A data processing method and system for early warning of water accumulation in tunnels

Abstract

This invention relates to the field of tunnel water level safety technology, and discloses a data processing method and system for tunnel water accumulation early warning. Multiple sets of data are collected every T minutes, and the correlation coefficients of the current rainfall, tunnel drainage capacity, rainfall time, and water depth T minutes later are analyzed. When the absolute value of the correlation coefficient is greater than a preset correlation coefficient threshold, a multiple linear regression model is constructed; when the absolute value of the correlation coefficient is less than or equal to the preset correlation coefficient threshold, a deep learning model is constructed. Based on the real-time monitored rainfall, tunnel drainage capacity, and rainfall time data, the constructed model is used to predict the water depth T minutes later and issue a real-time early warning. The system includes a rainfall sensor, a water level sensor, a monitoring platform, a communication unit, a camera, a barrier gate, and an information board. This data processing method for tunnel water accumulation early warning effectively improves the accuracy and timeliness of tunnel water accumulation early warning, thereby ensuring traffic safety.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel water level safety, and in particular to a data processing method and system for early warning of tunnel water accumulation. Background Art

[0002] Urban tunnels are usually located at low points on the road, making them prone to waterlogging. Once water accumulates in the tunnel, vehicles may take on water and stall when wading through it, leading to traffic jams and even casualties.

[0003] To deal with the problem of tunnel waterlogging, the main warning methods currently used include installing water level gauges and water level warning lines, as well as installing warning systems that directly monitor the depth of tunnel waterlogging. These measures have improved the monitoring and warning capabilities of tunnel waterlogging to a certain extent, but there are still the following shortcomings:

[0004] Manual visual inspection error: The water level height is dependent on manual visual inspection, which is easily affected by the external environment, and the judgment process is highly dependent on the subjective judgment of personnel, which may lead to large errors. Insufficient timeliness of early warning: When the system issues an early warning, the depth of water accumulation often exceeds the safety threshold. At this time, some vehicles may have entered the tunnel, which may easily cause vehicle flooding accidents. Insufficient accuracy of the early warning system: The current prediction method cannot more accurately capture the various factors that affect the depth of water accumulation. Different data distribution and characteristics cause errors due to improper model selection.

[0005] Therefore, there is an urgent need for a data processing method and system for tunnel waterlogging warning to improve the reliability of waterlogging depth prediction results. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a data processing method and system for tunnel waterlogging warning, which effectively improves the accuracy and timeliness of tunnel waterlogging warning, thereby ensuring driving safety.

[0007] The present invention provides a data processing method for early warning of water accumulation in tunnels, comprising the following steps:

[0008] S1, collect multiple sets of data every T minutes, and preprocess the multiple sets of data;

[0009] The multiple sets of data include the current rainfall, tunnel drainage capacity, rainfall time, and water depth after T minutes;

[0010] S2. Taking the water depth after T minutes as the prediction target variable, and taking the rainfall at the current moment, the tunnel drainage capacity, and the rainfall time as independent variables; analyzing the correlation coefficient between the rainfall at the current moment, the tunnel drainage capacity, the rainfall time, and the water depth after T minutes;

[0011] S3, when the absolute value of the correlation coefficient between the current rainfall, tunnel drainage capacity, rainfall time and water depth after T minutes is greater than a preset correlation coefficient threshold, construct a multivariate linear regression model;

[0012] When the absolute value of the correlation coefficient between the rainfall at the current moment, the tunnel drainage capacity, the rainfall time, and the water depth after T minutes is less than or equal to a preset correlation coefficient threshold, a deep learning model is constructed;

[0013] S4, based on the real-time monitoring of the current rainfall, tunnel drainage capacity, and rainfall time data, using the constructed model, predict the depth of water accumulation after T minutes;

[0014] S5. Issue a real-time warning based on the predicted water depth T minutes later.

[0015] Furthermore, the preprocessing of the multiple sets of data includes: data cleaning, missing value filling and data standardization.

[0016] Furthermore, the preset correlation coefficient threshold is 0.5.

[0017] Furthermore, the construction of a multiple linear regression model specifically includes:

[0018] Construct a multiple linear regression equation: ;

[0019] in, x 1 i is the rainfall at the current moment; x 2 i For tunnel drainage capacity; x 3 i is the rainfall time; i is the depth of water accumulation after T minutes; , , , is the regression coefficient, is the error term;

[0020] Using the least squares method, we get a set of regression coefficients , , , and error term Minimize the error between the predicted value and the actual value;

[0021] Substitute the regression coefficients and error terms into the multiple linear regression equation to obtain the multiple linear regression model.

