Metal corrosion prediction method and device in material storage and transportation environment

Abstract

The invention discloses a metal corrosion prediction method and device in a material storage and transportation environment. The method comprises the steps of obtaining a metal corrosion historical data set of the material storage and transportation environment; the metal corrosion historical data set comprises a storage and transportation environment monitoring data set, a monitoring sample physical data set and a corrosion amount sequence data set; the corrosion amount sequence data set is a corrosion amount sequence obtained by measuring a monitoring sample at a plurality of historical moments; the storage and transportation environment monitoring data set comprises a set of environment measurement sequences obtained by measuring the storage and transportation environment at a plurality of historical moments; preprocessing the metal corrosion historical data set to obtain a preprocessed metal corrosion historical data set; and performing corrosion prediction processing on the preprocessed metal corrosion historical data set to obtain a corrosion prediction value of the material storage and transportation environment.

Description

Technical Field

[0001] The invention relates to the field of metal corrosion prediction, and in particular to a method and device for predicting metal corrosion in a material storage and transportation environment. Background Art

[0002] Rust refers to the interaction between metals and oxygen, moisture and other impurities when exposed to the atmosphere, which causes the metal surface to discolor or corrode through chemical or electrochemical reactions, forming visible corrosion products such as metal oxides and hydroxides. These corrosion products are collectively called rust, commonly known as rust or rust. Whether in the storage stage of equipment or in the service process, the prevention and management of metal corrosion is crucial. In this management process, how to accurately predict the occurrence of metal corrosion has become a core issue.

[0003] In the material storage and transportation environment, the constant changes in environmental conditions make the early prediction of metal corrosion more critical. How to achieve accurate prediction and timely warning of metal corrosion in the material storage and transportation environment has become a challenge that needs to be solved urgently. Summary of the invention

[0004] The present invention mainly solves the problem of how to accurately warn metal corrosion prediction in a material storage and transportation environment. The present invention discloses a metal corrosion prediction method and device in a material storage and transportation environment.

[0005] In a first aspect of the embodiment of the present application, a method for predicting metal corrosion in a material storage and transportation environment is disclosed, comprising:

[0006] S1, obtaining a metal corrosion history dataset of a material storage and transportation environment; the metal corrosion history dataset includes a storage and transportation environment monitoring dataset, a monitoring sample physical dataset, and a corrosion amount sequence dataset;

[0007] The monitoring sample physical data set includes a set of sample measurement sequences obtained by measuring the monitoring samples at several historical moments; the sample measurement sequence includes a resistance value sequence, a mass value sequence and a thickness value sequence;

[0008] The corrosion amount series data set is a corrosion amount series obtained by measuring the monitoring samples at several historical moments;

[0009] The storage and transportation environment monitoring data set includes a set of environmental measurement sequences obtained by measuring the storage and transportation environment at several historical moments; the environmental measurement sequence includes a temperature sequence, a humidity sequence, an oxygen content sequence, and a harmful gas content sequence; the environmental measurement sequence, the sample measurement sequence, and the corrosion amount sequence establish a one-to-one correspondence according to the collection time;

[0010] S2, preprocessing the metal corrosion history data set to obtain a preprocessed metal corrosion history data set;

[0011] S3, performing corrosion prediction processing on the pre-processed metal corrosion history data set to obtain a corrosion prediction value for the material storage and transportation environment.

[0012] The preprocessing of the metal corrosion history data set to obtain the preprocessed metal corrosion history data set includes:

[0013] S21, performing time registration processing on the metal corrosion history dataset to obtain a first dataset;

[0014] S22, performing data cleaning processing on the first data set to obtain a second data set;

[0015] S23, performing pattern matching processing on the second data set to obtain a preprocessed metal corrosion history data set.

