An online monitoring and management system and method for embankment safety based on internet of things

Abstract

The application relates to the technical field of embankment management, and discloses an online monitoring and management system and method for embankment safety based on Internet of Things, which is used to solve the problem that the original fixed risk threshold is no longer representative when embankment safety management is performed. The system comprises the following steps: the monitoring data is safely transmitted to a cloud platform through an Internet of Things communication mode; integrity verification and preprocessing are performed on the received data; the preprocessed monitoring data is obtained; a risk index of an embankment is calculated; first risk judgment is performed according to the risk index of the embankment; if it is judged that the current embankment is at risk, a threshold influence parameter is obtained; a threshold self-adaptive correction index is obtained through evaluation; if it is judged that the initial risk threshold needs to be adjusted, the initial risk threshold is adjusted according to the threshold self-adaptive correction index; an adjusted risk threshold is obtained; second risk judgment is performed on the risk index of the embankment and the adjusted risk threshold; and the accuracy of embankment safety management is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of dike management technology, and more specifically to an online monitoring and management system and method for dike safety based on the Internet of Things. Background Technology

[0002] As the core protective structure in flood control and disaster prevention systems, the safety status of dikes directly affects regional disaster prevention and mitigation capabilities and public safety levels. In recent years, with the development of the Internet of Things (IoT) and intelligent sensing technologies, online dike safety monitoring systems based on multi-source sensors have been gradually applied to key sections of dikes in rivers, lakes, and reservoirs to achieve real-time monitoring of dike seepage, displacement, settlement, and water level changes. By deploying devices such as piezometers, rain gauges, water level gauges, soil moisture sensors, and tilt sensors at key locations on the dike, the system can continuously collect monitoring data from the dike's interior and surrounding environment. Combined with wireless communication networks, this data is transmitted to a cloud platform, forming a visualized monitoring system for dike operation, thereby providing data support for flood control command and maintenance decisions.

[0003] Existing online monitoring systems for dikes generally use a method based on comparing thresholds of monitoring indicators to determine the safety status of the dike. Specifically, this involves real-time collection of various monitoring parameters such as seepage pressure, water level, humidity, displacement, and tilt angle, extracting corresponding risk characteristic indicators, and comparing them with set safety thresholds to identify typical risk scenarios such as abnormal seepage, structural instability, or localized leakage.

[0004] However, the above-mentioned technologies have at least the following technical problems: In real-world operating environments, external dynamic factors directly affect the stability and comparability of risk indicators, causing fixed risk thresholds set under specific environmental conditions to gradually lose their representativeness. If the system continues to use static thresholds for risk assessment, false alarms or missed alarms may occur. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides an online monitoring and management system and method for dike safety based on the Internet of Things, so as to solve the problems existing in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: An Internet of Things (IoT)-based online monitoring and management system for dike safety includes: a data acquisition and transmission module for deploying various types of sensors at key locations on the dike to collect real-time monitoring data of the dike's interior and surrounding environment, including pore water pressure, water level, tilt angle, horizontal displacement, and vertical displacement; a data transmission and encryption module for encrypting, encoding, and packaging the monitoring data, securely transmitting it to a cloud platform via IoT communication, and performing integrity verification and preprocessing on the received data to obtain preprocessed monitoring data; a first risk assessment module for calculating a dike risk index based on the preprocessed monitoring data, obtaining an initial risk threshold, and performing a first risk assessment based on the dike risk index and the initial risk threshold; and a risk threshold impact assessment module for obtaining a threshold if the first risk assessment indicates that the dike is at risk. The system comprises several risk assessment modules. The threshold impact parameters include the input signal change time, the output signal response time, and environmental monitoring parameters. An adaptive threshold correction index is obtained based on these parameters, and the system determines whether the initial risk threshold needs adjustment. A risk threshold adjustment module adjusts the initial risk threshold based on the adaptive threshold correction index if adjustment is deemed necessary. A second risk assessment module compares the levee risk index with the adjusted risk threshold. If the levee risk index is greater than or equal to the adjusted risk threshold, a risk warning is issued; otherwise, no risk warning is issued. A visualization monitoring module transmits monitoring data and warning status to a visualization platform.

[0007] Preferably, the steps of encrypting, encoding, and packaging the monitoring data are as follows: performing data encoding and packetization processing on the monitoring data, packaging the monitoring timestamp, sensor number, and monitoring value into a data packet according to a preset field format; performing lightweight encryption processing on each data packet to obtain an encrypted data packet; transmitting the encrypted data packet, and during the transmission of the encrypted data packet, obtaining the data packet loss rate in real time, subtracting the data packet loss rate from 1 to obtain the communication quality index, and triggering a packet retransmission mechanism when the communication quality index is lower than the quality threshold, retransmitting only the lost or failed data packets.

[0008] Preferably, the steps for obtaining the embankment risk index are as follows: 1) Obtain the pore water pressure inside the embankment, the water level outside the embankment, and the water level inside the embankment; 2) Calculate the seepage risk based on the pore water pressure inside the embankment, the water level outside the embankment, and the water level inside the embankment; 3) Obtain the current horizontal displacement, the previous horizontal displacement, the current inclination angle, the previous inclination angle, and the vertical distance between the sensor and the embankment surface at the monitoring point; 4) Calculate the displacement change rate based on the current horizontal displacement, the previous horizontal displacement, the current inclination angle, the previous inclination angle, and the vertical distance between the sensor and the embankment surface; 5) Obtain the current and previous vertical displacements of the top of the embankment; 6) Calculate the settlement rate based on the current and previous embankment displacements; 7) Multiply the absolute values ​​of the seepage risk, the displacement change rate, and the settlement rate, and take the square root to obtain the embankment risk index.

