A method for predicting pH value of concrete in submarine immersed tunnel

Abstract

This invention discloses a method for predicting the pH value of concrete in subsea immersed tunnels. First, monitoring data of the subsea immersed tunnel is acquired, including temperature, humidity, polarization resistance of the concrete structure, dielectric resistance, Cl ion concentration, corrosion current density, and pH value. The acquired monitoring data is preprocessed and normalized to form a time-series dataset. Then, the periodicity of the frequency components of the data is identified. Next, an LSTM model and a Transformer model are constructed, using the temperature, humidity, polarization resistance, dielectric resistance, Cl ion concentration, corrosion current density, and pH value of the previous time period as input variables, and the pH value of the next time period as the output variable for prediction. The LSTM and Transformer models are trained using the dataset and the obtained periodicity features. After model training, the prediction results of the LSTM and Transformer models are fused using a weighted average method to generate the final prediction result.

Description

Technical Field

[0001] The invention belongs to the technical field of intelligent monitoring of engineering structures, and in particular relates to a method for predicting the pH value of concrete in a submarine immersed tube tunnel. Background Art

[0002] In the health monitoring of modern large-scale infrastructure, especially the structural health monitoring of submarine immersed tube tunnels is particularly important. Submarine immersed tube tunnels are usually faced with extremely complex environmental conditions, including high water pressure, salt water erosion, uneven geological pressure, ocean currents, etc. These factors can cause the tunnel structure to sink, shift, and even crack, which seriously threatens the long-term stability and safety of the tunnel. Under these environmental conditions, the pH value monitoring of concrete is particularly critical, because the change of pH value directly affects the alkalinity of concrete and the durability of steel bars. Low pH value means that the alkalinity of concrete is reduced, resulting in an increased risk of steel corrosion. Traditional methods for monitoring the structural health of submarine immersed tube tunnels mainly rely on regular manual inspections and limited sensor monitoring point data. This method is not only inefficient and costly, but also unable to capture the dynamic changes of the tunnel structure in real time, making it difficult to detect and predict potential problems in the early stage. With the continuous expansion of the project scale and the improvement of the safety requirements for submarine immersed tube tunnels, pH value, as an important indicator of concrete durability, has become an indispensable part of tunnel health prediction, which can provide more scientific support for structural protection and early warning.

[0003] In this context, the intelligent health prediction system based on deep learning has gradually become an effective way to solve this problem. Deep learning technology, especially the long short-term memory network (LSTM) and the Transformer model, has shown excellent performance in time series data analysis with its powerful data processing capabilities and the ability to capture complex nonlinear relationships. The LSTM model is good at processing long-term dependencies in time series and can effectively deal with the long-term trends and historical dependencies in the structural monitoring data of submarine immersed tube tunnels; while the Transformer model has significant advantages in processing multi-dimensional data through the self-attention mechanism, especially in capturing the complex interactive relationships between different environmental and dynamic parameters. Therefore, deep learning technology provides strong technical support for the prediction of the health of submarine immersed tube tunnels. This technical background has laid a solid foundation for the proposal of the present invention. Summary of the invention

[0004] The present invention mainly aims to overcome the deficiencies of the prior art and provides a method for predicting the pH value of concrete in a submarine immersed tube tunnel.

[0005] The present invention is achieved through the following technical solutions:

[0006] A method for predicting the pH value of concrete in a submarine immersed tube tunnel comprises the following steps:

[0007] Step 1, obtaining monitoring data of the submarine immersed tube tunnel, the data including: temperature, humidity, polarization resistance of the concrete structure, dielectric resistance of the concrete structure, Cl ion concentration of the concrete structure, corrosion current density of the concrete structure and pH value of the concrete structure;

[0008] Step 2: preprocess and normalize the monitoring data obtained in step 1 to form a time series data set;

[0009] Step 3, perform wavelet transform on the monitoring data to decompose the data into different frequency components;

[0010] Step 4, further decomposing the monitoring data of each frequency component processed in step 3 through wavelet packet analysis to identify its periodic characteristics;

[0011] Step 5, construct an LSTM model and a Transformer model, take the temperature, humidity, polarization resistance of the concrete structure, dielectric resistance of the concrete structure, Cl ion concentration of the concrete structure, corrosion current density of the concrete structure and pH value of the concrete structure in the previous time period as input variables, and take the pH value of the concrete structure in the next time period as the predicted output variable; use the data set in step 2 and the periodic characteristics obtained in step 4 to train the LSTM model and the Transformer model; after the model training is completed, the prediction results of the LSTM model and the Transformer model are fused by the weighted average method to generate the final prediction result.

