A method and system for safety assessment of a dike

Abstract

The application discloses a kind of safety evaluation method and safety evaluation system of dike engineering, belong to dike engineering safety technical field.The method includes: obtaining the temperature variation data of the dike region to be monitored;According to temperature variation data, the temperature gradient distribution in the dike is calculated, the historical data of seepage field state are combined, the delay influence of temperature gradient distribution to seepage field state is analyzed;Distributed sensor network is deployed in the key area of dike, real-time monitoring the change of seepage field state, obtains high-precision seepage field state data;According to the piping signal extracted from high-precision seepage field state data, and combining temperature gradient distribution and seepage field state data, piping path is predicted through random forest model;Based on the prediction result of temperature variation data, seepage field state data and piping path, the probability of piping occurrence and early warning level are predicted.The application can more comprehensively evaluate piping risk by integrating multi-source data, reduce the misjudgment possibly brought by single index.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of embankment engineering safety, and particularly relates to a safety evaluation method and a safety evaluation system for embankment engineering. BACKGROUND

[0002] A dam refers to a water retaining structure built along the edge of a river, a canal, a lake, a sea coast, or a flood discharge area, a flood diversion area, a reclamation area, etc. It is mainly used to effectively resist floods during flood periods, protect the safety of life and property and normal production and life of downstream residents, and is of great significance to regional economic development. Therefore, dam safety monitoring has received high attention and emphasis. Due to the complex and variable seepage path of the dam, the traditional monitoring method mainly relies on manual patrol and investigation, which is not only low in efficiency, but also cannot be fully covered.

[0003] The prior art CN115713187A discloses a dam monitoring and analyzing system based on multi-sensor technology, which obtains dam data through soil humidity sensors, displacement sensors, crack sensors and temperature sensors, calculates and analyzes the dam data to obtain dam state information, realizes multi-dimensional monitoring of the soil water content, the wetting line, the deformation and the internal leakage of the dam, improves the efficiency of dam patrol and analysis processing, and more comprehensively and quickly obtains the dam state and feeds back to the management center and the maintenance personnel. In addition, the dam state information can be regularly and automatically detected to determine whether the dam has abnormal wetting line, dangerous water content, landslide and piping risk, find the dam safety hazard, trigger an alarm and send it to the management center and the dam maintenance personnel, realize man-machine interaction, and timely monitor the dam and prevent danger. Although this method can realize intelligent and real-time monitoring of the dam, it can only predict the severity of the piping risk, but cannot predict the time window of piping occurrence, nor can it make further prediction of the future change of piping. SUMMARY

[0004] The main purpose of the present application is to provide a safety evaluation method and a safety evaluation system for embankment engineering to solve the problem that the prior art cannot predict the time window of piping occurrence, nor can it make further prediction of the future change of piping.

[0005] In order to achieve the above-mentioned purpose, the present application provides a safety evaluation method for embankment engineering, comprising the following steps:

[0006] Obtaining temperature change data of the embankment area to be monitored;

[0007] According to the temperature change data, calculating the temperature gradient distribution inside the embankment, combining the historical data of the seepage field state, and analyzing the time delay effect of the temperature gradient distribution on the seepage field state;

[0008] A distributed sensor network is deployed in critical areas of the embankment to monitor real-time changes in the seepage field state and obtain high-precision seepage field state data.

[0009] Based on the piping signal extracted from the high-precision seepage field state data, combined with the temperature gradient distribution and the seepage field state data, a random forest model is used to predict the piping path.

[0010] Based on the temperature change data, the seepage field state data, and the prediction results of the piping path, the probability of piping occurrence and the warning level are predicted.

[0011] Further, the temperature change data of the embankment area to be monitored includes installing temperature sensors at different depths and positions of the embankment to form a temperature monitoring grid, monitoring the internal and surface temperatures of the embankment in real time, and obtaining the temperature change data of the embankment area.

[0012] Further, the historical data of the seepage field state includes seepage pressure, flow rate, and flow volume.

[0013] Further, the time delay effect of the temperature gradient distribution on the seepage field state is represented by the following formula:

[0014] ;

[0015] where R(t) represents the cumulative effect, K(τ) is a kernel function describing the effect of temperature gradient, and t is time.

