Temperature Time Delay Prediction and Elimination Method for Concrete Box Girder Bridges Based on BP Neural Network

Abstract

The present invention discloses a method for predicting and eliminating temperature time lag of a concrete box girder bridge based on a BP neural network, belonging to the technical field of bridge health detection. This method proposes a temperature time lag calculation index and constructs a relationship model between daily temperature difference and temperature time lag coefficient based on a BP neural network, which can clarify the temperature time lag under different sunshine temperature differences, solves the problem that the conventional time domain translation method is not applicable to the long-term monitoring data sequence with large temperature fluctuations, greatly improves the correlation between the bridge structure response monitoring data and the structural temperature data, and provides support for structural state diagnosis.

Description

Technical Field

[0001] The present invention belongs to the field of bridge health detection technology, and particularly relates to a method for predicting and eliminating temperature time lag of a concrete box girder bridge based on a BP neural network. Background Art

[0002] Concrete box girder bridges have the advantages of large flexural and torsional stiffness, good overall performance, strong load-bearing capacity, and mature construction technology, and are widely used in national and provincial trunk line bridges. However, with the increase of the service time of the bridge, concrete bridges are affected by environmental factors such as solar radiation, seasonal alternation, and temperature change for a long time, and the temperature field inside and on the surface of the concrete structure shows an uneven distribution state. The secondary stress inside the structure caused by temperature action often exceeds the stress generated by other loads, which may lead to safety and durability problems such as local cracking of concrete and instability of components. Therefore, the influence of temperature effect on concrete box girder bridges needs to be taken seriously. Analyzing the influence of temperature on the structure generally requires performing a correlation analysis of temperature-structure response, and then diagnosing the health status of the bridge structure according to the change of temperature. However, due to the existence of temperature time lag effect, that is, the change of structural response caused by temperature change (such as beam end displacement, main beam displacement, cable force, etc.) lags behind the temperature change, the correlation analysis between temperature and structural response is inaccurate, which is likely to lead to missed reports and misreports of bridge diseases.

[0003] Taking the end displacement of a concrete bridge as an example, due to the thermal expansion and contraction characteristics of reinforced concrete materials, an increase in temperature will cause an increase in beam end displacement (expansion), and a decrease in temperature will cause a decrease in beam end displacement (contraction). In areas with large day-night temperature fluctuations, the thermal expansion and contraction effect of concrete bridges may be more significant. If the temperature changes and the change of beam end displacement is abnormal, there may be diseases in the expansion joints.

[0004] Existing research mainly eliminates the temperature time lag effect through the time domain translation method, that is, subtracting the time of the peak and valley values of the structural response and temperature, taking the difference as the temperature time lag, and translating the bridge structural response data forward on the time scale. However, the time lag generated by different temperatures is not a fixed value. In long-term monitoring, the temperature time lag is non-linear. The method of translating a single time is applicable to short-term monitoring data sequences with small temperature fluctuations, but not applicable to long-term monitoring data sequences with large temperature fluctuations, resulting in poor elimination effect of the temperature time lag effect and poor fitting effect of the correlation between structural response and structural temperature. Summary of the Invention

[0005] Objective of the Invention: To solve the problems of temperature time lag effect in the analysis of the correlation between structural response and temperature, and that the existing temperature time lag translation method is applicable to short-term monitoring data sequences with small temperature fluctuations. In particular, the prior art does not consider that the temperature time lag is non-linear and is not applicable to long-term monitoring data sequences with large temperature fluctuations, resulting in poor elimination effect of the temperature time lag effect. The present invention proposes a method for predicting and eliminating the temperature time lag of a concrete box girder bridge based on a BP neural network.

[0006] Technical Solution: A method for predicting and eliminating the temperature time lag of a concrete box girder bridge based on a BP neural network, the steps of the method include:

[0007] (1) Collect the beam end displacement and structural temperature data, and align the time of the beam end displacement and structural temperature data sequences. Specifically:

[0008] Compare the start collection times t_dis(1) and t_tmp(1) of the beam end displacement and structural temperature, and select the maximum value max(t_dis(1), t_tmp(1)) as the common start collection time for the two monitoring items;

[0009] Compare the end collection times t_dis(end) and t_tmp(end) of the beam end displacement and structural temperature, and select the minimum value min(t_dis(end), t_tmp(end)) as the common end collection time for the two monitoring items;