[0016] The performing pattern matching processing on the second data set to obtain a preprocessed metal corrosion history data set includes:

[0017] S231, for each type of data set in the second data set, using each type of data sequence in the data set as a row vector, constructing an approximation matrix; using data acquisition information of each type of data sequence in the data set as a row vector, constructing an acquisition matrix; the row vector of the acquisition matrix is ​​the data acquisition information of a type of data sequence; using the approximation matrix as a dependent variable and the acquisition matrix as an independent variable, constructing a pattern matching model;

[0018] S232, solving the pattern matching model to obtain a matching model for each type of data set; the matching model is expressed as f(x)=xA, where x is an acquisition matrix of the input of the matching model, and matrix A is obtained by solving the pattern matching model;

[0019] S233, using the matching model of each type of data set, calculating and processing the data collection information of the data sequence of the corresponding type of data set to obtain a solution matrix;

[0020] S234, subtracting each data in the solution matrix from the data in the approximation matrix corresponding to the same row and column as the data in the solution matrix, and calculating the absolute value to obtain a difference value of the data;

[0021] S235, deleting the data whose difference value is greater than a set difference threshold from the second data set;

[0022] S236, executing S231 to S235 for all types of data sets in the second data set to obtain a preprocessed metal corrosion history data set.

[0023] The performing corrosion prediction processing on the pre-processed metal corrosion history data set to obtain the corrosion prediction value of the material storage and transportation environment includes:

[0024] Performing standard prediction processing on the storage and transportation environment monitoring data set and the monitoring sample physical data set to obtain a first prediction value sequence;

[0025] Feedback prediction processing is performed on the first prediction value sequence, the corrosion amount sequence data set and the monitoring sample physical data set to obtain the corrosion prediction value of the material storage and transportation environment.

[0026] The step of performing standard prediction processing on the storage and transportation environment monitoring data set and the monitoring sample physical data set to obtain a first prediction value sequence includes:

[0027] The storage and transportation environment monitoring data set is represented as an environment data matrix; the row vector of the environment data matrix is ​​an environment measurement sequence;

[0028] The monitoring sample physical data set is represented as a physical data matrix; the row vector of the physical data matrix is ​​a sample measurement sequence;

[0029] Obtain a standard environmental data set and a standard monitoring sample physical data set; the standard environmental data set includes standard values ​​for temperature, humidity, oxygen content, and harmful gas content; the standard monitoring sample physical data set includes standard values ​​for resistance, mass, and thickness;

[0030] Subtract the corresponding standard value from each row vector of the environmental data matrix and take the absolute value to obtain an environmental difference matrix;

[0031] Subtract the corresponding standard value from each row vector of the physical data matrix and take the absolute value to obtain a physical difference matrix;

[0032] The environmental difference matrix and the physical difference matrix are processed by standard prediction calculation to obtain a first prediction value sequence.

[0033] The expression for the standard prediction calculation process is:

[0034]

[0035] l i =L2(R i )log|S i |,

[0036] Among them, f ijis the element in the i-th row and j-th column of the environmental difference matrix, p ij is the element of the ith row and jth column of the physical difference matrix, m and n are the row and column dimensions of the environmental difference matrix, respectively, and f j and p j are the jth terms of the first characteristic sequence and the second characteristic sequence respectively, L2() represents the second-order Legendre polynomial, S i and R i denote the i-th item of the third characteristic sequence and the fourth characteristic sequence respectively, l i Represents the i-th item of the first predicted value sequence.

[0037] The feedback prediction processing is performed on the first prediction value sequence, the corrosion amount sequence data set and the monitoring sample physical data set to obtain the corrosion prediction value of the material storage and transportation environment, including:

[0038] The corrosion amount sequence of the corrosion amount sequence data set is represented as E; the jth element of the corrosion amount sequence is represented as E j ;

[0039] The monitoring sample physical data set is represented as a physical data matrix; the row vector of the physical data matrix is ​​a sample measurement sequence;

[0040] Acquire a standard monitoring sample physical data set; the standard monitoring sample physical data set includes a resistance standard value, a mass standard value, and a thickness standard value;

[0041] Subtract the corresponding standard value from each row vector of the physical data matrix and take the absolute value to obtain a physical difference matrix;

[0042] Performing column optimization processing on the physical difference matrix to obtain an optimal solution vector;

[0043] The physical difference matrix, the optimal solution vector and the corrosion amount sequence are used to perform prediction calculation and processing to obtain the first feedback sequence;

[0044] The expression of the prediction calculation process is:

[0045]

[0046] Among them, s i is the i-th element of the first feedback sequence, ω j is the preset j-th weight value, is the jth element of the optimal solution vector;

[0047] The first prediction value sequence and the first feedback sequence are subjected to fusion calculation processing to obtain a corrosion prediction value of the material storage and transportation environment;

[0048] The expression of the fusion calculation process is:

[0049]

[0050] Among them, g is the corrosion prediction value of the material storage and transportation environment, μ is the mean of the first prediction value sequence, ν is the variance of the first prediction value sequence, τ is the median value of the first feedback sequence, φ is the variance of the first feedback sequence, and ρ is the mean of the first feedback sequence.

