Real-time monitoring method for station water levels applied to flood prevention and early warning in rail transit

Abstract

This application relates to the field of water level monitoring technology, specifically to a real-time water level monitoring method for railway flood prevention and early warning. The method includes: acquiring water level data at different times for each monitoring point in a tunnel between adjacent stations during each monitoring cycle, as well as the rainfall during that monitoring cycle; when the water level data at a monitoring point meets preset trigger conditions, a risk assessment strategy is executed. This strategy includes: calculating a first assessment value for each monitoring cycle; calculating a second assessment value for each monitoring cycle; obtaining a risk coefficient for each monitoring cycle; evaluating the correlation characteristics between the risk coefficients and rainfall for multiple consecutive monitoring cycles; and determining the risk assessment value for each monitoring cycle to enable real-time monitoring and early warning of water accumulation risk in the tunnel section. This application can improve the early and accurate early warning of water accumulation risk in tunnel sections.

Description

Technical Field

[0001] This application relates to the field of water level monitoring technology, specifically to a method for real-time monitoring of station water levels for flood prevention and early warning in rail transit. Background Technology

[0002] With the rapid development of urban rail transit networks and the frequent occurrence of extreme weather events, flood prevention and safety issues in underground tunnels are becoming increasingly prominent. In particular, tunnels between adjacent stations, due to their relatively enclosed and low-lying characteristics, are highly susceptible to water accumulation or even flooding during heavy rainfall, seriously threatening train operation safety, the integrity of equipment and facilities, and the safety of passengers' lives and property.

[0003] Traditional methods for monitoring water levels in tunnel sections rely on determining when the water level at a single location exceeds a threshold to issue an early warning. However, because tunnel sections typically employ a concave structure in their longitudinal profile design, water naturally tends to converge towards low-lying areas, forming localized deep puddles. This makes traditional methods unable to detect this spatially uneven distribution of risk, resulting in significant early warning delays and missed alarms. Furthermore, the complex internal environment of tunnels makes the water surface susceptible to momentary strong disturbances such as piston winds and water level fluctuations caused by high-speed train passages, leading to brief spikes in water level data. Traditional methods are highly prone to this interference, resulting in frequent false alarms. Finally, there is a significant time lag between water level rises and rainfall, potentially causing severe delays in early warning and preventing timely detection of water accumulation risks. Summary of the Invention

[0004] To address the aforementioned technical issues, a method for real-time monitoring of station water levels for flood prevention and early warning in rail transit is provided to resolve existing problems.

[0005] The solution to the technical problem presented in this application is to provide a method for real-time monitoring of station water levels for flood prevention and early warning in rail transit, including the following steps: Obtain water level data at different times for each monitoring point in the tunnel between adjacent stations during each monitoring cycle, as well as the rainfall during that monitoring cycle; When the water level data at the monitoring point meets the preset trigger conditions, a risk assessment strategy is executed, the strategy including: For each monitoring period, the fluctuation and discrete characteristics of the water level data at the monitoring points are analyzed to evaluate the fluctuation characteristics of water level changes and calculate the first evaluation value for each monitoring period. The overall trend of water level changes at monitoring points, and the spatial distribution differences of this trend at different monitoring points, are used to assess the trend characteristics of water level changes and calculate the second assessment value for each monitoring period. Based on the first and second assessment values, the risk coefficient for each monitoring period is obtained. The correlation characteristics of the risk coefficients and rainfall for multiple consecutive monitoring periods are evaluated to determine the risk assessment value for each monitoring period, so as to monitor and warn of water accumulation risk in the tunnel in real time.

[0006] Preferably, the triggering condition is: the water level data of at least one monitoring point exceeds a preset activation threshold for multiple consecutive time periods.

