Dam safety automatic monitoring system

Abstract

The invention discloses an automatic dam safety monitoring system which comprises a data acquisition module, a server, a data analysis module, a monitoring management module and a dynamic adjustment module. The data acquisition module is used for acquiring physical data and environmental data of the dam in real time and sending the data to the server for storage; the data analysis module analyzes and evaluates the physical data and the environmental data to obtain an early warning value and sends the early warning value to the monitoring management module; after receiving the early warning value, the monitoring management module notifies management personnel of a priority response value obtained by data analysis and performs real-time early warning response; the dynamic adjustment module executes a data acquisition dynamic adjustment strategy by identifying the early warning value; the system can realize real-time acquisition, intelligent analysis and early warning of various parameters of the dam, and improves the efficiency and accuracy of dam safety management.

Description

Technical Field

[0001] This invention belongs to the field of dam safety monitoring technology, and in particular relates to an automated dam safety monitoring system. Background Technology

[0002] In recent years, with the continuous development of water conservancy and hydropower projects, the safety management of dams has become increasingly important. Dams are usually closely related to important infrastructure such as water supply, irrigation, power generation, and flood control. Good dam safety management can ensure the continuity and stability of these functions, avoid water supply interruptions, property losses, or power shortages caused by dam problems, and dams play a role in regulating water volume in flood control and disaster reduction. Effective safety management ensures that dams operate normally during natural disasters (such as rainstorms or floods), reduces the impact of floods, and protects downstream areas from disasters. Dam safety management is of great significance for protecting life and property, maintaining social and economic stability, preventing ecological damage, and effectively utilizing water resources.

[0003] Traditional dam monitoring relies heavily on manual inspections and periodic data recording, which suffers from problems such as delayed information updates and low data processing efficiency, making it difficult to meet the needs of modern management and emergency early warning. Therefore, there is an urgent need for a system that can automatically and in real time monitor the safety status of dams.

[0004] Therefore, an automated dam safety monitoring system is needed to solve the above problems. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an automated monitoring system for dam safety, which can realize real-time acquisition, intelligent analysis and early warning of various parameters of the dam, thereby improving the efficiency and accuracy of dam safety management.

[0006] To achieve the above-mentioned technical effects, the technical solution adopted by the present invention is as follows: An automated dam safety monitoring system includes: Data acquisition module, server, data analysis module, monitoring and management module, and dynamic adjustment module; The data acquisition module is used to collect physical and environmental data of the dam in real time and send them to the server for storage; The data analysis module analyzes and evaluates physical and environmental data to obtain early warning values ​​and sends them to the monitoring and management module; After receiving the warning value, the monitoring and management module notifies the management personnel of the priority response value obtained from data analysis and performs a real-time warning response. The dynamic adjustment module executes a dynamic adjustment strategy for data acquisition by identifying early warning values.

[0007] Preferably, the method by which the data analysis module analyzes and evaluates physical and environmental data to obtain early warning values ​​and sends them to the monitoring and management module is as follows: Physical quantity data and environmental quantity data are analyzed separately to obtain physical quantity characteristic values ​​and environmental quantity characteristic values. The physical quantity characteristic values ​​and environmental quantity characteristic values ​​are then processed to obtain dam monitoring characteristic values. A two-dimensional rectangular coordinate system is constructed with the acquisition time as the horizontal axis and the dam monitoring characteristic value as the vertical axis. The position of the dam monitoring characteristic value in the coordinate system is recorded as the monitoring point. A smooth curve is used to connect each monitoring point in sequence according to the acquisition time to obtain the curve of the dam monitoring characteristic value changing with time. Substitute the two adjacent warning coordinate points The calculated rate of change value ,in This represents the change in the dam monitoring characteristic value, where Gi represents the time corresponding to the i-th data collection moment. This represents the time interval between two adjacent data collection moments. The change rate value and the dam monitoring characteristic value are multiplied by their respective weighting coefficients, and then summed to obtain the warning value, which is then sent to the monitoring management module.

[0008] Preferably, the data acquisition module acquires monitoring data on the structural safety of the dam by installing various sensors on the dam, including physical data and environmental data; wherein the sensors include water level sensors, pressure sensors, displacement sensors, seepage sensors, strain sensors and meteorological monitoring instruments; physical data include: water level value, pressure value, displacement value, seepage value and strain value; environmental data include temperature value, rainfall and wind speed value.