[0010] Align the collection times of the beam end displacement and structural temperature through the determined start collection time and end collection time;

[0011] (2) Perform moving average processing on the beam end displacement and structural temperature data, and adjust the time interval of the moving average processing to solve the problem of inconsistent sampling frequencies of the beam end displacement and structural temperature, including translating the beam end displacement data sequence on the time scale;

[0012] (3) Perform cross-correlation analysis on the beam end displacement and structural temperature data sequences, including calculating the Pearson correlation coefficient, plotting the cross-correlation coefficient curve, finding the cycle time corresponding to the maximum correlation coefficient, and obtaining the temperature time lag in a day;

[0013] Find the time offset that maximizes the correlation between the two signals by calculating the cross-correlation function of the beam end displacement and structural temperature data sequences, and this offset represents the temperature time lag;

[0014] Keep the value of the beam end displacement unchanged, and translate the beam end displacement acquisition data points by the temperature time lag value lag on the time scale to eliminate the temperature time lag for each day;

[0015] (4) Record the highest temperature within the time when the processing unit is located 、Minimum temperature 、Temperature time lag , calculate the temperature time lag coefficient Qi:

[0016] , i = 1, 2, …, n

[0017]

[0018] (5)Statistical daily temperature difference within a certain time and the temperature time lag coefficient Qi, construct a temperature time lag data set, and use the daily temperature difference as the independent variable and the temperature time lag coefficient Q i as the dependent variable to build a three-layer BP neural network model;

[0019] For the daily temperature difference and the temperature time lag coefficient Q i The data set is divided into a training set and a test set, and the BP neural network is trained to obtain a temperature time lag neural network model that can predict different daily temperature differences;

[0020] (6)Input the daily temperature difference of the target date into the neural network model to calculate and predict the temperature time lag Q of the target date i .

[0021] Furthermore, step (3) is to determine the similarity between two data sequences through cross-correlation analysis, which is used to evaluate the correlation between the structural temperature time series and the beam end displacement time series at different time delays;

[0022] The calculation of the Pearson correlation coefficient is:

[0023]

[0024] In the formula, represents the Pearson correlation coefficient, is the amount of time series data, 、 are the values of the i-th sequence of the time series variables X and Y respectively, 、 are the means of the sequence variables X and Y;

[0025] Furthermore, the BP neural network model constructed in step (5) of this method includes an input layer, a hidden layer, and an output layer;

[0026] Forward propagation of information: Input the data into the model and calculate through the activation function;

[0027] Backward propagation of error: Calculate and update the weights and biases from the hidden layer to the input layer;

[0028] Model fitting output: The degree of model fitting is measured by an error function. The mean square error is used as the error function in the training of the BP neural network. The smaller the value of MSE, the higher the accuracy of the prediction model. Finally, the updated weights, thresholds, and biases are recalculated and iterated continuously. When the specified number of iterations or the optimal error expectation value is reached, the training is completed.

[0029] This method collects the data of beam-end displacement and structural temperature on a daily basis, and takes the data sequence within each day as the processing unit, including looping through the data sequence within a processing unit.

[0030] For long-term data sequences of one year or more, it includes looping through the data sequence daily to obtain a time-delay sequence of 365 days, and then subtracting the time-delay of each day to align the structural temperature and beam-end displacement data of each day, thereby obtaining an accurate linear relationship.

[0031] Furthermore, for the forward propagation of information in the BP neural network model package, first, the data is normalized, then input into the model, calculated through the activation function, and then the error is propagated backward to calculate and update the weights and biases from the hidden layer to the input layer.

[0032] This BP neural network model uses the mean square error as the error function in the training of the BP neural network. The smaller the value of MSE, the higher the accuracy of the prediction model. Finally, the updated weights, thresholds, and biases are recalculated and iterated continuously. When the specified number of iterations or the optimal error expectation value is reached, the training is completed.

[0033] On the other hand, the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for predicting and eliminating temperature time-delay of a concrete box girder bridge based on a BP neural network as described above.

[0034] A computer storage medium stores a computer program, and when the computer program is executed by a computer, it implements a method for predicting and eliminating temperature time-delay of a concrete box girder bridge based on a BP neural network as described above.