[0007] Preferably, the calculation of the first evaluation value for each monitoring cycle includes: For each monitoring point, the extreme points of water level data at all times within each monitoring cycle are obtained, and the difference between the corresponding water level data between two adjacent extreme points is taken as the relative difference; the dispersion and average value of all relative differences for each monitoring point within each monitoring cycle are calculated. The first evaluation value is positively correlated with the dispersion of all monitoring points and the average value.

[0008] Preferably, the calculation of the second evaluation value for each monitoring cycle includes: For each monitoring point, the slope estimate is obtained by using the trend estimation method based on the water level data at all times within each monitoring cycle; the overall level of the slope estimate for different monitoring points under each monitoring cycle is evaluated, and the overall trend value for each monitoring cycle is calculated. Analyze the differences in slope estimates at different monitoring points and calculate the trend difference for each monitoring period; The second assessment value is positively correlated with both the overall trend value and the trend difference.

[0009] Preferably, the overall trend value is the result of forward fusion of the slope estimates of all monitoring points in each monitoring period.

[0010] Preferably, the calculation process of the trend difference is as follows: the monitoring point located at the middle position in the interval tunnel is marked as the target point; for each monitoring period, the difference between the slope estimates between the target point and all other monitoring points is positively fused to serve as the trend difference for each monitoring period.

[0011] Preferably, the risk coefficient is the result of a positive fusion of the first assessment value and the second assessment value.

[0012] Preferably, the correlation feature is measured by the risk coefficient of each monitoring period and the maximum mutual information coefficient of rainfall.

[0013] Preferably, the risk assessment value is positively correlated with both the risk coefficient and the maximum mutual information coefficient.

[0014] Preferably, the real-time monitoring and early warning of water accumulation risk in the tunnel section includes: if the risk assessment value is greater than or equal to a preset threshold, then a flood control emergency warning is triggered; otherwise, a flood control emergency warning is not triggered.

[0015] This application has at least the following beneficial effects: This application introduces a trigger condition mechanism, activating the strategy only when data initially indicates a potential risk, significantly saving computational resources. Calculating the first evaluation value for each monitoring period quantifies the volatility of water level changes over time, enabling in-depth diagnosis of the dynamic stability of the water surface and capturing irregular water level changes to assess potential flooding risks. Calculating the second evaluation value for each monitoring period considers the overall rate of water level change and spatial gradient characteristics from a spatial perspective, providing early warning of the risk of rapid local water level rises due to structural characteristics. This yields the results for each monitoring period. The risk coefficient has the beneficial effect of comprehensively assessing the overall level of water accumulation risk from multiple dimensions, more comprehensively depicting the overall risk status of tunnel water level, and avoiding one-sided warnings caused by large fluctuations in water surface but no upward trend in water level, thereby improving the overall reliability of risk assessment. Determining the risk assessment value for each monitoring period to monitor and warn of water accumulation risk in the tunnel section in real time has the beneficial effect of considering the dynamic correlation between water accumulation risk and rainfall, capturing the time lag pattern between rainfall and water level response, greatly improving the credibility and urgency of the warning, and achieving early and accurate warning of water accumulation risk in the tunnel section. Attached Figure Description

[0016] The following section provides a more detailed description of the real-time station water level monitoring method for flood prevention and early warning in rail transit, based on the accompanying drawings.

[0017] Figure 1 A flowchart illustrating the steps of a real-time station water level monitoring method for flood prevention and early warning in rail transit, provided in an embodiment of this application. Figure 2 A flowchart illustrating the steps of the method for obtaining the second evaluation value provided in this application embodiment. Detailed Implementation

[0018] The following, in conjunction with the accompanying drawings and implementation examples, provides a more detailed description of the real-time station water level monitoring method proposed in this application for flood prevention and early warning in rail transit.

[0019] Please see Figure 1 The document illustrates a flowchart of a method for real-time monitoring of station water levels for flood prevention and early warning in rail transit, according to an embodiment of this application. The method includes the following steps: Step 1: Obtain the water level data of each monitoring point at different times in each monitoring cycle in the tunnel between adjacent stations, as well as the rainfall in that monitoring cycle.