[0016] Further, a distributed sensor network is deployed in critical areas of the embankment to monitor real-time changes in the seepage field state and obtain high-precision seepage data, including:

[0017] Pressure sensors are installed in critical areas including embankment slopes, embankment tops, and embankment feet to monitor changes in pore water pressure and determine seepage paths and intensities.

[0018] Flow sensors are installed at different depths inside the embankment to measure water flow speed and direction and identify potential leakage points.

[0019] Further, the piping signal includes abnormal increase in soil pore water pressure, sudden change in seepage speed, or sharp rise in seepage water volume.

[0020] Further, the input of the random forest model is ;

[0021] where X represents the input feature matrix, T represents the temperature gradient distribution data, F represents the seepage field state data, and S represents the piping signal data.

[0022] The output of the random forest model is ;

[0023] wherein H(x) represents the prediction result of piping path of the random forest model, the piping path prediction result includes potential area and time window of piping, M represents the number of decision trees, h i (x) represents the prediction result of the i-th decision tree.

[0024] Further, based on the temperature change data, the seepage field state data and the prediction result of the piping path, the probability of piping occurrence and the warning level are predicted, including: establishing a warning system based on time series analysis, the warning system including an autoregressive integrated moving average model, a principal component analysis model and a hidden Markov model;

[0025] The temperature change data, the seepage field state data and the prediction result of the piping path are taken as input quantities of the warning system, and the probability of piping occurrence and the warning level are calculated through designed fuzzy rules.

[0026] Further, the autoregressive integrated moving average model is used for time series analysis of the temperature change data, and the specific analysis includes: predicting the predicted temperature in the future predetermined time according to the temperature change data, and judging whether the predicted temperature is continuously greater than a preset threshold.

[0027] Further, the principal component analysis model is used for analysis of the seepage field state data, and the specific analysis includes: performing dimension reduction processing on the seepage field state data, extracting the most representative features as principal components to represent the seepage state of different positions of the dam, and judging the change of seepage pressure at the corresponding position according to the change of the score of the principal component.

[0028] Further, the hidden Markov model is used for time series analysis of the prediction result of the piping path, and the specific analysis includes: dividing the piping development process into three stages of incubation period, development period and acceleration period according to its time window, and learning the features and transition probability of the three stages according to historical data to predict the probability of stage migration.

[0029] Another aspect of the present application also provides a safety evaluation system for implementing the safety evaluation method of the aforementioned embankment engineering, comprising:

[0030] A temperature change acquisition module is configured to acquire temperature change data of an embankment region to be monitored.

[0031] A time delay influence analysis module is configured to calculate temperature gradient distribution inside the embankment according to the temperature change data, and analyze the time delay influence of the temperature gradient distribution on the seepage field state in combination with historical data of the seepage field state.

[0032] The seepage field state data acquisition module is used for deploying a distributed sensor network in a key area of the embankment, monitoring changes in the seepage field state in real time, and acquiring high-precision seepage field state data.

[0033] The piping path prediction module is used for predicting a piping path according to a piping signal extracted from the high-precision seepage field state data, in combination with the temperature gradient distribution and the seepage field state data, through a random forest model.

[0034] The piping early warning module is used for predicting a piping occurrence probability and an early warning level based on the temperature change data, the seepage field state data, and a prediction result of the piping path.

[0035] Compared with the prior art, the present application has the following beneficial effects:

[0036] 1. The present application can comprehensively evaluate piping risks, reduce possible misjudgments of a single index, and improve prediction accuracy by comprehensively processing multiple source data, including temperature change data, seepage field state data, piping signal data, and the like.

[0037] 2. The present application can discover the correlation and time lag between the temperature gradient and the seepage field state by analyzing the time delay effect of the temperature gradient on the seepage field state in combination with historical data of the seepage field state, comparing time series of the temperature gradient change and the seepage field state change, and capturing abnormalities at an early stage of piping to provide early warning of piping occurrence.

[0038] 3. The present application can timely identify potential piping risk areas and time windows by processing multiple input factors through a random forest model, which has the characteristics of strong noise resistance and not being prone to overfitting, and is suitable for processing complex data in engineering practice.