[0009] Preferably, the priority response value obtained from data analysis is notified to the management personnel and a real-time early warning response is issued. The specific steps are as follows: The dam is divided into several monitoring areas; personnel information of management personnel stored on the server is obtained, including name, length of service, contact information, number of early warning responses, early warning response value, and number of early warnings pending response; the personnel information is analyzed, and priority response values ​​are output through a neural network model; When the priority response value is greater than or equal to the set priority response threshold, the manager is classified as a Level 1 response; when the priority response value is less than the set priority response threshold, the manager is classified as a Level 2 response. Managers at both Level 1 and Level 2 response are identified. Dam monitoring characteristic values ​​are sorted from largest to smallest, and the monitoring area with the highest dam monitoring characteristic value is allocated. Managers at Level 1 response with the highest priority response value are selected as the response personnel for that dam monitoring area. Information on areas with rapidly changing dam monitoring characteristic values ​​and areas with stable but not drastic changes in dam monitoring characteristic values ​​are sent to the terminals of the managers at Level 1 and Level 2 response.

[0010] Preferably, the management strategy involves identifying early warnings and executing corresponding dynamic adjustments to data collection, specifically: A warning threshold range is set. Received warning values ​​are compared and analyzed against this range. When the warning value is less than the minimum value of the range, the dam's current state is recorded as safe, and low-frequency data acquisition (P1) continues. When the warning value is within the range, the acquisition frequency is switched to medium-frequency (P2). When the warning value exceeds the maximum value, the acquisition frequency is switched to high-frequency (P3), and a warning signal is immediately issued to notify relevant management personnel. Based on the real-time data, the warning values ​​are continuously monitored, and the acquisition frequency is adjusted accordingly. When a significant change in the dam's state is detected, the acquisition frequency is immediately switched to a higher frequency (P3) for more intensive sampling and monitoring of the trend. If the state is stable, the acquisition frequency is switched back to P1 or P2 to reduce the acquisition frequency. This process continues, repeating the above steps until the dam's monitoring characteristics return to a safe state.

[0011] Preferably, the specific method for setting the sampling frequency is as follows: Three acquisition frequencies are set: P1 for low-frequency acquisition, P2 for medium-frequency acquisition, and P3 for high-frequency acquisition. When the monitored parameters exceed the warning value or show obvious abnormalities, the high-frequency acquisition mode is entered to track data changes in real time.

[0012] Preferably, the server includes a central processing unit (CPU), a network interface card (NIC), and a solid-state drive (SSD); the CPU is connected to both the NIC and the SSD, and is used to analyze the dam and adjust various sensors.

[0013] Preferably, the central processing unit executes: The drone is controlled by a network card to perform a 3D scan of the dam, generating 3D data and image data of the dam; A 3D model of the dam was created based on 3D data and meshed. Cracks are identified by comparing grid images and marked as level 1, level 2, or level 3 risk grids according to their expansion. Instructions for adding, calibrating, or replacing sensors are generated based on the risk level.

[0014] Preferably, the analysis of personnel information and the output of priority response values ​​through a neural network model specifically includes: Let the management terminal be denoted as Rn, where n = 1, 2, 3...N, N is a positive integer, N represents the total number of management terminals, and n represents the serial number of any one of the management terminals; The neural network model calculates and outputs the manager's priority response value XRn by substituting the manager's length of service 1Rn, number of warning responses 2Rn, and warning response value 3Rn into the model. The neural network model includes an input layer, a hidden layer, and an output layer. The input layer receives the manager's length of service 1Rn, number of warning responses 2Rn, and warning response value 3Rn, and performs a weighted calculation through the hidden layer. ; Where f is the activation function and ωj is the connection weight; Finally, the priority response value XRn of the manager is output through the output layer.

