comparing the R2 values of the models.
It enables accurate prediction of corrosion risks in steel structure buildings, reduces reliance on manual monitoring, improves assessment efficiency and accuracy, and reduces the subjectivity of human intervention.
Smart Images
Figure CN121350809A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of corrosion prediction, in particular to a steel structure building corrosion risk assessment method and device, electronic equipment and medium. BACKGROUND
[0002] In the long-term use of steel structure buildings, corrosion problems are always one of the key factors affecting the safety, durability and service life of the structure. Once the steel corrodes, not only will its carrying capacity be weakened, but also cracks, deformation and other problems may occur, thereby affecting the overall stability of the building, and even causing structural failure or collapse. Therefore, the prevention and monitoring of corrosion has become an important research topic in the field of steel structures.
[0003] Traditional steel structure corrosion monitoring and evaluation methods usually rely on manual inspection, experience judgment and limited instrument monitoring. These methods include visual inspection, ultrasonic monitoring, X-ray monitoring and other physical methods. These monitoring methods can detect some local corrosion phenomena, but they have obvious limitations. First, corrosion is often hidden, and internal corrosion may have developed when there is no obvious damage on the surface of the structure. Second, traditional monitoring methods have a long cycle and often cannot detect corrosion problems in a timely manner, which is lagging. In addition, manual monitoring requires a lot of manpower and time, and cannot efficiently cover all corrosion points of large or complex structures, which is costly and its accuracy depends on the experience level of the monitoring personnel. In addition, traditional methods mainly diagnose based on the phenomenon of corrosion that has occurred, and lack the ability to predict the development of corrosion. Since corrosion is a long-term and complex process affected by various environmental factors (such as humidity, temperature, wind, acid rain, salt fog, etc.), traditional monitoring methods often cannot provide effective prediction of future corrosion trends, making it difficult to provide strong support for preventive maintenance of the structure.
[0004] Therefore, there is an urgent need for a method that can effectively predict and evaluate the corrosion of steel structure buildings. SUMMARY
[0005] The technical problem to be solved by the present application is to solve the problem of the lack of effective prediction means in the existing monitoring method of steel structure building corrosion risk.
[0006] To solve the above technical problems, the present application provides a steel structure building corrosion risk assessment method, which comprises:
[0007] S10, acquiring corrosion data on the steel structure building, the corrosion data including data of corrosion accumulation amount and data of time point corresponding to the corrosion accumulation amount, the data of corrosion accumulation amount being acquired by arranging a corrosion sensor;
[0008] S20, collecting meteorological data of the region according to the time coordinate corresponding to the corrosion data;
[0009] S30, aligning the time granularity of the data of the corrosion accumulation amount and the data of the time point corresponding to the corrosion accumulation amount to obtain corrosion data corresponding to the time coordinate;
[0010] S40, constructing a corrosion database about the corrosion data corresponding to the meteorological data and the time coordinate;
[0011] S50, preprocessing the data in the corrosion database, the preprocessing method comprising smoothing the corrosion data using a filter to obtain smoothed data, and performing feature engineering processing on the smoothed data, the feature engineering processing comprising converting the corrosion accumulation amount data into corrosion increment data and corrosion rate data;
[0012] S60, constructing a corrosion prediction model about the corrosion increment and a corrosion prediction model about the corrosion rate according to the preprocessed database;
[0013] S70, selecting the corrosion prediction model of the steel structure corrosion risk by comparing the R 2 value of the corrosion prediction model about the corrosion increment data and the corrosion prediction model about the corrosion rate data, the R 2 value being higher, R being the explanatory of the prediction model;
[0014] S80, predicting and evaluating the corrosion risk of the steel structure building by the corrosion prediction model of the steel structure corrosion risk.
[0015] Further, the preprocessing method further comprises:
[0016] adopting a rejection strategy to reject record points with abnormally negative growth of corrosion rate;
[0017] adopting a mean interpolation method to process missing values of the corrosion data;
[0018] cleaning the data after the feature engineering processing to remove negative corrosion increment data and corrosion rate data;
[0019] performing K-means clustering on the corrosion increment data and the corrosion rate data to classify similar corrosion rate data or corrosion increment change patterns;
[0020] adopting a max-min value scaling algorithm to adjust all data ranges to 0-1 to unify data scales and remove the influence of different dimensions.