[0039] 4. The present application can capture dynamic changes of each index by constructing an early warning system based on time series analysis, not only focusing on the current state, but also considering historical trends and future predictions, so that the output of the early warning level provides intuitive and easy-to-understand risk assessment results for decision makers, which is helpful for timely taking corresponding prevention and control measures. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 A flowchart of a safety evaluation method of an embankment engineering in a specific embodiment of the present application is shown;

[0041] Figure 2 A flowchart of random forest model training in a specific embodiment of the present application is shown;

[0042] Figure 3 A flowchart of piping path prediction through a random forest model in a specific embodiment of the present application is shown. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0044] To solve the problems in the prior art, the present application discloses a safety evaluation method for embankment engineering, as shown in Figure 1 The method comprises the following steps:

[0045] In step S101, temperature change data of the embankment area to be monitored is obtained.

[0046] Further, step S101 comprises: installing temperature sensors at different depths and positions of the embankment to form a temperature monitoring grid, and monitoring the internal and surface temperatures of the embankment in real time to obtain the temperature change data of the embankment area.

[0047] The acquisition of the temperature change data of the embankment area is an important part of flood control management. By deploying a network of temperature sensors, the internal and surface temperatures of the embankment can be monitored in real time. These sensors are usually installed at different depths and positions of the embankment to form a temperature monitoring grid. For example, on a 100-meter-long embankment, a set of sensors can be installed every 10 meters, with each set containing three measuring points: surface, middle and deep layers. The data collected by the sensors is transmitted to the data center via a wireless network to construct a temperature distribution map of the embankment.

[0048] This temperature-based piping warning method has the advantages of non-invasiveness and real-time. Compared with traditional regular patrols, an automated temperature monitoring system can work 24 hours a day without interruption, capturing subtle changes in the embankment state in a timely manner. This not only improves monitoring efficiency, but also gives management personnel more emergency response time, effectively reducing the risk of flood disasters.

[0049] In step S102, the temperature gradient distribution inside the embankment is calculated according to the temperature change data, and the delayed influence of the temperature gradient distribution on the seepage field state is analyzed in combination with historical data of the seepage field state.

[0050] Specifically, the calculation of the temperature gradient distribution inside the embankment is crucial to understanding the seepage dynamics. The temperature gradient reflects the direction and intensity of heat transfer and can be calculated from the data obtained by the temperature sensor network. For example, in a 100-meter-long embankment, a set of temperature sensors is arranged every 10 meters, with each set containing three measuring points: surface, middle and deep layers. Through the temperature data of these measuring points, the vertical and horizontal temperature gradients can be calculated. The temperature gradient formula is: where ΔT represents the temperature gradient, T2 and T1 are the temperatures of the two measurement points, and d is the distance between the two points. This formula can be used to obtain the temperature gradient distribution inside the dike.

[0051] The vertical temperature gradient can reflect the seepage conditions inside the dike. Under normal circumstances, the surface temperature of the dike is greatly affected by the external air temperature, while the deep temperature is relatively stable. If an abnormal increase or decrease in deep temperature is observed, it may indicate the presence of a seepage channel. For example, in a normal summer state, the surface temperature of a dike is 30°C, the deep temperature is 20°C, and the vertical temperature gradient is 1°C / m. If the deep temperature suddenly drops to 15°C and the vertical temperature gradient increases to 1.5°C / m, it may indicate that cold water is seeping in. The horizontal temperature gradient can be used to locate the location of seepage occurrence. If local temperature anomalies are observed on a certain horizontal section of the dike, it may indicate the presence of a seepage channel.

[0052] For example, on a certain horizontal plane in the middle of the dike, if the temperature of most of the area remains at 22°C, and the temperature of a 10-meter section suddenly drops to 18°C, forming a horizontal temperature gradient of 0.4°C / m, it may indicate that there is a seepage problem. By analyzing the time delay effect of temperature gradient on seepage field state based on historical data, the risk of piping can be better predicted. Historical data of seepage field state usually includes parameters such as seepage pressure, flow rate, and flow.

[0053] In this scheme, by comparing the time series of temperature gradient changes and seepage field state changes, the correlation and time lag between the two can be found. For example, by analyzing data from the past 5 years, it is found that significant changes in temperature gradient usually lead changes in seepage field state by 12 to 24 hours. Specifically, when the vertical temperature gradient of a certain place in the dike increases by 50% within 6 hours, the seepage pressure at that place increases by 30% within the next 18 hours. This correlation provides a valuable time window for the early warning system. The time delay effect of temperature gradient distribution on seepage field state is represented by the following formula: where R(t) represents the cumulative effect, K(τ) is the kernel function describing the effect of temperature gradient, and t is the time.