[0015] Preferably, the specific analysis steps for the early warning response value 3Rn are as follows: After receiving an early warning signal through the administrator's terminal, the administrator needs to confirm receipt of the signal and respond accordingly. Upon completion of the response, the administrator sends an early warning response completion signal to the server via the terminal. The early warning response time is the time it takes for the administrator to receive the signal and confirm it on their terminal. A threshold for the response time is set. When the administrator's response time is less than the threshold, it is considered a Level 1 response time. When the administrator's response time is greater than or equal to the threshold, it is considered a Level 2 response time. The early warning response value CRn1 is calculated using a predefined formula. ; in This indicates the preset fixed value for the current alert response, where β represents the preset weight factor, β1 represents the corresponding weight factor set for the first-level response time of the manager, and β2 represents the corresponding weight factor set for the second-level response time of the manager, with β1 > β2. The manager's current alert response value 3Rn1 is sent to the server for storage, and the current alert response value CRn1 is summed with the historical alert response values ​​and the average is taken to obtain the current alert response value 3Rn. The manager's personnel information is then updated.

[0016] The beneficial effects of this invention are as follows: 1. By collecting physical and environmental data of the dam in real time, the system comprehensively and accurately monitors the structural safety of the dam and its external environment, ensuring the accuracy and timeliness of risk assessment. Early warning values ​​are dynamically calculated based on real-time monitoring characteristic values ​​and their rate of change, ensuring timely detection and rapid response to risk changes, avoiding delayed processing. The curves showing the changes in monitoring values ​​over time clearly demonstrate the trend of risks, providing decision-makers with an intuitive understanding of risk changes for timely judgment and handling. The real-time collected data is stored on a server and updated promptly as monitoring values ​​change, effectively improving the accuracy and efficiency of dam safety monitoring while optimizing resource utilization, achieving efficient and real-time monitoring and early warning.

[0017] 2. Through continuous monitoring and real-time data feedback, monitoring strategies can be dynamically adjusted to ensure optimal resource utilization and guarantee that the dam monitoring system can respond quickly to risks at critical moments, avoiding delayed detection or inappropriate responses. By dynamically adjusting the acquisition frequency, energy consumption and storage costs can be minimized while ensuring safe monitoring, improving the overall operating efficiency of the system. The multi-level notification mechanism of the dynamic adjustment module ensures timely information transmission, provides a more comprehensive understanding of the dam's operating status, allows for the advance development of emergency plans, and ensures the long-term safety of the dam. It achieves precise monitoring, timely response, resource conservation, real-time early warning, and continuous risk assessment, enhancing the safety of the dam and making the entire monitoring process more efficient and accurate, providing a strong guarantee for the long-term stable operation of the dam.

[0018] 3. By prioritizing the response of management personnel, it ensures that when dam monitoring characteristic values ​​change drastically, efficient management personnel are given priority to handle the situation, thus improving response speed. Dynamically displaying the changing trends of various monitoring parameters helps management personnel quickly grasp the dam's status, enhancing decision support. Through real-time data analysis and a rapid response mechanism, potential dam risks are reduced, and safety management levels are improved. Recording the response history and performance of management personnel provides a basis for future personnel training and management, optimizes the composition and efficiency of the early warning response team, and ensures the efficiency and timeliness of dam safety management, which is of great significance for ensuring the safe operation of the dam. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the system framework structure of the present invention; Figure 2 This is a schematic diagram of the server structure in an embodiment of the present invention. Detailed Implementation

[0020] Example 1: like Figure 1 As shown, an automated dam safety monitoring system includes: Data acquisition module, server, data analysis module, monitoring and management module, and dynamic adjustment module; The data acquisition module is used to collect physical and environmental data of the dam in real time and send them to the server for storage; The data analysis module analyzes and evaluates physical and environmental data to obtain early warning values ​​and sends them to the monitoring and management module; After receiving the warning value, the monitoring and management module notifies the management personnel of the priority response value obtained from data analysis and performs a real-time warning response. The dynamic adjustment module executes a dynamic adjustment strategy for data acquisition by identifying early warning values.