[0035] Beneficial effects: The method of the present invention proposes a temperature time-delay calculation index, constructs a relationship model between daily temperature difference and temperature time-delay coefficient based on a BP neural network, can clarify the temperature time-delay under different sunshine temperature differences, solves the problem that the conventional time-domain translation method is not applicable to long-term monitoring data sequences with large temperature fluctuations, greatly improves the correlation between bridge structure response monitoring data and structural temperature data, and provides support for structural state diagnosis. Description of the Drawings

[0036] Figure 1 is the flow chart of the method of the present invention;

[0037] Figure 2 is the schematic diagram of the three - layer BP neural network structure. Detailed implementation manners

[0038] To describe in detail the technical solution provided by the present invention, the following further introduction is made with reference to the accompanying drawings.

[0039] Combined with Figure 1 , the present invention provides a method for predicting and eliminating the temperature time - lag of a concrete box - girder bridge based on a BP neural network. Taking the analysis of the correlation between the beam - end displacement and temperature of a certain bridge in Huai'an as an example, it should be noted that since the layout of the beam - end displacement sensors of a certain bridge in Huai'an is the connection between the approach - bridge main girder and the main - bridge main girder, when the value of the beam - end displacement sensor becomes smaller, it reflects the elongation of the main girder, and the calculated regression relationship shows a negative correlation. At present, front - line technicians manually calculate the temperature time - lag of a certain day. On the one hand, the number of manual calculations is limited, the efficiency is low, and the time - lag is basically estimated by guesswork, with strong subjectivity. On the other hand, the temperature time - lag is different every day. The method for eliminating the temperature time - lag mainly includes:

[0040] S1. Taking the data sequence of each day as a processing unit, perform the loop operation from 2 to 9.

[0041] The loop operation is for long - term data of more than one year. The temperature time - lag is non - linear, and the method of translating a single time is not applicable to the long - term monitoring data sequence with large temperature fluctuations. Therefore, loop through 365 days of a year here to obtain a time - lag sequence of 365 days, and then subtract the time - lag of each day to align the structural temperature and beam - end displacement data of each day, so as to obtain an accurate linear relationship. The purpose of counting one year is that the seasonal and daily temperature differences between years are basically the same, and the linear relationship of previous years can be used to infer future correlations.

[0042] S2. Compare the start acquisition times t_dis(1) and t_tmp(1) of the beam - end displacement and the structural temperature, and select the maximum value max(t_dis(1), t_tmp(1)) as the common start acquisition time of the two monitoring items;

[0043] S3. Compare the end acquisition times t_dis(end) and t_tmp(end) of the beam - end displacement and the structural temperature, and select the minimum value min(t_dis(end), t_tmp(end)) as the common end acquisition time of the two monitoring items;

[0044] S4. Align the acquisition times of the beam - end displacement and the structural temperature through steps S2 and S3;

[0045] S5. At intervals of 5 minutes, perform a moving average process on the beam end displacement and the structural temperature to solve the problem of inconsistent sampling frequencies between the structural response and the structural temperature;

[0046] S6. Use 1 minute as the initial value of the loop, 2 hours as the final value of the loop, and 1 minute as the increment to translate the beam end displacement data sequence on the time scale;

[0047] S7. Perform a cross - correlation analysis on the beam end displacement and the structural temperature data sequences, calculate the Pearson correlation coefficient cor, plot the cross - correlation coefficient curve, and find the loop time corresponding to the maximum correlation coefficient cor_max, which is the temperature time lag lag in a day;

[0048] Cross - correlation analysis is a statistical method used to determine the similarity between two data sequences. It is usually used in time - series analysis to evaluate the correlation between two data sequences at different time lags. In the present invention, it is for the structural temperature time series and the beam end displacement time series.

[0049] It is further pointed out that the Pearson correlation coefficient is a method for measuring the linear correlation between two variables, and the linear correlation degree of variables can be evaluated through the covariance matrix of the data. The Pearson correlation coefficient is:

[0050]

[0051] In the formula, r XY is the Pearson correlation coefficient, n represents the amount of time - series data, Xi and Yi are the values of the i - th sequence of the time - series variables X and Y respectively, 、 are the means of the sequence variables X and Y.