[0009] Preferably, the step of making the first risk judgment based on the levee risk index and the initial risk threshold is as follows: compare the levee risk index with the initial risk threshold. If the levee risk index is greater than or equal to the initial risk threshold, the first risk judgment is that the levee is at risk. If the levee risk index is less than the initial risk threshold, the first risk judgment is that the levee is not at risk, and continue to collect monitoring data of the levee's interior and surrounding environment through sensors.

[0010] Preferably, the step of obtaining the threshold adaptive correction index is as follows: Within the current detection period, monitoring data is acquired through a sensor, and a data stability offset coefficient is obtained based on the monitoring data; within the current detection period, a response time series is acquired through a sensor, and a sensor time response offset coefficient is obtained based on the response time series; within the current detection period, environmental monitoring parameters are acquired, and an environmental disturbance consistency coefficient is obtained based on the environmental monitoring parameters; the data stability offset coefficient, sensor time response offset coefficient, and environmental disturbance consistency coefficient are normalized, and the threshold adaptive correction index is calculated based on the normalized data stability offset coefficient, sensor time response offset coefficient, and environmental disturbance consistency coefficient. The specific steps are as follows: In the formula, This is expressed as the threshold adaptive correction index. This represents the data stability offset coefficient after normalization. This is represented as the normalized sensor aging response offset coefficient. This is expressed as the environmental disturbance consistency coefficient after normalization. , , The weighting coefficients are represented as the normalized data stability offset coefficient, the normalized sensor time response offset coefficient, and the normalized environmental disturbance consistency coefficient. The steps for obtaining the data stability offset coefficient are as follows: within the current monitoring period, monitoring data is acquired through sensors to construct a monitoring data sequence; the sequence mean and sequence standard deviation are calculated for the monitoring data sequence of each type of sensor; the average rate of change of adjacent sampling points is calculated for the monitoring data sequence of each type of sensor, denoted as the sequence dynamic change rate; the stability offset is calculated based on the sequence mean, sequence standard deviation, and sequence dynamic change rate; the stability offset of each type of monitoring data sequence is obtained, and the stability offset of each type of monitoring data sequence is fused using a product-type composite method to obtain the data stability offset coefficient.

[0011] Preferably, the step of obtaining the sensor time response offset coefficient is as follows: Within the current detection cycle, obtain response time sequences from various sensors deployed along the embankment. For each type of sensor, record the time of input signal change and the time of output signal response to obtain the response time sequence. Simultaneously, obtain the corresponding reference response time and the reference output amplitude during sensor calibration from the sensor's factory or periodic calibration data. Obtain the time for the response signal to reach steady state corresponding to each response time data in the response time sequence of each type of sensor. Interpolate the time for the response signal to reach steady state with the response time data to obtain the response delay for each sampling. Calculate the average response delay for each response delay. For the response time sequence of each type of sensor, calculate the average response delay... The response delay offset rate is obtained by calculating the difference between the response time and the reference response time and then by calculating the ratio between the reference response time and the reference response time. Within the detection period, the peak value of each signal disturbance in the monitoring data sequence is calculated by calculating the difference between the peak value and the corresponding baseline value to obtain the amplitude of each response event. The amplitude of each effective response event within the detection period is averaged to obtain the sequence average amplitude. For the response time sequence of each type of sensor, the amplitude attenuation ratio is obtained by subtracting the ratio of the sequence average amplitude to the reference output amplitude from 1. For each type of sensor, the sensor aging response offset is calculated based on the response delay offset rate and the amplitude attenuation ratio. The aging response offset of each type of sensor is obtained and fused using a product-type composite method to obtain the sensor aging response offset coefficient.

[0012] Preferably, the steps for obtaining the environmental disturbance consistency coefficient are as follows: During the current monitoring period, environmental monitoring sensors are deployed around the dike to acquire environmental monitoring parameters and construct an environmental monitoring sequence; the mean and standard deviation are calculated for each type of environmental monitoring sequence to obtain the environmental monitoring mean and standard deviation; for each type of environmental monitoring sequence, the average rate of change of adjacent sampling points is calculated; the ratio of the environmental monitoring standard deviation to the environmental monitoring mean is calculated to obtain the proportion of disturbance fluctuation intensity relative to the mean; the ratio of the interval between two adjacent sampling times to the average rate of change is calculated to obtain the proportion of disturbance change rate relative to the mean; the reciprocal of the sum of 1 and the proportion of disturbance fluctuation intensity relative to the proportion of disturbance change rate relative to the mean is obtained to obtain the time series consistency index; the time series consistency index of each type of environmental monitoring sequence is obtained, and the time series consistency index of each type of environmental monitoring sequence is combined using a product-type composite method to obtain the environmental disturbance consistency coefficient.

[0013] Preferably, the step of determining whether the initial risk threshold needs to be adjusted based on the threshold adaptive correction index is as follows: compare the threshold adaptive correction index with the threshold to be adjusted; if the threshold adaptive correction index is greater than or equal to the threshold to be adjusted, then it is determined that the initial risk threshold needs to be adjusted; if the threshold adaptive correction index is less than the threshold to be adjusted, then it is determined that the initial risk threshold does not need to be adjusted, and no adjustment is made to the risk threshold.