[0051] In a second aspect of an embodiment of the present invention, a device for predicting metal corrosion in a material storage and transportation environment is disclosed, the device comprising:

[0052] A memory storing executable program code;

[0053] a processor coupled to the memory;

[0054] The processor calls the executable program code stored in the memory to execute the metal corrosion prediction method in the material storage and transportation environment.

[0055] According to a third aspect of an embodiment of the present invention, a computer storable medium is disclosed, wherein the computer storable medium stores computer instructions, and when the computer instructions are called by a computer, they are used to execute the metal corrosion prediction method in the material storage and transportation environment.

[0056] According to a fourth aspect of an embodiment of the present invention, an information data processing terminal is disclosed, and the information data processing terminal is used to implement the metal corrosion prediction method in the material storage and transportation environment.

[0057] The beneficial effects of the present invention are:

[0058] The present invention realizes accurate early warning of metal corrosion prediction in a material storage and transportation environment, thereby improving the success rate of metal corrosion prevention in material storage and transportation.

[0059] The present invention collects storage and transportation environment monitoring data sets, monitoring sample physical data sets and corrosion amount sequence data sets, and retains data with the same data pattern by performing pattern matching processing on the collected data, thereby ensuring the accuracy of subsequent metal corrosion prediction.

[0060] In the process of metal corrosion prediction, the present invention establishes a distribution prediction model to ensure the stability and convergence of the prediction process. First, standard prediction processing is performed on the storage and transportation environment monitoring data set and the monitoring sample physical data set to obtain a first prediction value sequence; feedback prediction processing is performed on the first prediction value sequence, the corrosion amount sequence data set and the monitoring sample physical data set to obtain the corrosion prediction value of the material storage and transportation environment.

[0061] During the standard prediction processing, a standard prediction calculation processing model is specially established. By performing difference feature extraction processing on the environmental difference matrix and the physical difference matrix, the first prediction value sequence is obtained. During the feedback prediction processing, the first prediction value sequence and the first feedback sequence are fused to achieve accurate prediction of the corrosion prediction value of the material storage and transportation environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a flow chart for implementing the method of the present invention. DETAILED DESCRIPTION

[0063] In order to better understand the content of the present invention, an embodiment is given here.

[0064] Figure 1 It is a flow chart for implementing the method of the present invention.

[0065] In a first aspect of the embodiment of the present application, a method for predicting metal corrosion in a material storage and transportation environment is disclosed, comprising:

[0066] S1, obtaining a metal corrosion history dataset of a material storage and transportation environment; the metal corrosion history dataset includes a storage and transportation environment monitoring dataset, a monitoring sample physical dataset, and a corrosion amount sequence dataset;

[0067] The monitoring sample physical data set includes a collection of sample measurement sequences obtained by measuring the monitoring samples at several historical moments; the sample measurement sequence includes a resistance value sequence, a mass value sequence and a thickness value sequence, specifically, the measurement results of the resistance value, mass value and thickness value of the flat resistance probe of the corrosion sensor;

[0068] The corrosion amount sequence data set is a corrosion amount sequence obtained by measuring the monitoring samples at several historical moments, and the monitoring samples are probes of the corrosion sensors;

[0069] The storage and transportation environment monitoring data set includes a set of environmental measurement sequences obtained by measuring the storage and transportation environment at several historical moments; the environmental measurement sequence includes a temperature sequence, a humidity sequence, an oxygen content sequence, and a harmful gas content sequence; the environmental measurement sequence, the sample measurement sequence, and the corrosion amount sequence establish a one-to-one correspondence according to the collection time;

[0070] S2, preprocessing the metal corrosion history data set to obtain a preprocessed metal corrosion history data set;

[0071] S3, performing corrosion prediction processing on the pre-processed metal corrosion history data set to obtain a corrosion prediction value for the material storage and transportation environment.