[0020] Because energy efficiency is considered in the longitudinal profile design of railway tracks, the tunnels between adjacent stations are typically concave in shape, utilizing downhill acceleration when leaving the station and uphill deceleration when entering the station to save energy. However, with frequent extreme weather events and increased urban flooding risks, these concave tunnels, due to their energy-saving design, are vulnerable to water accumulation and retention during heavy rainfall. Excessive water accumulation can directly lead to train stoppages, equipment damage, and other safety incidents.

[0021] Based on the above analysis, multiple monitoring points are set up in the tunnel between two adjacent stations. Ultrasonic liquid level sensors are installed at the same height above the track surface at different monitoring points to collect water level data at each monitoring point at different times in real time. In this embodiment, the interval between adjacent monitoring points is 2m, and the data acquisition time interval is 0.1s. As for other implementation methods, the implementer can set them according to the actual situation.

[0022] When there is no water accumulation, the water level data is 0; when the water level data of at least one monitoring point exceeds the preset trigger threshold for multiple consecutive times, a risk assessment strategy is executed. In this embodiment, an early warning analysis is performed if the water level data continuously exceeds a preset threshold for 30 seconds. The preset threshold is set to 0. In other implementation methods, the implementer can set the threshold according to the actual situation.

[0023] Multiple consecutive moments are treated as a monitoring cycle to obtain water level data at each monitoring point at each moment in each monitoring cycle; real-time rainfall in the area where the tunnel is located in the monitored section is extracted by meteorology, and the rainfall is aligned to the same monitoring cycle by time, thus obtaining the rainfall for each monitoring cycle; In this embodiment, the monitoring period is 5 minutes. As for other implementation methods, the implementer can set the duration according to the actual situation.

[0024] Thus, we obtained the water level data of each monitoring point at different times during each monitoring cycle, as well as the rainfall in the area where the tunnel is located during each monitoring cycle.

[0025] Step 2: For each monitoring period, analyze the fluctuation and discrete characteristics of the water level data at the monitoring point to evaluate the fluctuation characteristics of water level changes and calculate the first evaluation value for each monitoring period.

[0026] Furthermore, when the risk of water accumulation on the track is high, the water level data will show a significant upward trend. Simultaneously, external disturbances, such as water flow, ripples generated by train movement, and large floating objects, can cause irregular fluctuations in the water level, with the irregularities becoming more pronounced at higher water levels. Therefore, by analyzing the fluctuations in water level data within the monitoring period, a first assessment value is calculated, specifically: For each monitoring point, obtain the extreme points of water level data at all times within each monitoring cycle; In this embodiment, the AMPD (Automatic multiscale-based peak detection) algorithm is used to obtain extreme points, including maximum and minimum points. The AMPD algorithm is a well-known technology and will not be described in detail here.

[0027] The difference between the corresponding water level data of two adjacent extreme points is taken as the relative difference; In this embodiment, the absolute value of the difference between the corresponding water level data between two adjacent extreme points is taken as the relative difference.

[0028] Calculate the dispersion and average value of all relative differences at each monitoring point within each monitoring period; In this embodiment, the degree of dispersion is measured by calculating the permutation entropy of all relative differences within each monitoring period. The embedding dimension of the permutation entropy is set to 5 and the delay time is set to 1. As other implementations, implementers can set it according to the actual situation. The calculation of permutation entropy is a well-known technique and will not be described in detail here. As other implementations, implementers may also use other methods of the prior art, such as fractal dimension, etc. This embodiment does not impose any special restrictions on this.

[0029] The first evaluation value of each monitoring cycle is positively correlated with the dispersion of all monitoring points and the average value; It should be noted that a positive correlation means that the dependent variable increases as the independent variable increases.