[0054] The time delay effect of temperature gradient on seepage field state may be due to the thermodynamic properties of soil. When seepage occurs, the water flow will gradually change the temperature distribution of the soil. However, due to the heat capacity and heat conduction characteristics of the soil, this temperature change takes a certain amount of time to fully manifest itself. Therefore, changes in temperature gradient often precede significant changes in seepage field state. By utilizing this time delay effect, a more accurate early warning model can be constructed.

[0055] In addition, temperature gradient analysis can also help identify the development trend of seepage. If an abnormal area of temperature gradient is observed to gradually expand or increase in intensity, it may mean that the seepage channel is expanding. For example, a certain embankment monitoring found a 5-meter-wide temperature anomaly area with a temperature gradient of 0.5°C / meter. In the next week, this abnormal area expanded to 10 meters wide, and the temperature gradient increased to 0.8°C / meter. This trend of change indicates that the seepage problem is worsening, and timely intervention measures need to be taken.

[0056] Step S103, deploy a distributed sensor network in the key area of the embankment to monitor the changes in the seepage field in real time and obtain high-precision seepage field data.

[0057] Specifically, the deployment of a distributed sensor network in the key area of the embankment is an effective method for monitoring changes in the seepage field. This method can obtain high-precision seepage data in real time, providing key support for embankment safety management. In practical applications, various types of sensors can be selected, such as pressure sensors, flow sensors, etc., which are arranged at different positions and depths of the embankment. For example, pressure sensors are installed on the embankment slope, embankment top, and embankment foot to monitor changes in pore water pressure and determine the seepage path and intensity. Flow sensors are installed at different depths inside the embankment to measure water flow velocity and direction and identify potential leakage points.

[0058] These sensors form a network through wireless communication technology and transmit the collected data to the central control system in real time. Data transmission can use low-power wide-area network (LPWAN) technology such as LoRa or NB-IoT to ensure stable communication in harsh environments. The central control system uses high-performance computing devices and professional software to process and analyze the massive data received. During data analysis, various advanced algorithms can be applied. For example, time series analysis can be used to identify trends in seepage parameters; spatial interpolation techniques can be used to construct a three-dimensional seepage field model inside the embankment; and machine learning algorithms can be used to predict future changes in seepage state. These analysis results provide engineers with intuitive and comprehensive assessments of the embankment seepage situation.

[0059] The acquisition of high-precision seepage field data is of great significance for embankment safety management. Through real-time monitoring, abnormal seepage phenomena such as piping or leakage can be detected in a timely manner, so that appropriate protective measures can be taken. For example, when the pore water pressure in a certain area detected by the pressure sensor suddenly rises and exceeds the preset threshold, the system will immediately issue an alarm to remind the staff to conduct on-site inspection and treatment. In addition, the long-term accumulation of high-precision seepage data can also be used to optimize embankment design and maintenance strategies. By analyzing the seepage characteristics under different seasons and water level conditions, engineers can better understand the hydraulic performance of the embankment, so as to adopt more reasonable design parameters and construction methods in future projects. For example, it may be found that a certain specific embankment structure is more prone to seepage problems under certain hydrological conditions, which will guide future embankment reconstruction work. In summary, the use of distributed sensor networks to monitor embankment seepage field can not only improve monitoring accuracy and real-time performance, but also provide reliable data support for embankment safety management and engineering optimization, thereby significantly improving the overall safety and operational efficiency of the embankment.

[0060] Step S104, according to the piping signal extracted from the high-precision seepage field data, and combining the temperature gradient distribution and the seepage field data, the piping path is predicted by the random forest model.

[0061] Specifically, the present scheme mainly analyzes high-precision seepage data, temperature gradient distribution and seepage field data to predict the occurrence of piping. Piping is a major threat to embankment safety, and its occurrence is often accompanied by significant changes in seepage field conditions.

[0062] First, extracting piping signals from high-precision seepage data is a key step. These signals may include abnormal increases in soil pore water pressure, sudden changes in seepage velocity, or sharp increases in seepage water volume. For example, in a certain embankment monitoring, it was found that the pore water pressure in a local area increased from 20 kPa to 35 kPa in a short period of time, which may indicate the early signs of piping.