[0021] Further, the meteorological data comprises air temperature, precipitation, humidity, wind speed, salt fog, sulfur dioxide, and nitrogen oxides.
[0022] Further, the method of constructing the prediction model comprises:
[0023] Select the partial time series data from the beginning to 70%-90% as the training set, and select the remaining 10%-30% of the partial time series data as the test set.
[0024] Further, the filter is Savitzky-Golay, and the filtering process of the Savitzky-Golay filter is
[0025] P(x)=a0+a1x+a2x 2 +...+a k x k
[0026] Wherein, k is the order of the polynomial, x is the data point in the sliding window, the polynomial coefficients a0, a1,..., a k , are determined by minimizing the sum of squared fitting errors, and the data point at the center of the sliding window is replaced with the result of polynomial fitting, and the convolution form of the Savitzky-Golay filter is:
[0027]
[0028] Wherein, C j is the convolution coefficient of the filter, y i is the original data of the corrosion sensor, M is half of the window size, is the filtered data point.
[0029] According to another aspect of the present application, there is provided a device for evaluating the corrosion risk of a steel structure building, the device comprising:
[0030] A data acquisition module, configured to acquire corrosion data on the steel structure building, the corrosion data comprising data of corrosion accumulation amount and data of time points corresponding to the corrosion accumulation amount, the data of corrosion accumulation amount being acquired by arranging a corrosion sensor;
[0031] A meteorological data collection module, configured to collect meteorological data of the region where the corrosion data is located according to the time coordinates corresponding to the corrosion data;
[0032] A data corresponding module, configured to align the time granularities of the data of corrosion accumulation amount and the data of time points corresponding to the corrosion accumulation amount respectively to obtain corrosion data corresponding to the time coordinates;
[0033] A database construction module, configured to construct a corrosion database about the meteorological data and the corrosion data corresponding to the time coordinates;
[0034] A data preprocessing module is configured to preprocess data in the corrosion database, and the preprocessing method comprises the following steps: smoothing the corrosion data by using a filter to obtain smoothed data, and performing feature engineering processing on the smoothed data, wherein the feature engineering processing is converting the corrosion cumulative amount data into corrosion increment data and corrosion speed data.
[0035] A prediction model construction module is configured to construct a corrosion prediction model for corrosion increment and a corrosion prediction model for corrosion speed according to the preprocessed database.
[0036] A prediction model selection module is configured to select a prediction model for the corrosion risk of the steel structure by comparing the R 2 value of the corrosion prediction model for the corrosion increment data and the R 2 value of the corrosion prediction model for the corrosion speed data, wherein a prediction model with a higher R 2 value is selected, and R represents the interpretability of the prediction model.
[0037] A risk assessment module is configured to predict and assess the corrosion risk of the steel structure building by using the prediction model for the corrosion risk of the steel structure.
[0038] According to another aspect of the present application, an electronic device is provided, comprising at least one processor and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any one of the steel structure building corrosion risk assessment methods in the embodiments of the present application.
[0039] According to another aspect of the present application, a non-transitory computer readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to perform any one of the steel structure building corrosion risk assessment methods in the embodiments of the present application.
[0040] Compared with the prior art, the steel structure building corrosion risk assessment method in the embodiments of the present application has the following beneficial effects:
[0041] The present application constructs corrosion prediction models for corrosion increment and corrosion speed, compares the R 2 values (i.e., the square of the interpretability of the model) of the two models, and selects a prediction model with a higher R 2The model with a higher value is used as the final steel structure corrosion risk prediction model, and this method can more accurately reflect the corrosion condition of the steel structure building, thereby improving the prediction accuracy and realizing effective prediction of the steel structure risk. The present application not only considers the current corrosion data, but also combines meteorological data and performs time granularity alignment and feature engineering processing, so that the prediction model can more comprehensively consider the factors affecting corrosion, thereby enhancing the prediction accuracy. The present application realizes accurate prediction and evaluation of the corrosion risk of the steel structure building by constructing a prediction model, so that the corrosion risk evaluation of the steel structure building is free from relying on a large amount of manual monitoring and data recording, reduces the need for manual participation, reduces the subjectivity and uncertainty of manual evaluation, and improves the evaluation efficiency of the steel structure building risk. BRIEF DESCRIPTION OF DRAWINGS
[0042] Figure 1 is a flowchart of the steel structure building corrosion risk evaluation method provided by the embodiment of the present application;
[0043] Figure 2 is a database of corrosion data measured by each corrosion sensor provided by the embodiment of the present application;
[0044] Figure 3 is a comparison chart of the corrosion prediction model training result and the actual value provided by the embodiment of the present application;
[0045] Figure 4 is a comparison chart of the corrosion model prediction result and the real value provided by the embodiment of the present application;
[0046] Figure 5 is a prediction corrosion amount-time curve of each sensor and a corrosion risk division chart provided by the embodiment of the present application;
[0047] Figure 6 is a schematic diagram of the steel structure building corrosion risk evaluation device provided by the embodiment of the present application;
[0048] Figure 7 is a block diagram of an electronic device for implementing the embodiment of the present application.