[0030] In this embodiment, the specific process of forward fusion is as follows: the average value is normalized using the maximum-minimum normalization method, the product of the normalized average value and the degree of dispersion is calculated, and the average of the products of all monitoring points in each monitoring period is used as the first evaluation value; as another implementation method, the implementer may also use the sum of the normalized average value and the degree of dispersion, and the average of the sum of all monitoring points in each monitoring period, as the first evaluation value. The maximum-minimum normalization method is a well-known technology and will not be described in detail here.

[0031] It should be noted that the greater the relative difference, the larger the drop in water level during a single event, and the more violent the fluctuations; the greater the dispersion, the more chaotic and irregular the water level fluctuations are. Continuous high-risk water accumulation, due to the long-term complex turbulent state of the water surface, exhibits random and disordered fluctuations, reflecting the more significant irregular fluctuation characteristics of the water level data; the larger the average value, the higher the average intensity of the water level fluctuations; and the larger the obtained first assessment value, the higher the amplitude and irregularity of the water level fluctuations, reflecting the more unstable the water level state of the track.

[0032] Thus, the first evaluation value for each monitoring cycle is obtained.

[0033] Step 3: By analyzing the overall trend of water level data at the monitoring points and the spatial distribution differences of this trend at different monitoring points, the trend characteristics of water level changes are assessed, and a second assessment value is calculated for each monitoring period.

[0034] Furthermore, once water accumulates on the track, the overall water level change is influenced by both the rate of water accumulation and the drainage efficiency of the drainage system, resulting in different trends. The higher the rate of water accumulation or the greater the risk of drainage blockage due to floating debris, the higher the risk of a continuous increase in water level.

[0035] Secondly, to improve the energy efficiency of train operation, the tunnels between adjacent stations are usually designed with a concave structure. This facilitates the train's acceleration downhill when leaving the station and deceleration uphill when entering. The concave structure means that the track is lower in the middle and higher at both ends. Due to gravity and the gradient design, water will naturally converge towards the lowest point. Therefore, the rate of water level change exhibits a certain spatial gradient characteristic. Thus, in situations with a high risk of flooding, the rate of water level change at the monitoring point in the middle will be faster than that at the monitoring points on both sides.

[0036] Based on the above analysis, a second evaluation value is calculated by analyzing the overall water level change trend along the track and the spatial distribution differences in water level change trends at different monitoring points. The flowchart of the method for obtaining the second evaluation value provided in this embodiment is as follows: Figure 2 As shown, it specifically includes: For each monitoring point, the slope estimate is obtained by using the trend estimation method based on the water level data at all times within each monitoring cycle; In this embodiment, Sen's trend estimation method is used to obtain the slope estimate. Sen's trend estimation method is a well-known technique and will not be described in detail here.

[0037] It should be noted that a positive slope estimate indicates that the water level data is gradually increasing, while a negative slope indicates that the water level data is gradually decreasing.

[0038] The slope estimates of all monitoring points in each monitoring period are positively fused to form the overall trend value for each monitoring period. In this embodiment, the specific process of forward fusion is as follows: the mean of the slope estimates of all monitoring points in each monitoring period is taken as the overall trend value. In other implementation methods, the sum of the slope estimates of all monitoring points in each monitoring period can also be taken as the overall trend value.

[0039] Mark the monitoring point located in the middle of the tunnel section as the target point; For each monitoring period, the differences in slope estimates between the target point and all other monitoring points are positively fused to serve as the trend difference for each monitoring period. In this embodiment, the specific process of forward fusion is as follows: the average of the differences between the slope estimates between the target point and all other monitoring points is used as the trend difference for each monitoring period. In other implementations, the sum of the differences between the slope estimates between the target point and all other monitoring points can also be used as the trend difference for each monitoring period.