[0052] r XY is the Pearson correlation coefficient between two variables X and Y, ranging from - 1 to 1. The larger the absolute value, the more correlated the two variables are. When the value is 0, the two variables are not correlated. If the correlation coefficient value is greater than 0, then the two variables are positively correlated. If the correlation coefficient value is less than 0, then the two variables are negatively correlated. The Pearson correlation coefficient gives the degree of similarity between two time - series data at different time offsets, thus helping us find the time lag between two sets of time - series data.

[0053] In the analysis of bridge beam end displacement and structural temperature, cross-correlation analysis can be used to determine the response lag time of the beam end displacement to changes in structural temperature. By calculating the cross-correlation function of the beam end displacement and structural temperature data series, the time offset that maximizes the correlation between the two signals can be found, and this offset represents the temperature time lag. Specifically, starting from 0 minutes and stepping by 1 minute, the time delay of the beam end displacement data is gradually increased on the time scale until 2 hours. Each value on the abscissa represents the time by which the beam end displacement data is delayed relative to the structural temperature data.

[0054] S8. Keep the value of the beam end displacement unchanged, and translate the beam end displacement acquisition data points by the temperature time lag value lag on the time scale to eliminate the temperature time lag for each day.

[0055] S9. Record the highest temperature t imax and the lowest temperature t imin for each day, as well as the temperature time lag lag i , and calculate the temperature time lag coefficient Q i :

[0056] , i = 1, 2, …, n (1)

[0057] (2)

[0058] It should be noted that in this step, the present invention constructs statistical indicators and the temperature time lag coefficient Qi. In the stage of unprocessed data, the linear relationship between the beam end displacement and the temperature time lag is not obvious, and it is impossible to assist in disease diagnosis based on their relationship, so cross-correlation analysis is required.

[0059] S10. Statistically analyze the daily temperature difference a i and the temperature time lag coefficient Q i for 365 days, construct a temperature time lag data set, use the daily temperature difference a i as the independent variable and the temperature time lag coefficient Q i as the dependent variable, and build a three-layer BP neural network model.

[0060] S11. Divide the data set of the daily temperature difference a i and the temperature time lag coefficient Q i into a training set and a test set according to 9:1, train the BP neural network, and obtain a temperature time lag neural network model that can predict different daily temperature differences.

[0061] S12. Input the daily temperature difference a i of the target date into the neural network model, and the temperature time lag Q i of the target date can be calculated and predicted.

[0062] Among them, the Matlab code for aligning the beam-end displacement and the structural temperature acquisition time can be as follows:

[0063] t1 = max(t_dis(1),t_tmp(1));

[0064] t2 = min(t_dis(end),t_tmp(end));

[0065] ind_DIS = t_dis >= t1 & t_dis <= t2;

[0066] ind_TMP = t_tmp >= t1 & t_tmp <= t2;

[0067] t_dis = t_dis(ind_DIS);

[0068] dis = disf(ind_DIS);

[0069] t_tmp = t_tmp(ind_TMP);

[0070] tmp = tmp(ind_TMP);

[0071] In summary, to solve the temperature time-delay effect existing in the structural response-temperature correlation analysis, and the problem that the existing temperature time-delay translation methods are applicable to short-term monitoring data sequences with small temperature fluctuations, but not applicable to long-term monitoring data sequences with large temperature fluctuations, and the elimination effect of the temperature time-delay effect is poor. The present invention proposes a temperature time-delay calculation index, constructs a relationship model between the daily temperature difference and the temperature time-delay coefficient based on the BP neural network, can clarify the temperature time-delay under different sunshine temperature differences, and through the optimization and improvement of the BP algorithm, realizes the calculation, elimination and prediction of the nonlinear temperature time-delay, solves the problem that the conventional time-domain translation method is not applicable in long-term monitoring data sequences with large temperature fluctuations, greatly improves the correlation between the bridge structural response monitoring data and the structural temperature data, provides support for structural state diagnosis, and completely solves the problem of the current subjective determination of the temperature time-delay by humans, and has great practicality.

[0072] To enable those skilled in the art to better understand the present invention, the three-layer BP neural network model and its algorithm constructed by the present invention are further introduced.

[0073] The BP (Back Propagation) neural network model is a multi-layer feedforward network trained by the error backpropagation algorithm. Through a three-layer BP neural network, the present invention can fully approximate any complex non-linear relationship and has a high degree of self-learning and self-adaptive capabilities. Its basic idea for temperature time-delay elimination is: using the daily temperature difference ai as the independent variable and the temperature time-delay coefficient Qi as the dependent variable, and fitting an explicit expression of the independent variable with respect to the dependent variable through regression analysis.