[0021] Preferably, the method by which the data analysis module analyzes and evaluates physical and environmental data to obtain early warning values ​​and sends them to the monitoring and management module is as follows: Physical quantity data and environmental quantity data are analyzed separately to obtain physical quantity characteristic values ​​and environmental quantity characteristic values. The physical quantity characteristic values ​​and environmental quantity characteristic values ​​are then processed to obtain dam monitoring characteristic values. A two-dimensional rectangular coordinate system is constructed with the acquisition time as the horizontal axis and the dam monitoring characteristic value as the vertical axis. The position of the dam monitoring characteristic value in the coordinate system is recorded as the monitoring point. A smooth curve is used to connect each monitoring point in sequence according to the acquisition time to obtain the curve of the dam monitoring characteristic value changing with time. Substitute the two adjacent warning coordinate points The calculated rate of change value ,in This represents the change in the dam monitoring characteristic value, where Gi represents the time corresponding to the i-th data collection moment. This represents the time interval between two adjacent data collection moments. The change rate value and the dam monitoring characteristic value are multiplied by their respective weighting coefficients, and then summed to obtain the warning value, which is then sent to the monitoring management module.

[0022] Preferably, the data acquisition module acquires monitoring data on the structural safety of the dam by installing various sensors on the dam, including physical data and environmental data; wherein the sensors include water level sensors, pressure sensors, displacement sensors, seepage sensors, strain sensors and meteorological monitoring instruments; physical data include: water level value, pressure value, displacement value, seepage value and strain value; environmental data include temperature value, rainfall and wind speed value.

[0023] Preferably, the priority response value obtained from data analysis is notified to the management personnel and a real-time early warning response is issued. The specific steps are as follows: The dam is divided into several monitoring areas; personnel information of management personnel stored on the server is obtained, including name, length of service, contact information, number of early warning responses, early warning response value, and number of early warnings pending response; the personnel information is analyzed, and priority response values ​​are output through a neural network model; When the priority response value is greater than or equal to the set priority response threshold, the manager is classified as a Level 1 response; when the priority response value is less than the set priority response threshold, the manager is classified as a Level 2 response. Managers at both Level 1 and Level 2 response are identified. Dam monitoring characteristic values ​​are sorted from largest to smallest, and the monitoring area with the highest dam monitoring characteristic value is allocated. Managers at Level 1 response with the highest priority response value are selected as the response personnel for that dam monitoring area. Information on areas with rapidly changing dam monitoring characteristic values ​​and areas with stable but not drastic changes in dam monitoring characteristic values ​​are sent to the terminals of the managers at Level 1 and Level 2 response.

[0024] Preferably, the management strategy involves identifying early warnings and executing corresponding dynamic adjustments to data collection, specifically: A warning threshold range is set. Received warning values ​​are compared and analyzed against this range. When the warning value is less than the minimum value of the range, the dam's current state is recorded as safe, and low-frequency data acquisition (P1) continues. When the warning value is within the range, the acquisition frequency is switched to medium-frequency (P2). When the warning value exceeds the maximum value, the acquisition frequency is switched to high-frequency (P3), and a warning signal is immediately issued to notify relevant management personnel. Based on the real-time data, the warning values ​​are continuously monitored, and the acquisition frequency is adjusted accordingly. When a significant change in the dam's state is detected, the acquisition frequency is immediately switched to a higher frequency (P3) for more intensive sampling and monitoring of the trend. If the state is stable, the acquisition frequency is switched back to P1 or P2 to reduce the acquisition frequency. This process continues, repeating the above steps until the dam's monitoring characteristics return to a safe state.

[0025] Preferably, the specific method for setting the sampling frequency is as follows: Three acquisition frequencies are set: P1 for low-frequency acquisition, P2 for medium-frequency acquisition, and P3 for high-frequency acquisition. When the monitored parameters exceed the warning value or show obvious abnormalities, the high-frequency acquisition mode is entered to track data changes in real time.

[0026] like Figure 2 As shown, the server includes a central processing unit (CPU), a network interface card (NIC), and a solid-state drive (SSD); the CPU is connected to both the NIC and the SSD, and is used to analyze the dam and adjust various sensors.

[0027] Preferably, the central processing unit executes: The drone is controlled by a network card to perform a 3D scan of the dam, generating 3D data and image data of the dam; A 3D model of the dam was created based on 3D data and meshed. Cracks are identified by comparing grid images and marked as level 1, level 2, or level 3 risk grids according to their expansion. Instructions for adding, calibrating, or replacing sensors are generated based on the risk level.