[0014] Preferably, the steps for obtaining the adjusted risk threshold are as follows: The data stability offset coefficient and the sensor time response offset coefficient are added together to obtain the internal offset composite amount; 1 is subtracted from the environmental disturbance consistency coefficient to obtain the external disturbance deviation amount; the levee risk index of the current period and the levee risk index of the previous period are obtained; a direction factor is calculated based on the levee risk index of the current period, the levee risk index of the previous period, the internal offset composite amount, and the external disturbance deviation amount; the threshold adaptive correction index is compared with the initial risk threshold to obtain the adjustment range; if the direction factor is +1, the initial risk threshold is compared with the adjustment range to obtain the adjusted risk threshold; if the direction factor is -1, the initial risk threshold is multiplied by the adjustment range to obtain the adjusted risk threshold; if the direction factor is 0, no adjustment is performed.

[0015] Preferably, an online monitoring and management method for dike safety based on the Internet of Things (IoT) includes the following steps: Step 1: Deploying various types of sensors at key locations on the dike to collect real-time monitoring data of the dike's interior and surrounding environment. This monitoring data includes pore water pressure, water level, tilt angle, horizontal displacement, and vertical displacement. Step 2: Encrypting, encoding, and packaging the monitoring data, and securely transmitting it to a cloud platform via IoT communication. The received data undergoes integrity verification and preprocessing to obtain preprocessed monitoring data. Step 3: Calculating the dike risk index based on the preprocessed monitoring data, obtaining an initial risk threshold, and performing a first risk assessment based on the dike risk index and the initial risk threshold. Step 4: If the first risk assessment indicates that the dike is at risk, obtaining the threshold influence parameters. The threshold influence parameters include the input signal change time, the output signal response time, and environmental monitoring parameters. An adaptive threshold correction index is obtained based on the threshold influence parameters. The initial risk threshold is then adjusted based on this adaptive correction index. Step 5: If the initial risk threshold needs adjustment, it is adjusted according to the adaptive threshold correction index to obtain the adjusted risk threshold. Step 6: The levee risk index is compared with the adjusted risk threshold. If the levee risk index is greater than or equal to the adjusted risk threshold, a second risk assessment indicates that the levee is at risk, and a risk warning is issued. If the levee risk index is less than the adjusted risk threshold, a second risk assessment indicates that the levee is not at risk, and no risk warning is issued. Step 7: The monitoring data and warning status are transmitted to the visualization platform.

[0016] The technical effects and advantages of this invention are as follows: Monitoring data is securely transmitted to a cloud platform via IoT communication. The received data undergoes integrity verification and preprocessing to obtain preprocessed monitoring data. A levee risk index is then calculated, and an initial risk assessment is performed based on this index. If the levee is deemed to be at risk, threshold impact parameters are obtained, and an adaptive threshold correction index is evaluated. If the initial risk threshold needs adjustment, it is adjusted according to the adaptive threshold correction index to obtain the adjusted risk threshold. A second risk assessment is then performed using the levee risk index and the adjusted risk threshold, effectively improving the accuracy of levee safety management. Attached Figure Description

[0017] Figure 1 This application provides a structural diagram of an Internet of Things-based online monitoring and management system for dike safety.

[0018] Figure 2 A flowchart illustrating an online monitoring and management method for the safety of a dike based on the Internet of Things (IoT) is provided for embodiments of this application. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. In addition, the forms of the various structures described in the following embodiments are merely illustrative. The Internet of Things-based online monitoring and management system and method for dike safety involved in the present invention are not limited to the structures described in the following embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] This invention provides an Internet of Things-based online monitoring and management system for dike safety, such as... Figure 1 As shown, the system includes: The data acquisition module is used to deploy various types of sensors at key parts of the levee. These sensors include piezometers, rain gauges, water level gauges, soil moisture sensors, and tilt sensors. The sensors collect real-time monitoring data of the levee's interior and surrounding environment, including pore water pressure, water level, tilt angle, horizontal displacement, and vertical displacement. The data transmission and encryption module is used to encrypt, encode and package the monitoring data, and securely transmit the monitoring data to the cloud platform through IoT communication. It also performs integrity verification and preprocessing on the received data to obtain preprocessed monitoring data. Preprocessing includes classification and storage, noise reduction and outlier correction. In this embodiment, it should be specifically explained that the steps of encrypting, encoding, and packaging the monitoring data are as follows: The monitoring data is encoded and packaged, and the monitoring timestamp, sensor number and monitoring value are packaged into a data packet according to the preset field format. Each data packet is lightly encrypted to obtain an encrypted data packet. The encryption algorithm adopts a symmetric encryption mechanism based on a hash function, and a dynamic key is generated with the node number and timestamp as input. It should be noted that the use of a symmetric encryption mechanism based on a hash function for data encryption is an existing technology, and this embodiment will not describe its specific steps in detail. The encrypted data packets are transmitted. During the transmission of the encrypted data packets, the data packet loss rate is obtained in real time. The communication quality index is obtained by subtracting the data packet loss rate from 1. When the communication quality index is lower than the quality threshold, the packet retransmission mechanism is triggered. Only the lost or failed data packets are retransmitted to ensure the reliability of data transmission and reduce network load. It should be noted that the packet retransmission mechanism is an existing technology, and its specific steps are not described in detail in this embodiment.

[0021] By introducing a dynamic key generation mechanism and an integrity verification mechanism into the data transmission and encryption module, the confidentiality and tamper resistance of the monitoring data during transmission are achieved; and by using communication quality index detection and packet retransmission strategies, reliable transmission in weak signal or high interference environments is achieved.