[0040] The second assessment value for each monitoring period is positively correlated with the overall trend value and the trend difference. In this embodiment, the normalized result of the sum of the overall trend value and the trend difference is calculated and positively mapped to serve as the second evaluation value for each monitoring period. The normalization process uses the maximum-minimum normalization method. The specific process of positive mapping is as follows: an exponential function is used for positive mapping. Let the normalized result be denoted as... ,but The result is used as the second evaluation value, where, Let be an exponential function with the natural constant as its base. Through the process of positive mapping, the result of the positive mapping is made to be greater than 0.

[0041] It should be noted that the larger the overall trend value, the faster the water level rises in the entire tunnel section during the monitoring period, and the higher the overall risk of water accumulation. The larger the trend difference, the higher the rate of water level change at the center and other monitoring points, reflecting a more significant convergence effect of water accumulation in the middle. The larger the second assessment value, the more obvious the gradient pattern of rising overall water level and decreasing rate of change from the middle to both sides during the monitoring period, indicating a higher risk of water accumulation.

[0042] This yields the second evaluation value for each monitoring cycle.

[0043] Step 4: Based on the first and second assessment values, obtain the risk coefficient for each monitoring period, and assess the correlation characteristics of the risk coefficients and rainfall for multiple consecutive monitoring periods to determine the risk assessment value for each monitoring period, so as to monitor and warn of the water accumulation risk in the tunnel in real time.

[0044] Furthermore, the higher the risk of water accumulation on the track, the more pronounced the irregular fluctuations in water level data, the overall rate of water level increase, and its spatial gradient characteristics become. Therefore, the risk coefficient is determined using the first and second assessment values, specifically as follows: The first and second assessment values ​​are positively integrated to obtain the risk coefficient for each monitoring period; In this embodiment, the specific process of forward fusion is as follows: the first evaluation value and the second evaluation value are normalized using the maximum-minimum normalization method respectively. A weighted sum of the normalized first evaluation value and the normalized second evaluation value is then calculated using a preset first weight and a preset second weight, which serves as the risk coefficient for each monitoring period. The sum of the preset first weight and the preset second weight is 1. In this embodiment, considering that the overall water level trend has a higher contribution to reflecting the risk of water accumulation, the preset first weight is less than the preset second weight. Specifically, the preset first weight is 0.4 and the preset second weight is 0.6. In other implementation methods, the implementer can set these values ​​according to the actual situation.

[0045] It should be noted that the higher the risk coefficient, the greater the likelihood of water accumulation risk during the monitoring period.

[0046] Secondly, the rise in water level is mainly affected by local heavy rainfall, and there is a certain lag between water level changes and rainfall. For example, after heavy rainfall begins, surface runoff in the catchment area needs time to collect and flow into the track section, or groundwater infiltration needs time to saturate and manifest as a rise in water level. Therefore, by analyzing the correlation between the risk coefficient and rainfall, the risk assessment value is determined, specifically as follows: Calculate the risk coefficient and the maximum mutual information coefficient of rainfall for each monitoring period and several preceding monitoring periods; In this embodiment, the risk coefficient and the maximum mutual information coefficient of rainfall for each monitoring cycle and the preceding 30 monitoring cycles are calculated. The calculation of the maximum mutual information coefficient is a well-known technique and will not be described in detail here.

[0047] The risk assessment value, risk coefficient, and maximum mutual information coefficient for each monitoring period are all positively correlated. In this embodiment, the maximum mutual information coefficient is used as the gain weight, and its sum with the value 1 is calculated. The product of the sum and the risk coefficient is used as the risk assessment value for each monitoring period.

[0048] It should be noted that the larger the maximum mutual information coefficient, the more synchronous and correlated the fluctuation of track water accumulation risk is with the fluctuation of rainfall, reflecting that the track water accumulation status is more significantly affected by rainfall; the larger the risk assessment value, the higher the water accumulation risk during the monitoring period, and the more significant the impact of rainfall on the risk, which may require emergency measures to deal with the water accumulation problem, and the more likely a flood prevention warning should be triggered.