[0049] In the figure, 10, data acquisition module; 20, meteorological data collection module; 30, data corresponding module; 40, database construction module; 50, data preprocessing module; 60, prediction model construction module; 70, prediction model selection module; 80, risk evaluation module; 600, electronic device; 601, calculation unit; 602, ROM; 603, RAM; 604, bus; 605, I / O interface; 606, input unit; 607, output unit; 608, storage unit; 609, communication unit. DETAILED DESCRIPTION
[0050] The exemplary embodiments of this application are described herein with reference to the accompanying drawings, which are meant to be exemplary in nature, and include various details intended to facilitate understanding of the application. Accordingly, one skilled in the art should realize that they can be various changes and modifications that can be made to the embodiments described and illustrated herein without departing from the scope of the present application. Also, for the purpose of clarity and the brevity, the description below omits the description of well-known functions and constructions.
[0051] As shown in the optional embodiment of the present application, the risk assessment method for predicting corrosion of steel structure buildings comprises: Figures 1-5
[0052] S10, obtaining corrosion data on the steel structure building, the corrosion data comprising data of corrosion accumulation amount and data of time point corresponding to the corrosion accumulation amount, the data of corrosion accumulation amount being obtained by arranging corrosion sensors;
[0053] S20, collecting meteorological data of the region according to the time coordinate corresponding to the corrosion data;
[0054] S30, aligning the time granularity of the data of corrosion accumulation amount and the data of time point corresponding to the corrosion accumulation amount respectively to obtain corrosion data corresponding to the time coordinate;
[0055] S40, constructing a corrosion database about the meteorological data and the corrosion data corresponding to the time coordinate;
[0056] S50, preprocessing the data in the corrosion database, the preprocessing method comprising smoothing the corrosion data by using a filter to obtain smoothed data, and performing feature engineering processing on the smoothed data, the feature engineering processing being converting the corrosion accumulation amount data into corrosion increment data and corrosion speed data;
[0057] S60, constructing a corrosion prediction model about corrosion increment and a corrosion prediction model about corrosion speed respectively according to the preprocessed database;
[0058] S70, selecting the corrosion prediction model of the steel structure corrosion risk by comparing the R 2 value of the corrosion prediction model about the corrosion increment data and the R 2 value of the corrosion prediction model about the corrosion speed data, the R
[0059] S80, predicting and assessing the corrosion risk of the steel structure building by using the corrosion prediction model of the steel structure corrosion risk.
[0060] Specifically, in S10, the corrosion sensor is a sensor that can monitor the corrosion condition of the surface of a metal material or equipment in real time. It monitors corrosion through electrochemical, physical, chemical, and other methods. Corrosion data refers to the data generated by the chemical or electrochemical corrosion reaction on the surface of the steel structure due to environmental factors (such as humidity, temperature, atmospheric pollutants, etc.) during the use of the steel structure building. These data reflect the degree, speed, and changes over time of the corrosion of the steel structure. Corrosion accumulation refers to the mass or volume of corrosion material accumulated on the steel structure over a certain period of time due to corrosion. It reflects the total amount of corrosion of the steel structure. The time point corresponding to the corrosion accumulation refers to the time marker when the corrosion accumulation is recorded, which is used to determine the time series and trend of the corrosion data.
[0061] Specifically, in S20, the time coordinate refers to the time period of the corrosion data record, and the meteorological data refers to the weather condition information of the region within a certain time period, including but not limited to temperature, humidity, precipitation, wind speed, wind direction, air pressure, solar radiation, etc. By collecting meteorological data matching the time coordinate of the corrosion data, the correlation between the corrosion process and meteorological factors can be more accurately analyzed, thereby improving the accuracy of corrosion risk assessment and prediction. For example, high temperature and high humidity environment may accelerate the corrosion process, while dry environment may slow down the corrosion.