[0022] The first risk assessment module is used to calculate the levee risk index based on the pre-processed monitoring data, obtain the initial risk threshold, and make the first risk assessment based on the levee risk index and the initial risk threshold. In this embodiment, it should be specifically explained that the steps for obtaining the levee risk index are as follows: The seepage risk is calculated based on the pore water pressure inside the dike, the water level outside the dike, and the water level inside the dike. The specific steps for obtaining these parameters are as follows: ; In the formula, This is represented as seepage risk. When the seepage risk is greater than 0, it means that the seepage pressure inside the dike exceeds the external water pressure difference, and there is a seepage risk. The greater the seepage risk, the higher the risk. This is expressed as the pore water pressure inside the dike. Expressed as water density, Expressed as gravitational acceleration, This represents the water level outside the dike. This represents the water level height inside the dike. The following steps are taken to obtain the current horizontal displacement, the previous horizontal displacement, the current tilt angle, the previous tilt angle, and the vertical distance between the sensor and the embankment surface at the monitoring point: ; In the formula, Expressed as the rate of change of displacement, This represents the horizontal displacement at the current moment. This represents the horizontal displacement at the previous moment. This is expressed as the vertical distance between the sensor and the surface of the embankment. This represents the tilt angle value at the current moment. This represents the tilt angle value at the previous moment. This is represented as the time interval between two adjacent sampling times; Obtain the current and previous vertical displacements of the embankment, and calculate the settlement rate based on these displacements. The specific steps are as follows: ; In the formula, Expressed as settlement rate, Represented as the vertical displacement at the current moment. This represents the vertical displacement at the previous moment. This is represented as the time interval between two adjacent sampling times; The risk index of the embankment is obtained by multiplying the absolute values ​​of seepage risk, displacement change rate and settlement rate and taking the square root. This enhances the synergistic sensitivity among the three physical quantities. If any risk indicator rises abnormally, the risk index of the embankment will increase significantly, thereby reflecting the overall safety status of the embankment.

[0023] In this embodiment, it should be specifically explained that the steps for the first risk assessment based on the levee risk index and the initial risk threshold are as follows: The levee risk index is compared with an initial risk threshold. If the levee risk index is greater than or equal to the initial risk threshold, the levee is initially deemed to be at risk. If the levee risk index is less than the initial risk threshold, the levee is deemed not to be at risk, and monitoring data of the levee's interior and surrounding environment continues to be collected via sensors. The initial risk threshold is obtained using an adaptive thresholding method, an algorithm that automatically adjusts the judgment threshold based on the dynamic changes in monitoring data. The basic principle is to identify stable and disturbed intervals in the monitoring environment by calculating statistical characteristics such as the average value, fluctuation amplitude, and trend of the current data in the continuously collected monitoring data sequence. When a significant deviation is detected in the overall distribution or rate of change of the monitoring data from historical characteristics, the original threshold is automatically corrected to better reflect the current operating status and environmental conditions.

[0024] The risk threshold impact determination module, if the first risk determination indicates that there is a risk to the current levee, obtains the threshold impact parameters. The threshold impact parameters include the time of change of the input signal, the time of response of the output signal, and environmental monitoring parameters. The threshold adaptive correction index is obtained based on the threshold impact parameters. The initial risk threshold needs to be adjusted based on the threshold adaptive correction index. In this embodiment, it should be specifically explained that the step of obtaining the threshold adaptive correction index is as follows: Within the current detection cycle, monitoring data is acquired through sensors, and the data stability offset coefficient is evaluated based on the monitoring data. Within the current detection cycle, the response time series is obtained through the sensor, and the sensor's time response offset coefficient is evaluated based on the response time series. Within the current monitoring cycle, environmental monitoring parameters are acquired, and the environmental disturbance consistency coefficient is obtained based on the environmental monitoring parameters. The data stability offset coefficient, sensor time response offset coefficient, and environmental disturbance consistency coefficient are normalized. Specifically, this embodiment uses vector normalization. The threshold adaptive correction index is calculated based on the normalized data stability offset coefficient, sensor time response offset coefficient, and environmental disturbance consistency coefficient. The specific steps for obtaining this index are as follows: ; In the formula, This is expressed as the threshold adaptive correction index. This represents the data stability offset coefficient after normalization. The greater the volatility and the worse the stability of the monitored data, the lower the reliability of the data used for the current risk assessment is considered. In this case, a more significant adaptive adjustment to the risk threshold is needed to avoid misjudgments caused by data instability. This represents the sensor's time-response offset coefficient after normalization. The more significant the sensor response delay or output amplitude attenuation, the more severe the sensor performance degradation is considered. To maintain the reliability of risk identification, a more significant adaptive correction to the risk threshold is needed. This represents the environmental disturbance consistency coefficient after normalization. The more stable and consistent the external environment, the less interference the environment has on the monitoring data, and the risk threshold can remain stable or only needs slight adjustment. Conversely, when the environmental disturbance consistency coefficient decreases and environmental fluctuations intensify, the threshold correction range will be increased accordingly to prevent misjudgments caused by sudden external disturbances. , , This represents the weighting coefficients of the normalized data stability offset coefficient, the normalized sensor time response offset coefficient, and the normalized environmental disturbance consistency coefficient. , , , The weighting coefficients are obtained through the Analytic Hierarchy Process (AHP), a decision analysis method based on a multi-level structural model used to determine the relative importance of multiple influencing factors and calculate their weights. Its basic principle is as follows: First, a hierarchical structural model is established based on the problem's objectives, criteria, and indicators; then, through pairwise comparisons, experts or the system calculate the importance ratio of each indicator relative to the objective based on a judgment matrix; finally, the judgment matrix is ​​checked for consistency to ensure the reasonableness of the comparison results, and the weighting coefficients of each factor are obtained using the eigenvalue method.