[0012] In the above technical solution, in step 1, an adaptive sampling technique is used to dynamically adjust the sampling frequency of the data based on the change rate of the historical monitoring data, so as to determine the appropriate sampling intervals for each type of monitoring data; the faster the change rate of the historical monitoring data, the faster the corresponding sampling frequency is adjusted.

[0013] In the above technical solution, in step 2, the preprocessing includes missing value processing and outlier detection and correction to ensure data quality and integrity.

[0014] In the above technical solution, in step 2, the missing value processing includes: 1) if the missing rate of monitoring data at a certain time point exceeds the set threshold, the data at that time point will be deleted; 2) if continuous missing values ​​appear within the sampling period, choose to use the first valid value in front or behind to fill the missing data.

[0015] In the above technical solution, in step 2, outlier detection and correction: outlier detection is performed using a Z-score-based method, and the identified abnormal data is corrected.

[0016] In the above technical solution, in step 2, the normalization process is to scale all data to the interval [0, 1] to ensure that all data have the same scale.

[0017] In the above technical solution, the data set obtained in step 2 is encrypted and stored in a cloud database, and encrypted and managed using blockchain technology to ensure that the data cannot be tampered with and is traceable.

[0018] In the above technical solution, the mathematical expression of the periodic characteristic is: ;

[0019] in, For the target data, is the wavelet packet basis function, For time, is the index of the discrete time point.

[0020] In the above technical solution, in step 5, when the prediction results of the LSTM model and the Transformer model are fused by the weighted average method, the weights of the two models are allocated according to the inverse proportion of the errors of the two models, and the model with smaller error is allocated a higher weight.

[0021] The advantages and beneficial effects of the present invention are:

[0022] The present invention records and processes multi-dimensional key data including temperature, humidity, polarization resistance of concrete structure, dielectric resistance of concrete structure, Cl ion concentration of concrete structure, corrosion current density of concrete structure and pH value of concrete structure by deploying high-precision sensor network in large-scale submarine immersed tube tunnel construction projects. These data are optimized by time series processing and adaptive sampling technology to ensure the continuity and representativeness of the data. In the data processing link, the system improves the quality and integrity of the data by deleting high missing rate data points, repairing a small amount of missing data by forward and backward filling method, and detecting and correcting outliers by Z-score method. Through wavelet transform, the time series data is decomposed into different frequency components, and then through wavelet packet analysis, the monitoring data of each frequency component is further decomposed to identify its periodic characteristics. Finally, by weighted averaging the prediction results of the two deep learning models LSTM and Transformer, a high-precision prediction of the pH value of the tunnel concrete structure is achieved. The predicted pH value can be used to predict and evaluate the durability of the tunnel concrete structure and the corrosion risk of steel bars. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The present invention is a flow chart of the method for predicting the pH value of concrete in a submarine immersed tube tunnel.

[0024] For ordinary technicians in this field, other relevant drawings can be obtained based on the above drawings without any creative work. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is further described below in conjunction with specific embodiments.

[0026] The present invention designs a method for predicting the pH value of concrete in a submarine immersed tube tunnel. Figure 1 , the specific steps are as follows.

[0027] Step 1: Deploy multiple high-precision sensor networks in the target submarine immersed tube tunnel project to monitor key environmental and structural data, including the following monitoring data.

[0028] Temperature (T): Calculated by temperature sensor, with a data range of 15-40°C. The temperature sensor is arranged on the top side wall of the carriageway of the immersed tube tunnel.

[0029] Humidity (H): Measured by humidity sensors, the humidity range is 50-100%. The humidity sensors are arranged on the top side wall of the carriageway of the immersed tube tunnel.

[0030] Polarization resistance of concrete structure ( R p ): recorded by electrochemical impedance sensor, embedded in the side of the immersed tube tunnel structure, the data range is recommended to be 50-500Ω, the recording frequency is once an hour, and the sensor accuracy should reach ±5Ω.

[0031] Dielectric resistance of concrete structure ( R m ): The recommended recording range is 10Ω to 100Ω, recorded once every hour, and the recommended sensor accuracy of the medium resistance detection sensor is ±1Ω, embedded in the side of the immersed tube tunnel structure.

[0032] Cl ion concentration in concrete structure ( C cl ): The recording range is 0.001-1.0M, and it is recorded once an hour through an ion selective electrode sensor with an accuracy of ±0.001M, which is embedded in the side of the immersed tube tunnel structure.

[0033] Corrosion current density of concrete structure ( I corr ): The recording range is recommended to be 0.001-1.5mA / cm², recorded once every hour, using an electrochemical corrosion sensor with an accuracy of ±0.01mA / cm², embedded in the side of the immersed tube tunnel structure.