[0063] In the present scheme, as shown in Figure 2 Before using the random forest model for prediction, the random forest model needs to be trained for piping path identification, and the training process includes the following steps:

[0064] Select a random forest model;

[0065] In this embodiment, the random forest model can contain 100 decision trees, each tree using a randomly selected subset of features. Through cross-validation, the best parameter combination is selected, such as setting the depth of the tree to 8 and the minimum number of samples per node to 5.

[0066] Prepare historical temperature gradient distribution data, seepage field state data, piping signal data, and their corresponding piping results (including piping areas and time windows);

[0067] Then, the historical temperature gradient distribution data, seepage field state data, and piping signal data are used as inputs, and the corresponding piping results are used as outputs to train the random forest model, obtaining a trained random forest model.

[0068] After training, the model can be used for prediction of new data. Suppose the monitoring data of a certain section of the embankment is input: the temperature gradient is 0.05℃ / m, the pore water pressure increases by 15%, and the frequency of the detected acoustic signal is 60kHz. The model may output a 90% probability that the area will experience piping within the next 24 hours and provide possible seepage paths. The advantage of this method is that it can consider multiple factors and capture the complex relationships between them. For example, the combination of temperature anomalies and pressure changes may be more accurate in predicting piping than a single indicator. At the same time, the random forest model has strong noise resistance and is not prone to overfitting, making it suitable for handling complex data in engineering practice. Through this method, potential piping risk areas can be identified in a timely manner, providing a scientific basis for engineering decision-making. For example, after predicting high-risk areas, the frequency of patrols can be increased, or temporary reinforcement measures can be taken. In the long term, this prediction model can also guide the optimization of embankment design and maintenance strategies, improving the overall safety and economy of the project.

[0069] Specifically, the input of the random forest model is ; where X represents the input feature matrix, T represents the temperature gradient distribution data, F represents the seepage field state data, and S represents the piping signal data. The output of the random forest model is ; where H(x) represents the prediction result of the piping path of the random forest model, M represents the number of decision trees, and h i (x) represents the prediction result of the i-th decision tree. The random forest obtains the final result through the average prediction of multiple decision trees, i.e., the final prediction result of the piping path, including the potential area of piping and the time window. The prediction of the piping path by the random forest model is shown in Figure 3 , including the following steps: extracting piping signals; analyzing temperature gradients; studying seepage field states; determining potential areas; and determining time windows.

[0070] Step S105, based on the temperature change data, the seepage field state data, and the prediction result of the piping path, predicting the probability of piping occurrence and the warning level.

[0071] Further, step S105 includes establishing a warning system based on time series analysis, which includes an autoregressive integrated moving average model, a principal component analysis model, and a hidden Markov model;

[0072] The temperature change data, the seepage field state data, and the prediction results of the piping path are used as input quantities of the early warning system, and the occurrence probability of piping and the early warning level are calculated through the designed fuzzy rules.

[0073] Specifically, establishing an early warning system based on time series analysis is a key link in piping risk assessment. The system takes temperature change data, seepage field state data, and piping path prediction results as inputs, and outputs piping occurrence probability and early warning level through comprehensive analysis, providing a scientific basis for flood control decision-making.

[0074] The time series analysis method can use the autoregressive integrated moving average model (ARIMA). ARIMA is used for time series analysis of temperature change data. The specific analysis includes predicting the predicted temperature in the future predetermined time according to the temperature change data, and determining whether the predicted temperature continues to be greater than the preset threshold. This model can capture the trend, seasonality, and random fluctuations of the data, and is suitable for processing non-stationary time series data. Taking temperature change data as an example, suppose the temperature data of a dam monitoring point for 30 consecutive days shows a gradual upward trend, from 20°C to 25°C. The ARIMA model can predict the temperature change in the next 7 days. If the predicted temperature continues to rise to 28°C, it may indicate that seepage activity is intensifying.

[0075] The analysis of seepage field state data can use the principal component analysis (PCA) model. The specific analysis includes: reducing the dimension of the seepage field state data, extracting the most representative features as principal components to represent the seepage state at different positions of the dam, and judging the change of seepage pressure at the corresponding position according to the change of the scores of the principal components. For example, PCA analysis of pressure data from 100 monitoring points inside the dam shows that the first three principal components can explain 85% of the data variance. These principal components may represent the seepage state of the upstream, midstream, and downstream of the dam, respectively. If the score of the first principal component continues to increase, it may indicate that the upstream seepage pressure is increasing.