[0025] In this embodiment, it should be specifically explained that the steps for obtaining the data stability offset coefficient are as follows: Within the current detection cycle, monitoring data is acquired through sensors to construct a monitoring data sequence. The detection sequence data includes pore water pressure sequence, water level height sequence, tilt angle sequence, horizontal displacement sequence, and vertical displacement sequence. Each sequence collects N data points at a fixed sampling time interval. The mean and standard deviation of the monitoring data sequences for each type of sensor are calculated. The average rate of change of adjacent sampling points is calculated for the monitoring data sequence of each type of sensor, and is denoted as the sequence dynamic rate of change. The larger the sequence dynamic rate of change value, the faster the change and the worse the stability. The stability shift is calculated based on the sequence mean, sequence standard deviation, and sequence dynamic change rate. The specific steps for obtaining this shift are as follows: ; In the formula, This is represented as a stability offset. Expressed as the standard deviation of the sequence, Represented as the series average, Expressed as the rate of change of the sequence. This is represented as the time interval between two adjacent sampling times; The stability offset of each type of monitoring data sequence is obtained, and the stability offset of each type of monitoring data sequence is fused using a product-type composite method to obtain the data stability offset coefficient. The specific steps are as follows: ; In the formula, Represented as the data stability offset coefficient, This represents the total number of sensor categories. It represents the stability offset of the monitoring data sequence of the j-th type of sensor.

[0026] In this embodiment, it should be specifically explained that the steps for obtaining the sensor time response offset coefficient are as follows: During the current testing cycle, response time series are obtained from various sensors (such as piezometers, water level gauges, tilt sensors, etc.) deployed on the levee. For each type of sensor, the time of change of its input signal and the time of response of its output signal are recorded to obtain the response time series. At the same time, the corresponding reference response time and the reference output amplitude during sensor calibration are obtained from the sensor's factory or periodic calibration data. The time for the response signal to reach steady state corresponding to each response time data in the response time sequence of each type of sensor is obtained. The time for the response signal to reach steady state is interpolated with the response time data to obtain the response delay for each sampling. The mean of each response delay is calculated to obtain the average response delay. For the response time series of each type of sensor, the difference between the average response delay and the reference response time is calculated, and then the ratio of the difference to the reference response time is calculated to obtain the response delay offset rate. Within the detection period, the difference between the peak value of each signal disturbance in the monitoring data sequence and the corresponding baseline value is calculated to obtain the amplitude of each response event. The amplitude of each effective response event within the detection period is averaged to obtain the sequence average amplitude. The sequence average amplitude is used to characterize the overall response intensity of this type of monitoring data within the current detection period. The larger the value, the more stable the amplitude of the sensor output signal; the smaller the value, the lower the sensor sensitivity or the response decay. For the response time series of each type of sensor, the amplitude attenuation ratio is obtained by subtracting the ratio of the average amplitude of the series to the reference output amplitude from 1. For each type of sensor, the sensor's time-response offset is calculated based on the response delay offset rate and amplitude attenuation ratio. The specific steps for obtaining this are as follows: ; In the formula, This is expressed as the sensor's time-response offset. This is expressed as the response delay offset. It is expressed as amplitude attenuation ratio, with the numerator being the amplitude attenuation ratio and the denominator ensuring a smooth result without extreme values; Obtain the time-response offset of each type of sensor, and fuse the time-response offsets of each type of sensor using a product-type composite method to obtain the sensor time-response offset coefficient. The specific steps are as follows: ; In the formula, This is expressed as the sensor's time-response offset coefficient. This represents the total number of sensor categories. It is represented as the time-response offset of the j-th type of sensor.

[0027] In this embodiment, it should be specifically explained that the steps for obtaining the environmental disturbance consistency coefficient are as follows: During the current monitoring period, environmental monitoring sensors are deployed around the dike to acquire environmental monitoring parameters and construct an environmental monitoring sequence, which includes rainfall, wind speed, temperature and external water level change sequences. For each type of environmental monitoring sequence, the mean and standard deviation are calculated to obtain the environmental monitoring mean and environmental monitoring standard deviation. For each type of environmental monitoring sequence, calculate the average rate of change of adjacent sampling points. The larger the value, the faster the disturbance changes and the more unstable the environment. The ratio of the standard deviation of environmental monitoring to the mean of environmental monitoring is used to calculate the proportion of the disturbance fluctuation intensity relative to the mean. The ratio of the interval between two adjacent sampling times to the average rate of change is used to calculate the proportion of the disturbance change rate relative to the mean. The time series consistency index is obtained by summing the ratio of 1 to the relative mean of the disturbance fluctuation intensity and the ratio of the relative mean of the disturbance change rate. The time series consistency index of each type of environmental monitoring sequence is obtained, and the time series consistency index of each type of environmental monitoring sequence is combined using a product-type composite method to obtain the environmental disturbance consistency coefficient.