[0072] The preprocessing of the metal corrosion history data set to obtain the preprocessed metal corrosion history data set includes:

[0073] S21, performing time registration processing on the metal corrosion history dataset to obtain a first dataset;

[0074] S22, performing data cleaning processing on the first data set to obtain a second data set;

[0075] S23, performing pattern matching processing on the second data set to obtain a preprocessed metal corrosion history data set;

[0076] The performing pattern matching processing on the second data set to obtain a preprocessed metal corrosion history data set includes:

[0077] S231, for each type of data set in the second data set, using each type of data sequence in the data set as a row vector, constructing an approximation matrix; using data acquisition information of each type of data sequence in the data set as a row vector, constructing an acquisition matrix; the row vector of the acquisition matrix is ​​the data acquisition information of a type of data sequence; using the approximation matrix as a dependent variable and the acquisition matrix as an independent variable, constructing a pattern matching model;

[0078] S232, solving the pattern matching model to obtain a matching model for each type of data set; the matching model is expressed as f(x)=xA, where x is an acquisition matrix of the input of the matching model, and matrix A is obtained by solving the pattern matching model;

[0079] S233, using the matching model of each type of data set, calculating and processing the data collection information of the data sequence of the corresponding type of data set to obtain a solution matrix;

[0080] S234, subtracting each data in the solution matrix from the data in the approximation matrix corresponding to the same row and column as the data in the solution matrix, and calculating the absolute value to obtain a difference value of the data;

[0081] S235, deleting the data whose difference value is greater than a set difference threshold from the second data set;

[0082] S236, executing S231 to S235 for all types of data sets in the second data set to obtain a preprocessed metal corrosion history data set.

[0083] The performing corrosion prediction processing on the pre-processed metal corrosion history data set to obtain the corrosion prediction value of the material storage and transportation environment includes:

[0084] Performing standard prediction processing on the storage and transportation environment monitoring data set and the monitoring sample physical data set to obtain a first prediction value sequence;

[0085] Feedback prediction processing is performed on the first prediction value sequence, the corrosion amount sequence data set and the monitoring sample physical data set to obtain the corrosion prediction value of the material storage and transportation environment.

[0086] The step of performing standard prediction processing on the storage and transportation environment monitoring data set and the monitoring sample physical data set to obtain a first prediction value sequence includes:

[0087] The storage and transportation environment monitoring data set is represented as an environment data matrix; the row vector of the environment data matrix includes a temperature sequence, a humidity sequence, an oxygen content sequence, and a harmful gas content sequence;

[0088] The monitoring sample physical data set is represented as a physical data matrix; the row vector of the physical data matrix includes a resistance value sequence, a mass value sequence and a thickness value sequence;

[0089] Obtain a standard environmental data set and a standard monitoring sample physical data set; the standard environmental data set includes standard values ​​for temperature, humidity, oxygen content, and harmful gas content; the standard monitoring sample physical data set includes standard values ​​for resistance, mass, and thickness;

[0090] Subtract the corresponding standard value from each row vector of the environmental data matrix and take the absolute value to obtain an environmental difference matrix;

[0091] Subtract the corresponding standard value from each row vector of the physical data matrix and take the absolute value to obtain a physical difference matrix;

[0092] Performing standard prediction calculation processing on the environmental difference matrix and the physical difference matrix to obtain a first prediction value sequence;

[0093] The expression for the standard prediction calculation process is:

[0094]

[0095] l i =L2(R i )log|S i |,

[0096] Among them, f ij is the element in the i-th row and j-th column of the environmental difference matrix, p ij is the element of the ith row and jth column of the physical difference matrix, m and n are the row and column dimensions of the environmental difference matrix, respectively, and f j and p jare the jth terms of the first characteristic sequence and the second characteristic sequence respectively, L2() represents the second-order Legendre polynomial, S i and R i denote the i-th item of the third characteristic sequence and the fourth characteristic sequence respectively, l i Represents the i-th item of the first predicted value sequence.