[0062] Specifically, in S50, through the filter and feature engineering processing, the noise and outliers in the data can be removed, and the feature engineering converts the corrosion accumulation data into corrosion increment or corrosion speed.
[0063] Specifically, in S60 and S70, by constructing a corrosion prediction model, historical corrosion data can be used to predict future corrosion conditions. By comparing the explanatory power of the prediction model of the corrosion increment data and the prediction model of the corrosion speed data, the optimal model is selected as the prediction model of the steel structure corrosion risk. Because in the prediction model, different features have different contributions to the prediction results. By evaluating the importance of the features, it can be understood which features are more important in influencing the prediction model.
[0064] Specifically, the following embodiments are combined Figures 1-5 Further explanation:
[0065] (1) For a certain coastal thermal power plant steel structure building in the south, multiple sensors are arranged on each surface and at different heights of the building.
[0066] (2) Considering the special location of the power plant, far away from multiple weather monitoring stations. The simulation climate data of the Panggu climate model is used as the environmental data of the building. The accuracy of the Panggu climate model can reach 0.25 longitude / 0.25 latitude, which is closer to the specific corrosion environment compared to the data measured by the weather station more than 60 kilometers away and with huge geographical differences.
[0067] (3) Considering that the power plant emissions contain sulfur dioxide, nitrogen oxides and other substances closely related to corrosion, resulting in a huge difference between the surrounding environment and the air quality data of the monitoring station. The emission data is also added to the environmental data, which is more in line with the actual corrosion environment.
[0068] (4) Due to the insufficient measurement accuracy of the sensor, there is obvious high-frequency noise in the hourly corrosion reading. For the corrosion measurement value, the highest value in a day is selected to weaken the influence caused by the accuracy of the sensor itself. Then the corrosion amount data is cleaned and the missing values are supplemented using the mean method.
[0069] (5) To further remove high-frequency noise from the data, the Savitzky-Golay filter is used to filter the corrosion amount reading. The window length is set to 7 to remove high-frequency noise while preserving the overall trend and local characteristics.
[0070] (6) Feature engineering is performed on the data filtered in (5) to convert the corrosion cumulative amount data into corrosion increment or corrosion speed to reflect short-term time series changes. At the same time, the feature engineering data is cleaned again to remove negative corrosion increment and corrosion speed to comply with the assumption that "corrosion is positively correlated with time under the influence of the environment".
[0071] (7) Considering that the size of the building or the orientation and other factors cause different corrosion conditions, the corrosion increment or corrosion speed measured by each sensor is clustered using K-means clustering to classify similar corrosion speed or corrosion increment change patterns. It can be considered that the steel structure position of the corrosion sensor under each class has similar environmental-corrosion relationship.
[0072] (8) Normalize the data using the max-min scaling algorithm to adjust all data ranges to 0-1 to unify the data scale and remove the influence caused by different dimensions.
[0073] (9) Based on the processed feature data of (8), the retained feature time series is serialized, the first 80% of the time series data is taken as the training set, and the last 20% is taken as the test level to train the neural network model. Because the corrosion data depends on the occurred corrosion and the past environmental factors, and depends on the future environmental data, a long short-term memory neural network is used to capture the change mode. At the same time, considering that the relationship between corrosion and environment is complex, and long-term relationship and short-term relationship exist at the same time, a multi-layer long short-term memory neural network is designed to capture the complex time series change mode.
[0074] (10) The R of the corrosion rate and the corrosion increment model 2 The numerical value is selected to predict the corrosion amount by using the model with better change explanation. The R of the corrosion increment model 2 can reach 0.84, which shows that the model for predicting the corrosion increment of the group of sensors has good explanation. The data is significantly higher than the R of the predicted corrosion rate 2 The numerical value is 0.62. Accordingly, the predicted model of the corrosion increment is used to predict the future corrosion increment of the position arranged by the group of sensors.
[0075] In an optional embodiment of the present application, the pre-processing method further comprises:
[0076] The record points with abnormally negative growth of the corrosion rate are removed by using the rejection strategy;
[0077] The missing values of the corrosion data are processed by using the mean interpolation method;
[0078] The data after the feature engineering processing is cleaned to remove the negative corrosion increment data and the corrosion speed data;
[0079] The corrosion increment data and the corrosion speed data are K-means clustered to classify the similar corrosion speed data or corrosion increment change mode;
[0080] The maximum-minimum value scaling algorithm is used to adjust all data ranges to 0-1, so as to unify the data scale and remove the influence caused by different dimensions.