[0028] In this embodiment, it should be specifically explained that the step of determining whether the initial risk threshold needs to be adjusted based on the threshold adaptive correction index is as follows: The adaptive correction index of the threshold is compared with the threshold that needs to be adjusted. If the adaptive correction index of the threshold is greater than or equal to the threshold that needs to be adjusted, the initial risk threshold is determined to need to be adjusted. If the adaptive correction index of the threshold is less than the threshold that needs to be adjusted, the initial risk threshold is determined not to need to be adjusted, and the risk threshold is not adjusted. The threshold that needs to be adjusted is obtained through the adaptive threshold method.

[0029] The risk threshold adjustment module adjusts the initial risk threshold according to the threshold adaptive correction index if it determines that the initial risk threshold needs to be adjusted, thus obtaining the adjusted risk threshold. In this embodiment, it should be specifically explained that the steps for obtaining the adjusted risk threshold are as follows: The data stability offset coefficient and the sensor time response offset coefficient are added together to obtain the internal offset composite amount. The environmental disturbance consistency coefficient is subtracted from 1 to obtain the external disturbance deviation amount. Obtain the levee risk index for the current period and the levee risk index for the previous period. Calculate the direction factor based on the current period's levee risk index, the previous period's levee risk index, the internal offset composite amount, and the external disturbance deviation amount. The specific steps are as follows: ; In the formula, This is represented as a directional factor. If the rate of risk increase exceeds the difference between internal and external volatility, it indicates a genuine increase in risk, and the threshold should be lowered. If internal or external volatility dominates the change, it indicates that the risk change is influenced by noise or the environment, and the threshold should be increased. If the difference between the two is small, the system is considered stable and no adjustment is needed. This represents the levee risk index for the current period. This represents the levee risk index for the previous period. This is represented as the internal offset composite amount. This is expressed as the deviation from external disturbances; It should be specifically noted that the sign function Used to determine the sign of the difference within the parentheses, its output is only +1, -1, or 0, representing different threshold adjustment directions: when the result is +1, it indicates that the upward trend of risk is stronger than the change in disturbance, and the system lowers the risk threshold to improve sensitivity; when the result is -1, it indicates that the change in disturbance is dominant, and the system raises the risk threshold to reduce false alarms; when the result is 0, it indicates that the difference between the risk change and the disturbance is close, and the system keeps the threshold unchanged.

[0030] The adjustment range is obtained by calculating the ratio of the threshold adaptive correction index to the initial risk threshold; If the direction factor is +1, it indicates a clear upward trend in risk, and the threshold should be tightened. The ratio of the initial risk threshold to the adjustment range is calculated to obtain the adjusted risk threshold. If the direction factor is -1, it indicates that disturbance or noise is dominant, and the threshold should be relaxed. The product of the initial risk threshold and the adjustment range is calculated to obtain the adjusted risk threshold. If the direction factor is 0, it indicates that the current state is stable, and no adjustment is made.

[0031] The second risk assessment module compares the levee risk index with the adjusted risk threshold. If the levee risk index is greater than or equal to the adjusted risk threshold, the second risk assessment indicates that the levee is at risk and a risk warning is issued to remind relevant personnel to carry out timely levee maintenance. If the levee risk index is less than the adjusted risk threshold, the second risk assessment indicates that the levee is not at risk and no risk warning is issued. The visualization monitoring module is used to transmit monitoring data and early warning status to the visualization platform, facilitating rapid decision-making by managers.

[0032] In this embodiment, it should be specifically explained that, as Figure 2 As shown, an online monitoring and management method for dike safety based on the Internet of Things includes the following steps: Step 1: Deploy various types of sensors at key locations on the embankment to collect real-time monitoring data on the embankment's interior and surrounding environment. The monitoring data includes pore water pressure, water level, tilt angle, horizontal displacement, and vertical displacement. Step 2: Encrypt, encode, and package the monitoring data, and securely transmit the monitoring data to the cloud platform via IoT communication. Perform integrity verification and preprocessing on the received data to obtain preprocessed monitoring data. Step 3: Calculate the embankment risk index based on the pre-processed monitoring data, obtain the initial risk threshold, and make the first risk assessment based on the embankment risk index and the initial risk threshold. Step 4: If the first risk assessment indicates that there is a risk to the current levee, then obtain the threshold impact parameters. The threshold impact parameters include the time of change of the input signal, the time of response of the output signal, and the environmental monitoring parameters. Based on the threshold impact parameters, the threshold adaptive correction index is obtained. Based on the threshold adaptive correction index, it is determined whether the initial risk threshold needs to be adjusted. Step 5: If the initial risk threshold needs to be adjusted, then adjust the initial risk threshold according to the threshold adaptive correction index to obtain the adjusted risk threshold. Step 6: Compare the levee risk index with the adjusted risk threshold. If the levee risk index is greater than or equal to the adjusted risk threshold, the second risk assessment indicates that the levee is at risk and a risk warning is issued. If the levee risk index is less than the adjusted risk threshold, the second risk assessment indicates that the levee is not at risk and no risk warning is issued. Step 7: Transmit the monitoring data and early warning status to the visualization platform.