[0028] Preferably, the analysis of personnel information and the output of priority response values ​​through a neural network model specifically includes: Let the management terminal be denoted as Rn, where n = 1, 2, 3...N, N is a positive integer, N represents the total number of management terminals, and n represents the serial number of any one of the management terminals; The neural network model calculates and outputs the manager's priority response value XRn by substituting the manager's length of service 1Rn, number of warning responses 2Rn, and warning response value 3Rn into the model. The neural network model includes an input layer, a hidden layer, and an output layer. The input layer receives the manager's length of service 1Rn, number of warning responses 2Rn, and warning response value 3Rn, and performs a weighted calculation through the hidden layer. ; Where f is the activation function and ωj is the connection weight; Finally, the priority response value XRn of the manager is output through the output layer.

[0029] Preferably, the specific analysis steps for the early warning response value 3Rn are as follows: After receiving an early warning signal through the administrator's terminal, the administrator needs to confirm receipt of the signal and respond accordingly. Upon completion of the response, the administrator sends an early warning response completion signal to the server via the terminal. The early warning response time is the time it takes for the administrator to receive the signal and confirm it on their terminal. A threshold for the response time is set. When the administrator's response time is less than the threshold, it is considered a Level 1 response time. When the administrator's response time is greater than or equal to the threshold, it is considered a Level 2 response time. The early warning response value CRn1 is calculated using a predefined formula. ; in This indicates the preset fixed value for the current alert response, where β represents the preset weight factor, β1 represents the corresponding weight factor set for the first-level response time of the manager, and β2 represents the corresponding weight factor set for the second-level response time of the manager, with β1 > β2. The manager's current alert response value 3Rn1 is sent to the server for storage, and the current alert response value CRn1 is summed with the historical alert response values ​​and the average is taken to obtain the current alert response value 3Rn. The manager's personnel information is then updated.

[0030] Example 2: This embodiment provides a specific deployment method for an automated dam safety monitoring system, the process of which includes: First, set up the system architecture, including: data acquisition module, server, data analysis module, monitoring and management module, and dynamic adjustment module.

[0031] The data acquisition module collects physical and environmental data of the dam in real time and sends them to the server for storage. Various sensors are installed at key parts of the dam (dam body, foundation, spillway, etc.) to monitor the dam's structural safety, environmental conditions, and operational status. These sensors include: water level sensors, pressure sensors, displacement sensors, seepage sensors, strain sensors, and meteorological monitoring instruments. Physical data includes: water level, pressure, displacement, seepage, and strain values. Environmental data includes: temperature, rainfall, and wind speed. It should be noted that pressure sensors measure the internal and external water pressure of the dam body in Pa, typically expressed in kilopascals (kPa) or megapascals (MPa); displacement sensors monitor the horizontal or vertical displacement of the dam body in millimeters; seepage sensors monitor the seepage of the dam body in cubic meters per second or liters per second; and strain sensors measure the minute deformations of the dam structure under stress, typically expressed as micro-strain.

[0032] The data analysis module processes the collected data to assess the dam, obtains early warning values, and sends them to the monitoring and management module; the specific steps are as follows: Nearly one year of dam monitoring data was extracted from the server. This data included water level, pressure, displacement, seepage, strain, temperature, rainfall, and wind speed. The water level, pressure, displacement, seepage, and strain values ​​were compared with their corresponding standard values: if the water level was greater than the standard value, the difference was calculated; if the pressure was greater than the standard value, the difference was calculated; if the displacement was greater than the standard value, the difference was calculated; if the seepage was greater than the standard value, the difference was calculated; and if the strain was greater than the standard value, the difference was calculated. The resulting differences in water level, pressure, displacement, seepage, and strain were then multiplied by their respective weighting coefficients using a linear weighted average method and summed. The results were output and marked as characteristic values ​​of the physical quantities. If no corresponding difference was found, it was not included in the calculation. The higher the water level value and the greater the deviation from the average water level, the larger the physical quantity characteristic value; the higher the pressure value and the greater the deviation from the average pressure, the larger the physical quantity characteristic value; the higher the displacement value and the greater the deviation from the average displacement, the larger the physical quantity characteristic value; the higher the seepage value and the greater the deviation from the average seepage, the larger the physical quantity characteristic value; the higher the strain value and the greater the deviation from the average strain, the larger the physical quantity characteristic value; the larger the physical quantity characteristic value, the greater the potential safety risk of the dam.