[0081] Among them, the elimination strategy is a data cleaning method for identifying and removing data points that do not meet expectations or are abnormal. By eliminating abnormal negative growth corrosion rate record points, abnormal values caused by measurement errors or data recording problems can be excluded. The mean interpolation method is a method for handling missing data, which uses the average value of existing data to fill in missing values. By mean interpolation, the missing data points can be reasonably estimated, thereby improving the stability and accuracy of the model. The average value of each feature (such as corrosion rate, corrosion increment) is calculated, and then the average value is used to replace the corresponding missing value. K-means clustering is an unsupervised learning algorithm that divides data points into K clusters, so that the data points within each cluster are as similar as possible. By performing K-means clustering on corrosion increment data and corrosion speed data, subsets with similar corrosion speed or corrosion increment change patterns can be identified, which helps better understand the distribution and characteristics of the data. The max-min value scaling algorithm is a data standardization method that scales data to a specified range (usually 0 to 1) to unify data scales and eliminate the influence of different dimensions on model training.
[0082] The preprocessing method in the present application significantly improves the quality of corrosion data, enhances the consistency of corrosion data, unifies the data scale and eliminates the dimensional influence, providing a guarantee for subsequent data analysis and construction of corrosion prediction model.
[0083] In an optional embodiment of the present application, the meteorological data includes temperature, precipitation, humidity, wind speed, salt fog, sulfur dioxide, and nitrogen oxides.
[0084] In an optional embodiment of the present application, the method for constructing the prediction model comprises:
[0085] Selecting a portion of time series data from the beginning to 70%-90% as the training set, and selecting the remaining 10%-30% of the time series data as the test set.
[0086] Among them, the training set should contain corrosion data from the beginning to 70%-90%. For example, if the total time series has 1000 data points, 70% is 700, and 90% is 900. Preferably, the first 80% of the time series data is used as the training set, and the last 20% is used as the test set to train the neural network model. Because corrosion data depends on both past environmental factors and future environmental data, a long short-term memory neural network is used to capture this change pattern. At the same time, given the complex relationship between corrosion and environment, both long-term and short-term relationships exist, so a multi-layer long short-term memory neural network is designed to capture complex time series change patterns.
[0087] In an optional embodiment of the present invention, the filter is a Savitzky-Golay filter, and the filtering process of the Savitzky-Golay filter is as follows:
[0088] P(x) = a0 + a1x + a2x 2 +...+a k x k
[0089] Where k is the order of the polynomial, and x is the data point within the sliding window, the polynomial coefficients a0, a1, ..., a are determined by minimizing the sum of squared fitting errors. k The data points at the center of the sliding window are replaced with the results of polynomial fitting. The convolutional form of the Savitzky-Golay filter is as follows:
[0090]
[0091] Among them, C j y represents the convolution coefficients of the filter. i This represents the raw data from the corrosion sensor, where M is half the window size. These are the filtered data points.
[0092] Specifically, the Savitzky-Golay filter is a digital filter that smooths data by performing a polynomial fit within a local window around the data points. This method effectively removes noise while preserving the high-frequency characteristics of the data, such as peaks and valleys. The Savitzky-Golay filter can smooth the raw data from corrosion sensors.
[0093] like Figure 6 As shown, according to another aspect of the present invention, an apparatus for assessing the corrosion risk of steel structure buildings is provided, the apparatus comprising:
[0094] Data acquisition module 10 is used to acquire corrosion data on steel structure buildings. The corrosion data includes data on corrosion accumulation and data on time points corresponding to the corrosion accumulation. The corrosion accumulation data is acquired by deploying corrosion sensors.
[0095] Meteorological data collection module 20, which is used to collect meteorological data of the area according to the time coordinates corresponding to the corrosion data;
[0096] The data correspondence module 30 is used to align the time granularity of the corrosion accumulation data and the time point data corresponding to the corrosion accumulation to obtain corrosion data corresponding to the time coordinate.