[0074] The calculation process of the BP algorithm is first the forward propagation of information, where data is input into the model and calculated through the activation function. Then it is the backpropagation of errors, calculating and updating the weights and biases from the hidden layer to the input layer. Usually, an error function is used to measure the degree of model fitting. The present invention uses the Mean-Square Error (MSE) as the error function in BP neural network training. The smaller the value of MSE, the higher the accuracy of the prediction model. Finally, the updated weights, thresholds, and biases are recalculated and iterated continuously. When the specified number of iterations or the optimal error expectation value is reached, the training is completed. The process for constructing the model is as follows:

[0075] The following uses Figure 2 the three-layer neural network structure shown to illustrate the calculation process of the BP algorithm.

[0076] (1) First is the forward propagation of information. To avoid the BP neural network having a long training time, inconsistent input data dimensions, or a large gap between input data, it is necessary to normalize the input and output data according to the following formula:[[]]END]]

[0077] (3)

[0078] represents any value in the training samples, represents the mean of the training samples, represents the variance of the training samples, represents the data after normalizing the original training samples.

[0079] Thus, the normalized input samples are obtained. From Figure 2 it can be seen that the first layer is the input layer, which contains three neurons and the bias term ; the second layer is the hidden layer, which contains two neurons and the bias term ; the third layer output layer has only one neuron , and in the figure is the weight of the connection between layers. Taking the sigmoid function as an example for the activation functions from the input layer to the hidden layer and from the hidden layer to the output layer, their formulas are as follows:

[0080] (4)

[0081] The input to the hidden layer is calculated as:

[0082] (5)

[0083] (6)

[0084] The output of the hidden layer is calculated as:

[0085] (7)

[0086] (8)

[0087] The input to the output layer is calculated as

[0088] (9)

[0089] The output of the output layer is calculated as

[0090] (10)

[0091] (2) Then comes the error backpropagation. Usually, an error function is used to measure the degree of model fitting. In this paper, the Mean-Square Error (MSE) is adopted as the error function in the training of the BP neural network. The smaller the value of MSE, the higher the accuracy of the prediction model.

[0092] (11)

[0093] represents the number of training samples, represents the neural network prediction data of the j-th sample, represents the true value of the j-th sample. The training error of this forward information propagation is obtained as:

[0094] (12)

[0095] Calculate the weight and bias updates from the output layer to the hidden layer:

[0096] (13)

[0097] (14)

[0098] Among them:

[0099] , ,

[0100] , ,

[0101] It can be obtained that:

[0102] (15)

[0103] (16)

[0104] Similarly, it can be obtained that

[0105] (17)

[0106] Update the weights and biases:

[0107] (18)

[0108] (19)

[0109] where is the step size, which determines the length of each step along the negative gradient direction during the error backpropagation process.

[0110] Calculate the weight and bias updates from the hidden layer to the input layer:

[0111] (20)

[0112] (21)

[0113] where:

[0114] ,

[0115] , , ,

[0116] It can be obtained that:

[0117] (22)

[0118] (23)

[0119] Similarly, it can be obtained that:

[0120] (24)

[0121] (25)

[0122] (26)

[0123] Update weights and biases:

[0124] (27)

[0125] (28)

[0126] Thus, a single backpropagation of the error is completed. Finally, the updated weights, thresholds, and biases are recalculated and iterated continuously. When the specified number of iterations or the optimal error expectation value is reached, the training is completed.

[0127] For the application of the three-layer BP neural network model, when restoring the BP neural network prediction value to the actual value, it is necessary to perform denormalization on the output data. The denormalization formula is .