[0034] pH value of concrete structure: The recording range is 0-14, recorded once an hour by a pH electrode sensor, the sensor accuracy should reach ±0.01, and it is embedded in the side of the immersed tube tunnel structure.

[0035] These monitoring data are stored in a structured manner in the database. The system processes these data in a time series and associates each record with a specific timestamp to form a highly structured data set. , the data monitored by each sensor is , which the system stores as:

[0036] ;

[0037] Furthermore, for the optimization of sampling intervals, adaptive sampling technology is used to dynamically adjust the sampling frequency of data based on the change rate of historical monitoring data, so as to determine the appropriate sampling intervals for each type of monitoring data. The faster the change rate of historical monitoring data, the faster the corresponding sampling frequency is adjusted to enhance the sensitivity and real-time performance of data collection.

[0038] For example, when the polarization resistance of a concrete structure is detected ( R p ) drops sharply in a short period of time, the system will automatically start the high-frequency sampling mode. The specific adjustment formula is as follows:

[0039] ;

[0040] in, is the adjusted sampling frequency, is the initial sampling frequency, is the adjustment coefficient, is the polarization resistance of the concrete structure ( R p ) rate of change. For example: The initial frequency is once an hour, when the system detects R p Rapidly drop from 150Ω to 100Ω, k =0.5, the new sampling frequency can be adjusted to once every 5 minutes.

[0041] Step 2: preprocess and normalize the monitoring data obtained in step 1 to form a time series data set.

[0042] Preprocessing, including missing value handling and outlier detection and correction, to ensure data quality and integrity:

[0043] 1. Missing value processing: If the missing rate of monitoring data at a certain time point exceeds 30%, the data at that time point will be deleted to avoid significant impact on the overall data set. For a small number of consecutive missing data (such as missing 3-5 points), forward or backward filling methods are used to fill in. Specifically, if If the data at a certain moment is missing, you can use the previous valid data or filling: ;

[0044] 2. Outlier detection and correction: The Z-score-based method is used for outlier detection. The outlier judgment formula is: ;

[0045] in is the sample mean, is the sample standard deviation; if , then the point is considered as an outlier and is corrected.

[0046] Example description: Assume that the original temperature data is: , the processed data is: Through these processing, the raw data are converted into a high-quality data set suitable for concrete pH value prediction model training and analysis, ensuring the accuracy and reliability of subsequent concrete pH value prediction model predictions.

[0047] Normalization processing is to scale all data to the interval [0, 1] to ensure that all data have the same scale, thereby improving the performance and stability of the concrete pH value prediction model.

[0048] Furthermore, the data set obtained in step 2 can be encrypted and stored in a cloud database, and encrypted and managed using blockchain technology to ensure that the data cannot be tampered with and is traceable.

[0049] In addition, the data storage adopts the NoSQL database structure, which can efficiently process and query complex time series data. For example, each data record can be represented as: ;

[0050] The system will regularly back up data to an offsite server to prevent data loss. The data processing status is updated in real time through the cloud console, and the status is marked as "processing", "processed" or "error".

[0051] Step 3: Perform wavelet transform on the monitoring data to decompose the data into different frequency components.

[0052] For example, temperature data can be transformed by wavelet transform to obtain high-frequency and low-frequency components:

[0053] ;

[0054] in, is the wavelet function, is the temperature data, is the scale factor, is the time shift parameter.

[0055] Step 4: further decompose the monitoring data of each frequency component processed in step 3 through wavelet packet analysis to identify its periodic characteristics.

[0056] The mathematical expression of the periodic characteristics is: ;

[0057] in, For the target data, is the wavelet packet basis function, For time, is the index of the discrete time point.

[0058] Through this decomposition, the periodic characteristics implicit in the data are identified, and the potential impact of these periodic changes on the pH value of concrete can be predicted.

[0059] Step 5, construct a long short-term memory network LSTM model and a Transformer model, take the temperature, humidity, polarization resistance of the concrete structure, dielectric resistance of the concrete structure, Cl ion concentration of the concrete structure, corrosion current density of the concrete structure and pH value of the concrete structure in the previous time period as input variables, and take the pH value of the concrete structure in the next time period as the predicted output variable; use the data set in step 2 and the periodic characteristics obtained in step 4 to train the LSTM model and the Transformer model; after the model training is completed, the prediction results of the LSTM model and the Transformer model are fused by the weighted average method to generate the final prediction result.

[0060] Specifically:

[0061] In this step, the LSTM network can well capture the long-term dependencies in the input data; the Transformer model can process multi-dimensional input features through the self-attention mechanism, combine the sequential information in the time series, and predict the future pH value of concrete.

[0062] During model training, the mean square error (MSE) is used as the loss function: ;

[0063] in is the true value, is the predicted value.