[0022] Furthermore, the specific steps of constructing the deep learning model include:

[0023] Collect multiple sets of data every T minutes on rainy days, including the current rainfall x1 i ; Tunnel drainage capacity x2 i ; Rainfall time x3 i ; Water depth y after T minutes i , forming a data set , n represents the number of data groups;

[0024] Divide the data set into training set, validation set and test set according to the preset ratio;

[0025] Build a deep learning model based on TensorFlow or PyTorch, train the deep learning model based on the data in the training set using a sliding window method, adjust the hyperparameters through the validation set, and use the test set to evaluate the performance of the model to obtain the final deep learning prediction network used for prediction.

[0026] Furthermore, the S5 specifically includes:

[0027] Determine a safe water depth threshold based on historical data and tunnel design standards;

[0028] When the predicted water depth after T minutes exceeds the safe water depth threshold, the warning function is triggered and the warning information is prompted during the visual display.

[0029] The present invention also provides a data processing system for early warning of waterlogging in tunnels, which is used in the above-mentioned data processing method for early warning of waterlogging in tunnels, and comprises: a rainfall sensor, a liquid level sensor, a monitoring platform, and a communication unit;

[0030] The rain sensor is arranged on the road surface section outside the tunnel with no obstruction on the top, and is used to detect the real-time rainfall and upload the rainfall data to the monitoring platform through the communication unit;

[0031] The liquid level sensor is arranged at the lowest point of the road surface section inside the tunnel and close to the roadside, and is used to detect the depth of water accumulation at the lowest point of the tunnel, and upload the water accumulation depth data to the monitoring platform through the communication unit;

[0032] The monitoring platform is set up at the tunnel management station, and is used to process the received rainfall data and water depth data, monitor and predict the water depth after T minutes, and generate corresponding warning information and operation instructions according to the monitoring and prediction results.

[0033] Furthermore, a data processing system for early warning of water accumulation in tunnels also includes: a camera, a gate, and an information board;

[0034] The gate is arranged at a preset distance outside the tunnel entrance and is used to receive an operation instruction from the monitoring platform to open or close;

[0035] The camera is arranged on the roadside on the right side of the driving direction at the gate, and is used to monitor the vehicle passing information at the gate in real time, and upload the vehicle passing information data to the monitoring platform through the communication unit;

[0036] The information board is arranged on the roadside on the right side of the driving direction at the gate, and is used to publish and display early warning information.

[0037] The embodiments of the present invention have the following technical effects:

[0038] 1. The present invention realizes real-time monitoring of the outside and inside of the tunnel by setting up rainfall sensors and liquid level sensors, ensuring that the data of rainfall and water depth can be collected and updated in time, thereby improving the accuracy and timeliness of early warning.

[0039] 2. The present invention provides flexibility for data processing by judging the correlation coefficient of the current rainfall, tunnel drainage capacity, rainfall time and water depth after T minutes, and selecting different models according to the correlation coefficient. It can more accurately capture various factors affecting the water depth, making the prediction results more reliable. By dynamically selecting models (linear regression or deep learning), the present invention can adapt to different data distributions and characteristics and reduce errors caused by improper model selection.

[0040] 3. The present invention sets a safe water depth threshold. When the predicted water depth exceeds the threshold, an early warning will be triggered. This early warning mechanism can effectively reduce potential safety hazards and ensure the safety of pedestrians in the tunnel. Early warning information is issued through the information board and combined with the operating instructions of the monitoring platform to ensure the effectiveness of information transmission and help drivers obtain important information in a timely manner, thereby adjusting driving strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 is a flow chart of a data processing method for early warning of tunnel waterlogging provided by an embodiment of the present invention;

[0043] Figure 2 It is a schematic diagram of the installation of a road surface waterlogging warning device for a tunnel waterlogging warning data processing system provided by an embodiment of the present invention;

[0044] 1- Tunnel pavement, 2- Liquid level sensor, 3- Rainfall sensor, 4- Information board, 5- Camera, 6- Gate. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0046] Figure 1 is a flow chart of a data processing method for tunnel waterlogging warning provided by an embodiment of the present invention. Figure 1 , including the following steps:

[0047] S1, collect multiple sets of data every T minutes, and preprocess the multiple sets of data;