Claims

1. A method for predicting and eliminating temperature lag of concrete box girder bridge based on BP neural network, characterized in that: The method steps include: (1) Collect beam end displacement and structural temperature data, and align the beam end displacement and structural temperature data sequence time, specifically: Compare the start collection time t_dis(1) and t_tmp(1) of the beam end displacement and the structural temperature, and select the maximum value max(t_dis(1), t_tmp(1)) as the common start collection time of the two monitoring items; Compare the end acquisition time t_dis(end) and t_tmp(end) of the beam end displacement and the structural temperature, and select the minimum value min(t_dis(end), t_tmp(end)) as the common end acquisition time of the two monitoring items; The beam end displacement and structure temperature collection times are aligned through the determined start collection time and end collection time; (2) Perform sliding average processing on the beam end displacement and structural temperature data. The time interval of the sliding average processing is adjusted to solve the problem of inconsistent sampling frequency of beam end displacement and structural temperature, including translating the beam end displacement data series on the time scale; (3) Perform cross-correlation analysis on the beam end displacement and structural temperature data series, including calculating the Pearson correlation coefficient, drawing the cross-correlation coefficient curve, finding the cycle time corresponding to the maximum correlation coefficient, and obtaining the temperature lag in a day; By calculating the cross-correlation function of the beam end displacement and structure temperature data series, the time offset that maximizes the correlation between the two signals is found, and the offset represents the temperature lag; Keep the beam end displacement value unchanged, and shift the beam end displacement data points on the time scale by the temperature lag value lag to eliminate the temperature lag every day; (4) Record the highest temperature of the processing unit during the period of time , minimum temperature , Temperature hysteresis , calculate the temperature hysteresis coefficient Qi: ,i=1,2,…,n ； (5) Counting daily temperature differences within a certain period of time and temperature lag coefficient Qi, construct the temperature lag data set, and convert the daily temperature difference As an independent variable, the temperature hysteresis coefficient Q i As the dependent variable, a three-layer BP neural network model was built; Temperature difference between days and temperature hysteresis coefficient Q i The data set is divided into a training set and a test set, and the BP neural network is trained to obtain a temperature time-lag neural network model that can predict different daily temperature differences; (6) The daily temperature difference on the target date Input neural network model to calculate and predict temperature lag Q on target date i .

2. The method for predicting and eliminating temperature lag of concrete box girder bridge according to claim 1 is characterized in that: Step (3) is to determine the similarity between the two data series through cross-correlation analysis, which is used to evaluate the correlation between the structural temperature time series and the beam end displacement time series at different time delays; The calculation of Pearson's correlation coefficient is: ； In the formula, represents the Pearson correlation coefficient, is the amount of time series data, , are the values ​​of the i-th sequence of time series variables X and Y, respectively. , is the mean of the sequence variables X and Y.

3. The method for predicting and eliminating temperature lag of concrete box girder bridge according to claim 1 is characterized in that: The BP neural network model constructed in step (5) includes an input layer, a hidden layer and an output layer; Information forward propagation: data is input into the model and calculated through the activation function; Error back propagation: Calculate and update the weights and biases from the hidden layer to the input layer; Model fitting output: The error function is used to measure the degree of model fitting. The mean square error is used as the error function in BP neural network training. The smaller the MSE value, the higher the accuracy of the prediction model. Finally, the updated weights, thresholds and biases are recalculated and iterated continuously. When the planned number of iterations or the optimal error expectation value is reached, the training is completed.

4. The method for predicting and eliminating temperature lag of concrete box girder bridge according to claim 1, characterized in that: The method collects the beam end displacement and structure temperature data in units of days, and regards the data sequence within each day as a processing unit, including looping the data sequence within a processing unit; For long-term data series of one year or more, the data series is cycled daily to obtain a 365-day time-lag series, and then the corresponding time lag is subtracted to align the structural temperature and beam end displacement data for each day to obtain an accurate linear relationship.

5. The method for predicting and eliminating temperature lag of concrete box girder bridge according to claim 3 is characterized in that: For the forward propagation of information in the BP neural network model package, the data is first normalized and then input into the model, calculated through the activation function, and then the error is back-propagated to calculate and update the weights and biases from the hidden layer to the input layer; The BP neural network model uses mean square error as the error function in BP neural network training. The smaller the MSE value, the higher the accuracy of the prediction model. Finally, the updated weights, thresholds and biases are recalculated and iterated continuously. When the planned number of iterations or the optimal error expected value is reached, the training is completed.

6. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the temperature lag prediction and elimination method of concrete box girder bridge based on BP neural network as described in any one of claims 1 to 5.

7. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a method for predicting and eliminating temperature lag of a concrete box girder bridge based on a BP neural network as described in any one of claims 1 to 5 is implemented.

Images