[0076] The time series analysis of the piping path prediction results can use the hidden Markov model (HMM). This model can describe the transition process of system states. Suppose the piping development process is divided into three stages according to its time window: latent period, development period, and acceleration period. HMM can learn the characteristics and transition probabilities of these three stages from historical data to predict the probability of their stage migration. If the model predicts that the probability of the system transitioning from the latent period to the development period exceeds 60%, it will be an important early warning signal. Based on the above analysis results, a fuzzy logic-based early warning level evaluation system can be constructed.

[0077] The system takes the outputs of ARIMA, PCA and HMM as input variables, and calculates the piping occurrence probability and early warning level through the set fuzzy rules. For example:

[0078] 1. If the temperature prediction shows a stable trend, the principal component score of seepage pressure remains relatively stable, and the piping state is in the incubation period, the system outputs a piping occurrence probability of 20%, and gives a green early warning level, i.e. relatively safe;

[0079] 2. If the temperature prediction shows an upward trend or the principal component score of seepage pressure increases, and the piping state is in the incubation period, the system outputs a piping occurrence probability of 50%, and gives a blue early warning level, i.e. with certain hidden dangers;

[0080] 3. If the temperature prediction shows an upward trend, the principal component score of seepage pressure increases, and the piping state is in the development period, the system may output a piping occurrence probability of 80%, and give an orange early warning level;

[0081] 4. If the temperature prediction shows an upward trend, the principal component score of seepage pressure increases, and the piping state is in the acceleration period, the system may output a piping occurrence probability of 100%, and give a red early warning level.

[0082] It should be noted that the piping occurrence probability and the corresponding early warning level defined by the fuzzy rules, so the fuzzy rules can be set artificially according to actual needs to achieve different early warning effects, which are within the protection scope of the present application.

[0083] The embodiment of the present application also provides a safety evaluation system for implementing the safety evaluation method of the dike engineering.

[0084] A temperature change acquisition module is configured to acquire temperature change data of a dike region to be monitored.

[0085] A time delay influence analysis module is configured to calculate a temperature gradient distribution inside the dike according to the temperature change data, and analyze a time delay influence of the temperature gradient distribution on the seepage field state in combination with historical data of the seepage field state.

[0086] A seepage field state data acquisition module is configured to deploy a distributed sensor network in a key region of the dike, monitor changes in the seepage field state in real time, and acquire high-precision seepage field state data.

[0087] A piping path prediction module is configured to predict a piping path through a random forest model according to a piping signal extracted from the high-precision seepage field state data, in combination with the temperature gradient distribution and the seepage field state data.

[0088] A pipe gushing early warning module is configured to predict a probability of pipe gushing occurrence and an early warning level based on the temperature change data, the seepage field state data and a prediction result of the pipe gushing path.

[0089] The above-described embodiments are merely illustrative of the present application and are described in more detail and specifically, but should not be understood as a limitation on the scope of the patent of the present application. The present application can also be implemented in other specific ways or other specific forms without departing from the spirit or essential characteristics of the present application. Therefore, the described embodiments should be considered illustrative rather than limiting in any aspect. The scope of the present application should be explained by the appended claims, and any changes equivalent to the intent and scope of the claims should be included in the scope of the present application.

Claims

1. A safety assessment method for dike engineering, characterized in that, Includes the following steps: Acquire temperature change data for the levee area to be monitored; Based on the temperature change data, the temperature gradient distribution inside the dike is calculated. Combined with historical data on the seepage field, the time-delay effect of the temperature gradient distribution on the seepage field is analyzed. The time-delay effect of the temperature gradient distribution on the seepage field is expressed by the following formula: Where R(t) represents the cumulative effect, K(τ) is the kernel function describing the effect of the temperature gradient, and t is time; Deploying a distributed sensor network in key areas of the dike enables real-time monitoring of changes in the seepage field and acquisition of high-precision seepage field data. Based on the piping signal extracted from the high-precision seepage field data, and combined with the temperature gradient distribution and the seepage field data, the piping path is predicted using a random forest model. Establish an early warning system based on time series analysis, wherein the early warning system includes an autoregressive integral moving average model, a principal component analysis model, and a hidden Markov model; The autoregressive integral moving average model is used for time series analysis of temperature change data, specifically including: predicting the predicted temperature within a predetermined time period based on the temperature change data, and determining whether the predicted temperature continues to be greater than a preset threshold. The principal component analysis model is used to analyze seepage field data, specifically including: performing dimensionality reduction on the seepage field data, extracting the most representative features as principal components to represent the seepage state at different locations of the dam, and judging the change in seepage pressure at the corresponding location based on the change in the score of the principal components. The Hidden Markov Model is used for time series analysis of the prediction results of piping path. The specific analysis includes: dividing the piping development process into three stages according to its time window: incubation period, development period and acceleration period, and learning the characteristics and transfer probability of these three stages based on historical data to predict the probability of its occurrence stage migration. The temperature change data, the seepage field data, and the predicted piping path are used as inputs to the early warning system. The probability of piping occurrence and the early warning level are calculated using designed fuzzy rules.