[0033] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0034] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A levee safety online monitoring and management system based on the Internet of Things, characterized in that, The system includes; The data acquisition and transmission module is used to deploy various types of sensors at key parts of the embankment. The sensors collect real-time monitoring data of the embankment's interior and surrounding environment, including pore water pressure, water level, tilt angle, horizontal displacement, and vertical displacement. The data transmission and encryption module is used to encrypt, encode and package the monitoring data, and securely transmit the monitoring data to the cloud platform through IoT communication. It also performs integrity verification and preprocessing on the received data to obtain preprocessed monitoring data. The first risk assessment module is used to calculate the levee risk index based on the pre-processed monitoring data, obtain the initial risk threshold, and make the first risk assessment based on the levee risk index and the initial risk threshold. The risk threshold impact determination module, if the first risk determination indicates that there is a risk to the current levee, obtains the threshold impact parameters. The threshold impact parameters include the time of change of the input signal, the time of response of the output signal, and environmental monitoring parameters. The threshold adaptive correction index is obtained based on the threshold impact parameters. The initial risk threshold needs to be adjusted based on the threshold adaptive correction index. The risk threshold adjustment module adjusts the initial risk threshold according to the threshold adaptive correction index if it determines that the initial risk threshold needs to be adjusted, thus obtaining the adjusted risk threshold. The second risk assessment module compares the levee risk index with the adjusted risk threshold. If the levee risk index is greater than or equal to the adjusted risk threshold, the second risk assessment indicates that the levee is at risk and a risk warning is issued. If the risk index of the dike is less than the adjusted risk threshold, the second risk assessment will determine that there is no risk to the current dike and no risk warning will be issued. The visualization monitoring module is used to transmit monitoring data and early warning status to the visualization platform; The steps for obtaining the threshold adaptive correction index are as follows: Within the current detection cycle, monitoring data is acquired through sensors, and the data stability offset coefficient is evaluated based on the monitoring data. Within the current detection cycle, the response time series is obtained through the sensor, and the sensor's time response offset coefficient is evaluated based on the response time series. Within the current monitoring cycle, environmental monitoring parameters are acquired, and the environmental disturbance consistency coefficient is obtained based on the environmental monitoring parameters. The data stability offset coefficient, sensor time response offset coefficient, and environmental disturbance consistency coefficient are normalized. Based on these normalized coefficients, the threshold adaptive correction index is calculated. The specific steps are as follows: ； In the formula, This is expressed as the threshold adaptive correction index. This represents the data stability offset coefficient after normalization. This is represented as the normalized sensor aging response offset coefficient. This is expressed as the environmental disturbance consistency coefficient after normalization. , , The weighting coefficients are represented as the normalized data stability offset coefficient, the normalized sensor time response offset coefficient, and the normalized environmental disturbance consistency coefficient. The steps for obtaining the data stability offset coefficient are as follows: Within the current detection cycle, monitoring data is acquired through sensors, and a monitoring data sequence is constructed. The mean and standard deviation of the monitoring data sequences for each type of sensor are calculated. The average rate of change of adjacent sampling points is calculated for the monitoring data sequence of each type of sensor, and is denoted as the dynamic rate of change of the sequence. The stability offset is calculated based on the sequence mean, sequence standard deviation, and sequence dynamic change rate. The stability offset of each type of monitoring data sequence is obtained, and the stability offset of each type of monitoring data sequence is fused using a product-type composite method to obtain the data stability offset coefficient.

2. The online monitoring and management system for dike safety based on the Internet of Things as described in claim 1, characterized in that: The steps for encrypting, encoding, and packaging the monitoring data are as follows: The monitoring data is encoded and packaged, and the monitoring timestamp, sensor number and monitoring value are packaged into a data packet according to the preset field format. Each data packet is lightly encrypted to obtain the encrypted data packet; The encrypted data packets are transmitted. During the transmission of the encrypted data packets, the data packet loss rate is obtained in real time. The communication quality index is obtained by subtracting the data packet loss rate from 1. When the communication quality index is lower than the quality threshold, the packet retransmission mechanism is triggered, and retransmission is only performed on the data packets that are lost or fail to be verified.

3. The online monitoring and management system for dike safety based on the Internet of Things as described in claim 1, characterized in that, The steps for obtaining the embankment risk index are as follows: The pore water pressure inside the dike, the water level outside the dike, and the water level inside the dike are obtained. The seepage risk is calculated based on the pore water pressure inside the dike, the water level outside the dike, and the water level inside the dike. The current horizontal displacement, the previous horizontal displacement, the current tilt angle, the previous tilt angle, and the vertical distance between the sensor and the embankment surface are obtained at the monitoring point. The displacement change rate is calculated based on the current horizontal displacement, the previous horizontal displacement, the current tilt angle, the previous tilt angle, and the vertical distance between the sensor and the embankment surface. Obtain the current vertical displacement and the previous vertical displacement of the top of the dike, and calculate the settlement rate based on the current dike displacement and the previous dike displacement. The risk index of the embankment is obtained by multiplying the absolute values ​​of seepage risk, displacement change rate, and settlement rate, and then taking the square root.

4. The online monitoring and management system for dike safety based on the Internet of Things as described in claim 1, characterized in that, The steps for making the first risk assessment based on the levee risk index and the initial risk threshold are as follows: The risk index of the levee is compared with the initial risk threshold. If the risk index is greater than or equal to the initial risk threshold, the levee is deemed to be at risk in the first risk assessment. If the risk index is less than the initial risk threshold, the levee is deemed not to be at risk in the first risk assessment, and monitoring data of the levee's interior and surrounding environment are collected by sensors.