[0097] The feedback prediction processing is performed on the first prediction value sequence, the corrosion amount sequence data set and the monitoring sample physical data set to obtain the corrosion prediction value of the material storage and transportation environment, including:

[0098] The corrosion amount sequence of the corrosion amount sequence data set is represented as E; the jth element of the corrosion amount sequence is represented as E j ;

[0099] The monitoring sample physical data set is represented as a physical data matrix; the row vector of the physical data matrix includes a resistance value sequence, a mass value sequence and a thickness value sequence;

[0100] Acquire a standard monitoring sample physical data set; the standard monitoring sample physical data set includes a resistance standard value, a mass standard value, and a thickness standard value;

[0101] Subtract the corresponding standard value from each row vector of the physical data matrix and take the absolute value to obtain a physical difference matrix;

[0102] Performing column optimization processing on the physical difference matrix to obtain an optimal solution vector;

[0103] The physical difference matrix, the optimal solution vector and the corrosion amount sequence are used to perform prediction calculation and processing to obtain the first feedback sequence;

[0104] The expression of the prediction calculation process is:

[0105]

[0106] Among them, s i is the i-th element of the first feedback sequence, ω j is the preset j-th weight value, is the jth element of the optimal solution vector;

[0107] The first prediction value sequence and the first feedback sequence are fused and calculated to obtain a corrosion prediction value of the material storage and transportation environment;

[0108] The expression of the fusion calculation process is:

[0109]

[0110] Among them, g is the corrosion prediction value of the material storage and transportation environment, μ is the mean of the first prediction value sequence, ν is the variance of the first prediction value sequence, τ is the median value of the first feedback sequence, φ is the variance of the first feedback sequence, and ρ is the mean of the first feedback sequence.

[0111] ω j The value of can be set to 1 / 2n;

[0112] The column optimal processing is to take out the largest number in each column to form the optimal solution vector z + ;

[0113] The performing pattern matching processing on the second data set to obtain a preprocessed metal corrosion history data set includes:

[0114] For each type of data attribute of the second data set, data collection information of the data is used as a known independent variable, data values ​​of the data are used as a known dependent variable, and a curve to be approximated is constructed using the known independent variables and the known dependent variables;

[0115] Performing curve fitting on the curve to be approximated by using a function approximation method to obtain the best consistent approximation polynomial f(Ix) of the class data attribute;

[0116] Using the best consistent approximation polynomial f(Ix), the known independent variable is calculated to obtain an approximate dependent variable;

[0117] Determine whether the absolute value of the difference between the approximate dependent variable and the corresponding known dependent variable is greater than a set first regression discrimination threshold; if it is greater than the first regression discrimination threshold, delete the data from the second data set; if it is less than or equal to the first regression discrimination threshold, do not process the data;

[0118] Performing fusion processing on all the data after the execution pattern matching of the second data set to obtain a preprocessed metal corrosion history data set;

[0119] The curve fitting of the curve to be approximated by the function approximation method may be performed by using the best consistent linear approximation method. The best consistent approximation polynomial f(Ix) is expressed as:

[0120] f(Ix)=α P1 (Ix) P1 +α P1-1 (Ix) P1-1 +…+α2(Ix) 2 +α1(Ix)+α0,

[0121] Where P1 is the order of the best consistent approximation polynomial f(Ix), α0, α1, α2, …, αP1 are the coefficients of the optimal consistent approximation polynomial f(Ix);

[0122] The difference threshold is expressed as 0.6.

[0123] The pattern matching model is expressed as follows:

[0124] min|vA-R|,

[0125] subject to AA T =I A ,

[0126] Among them, I A represents the identity matrix with the row dimension of matrix A as the dimension, matrix A represents the matrix to be solved, v represents the acquisition matrix, and R represents the approximation matrix;

[0127] The algorithm for solving the pattern matching model may adopt a genetic algorithm and a particle filter algorithm;

[0128] The corrosion amount sequence is obtained by measuring the corrosion sensor;

[0129] The harmful gases include sulfur dioxide, hydrogen sulfide and chlorides;

[0130] The storage and transportation environment monitoring data set and the monitoring sample physical data set obtained can be collected by sensors. The storage and transportation environment monitoring data set can be collected by environmental sensors.

[0131] The data collection information is a data sequence number value or a data collection time value or a frequency value corresponding to the data;

[0132] The time registration process is to unify different types of data to the same time reference; the time registration process can adopt the internal push / extrapolation method, the Lagrange three-point interpolation method, etc.;

[0133] The data cleaning process includes filling missing values, smoothing noise data, and smoothing or deleting outliers; the smoothed noise data is obtained by first determining the noise data, and then smoothing the noise data according to the data before and after the noise data; the noise data is a value whose value is less than the detection sensitivity of the sensor of the observed data, or greater than the measurement upper limit of the sensor of the observed data. The outlier point can be determined by the Kalman filter method. The filling value of the missing value can be determined by averaging the measured values ​​within a certain sampling interval before and after the missing value.