2. The safety assessment method for dike engineering according to claim 1, characterized in that, The acquisition of temperature change data for the monitored dike area includes: Temperature sensors are installed at different depths and locations within the dike to form a temperature monitoring grid, allowing for real-time monitoring of the internal and surface temperatures of the dike and obtaining temperature change data for the dike area.

3. The safety assessment method for dike engineering according to claim 1, characterized in that, The historical data of the seepage field includes seepage pressure, flow velocity, and flow rate.

4. The safety assessment method for dike engineering according to claim 1, characterized in that, Deploying a distributed sensor network in key areas of the dike allows for real-time monitoring of changes in the seepage field, acquiring high-precision seepage data, including: Pressure sensors are installed in key areas, including the levee slope, levee top, and levee toe, to monitor changes in pore water pressure and determine seepage path and intensity. Flow sensors are installed at different depths inside the dike to measure the water flow velocity and direction, and to identify potential seepage points.

5. The safety assessment method for dike engineering according to claim 1, characterized in that, The piping signals include an abnormal increase in soil pore water pressure, a sudden change in seepage velocity, or a sharp increase in the amount of seepage water.

6. The safety assessment method for dike engineering according to claim 1, characterized in that, The input to the random forest model is ; Where X represents the input feature matrix, T represents the temperature gradient distribution data, F represents the seepage field state data, and S represents the piping signal data; The output of the random forest model is ; Where H(x) represents the piping path prediction result of the random forest model, which includes the potential area and time window of the piping, M represents the number of decision trees, and h i (x) represents the prediction result of the i-th decision tree.

7. A safety assessment system for implementing the safety assessment method for dike engineering according to any one of claims 1 to 6, characterized in that, include: The temperature change acquisition module is used to acquire temperature change data of the levee area to be monitored. The delay effect analysis module is used to calculate the temperature gradient distribution inside the dike based on the temperature change data, and analyze the delay effect of the temperature gradient distribution on the seepage field state by combining historical data of the seepage field state; the delay effect of the temperature gradient distribution on the seepage field state is expressed by the following formula: Where R(t) represents the cumulative effect, K(τ) is the kernel function describing the effect of the temperature gradient, and t is time; The seepage field data acquisition module is used to deploy a distributed sensor network in key areas of the dike to monitor changes in the seepage field in real time and acquire high-precision seepage field data. The piping path prediction module is used to predict the piping path using a random forest model based on the piping signal extracted from the high-precision seepage field data and in combination with the temperature gradient distribution and the seepage field data. The piping early warning module is used to establish an early warning system based on time series analysis. The temperature change data, the seepage field data, and the predicted piping path are used as inputs to the early warning system. The probability of piping occurrence and the early warning level are calculated through designed fuzzy rules. The early warning system includes an autoregressive integral moving average model, a principal component analysis model, and a hidden Markov model. The autoregressive integral moving average model is used for time series analysis of temperature change data, specifically including: predicting the predicted temperature within a predetermined time period based on the temperature change data, and determining whether the predicted temperature continues to be greater than a preset threshold. The principal component analysis model is used to analyze seepage field data, specifically including: performing dimensionality reduction on the seepage field data, extracting the most representative features as principal components to represent the seepage state at different locations of the dam, and judging the change in seepage pressure at the corresponding location based on the change in the score of the principal components. The Hidden Markov Model is used for time series analysis of the prediction results of piping path. The specific analysis includes: dividing the piping development process into three stages according to its time window: incubation period, development period and acceleration period, and learning the characteristics and transition probabilities of these three stages based on historical data to predict the probability of its occurrence stage migration.

Images