[0097] A database construction module 40, which constructs a corrosion database about the corrosion data corresponding to the meteorological data and time coordinates;
[0098] A data preprocessing module 50, which is used for preprocessing the data in the corrosion database, and the preprocessing method comprises smoothing the corrosion data by using a filter to obtain filtered smoothed data, and performing feature engineering processing on the smoothed data, the feature engineering processing being converting the corrosion cumulative amount data into corrosion increment data and corrosion speed data;
[0099] A prediction model construction module 60, which is used for constructing a corrosion prediction model about corrosion increment and a corrosion prediction model about corrosion speed respectively according to the preprocessed database;
[0100] A prediction model selection module 70, which is used for selecting the prediction model of the steel structure corrosion risk by comparing the R 2 values in the corrosion prediction model about the corrosion increment data and the corrosion prediction model about the corrosion speed data, and selecting the model with a higher R 2 value as the prediction model of the steel structure corrosion risk, R being the interpretability of the prediction model;
[0101] A risk assessment module 80, which is used for predicting and assessing the corrosion risk of the steel structure building by using the prediction model of the steel structure corrosion risk.
[0102] The present application can more accurately reflect the corrosion condition of the steel structure building by constructing the corrosion prediction models about the corrosion increment and the corrosion speed, comparing the R 2 values (i.e., the square of the interpretability of the model) of the two models, and selecting the model with a higher R 2 value as the final prediction model of the steel structure corrosion risk, thereby improving the prediction accuracy and realizing the effective prediction of the steel structure risk. The present application not only considers the current corrosion data, but also combines the meteorological data and performs time granularity alignment and feature engineering processing, so that the prediction model can more comprehensively consider the factors affecting corrosion, thereby enhancing the prediction accuracy. The present application realizes the accurate prediction and assessment of the corrosion risk of the steel structure building by constructing the prediction model, so that the corrosion risk assessment of the steel structure building is free from relying on a large amount of manual monitoring and data recording, reduces the need for manual participation, reduces the subjectivity and uncertainty of manual assessment, and improves the efficiency of the risk assessment of the steel structure building.
[0103] According to the embodiments of the present application, the present application also provides an electronic device, a readable storage medium and a computer program product.
[0104] Figure 7 A schematic block diagram of an example electronic device 600 that can be used to implement embodiments of the present application is shown. The electronic device 600 is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the inventiveness in the present document as described and / or claimed.
[0105] As shown in Figure 7 The electronic device 600 includes a computing unit 601 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. Various programs and data required for the operation of the device 600 can also be stored in the RAM 603. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.
[0106] Various components in the electronic device 600 are connected to the I / O interface 605, including an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, a speaker, etc.; the storage unit 608, such as a magnetic disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the electronic device 600 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.
[0107] The computing unit 601 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs various methods and processes described above, such as a risk assessment method for corrosion prediction of steel structure buildings. For example, in some embodiments, a risk assessment method for corrosion prediction of steel structure buildings can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded onto the RAM 603 and executed by the computing unit 601, one or more steps of a risk assessment method for corrosion prediction of steel structure buildings described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform a risk assessment method for corrosion prediction of steel structure buildings by any other suitable means, such as by means of firmware.
[0108] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.
[0109] Program code for carrying out methods of the present application can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces a means for implementing the functions / acts specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine and partially on a remote machine or entirely on a remote machine or server.
[0110] In the context of the present application, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0111] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.
[0112] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.
[0113] The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, a server of a distributed system, or a server combined with a blockchain.
[0114] It should be understood that the various forms of flow shown above can be re-ordered, added to, or have steps deleted, using the flow. For example, the steps described in the present application can be performed in parallel, in series, or in a different order, as long as the desired results of the technology disclosed in the present application can be achieved, which is not limited herein.
[0115] The above detailed description does not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the principles of the present application shall be included in the protection scope of the present application.