[0134] The material storage and transportation environment is the environment in which the materials are transported in a closed packaging state.

[0135] The corrosion prediction value of the material storage and transportation environment is used to characterize the possibility of metal corrosion in the material storage and transportation environment at the prediction time. The larger the value, the higher the probability of corrosion occurring.

[0136] In a second aspect of an embodiment of the present invention, a device for predicting metal corrosion in a material storage and transportation environment is disclosed, the device comprising:

[0137] A memory storing executable program code;

[0138] a processor coupled to the memory;

[0139] The processor calls the executable program code stored in the memory to execute the metal corrosion prediction method in the material storage and transportation environment.

[0140] According to a third aspect of an embodiment of the present invention, a computer storable medium is disclosed, wherein the computer storable medium stores computer instructions, and when the computer instructions are called by a computer, they are used to execute the metal corrosion prediction method in the material storage and transportation environment.

[0141] According to a fourth aspect of an embodiment of the present invention, an information data processing terminal is disclosed, and the information data processing terminal is used to implement the metal corrosion prediction method in the material storage and transportation environment.

[0142] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for predicting metal corrosion in a material storage and transportation environment, characterized in that: include: S1, obtaining a metal corrosion history dataset of a material storage and transportation environment; the metal corrosion history dataset includes a storage and transportation environment monitoring dataset, a monitoring sample physical dataset, and a corrosion amount sequence dataset; The monitoring sample physical data set includes a set of sample measurement sequences obtained by measuring the monitoring samples at several historical moments; the sample measurement sequence includes a resistance value sequence, a mass value sequence and a thickness value sequence; The corrosion amount series data set is a corrosion amount series obtained by measuring the monitoring samples at several historical moments; The storage and transportation environment monitoring data set includes a set of environmental measurement sequences obtained by measuring the storage and transportation environment at several historical moments; the environmental measurement sequence includes a temperature sequence, a humidity sequence, an oxygen content sequence, and a harmful gas content sequence; the environmental measurement sequence, the sample measurement sequence, and the corrosion amount sequence establish a one-to-one correspondence according to the collection time; S2, preprocessing the metal corrosion history data set to obtain a preprocessed metal corrosion history data set; S3, performing corrosion prediction processing on the pre-processed metal corrosion history data set to obtain a corrosion prediction value for the material storage and transportation environment.

2. The metal corrosion prediction method in a material storage and transportation environment according to claim 1, characterized in that: The preprocessing of the metal corrosion history data set to obtain the preprocessed metal corrosion history data set includes: S21, performing time registration processing on the metal corrosion history dataset to obtain a first dataset; S22, performing data cleaning processing on the first data set to obtain a second data set; S23, performing pattern matching processing on the second data set to obtain a preprocessed metal corrosion history data set.

3. The metal corrosion prediction method in a material storage and transportation environment according to claim 2, characterized in that: The performing pattern matching processing on the second data set to obtain a preprocessed metal corrosion history data set includes: S231, for each type of data set in the second data set, using each type of data sequence in the data set as a row vector, constructing an approximation matrix; using data acquisition information of each type of data sequence in the data set as a row vector, constructing an acquisition matrix; the row vector of the acquisition matrix is ​​the data acquisition information of a type of data sequence; using the approximation matrix as a dependent variable and the acquisition matrix as an independent variable, constructing a pattern matching model; S232, solving the pattern matching model to obtain a matching model for each type of data set; the matching model is expressed as f(x)=xA, where x is an acquisition matrix of the input of the matching model, and matrix A is obtained by solving the pattern matching model; S233, using the matching model of each type of data set, calculating and processing the data collection information of the data sequence of the corresponding type of data set to obtain a solution matrix; S234, subtracting each data in the solution matrix from the data in the approximation matrix corresponding to the same row and column as the data in the solution matrix, and calculating the absolute value to obtain a difference value of the data; S235, deleting the data whose difference value is greater than a set difference threshold from the second data set; S236, executing S231 to S235 for all types of data sets in the second data set to obtain a preprocessed metal corrosion history data set.