[0048] The multiple sets of data include the current rainfall, tunnel drainage capacity, rainfall time, and water depth after T minutes;

[0049] Exemplarily, the amount of water flowing into the tunnel is calculated based on rainfall data. For example, if the rainfall is R mm and the cross-sectional area of ​​the tunnel is A square meters, the inflow flow can be calculated by the following formula:

[0050] Inflow rate = R × A × 1 / 1000 (unit: cubic meters / hour);

[0051] The volume of water accumulated in T minutes can be calculated using the depth of water accumulated and the cross-sectional area of ​​the tunnel:

[0052] Water accumulation volume = tunnel cross-sectional area × water accumulation depth; Estimation of tunnel drainage capacity:

[0053] In a specific time period (such as T minutes), the tunnel drainage capacity can be estimated by combining the tunnel inflow flow and water accumulation changes. The following formula can be used:

[0054] Tunnel drainage capacity = change in water volume / T + inflow flow drainage capacity, where the change in water volume can be calculated by monitoring the change in water depth at different time points.

[0055] The preprocessing of the multiple sets of data includes: data cleaning, missing value filling and data standardization.

[0056] Specifically, data cleaning is the process of identifying and correcting errors and inconsistencies in a data set, including: removing duplicate values, checking whether there are duplicate records in the data set, and removing them to avoid interference with the analysis results; identifying erroneous data: identifying unreasonable values ​​(such as negative values ​​or values ​​outside a reasonable range) through logical checks and data range verification, and making corresponding corrections or eliminations; outlier processing: using statistical methods (such as box plots) to detect outliers and decide whether to correct, replace or delete them. There may be individual missing values ​​in the multiple sets of data collected, which can be supplemented by quadratic spline interpolation. At the same time, the process of converting data to a unified scale for effective comparison and analysis can use Z-score standardization and minimum-maximum standardization methods.

[0057] S2. Taking the water depth after T minutes as the prediction target variable, and taking the rainfall at the current moment, the tunnel drainage capacity, and the rainfall time as independent variables; analyzing the correlation coefficient between the rainfall at the current moment, the tunnel drainage capacity, the rainfall time, and the water depth after T minutes;

[0058] In some embodiments, the Pearson correlation coefficient between each independent variable (the rainfall at the current moment, the tunnel drainage capacity, and the rainfall time) and the predicted target variable (the water depth after T minutes) is calculated to determine the correlation between the rainfall at the current moment, the tunnel drainage capacity, the rainfall time, and the water depth after T minutes;

[0059] For two variables X and Y, the formula for the Pearson correlation coefficient r is: ;

[0060] Where Cov(X,Y) is the covariance of X and Y; is the standard deviation of the independent variable X; σY is the standard deviation of the dependent variable Y.

[0061] S3, when the absolute value of the correlation coefficient between the current rainfall, tunnel drainage capacity, rainfall time and water depth after T minutes is greater than a preset correlation coefficient threshold, construct a multivariate linear regression model;

[0062] When the absolute value of the correlation coefficient between the rainfall at the current moment, the tunnel drainage capacity, the rainfall time, and the water depth after T minutes is less than or equal to a preset correlation coefficient threshold, a deep learning model is constructed;

[0063] The preset correlation coefficient threshold is 0.5.

[0064] In some embodiments, the Pearson correlation coefficient between each independent variable (the rainfall at the current moment, the tunnel drainage capacity, and the rainfall time) and the predicted target variable (the water depth after T minutes) is analyzed, and the strength of the correlation between the two variables is determined based on the calculated correlation coefficient value.

[0065] When the absolute values ​​of the correlation coefficients of the rainfall at the current moment, the tunnel drainage capacity, the rainfall time, and the water depth after T minutes are all greater than the preset correlation coefficient threshold of 0.5, a multivariate linear regression model is constructed;

[0066] Furthermore, the construction of a multiple linear regression model specifically includes:

[0067] Construct a multiple linear regression equation: ;

[0068] in, x 1 i is the rainfall at the current moment; x 2 i For tunnel drainage capacity; x 3 i is the rainfall time; i is the depth of water accumulation after T minutes; , , , is the regression coefficient, is the error term;

[0069] Using the least squares method, we get a set of regression coefficients , , , and the error term Minimize the error between the predicted value and the actual value;

[0070] Substitute the regression coefficients and error terms into the multiple linear regression equation to obtain the multiple linear regression model.