5. The online monitoring and management system for dike safety based on the Internet of Things as described in claim 1, characterized in that: The steps for obtaining the sensor time response offset coefficient are as follows: During the current testing cycle, response time series are obtained from various sensors deployed on the embankment. For each type of sensor, the time of change of its input signal and the time of response of its output signal are recorded to obtain the response time series. At the same time, the corresponding reference response time and the reference output amplitude during sensor calibration are obtained from the sensor's factory or periodic calibration data. The time for the response signal to reach steady state corresponding to each response time data in the response time sequence of each type of sensor is obtained. The time for the response signal to reach steady state is interpolated with the response time data to obtain the response delay for each sampling. The mean of each response delay is calculated to obtain the average response delay. For the response time series of each type of sensor, the difference between the average response delay and the reference response time is calculated, and then the ratio of the difference to the reference response time is calculated to obtain the response delay offset rate. Within the detection period, the difference between the peak value of each signal disturbance in the monitoring data sequence and the corresponding baseline value is calculated to obtain the amplitude of each response event. The amplitude of each effective response event within the detection period is averaged to obtain the sequence average amplitude. For the response time series of each type of sensor, the amplitude attenuation ratio is obtained by subtracting the ratio of the average amplitude of the series to the reference output amplitude from 1. For each type of sensor, the sensor's time-response offset is calculated based on the response delay offset rate and the amplitude attenuation ratio. Obtain the time response offset of each type of sensor, and fuse the time response offsets of each type of sensor using a product-type composite method to obtain the sensor time response offset coefficient.

6. The online monitoring and management system for dike safety based on the Internet of Things as described in claim 1, characterized in that: The steps for obtaining the environmental disturbance consistency coefficient are as follows: During the current monitoring cycle, environmental monitoring sensors are deployed around the dike to obtain environmental monitoring parameters and construct an environmental monitoring sequence. For each type of environmental monitoring sequence, the mean and standard deviation are calculated to obtain the environmental monitoring mean and environmental monitoring standard deviation. For each type of environmental monitoring sequence, calculate the average rate of change of adjacent sampling points; The ratio of the standard deviation of environmental monitoring to the mean of environmental monitoring is used to calculate the proportion of the disturbance fluctuation intensity relative to the mean. The ratio of the interval between two adjacent sampling times to the average rate of change is used to calculate the proportion of the disturbance change rate relative to the mean. The time series consistency index is obtained by summing the ratio of 1 to the relative mean of the disturbance fluctuation intensity and the ratio of the relative mean of the disturbance change rate. The time series consistency index of each type of environmental monitoring sequence is obtained, and the time series consistency index of each type of environmental monitoring sequence is combined using a product-type composite method to obtain the environmental disturbance consistency coefficient.

7. The online monitoring and management system for dike safety based on the Internet of Things as described in claim 1, characterized in that: The step of determining whether the initial risk threshold needs adjustment based on the threshold adaptive correction index is as follows: The threshold adaptive correction index is compared with the threshold that needs to be adjusted. If the threshold adaptive correction index is greater than or equal to the threshold that needs to be adjusted, the initial risk threshold is determined to need to be adjusted. If the threshold adaptive correction index is less than the threshold that needs adjustment, then the initial risk threshold is determined not to need adjustment, and no adjustment is made to the risk threshold.

8. The online monitoring and management system for dike safety based on the Internet of Things as described in claim 1, characterized in that: The steps for obtaining the adjusted risk threshold are as follows: The data stability offset coefficient and the sensor time response offset coefficient are added together to obtain the internal offset composite amount. The environmental disturbance consistency coefficient is subtracted from 1 to obtain the external disturbance deviation amount. Obtain the levee risk index for the current period and the levee risk index for the previous period. Calculate the direction factor based on the levee risk index for the current period, the levee risk index for the previous period, the internal offset composite amount, and the external disturbance deviation amount. The adjustment range is obtained by calculating the ratio of the threshold adaptive correction index to the initial risk threshold; If the direction factor is +1, the ratio of the initial risk threshold to the adjustment range is calculated to obtain the adjusted risk threshold; if the direction factor is -1, the product of the initial risk threshold and the adjustment range is calculated to obtain the adjusted risk threshold. If the direction factor is 0, no adjustment is made.

9. A method for online monitoring and management of dike safety based on the Internet of Things (IoT), used to implement the online monitoring and management system for dike safety based on the IoT as described in any one of claims 1-8, characterized in that: Includes the following steps: Step 1: Deploy various types of sensors at key locations on the embankment to collect real-time monitoring data on the embankment's interior and surrounding environment. The monitoring data includes pore water pressure, water level, tilt angle, horizontal displacement, and vertical displacement. Step 2: Encrypt, encode, and package the monitoring data, and securely transmit the monitoring data to the cloud platform via IoT communication. Perform integrity verification and preprocessing on the received data to obtain preprocessed monitoring data. Step 3: Calculate the embankment risk index based on the pre-processed monitoring data, obtain the initial risk threshold, and make the first risk assessment based on the embankment risk index and the initial risk threshold. Step 4: If the first risk assessment indicates that there is a risk to the current levee, then obtain the threshold impact parameters. The threshold impact parameters include the time of change of the input signal, the time of response of the output signal, and the environmental monitoring parameters. Based on the threshold impact parameters, the threshold adaptive correction index is obtained. Based on the threshold adaptive correction index, it is determined whether the initial risk threshold needs to be adjusted. Step 5: If the initial risk threshold needs to be adjusted, then adjust the initial risk threshold according to the threshold adaptive correction index to obtain the adjusted risk threshold. Step 6: Compare the levee risk index with the adjusted risk threshold. If the levee risk index is greater than or equal to the adjusted risk threshold, the second risk assessment indicates that the levee is at risk and a risk warning is issued. If the risk index of the dike is less than the adjusted risk threshold, the second risk assessment will determine that there is no risk to the current dike and no risk warning will be issued. Step 7: Transmit the monitoring data and early warning status to the visualization platform.

Images