Claims
1. A method of assessing the corrosion risk of a steel construction building, characterized in that, The method comprises: S10, acquiring corrosion data on the steel structure building, the corrosion data comprising data of corrosion accumulation amount and data of time point corresponding to the corrosion accumulation amount, the data of corrosion accumulation amount being acquired by arranging a corrosion sensor; S20, collecting meteorological data of the region according to the time coordinates corresponding to the corrosion data; S30, aligning the time granularities of the data of corrosion accumulation amount and the data of time point corresponding to the corrosion accumulation amount to obtain corrosion data corresponding to the time coordinates; S40, constructing a corrosion database about the meteorological data and the corrosion data corresponding to the time coordinates; S50, preprocessing the data in the corrosion database, the preprocessing method comprising smoothing the corrosion data by using a filter to obtain smoothed data, and performing feature engineering processing on the smoothed data, the feature engineering processing being converting the corrosion accumulation amount data into corrosion increment data and corrosion speed data; S60, respectively constructing a corrosion prediction model about corrosion increment and a corrosion prediction model about corrosion speed according to the preprocessed database; S70, selecting the corrosion prediction model with a higher R value by comparing the R values of the corrosion prediction model based on the corrosion increment data and the corrosion prediction model based on the corrosion rate data 2 value of the corrosion prediction model based on the corrosion rate data, R is the explanatory power of the prediction model 2 value is higher, the prediction model of the steel structure corrosion risk, R is the explanatory power of the prediction model S80, predicting and evaluating the corrosion risk of the steel structure building by using the prediction model of the corrosion risk of the steel structure.
2. The evaluation method according to claim 1, characterized in that The preprocessing method further comprises: adopting an elimination strategy to eliminate record points with abnormally negative growth of corrosion rate; adopting a mean interpolation method to process missing values of the corrosion data; cleaning the data after the feature engineering processing to remove negative corrosion increment data and corrosion speed data; performing K-means clustering on the corrosion increment data and the corrosion speed data to classify them according to similar corrosion speed data or corrosion increment change patterns; adopting a max-min value scaling algorithm to adjust the range of all data to 0-1 to unify the data scale and remove the influence caused by different dimensions.
3. The evaluation method according to claim 1, characterized in that The meteorological data comprises air temperature, precipitation, humidity, wind speed, salt fog, sulfur dioxide and nitrogen oxides.
4. The evaluation method according to claim 2, characterized in that The method for constructing the prediction model comprises selecting part of time series data from the beginning to 70%-90% as a training set, and selecting the remaining 10%-30% of the time series data as a test set.
5. The evaluation method according to claim 2, characterized in that The filter is a Savitzky-Golay filter, and the filtering process of the Savitzky-Golay filter is P(x) = a0+ a1x + a2x2+... + anxn 2 +...+a k x k where, where k is the order of the polynomial, x is the data point within the sliding window, and the polynomial coefficients a0, a1,..., ak are determined by minimizing the sum of squared fitting errors k and replacing the data point at the center of the sliding window with the result of the polynomial fit, the convolution form of the Savitzky-Golay filter is: where C j is the convolution coefficient of the filter, y i is the original data of the corrosion sensor, M is half of the window size, is the filtered data point.
6. An apparatus for assessing corrosion risk of a steel structure building, characterized by, The device comprises: a data acquisition module, which is used to acquire corrosion data on the steel structure building, the corrosion data comprising data of corrosion accumulation amount and data of time point corresponding to the corrosion accumulation amount, the data of corrosion accumulation amount being acquired by arranging a corrosion sensor; a meteorological data collection module, which is used to collect meteorological data of the region according to the time coordinates corresponding to the corrosion data; a data correspondence module, which is used to align the time granularities of the data of corrosion accumulation amount and the data of time point corresponding to the corrosion accumulation amount to obtain corrosion data corresponding to the time coordinates; a database construction module, which is used to construct a corrosion database about the meteorological data and the corrosion data corresponding to the time coordinates; A data preprocessing module, configured to preprocess data in the corrosion database, the preprocessing method comprising smoothing the corrosion data using a filter to obtain smoothed data, and performing feature engineering on the smoothed data, the feature engineering comprising converting the corrosion cumulative amount data into corrosion increment data and corrosion speed data; A prediction model construction module, configured to construct a corrosion prediction model for corrosion increment and a corrosion prediction model for corrosion speed respectively according to the preprocessed database; a prediction model selection module for selecting a prediction model for a steel structure corrosion risk by comparing the size of R values of a corrosion prediction model for corrosion increment data and a corrosion prediction model for corrosion rate data 2 R values of the prediction model for the corrosion increment data are higher, R is an explanatory of the prediction model 2 values of the prediction model for the corrosion increment data are higher, R is an explanatory of the prediction model A risk assessment module, configured to predict and assess the corrosion risk of the steel structure building through the prediction model for the corrosion risk of the steel structure.
7. An electronic device, comprising: Comprise: At least one processor, and a memory connected in communication with the at least one processor; Wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method of any one of claims 1-6.
8. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to execute the method of any one of claims 1-6.