[0049] Furthermore, based on risk assessment values, the risk of water accumulation in the station's tunnels is monitored and early warnings are issued, specifically as follows: If the risk assessment value is greater than or equal to the preset threshold, a flood control emergency warning will be triggered; otherwise, a flood control emergency warning will not be triggered. In this embodiment, the process of obtaining the preset threshold is as follows: collect water level data when the track experienced flooding disasters in historical periods, calculate the risk assessment value according to the above process, and use it as the preset threshold.

[0050] When a flood control emergency warning is triggered, it indicates that there is serious water accumulation in the station tracks, and corresponding protective measures must be taken in a timely manner.

Claims

1. A method for real-time monitoring of station water levels for flood prevention and early warning in rail transit, characterized in that, The method includes the following steps: Obtain water level data at different times for each monitoring point in the tunnel between adjacent stations during each monitoring cycle, as well as the rainfall during that monitoring cycle; When the water level data at the monitoring point meets the preset trigger conditions, a risk assessment strategy is executed, the strategy including: For each monitoring period, the fluctuation and discrete characteristics of the water level data at the monitoring points are analyzed to evaluate the fluctuation characteristics of water level changes and calculate the first evaluation value for each monitoring period. The overall trend of water level changes at monitoring points, and the spatial distribution differences of this trend at different monitoring points, are used to assess the trend characteristics of water level changes and calculate the second assessment value for each monitoring period. Based on the first and second assessment values, the risk coefficient for each monitoring period is obtained, and the correlation characteristics of the risk coefficients and rainfall for multiple consecutive monitoring periods are evaluated to determine the risk assessment value for each monitoring period, so as to monitor and warn of the water accumulation risk in the tunnel in real time. The triggering condition is: the water level data of at least one monitoring point exceeds the preset activation threshold for multiple consecutive moments; The calculation of the first evaluation value for each monitoring cycle includes: For each monitoring point, the extreme points of water level data at all times within each monitoring cycle are obtained, and the difference between the corresponding water level data between two adjacent extreme points is taken as the relative difference; the dispersion and average value of all relative differences for each monitoring point within each monitoring cycle are calculated. The first evaluation value is positively correlated with the dispersion of all monitoring points and the average value; The calculation of the second evaluation value for each monitoring cycle includes: For each monitoring point, the slope estimate is obtained by using the trend estimation method based on the water level data at all times within each monitoring cycle; the overall level of the slope estimate for different monitoring points under each monitoring cycle is evaluated, and the overall trend value for each monitoring cycle is calculated. Analyze the differences in slope estimates at different monitoring points and calculate the trend difference for each monitoring period; The second assessment value is positively correlated with both the overall trend value and the trend difference. The calculation process for the trend difference is as follows: the monitoring point located at the middle position in the interval tunnel is marked as the target point; for each monitoring period, the difference between the slope estimates of the target point and all other monitoring points is positively fused to serve as the trend difference for each monitoring period. The risk coefficient is the result of a positive fusion of the first assessment value and the second assessment value; The correlation feature is measured by the risk coefficient of each monitoring period and the maximum mutual information coefficient of rainfall over multiple monitoring periods prior to that period.

2. The real-time station water level monitoring method for flood prevention and early warning in rail transit as described in claim 1, characterized in that, The overall trend value is the result of a positive fusion of the slope estimates of all monitoring points in each monitoring period.

3. The real-time station water level monitoring method for flood prevention and early warning in rail transit as described in claim 1, characterized in that, The risk assessment value is positively correlated with the risk coefficient and the maximum mutual information coefficient.

4. The real-time station water level monitoring method for flood prevention and early warning in rail transit as described in claim 1, characterized in that, The real-time monitoring and early warning of water accumulation risk in the tunnel section includes: if the risk assessment value is greater than or equal to a preset threshold, a flood control emergency warning is triggered; otherwise, a flood control emergency warning is not triggered.

Images