[0064] After the model training is completed, the test set is used for evaluation, the mean square error (MSE) is calculated, and the parameters are continuously adjusted to ensure the accuracy of the prediction results.

[0065] Finally, the prediction results of the LSTM model and the Transformer model are combined, and the weighted average method is used to further optimize the final prediction output. First, the error performance of the LSTM model and the Transformer model on the validation set is evaluated. Common evaluation indicators include mean square error (MSE), root mean square error (RMSE) or mean absolute error (MAE). The evaluation results of the example are as follows:

[0066] The MSE of the LSTM model is: ;

[0067] The MSE of the Transformer model is: ;

[0068] The weights of the two models are assigned according to the inverse ratio of their errors, that is, the model with smaller error is assigned higher weight. Assume that the inverse ratio of MSE is used as the weight calculation method:

[0069] The weights of the LSTM model are: ;

[0070] The weights of the Transformer model are:

[0071] ;

[0072] Substituting the values ​​in the above formula, we get:

[0073] ; ;

[0074] Finally, the prediction results based on the LSTM and Transformer models and As the predicted value of future data, the final weighted average prediction result y predicted for:

[0075] .

[0076] The present invention is described above by way of example. It should be noted that, without departing from the core of the present invention, any simple deformation, modification or other equivalent replacement that can be made by those skilled in the art without inventive effort falls within the protection scope of the present invention.

Claims

1. A method for predicting the pH value of concrete in a submarine immersed tube tunnel, characterized in that: The following steps are involved: Step 1, obtaining monitoring data of the submarine immersed tube tunnel, the data including: temperature, humidity, polarization resistance of the concrete structure, dielectric resistance of the concrete structure, Cl ion concentration of the concrete structure, corrosion current density of the concrete structure and pH value of the concrete structure; Step 2: preprocess and normalize the monitoring data obtained in step 1 to form a time series data set; Step 3, perform wavelet transform on the monitoring data to decompose the data into different frequency components; Step 4, further decomposing the monitoring data of each frequency component processed in step 3 through wavelet packet analysis to identify its periodic characteristics; Step 5, construct an LSTM model and a Transformer model, take the temperature, humidity, polarization resistance of the concrete structure, dielectric resistance of the concrete structure, Cl ion concentration of the concrete structure, corrosion current density of the concrete structure and pH value of the concrete structure in the previous time period as input variables, and take the pH value of the concrete structure in the next time period as the predicted output variable; use the data set in step 2 and the periodic characteristics obtained in step 4 to train the LSTM model and the Transformer model; after the model training is completed, the prediction results of the LSTM model and the Transformer model are fused by the weighted average method to generate the final prediction result.

2. The method for predicting the pH value of concrete in a submarine immersed tube tunnel according to claim 1, characterized in that: In step 1, an adaptive sampling technique is used to dynamically adjust the sampling frequency of the data based on the change rate of the historical monitoring data. The faster the change rate of the historical monitoring data, the faster the corresponding sampling frequency is adjusted.

3. The method for predicting the pH value of concrete in a submarine immersed tube tunnel according to claim 1, characterized in that: In step 2, the preprocessing includes missing value processing and outlier detection and correction.

4. The method for predicting the pH value of concrete in a submarine immersed tube tunnel according to claim 3 is characterized in that: In step 2, missing value processing includes: 1) if the missing rate of monitoring data at a certain time point exceeds the set threshold, the data at that time point will be deleted; 2) if there are consecutive missing values ​​within the sampling period, choose to use the first valid value before or after to fill the missing data.

5. The method for predicting the pH value of concrete in a submarine immersed tube tunnel according to claim 3, characterized in that: In step 2, outlier detection and correction: a Z-score-based method is used to detect outliers and correct the identified abnormal data.

6. The method for predicting the concrete pH value of a submarine immersed tube tunnel according to claim 1, characterized in that: In step 2, the normalization process is to scale all data to the interval [0, 1].

7. The method for predicting the pH value of concrete in a submarine immersed tube tunnel according to claim 1, characterized in that: The data set obtained in step 2 is encrypted and stored in the cloud database, and encrypted and managed using blockchain technology.

8. The method for predicting the pH value of concrete in a submarine immersed tube tunnel according to claim 1, characterized in that: The mathematical expression of the periodic characteristic is: ; in, For the target data, is the wavelet packet basis function, For time, is the index of the discrete time point.

9. The method for predicting the pH value of concrete in a submarine immersed tube tunnel according to claim 1, characterized in that: In step 5, when the prediction results of the LSTM model and the Transformer model are fused by the weighted average method, the weights of the two models are assigned according to the inverse ratio of the errors of the two models. The model with smaller error is assigned higher weight.

Images