[0071] When the absolute values ​​of the correlation coefficients of the rainfall at the current moment, the tunnel drainage capacity, the rainfall time, and the water depth after T minutes are all less than or equal to the preset correlation coefficient threshold of 0.5, a deep learning model is constructed;

[0072] Furthermore, the specific steps of constructing the deep learning model include:

[0073] Collect multiple sets of data every T minutes on rainy days, including the current rainfall x1 i ; Tunnel drainage capacity x2 i ; Rainfall time x3 i ; Water depth y after T minutes i , forming a data set , n represents the number of data groups;

[0074] Divide the data set into training set, validation set and test set according to the preset ratio;

[0075] Build a deep learning model based on TensorFlow or PyTorch, train the deep learning model based on the data in the training set using a sliding window method, adjust the hyperparameters through the validation set, and use the test set to evaluate the performance of the model to obtain the final deep learning prediction network used for prediction.

[0076] S4. Based on the real-time monitoring of the current rainfall, tunnel drainage capacity, and rainfall time data, the constructed model is used to predict the depth of water accumulation after T minutes.

[0077] S5. Provide real-time warning based on the predicted water depth T minutes later; specifically, include:

[0078] Determine a safe water depth threshold based on historical data and tunnel design standards;

[0079] When the predicted water depth after T minutes exceeds the safe water depth threshold, the warning function is triggered and the warning information is prompted during the visual display.

[0080] The present invention also provides a data processing system for early warning of waterlogging in tunnels, which is used to execute the above-mentioned data processing method for early warning of waterlogging in tunnels, and specifically comprises: a rainfall sensor, a liquid level sensor, a monitoring platform, and a communication unit;

[0081] The rain sensor is arranged on the road surface section outside the tunnel with no obstruction on the top, and is used to detect the real-time rainfall and upload the rainfall data to the monitoring platform through the communication unit;

[0082] The liquid level sensor is arranged at the lowest point of the road surface section inside the tunnel and close to the roadside, and is used to detect the depth of water accumulation at the lowest point of the tunnel, and upload the water accumulation depth data to the monitoring platform through the communication unit;

[0083] The monitoring platform is set up at the tunnel management station, and is used to process the received rainfall data and water depth data, monitor and predict the water depth after T minutes, and generate corresponding warning information and operation instructions according to the monitoring and prediction results.

[0084] A data processing system for early warning of water accumulation in tunnels, further comprising: a camera, a gate, and an information board;

[0085] The gate is arranged at a preset distance outside the tunnel entrance and is used to receive an operation instruction from the monitoring platform to open or close;

[0086] The camera is arranged on the roadside on the right side of the driving direction at the gate, and is used to monitor the vehicle passing information at the gate in real time, and upload the vehicle passing information data to the monitoring platform through the communication unit;

[0087] The information board is arranged on the roadside on the right side of the driving direction at the gate, and is used to publish and display early warning information.

[0088] Figure 2 1 is a schematic diagram of the installation of a road surface waterlogging warning device for a tunnel waterlogging warning data processing system provided by an embodiment of the present invention. The tunnel pavement 1, liquid level sensor 2, rainfall sensor 3, information board 4, camera 5, and gate 6 are arranged in the following positions: Figure 2 shown.

[0089] A data processing method and system for early warning of tunnel waterlogging includes the following steps:

[0090] Step 1: Rainfall sensor 3 detects rainfall at time t, and liquid level sensor 2 detects the water depth at the lowest point of the tunnel at time t and uploads it to the monitoring platform;

[0091] Step 2: After the monitoring platform receives the data, the data processing method for tunnel waterlogging warning predicts the tunnel waterlogging depth at time t+T, and publishes the waterlogging depth detected at time t and the predicted waterlogging depth at time t+T through the information board;

[0092] Step 3: The monitoring platform determines whether the predicted water depth at time t+T reaches the safe water depth threshold; when the predicted water depth at time t+T exceeds the safe water depth threshold, a corresponding warning message "the tunnel is about to close" is generated, and the information board 4 displays the warning message "the tunnel is about to close"; when the gate 6 receives the operation instruction of the monitoring platform, it closes the gate; and when the time reaches time t+T, the camera 5 detects the vehicle passing information at the gate 6 and transmits it to the monitoring platform. If no vehicle passes, the monitoring platform generates an operation instruction "close the gate" and generates a corresponding warning message "the tunnel is closed and vehicles are prohibited from entering"; the gate 6 executes the operation instruction, closes the gate, and the information board 4 displays the warning message "the tunnel is closed and vehicles are prohibited from entering".