4. The metal corrosion prediction method in a material storage and transportation environment according to claim 2, characterized in that: The performing corrosion prediction processing on the pre-processed metal corrosion history data set to obtain the corrosion prediction value of the material storage and transportation environment includes: Performing standard prediction processing on the storage and transportation environment monitoring data set and the monitoring sample physical data set to obtain a first prediction value sequence; Feedback prediction processing is performed on the first prediction value sequence, the corrosion amount sequence data set and the monitoring sample physical data set to obtain the corrosion prediction value of the material storage and transportation environment.

5. The metal corrosion prediction method in a material storage and transportation environment according to claim 4, characterized in that: The step of performing standard prediction processing on the storage and transportation environment monitoring data set and the monitoring sample physical data set to obtain a first prediction value sequence includes: The storage and transportation environment monitoring data set is represented as an environment data matrix; the row vector of the environment data matrix is ​​an environment measurement sequence; The monitoring sample physical data set is represented as a physical data matrix; the row vector of the physical data matrix is ​​a sample measurement sequence; Obtain a standard environmental data set and a standard monitoring sample physical data set; the standard environmental data set includes standard values ​​for temperature, humidity, oxygen content, and harmful gas content; the standard monitoring sample physical data set includes standard values ​​for resistance, mass, and thickness; Subtract the corresponding standard value from each row vector of the environmental data matrix and take the absolute value to obtain an environmental difference matrix; Subtract the corresponding standard value from each row vector of the physical data matrix and take the absolute value to obtain a physical difference matrix; The environmental difference matrix and the physical difference matrix are processed by standard prediction calculation to obtain a first prediction value sequence.

6. The metal corrosion prediction method in a material storage and transportation environment according to claim 5, characterized in that: The expression for the standard prediction calculation process is: l i =L2(R i )log|S i |, Among them, f ij is the element in the i-th row and j-th column of the environmental difference matrix, p ij is the element of the ith row and jth column of the physical difference matrix, m and n are the row and column dimensions of the environmental difference matrix, respectively, and f j and p j are the jth terms of the first characteristic sequence and the second characteristic sequence respectively, L2() represents the second-order Legendre polynomial, S i and R i denote the i-th item of the third characteristic sequence and the fourth characteristic sequence respectively, l i Represents the i-th item of the first predicted value sequence.

7. The metal corrosion prediction method in a material storage and transportation environment according to claim 6, characterized in that: The feedback prediction processing is performed on the first prediction value sequence, the corrosion amount sequence data set and the monitoring sample physical data set to obtain the corrosion prediction value of the material storage and transportation environment, including: The corrosion amount sequence of the corrosion amount sequence data set is represented as E; the jth element of the corrosion amount sequence is represented as E j ; The monitoring sample physical data set is represented as a physical data matrix; the row vector of the physical data matrix is ​​a sample measurement sequence; Acquire a standard monitoring sample physical data set; the standard monitoring sample physical data set includes a resistance standard value, a mass standard value, and a thickness standard value; Subtract the corresponding standard value from each row vector of the physical data matrix and take the absolute value to obtain a physical difference matrix; Performing column optimization processing on the physical difference matrix to obtain an optimal solution vector; The physical difference matrix, the optimal solution vector and the corrosion amount sequence are used to perform prediction calculation and processing to obtain the first feedback sequence; The expression of the prediction calculation process is: Among them, s i is the i-th element of the first feedback sequence, ω j is the preset j-th weight value, is the jth element of the optimal solution vector; The first prediction value sequence and the first feedback sequence are fused and calculated to obtain a corrosion prediction value of the material storage and transportation environment; The expression of the fusion calculation process is: Among them, g is the corrosion prediction value of the material storage and transportation environment, μ is the mean of the first prediction value sequence, ν is the variance of the first prediction value sequence, τ is the median value of the first feedback sequence, φ is the variance of the first feedback sequence, and ρ is the mean of the first feedback sequence.

8. A metal corrosion prediction device in a material storage and transportation environment, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the metal corrosion prediction method in the material storage and transportation environment according to any one of claims 1 to 7.

9. A computer storable medium, characterized in that: The computer storable medium stores computer instructions, and when the computer instructions are called by a computer, they are used to execute the metal corrosion prediction method in a material storage and transportation environment as described in any one of claims 1 to 7.

10. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the metal corrosion prediction method in the material storage and transportation environment as described in any one of claims 1 to 7.

Images