[0093] It should be noted that the terms used in the present invention are only for describing specific embodiments, rather than limiting the scope of the present application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of more restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0094] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A data processing method for early warning of tunnel waterlogging, characterized in that: The steps include: S1, collect multiple sets of data every T minutes, and preprocess the multiple sets of data; The multiple sets of data include the current rainfall, tunnel drainage capacity, rainfall time, and water depth after T minutes; S2, taking the water depth after T minutes as the prediction target variable, and taking the rainfall at the current moment, the tunnel drainage capacity, and the rainfall time as independent variables; Analyze the correlation coefficient of the rainfall at the current moment, the tunnel drainage capacity, the rainfall time, and the water depth after T minutes; S3, when the absolute value of the correlation coefficient between the current rainfall, tunnel drainage capacity, rainfall time and water depth after T minutes is greater than a preset correlation coefficient threshold, construct a multivariate linear regression model; The construction of the multivariate linear regression model specifically includes: Construct a multiple linear regression equation: ; in, x 1 i is the rainfall at the current moment; x 2 i For tunnel drainage capacity; x 3 i is the rainfall time; i is the depth of water accumulation after T minutes; , , , is the regression coefficient, is the error term; Using the least squares method, we get a set of regression coefficients , , , and error term , so that the error between the predicted value and the actual value is minimized; Substitute the regression coefficient and error term into the multiple linear regression equation to obtain the multiple linear regression model; When the absolute value of the correlation coefficient between the rainfall at the current moment, the tunnel drainage capacity, the rainfall time, and the water depth after T minutes is less than or equal to a preset correlation coefficient threshold, a deep learning model is constructed; The specific steps of constructing the deep learning model include: Collect multiple sets of data every T minutes on rainy days, including the current rainfall x1 i ; Tunnel drainage capacity x2 i ; Rainfall time x3 i ; Water depth y after T minutes i , forming a data set , n represents the number of data groups; Divide the data set into training set, validation set and test set according to the preset ratio; Build a deep learning model based on TensorFlow or PyTorch, train the deep learning model based on the data in the training set using a sliding window method, adjust the hyperparameters through the validation set, and use the test set to evaluate the performance of the model to obtain the final deep learning prediction network for prediction; The preset correlation coefficient threshold is 0.5; S4, based on the real-time monitoring of the current rainfall, tunnel drainage capacity, and rainfall time data, using the constructed model, predict the depth of water accumulation after T minutes; S5. Issue a real-time warning based on the predicted water depth T minutes later.

2. A data processing method for tunnel waterlogging early warning according to claim 1, characterized in that: The preprocessing of the multiple sets of data includes: data cleaning, missing value filling and data standardization.

3. The data processing method for early warning of tunnel waterlogging according to claim 1 is characterized in that: The S5 specifically includes: Determine a safe water depth threshold based on historical data and tunnel design standards; When the predicted water depth after T minutes exceeds the safe water depth threshold, the warning function is triggered and the warning information is prompted during the visual display.

4. A data processing system for early warning of waterlogging in tunnels, used to execute a data processing method for early warning of waterlogging in tunnels as claimed in any one of claims 1 to 3, characterized in that: include: Rainfall sensor, liquid level sensor, monitoring platform, communication unit; The rain sensor is arranged on the road surface section outside the tunnel with no obstruction on the top, and is used to detect the real-time rainfall and upload the rainfall data to the monitoring platform through the communication unit; The liquid level sensor is arranged at the lowest point of the road surface section inside the tunnel and close to the roadside, and is used to detect the depth of water accumulation at the lowest point of the tunnel, and upload the water accumulation depth data to the monitoring platform through the communication unit; The monitoring platform is set up at the tunnel management station, and is used to process the received rainfall data and water depth data, and monitor and predict the water depth after T minutes, and generate corresponding warning information and operation instructions according to the monitoring and prediction results.

5. A data processing system for early warning of tunnel waterlogging according to claim 4, characterized in that: Also includes: Cameras, barriers, information boards; The gate is arranged at a preset distance outside the tunnel entrance and is used to receive an operation instruction from the monitoring platform to open or close; The camera is arranged on the roadside on the right side of the driving direction at the gate, and is used to monitor the vehicle passing information at the gate in real time, and upload the vehicle passing information data to the monitoring platform through the communication unit; The information board is arranged on the roadside on the right side of the driving direction at the gate, and is used to publish and display early warning information.

Images