[0033] The temperature, rainfall, and wind speed values ​​are compared with their corresponding standard values. If the temperature, rainfall, or wind speed is greater than the standard value, the difference between the two values ​​is calculated. These differences are then used to calculate the environmental characteristic values ​​using a linear weighted average method. As temperature increases, the expansion effect increases, leading to changes in dam stress, reduced safety, and a larger environmental characteristic value. Increased rainfall leads to rising water levels and increased seepage, increasing the pressure on the dam, reducing safety, and also resulting in a larger environmental characteristic value. Increased wind speed intensifies the impact of waves on the dam, especially during storms, further increasing pressure on the dam, reducing safety, and also resulting in a larger environmental characteristic value.

[0034] The physical and environmental characteristic values ​​are sent to the image processor. The image processor constructs two line segments based on these values ​​in a certain proportion, draws rectangles from these line segments, extracts the area values ​​corresponding to the rectangles, marks them as dam monitoring characteristic values, and outputs them. A larger physical anomaly characteristic value indicates a higher dam risk, and a larger environmental anomaly characteristic value corresponds to a larger dam monitoring characteristic value. A threshold range for dam monitoring characteristic values ​​is set, and the dam monitoring characteristic values ​​are compared and analyzed against this range. When the dam monitoring characteristic value is less than the minimum value of the set threshold range, it indicates that the dam is currently in a stable state and there is no immediate risk. When the dam warning value is within the set threshold range, it indicates that there is a certain risk in the dam warning, and a warning label is generated and displayed on the corresponding screen.

[0035] A two-dimensional rectangular coordinate system is constructed with the data acquisition time as the x-axis and the dam monitoring characteristic value as the y-axis. The positions of the dam monitoring characteristic values ​​in the coordinate system are recorded as monitoring points. A smooth curve is used to connect each monitoring point sequentially according to the data acquisition time to obtain a curve of the dam monitoring characteristic value changing over time. The values ​​of two adjacent early warning coordinate points are then substituted into the coordinate system. The calculated rate of change value ,in This represents the change in the dam monitoring characteristic value, where Gi represents the time corresponding to the i-th data collection moment. The interval between two adjacent data collection times is used to calculate the warning value. The values ​​of the rate of change and the dam monitoring characteristic value are multiplied by their corresponding weighting coefficients and then summed to obtain the warning value. The larger the rate of change, the larger the warning value, indicating that the dam risk is increasing. The smaller the rate of change, the smaller the warning value, indicating that the dam risk is decreasing. The warning value is sent to the server for storage and also to the monitoring management module.

[0036] By collecting physical and environmental data of the dam in real time, the structural safety of the dam and its external environment can be monitored comprehensively and accurately, ensuring the accuracy and timeliness of risk assessment. Early warning values ​​are dynamically calculated based on real-time monitoring values ​​and the rate of change, ensuring timely detection and rapid response to risk changes and avoiding delayed processing. The curves showing the changes in monitoring values ​​over time clearly demonstrate the trend of risks, providing decision-makers with an intuitive understanding of risk changes for timely judgment and action. The real-time collected data is stored on a server and updated promptly as monitoring values ​​change, effectively improving the accuracy and efficiency of dam safety monitoring while optimizing resource utilization, achieving efficient and real-time monitoring and early warning.

[0037] After receiving the early warning value, the monitoring and management module notifies the management personnel of the priority response value obtained from data analysis and issues a real-time early warning; the specific steps are as follows: A two-dimensional Cartesian coordinate system is constructed using the real-time collected water level, pressure, displacement, seepage, strain, temperature, rainfall, and wind speed values. The system uses the acquisition time as the x-axis and the real-time collected values ​​as the y-axis. The positions of the water level, pressure, displacement, seepage, strain, temperature, rainfall, and wind speed values ​​in the coordinate system are denoted as water level observation point, pressure observation point, displacement observation point, seepage observation point, strain observation point, temperature observation point, rainfall observation point, and wind speed observation point. A smooth curve is used to connect each observation point sequentially according to the acquisition time to obtain a curve of the changes in water level, pressure, displacement, seepage, strain, temperature, rainfall, and wind speed values ​​over time, displaying the real-time monitoring data.

[0038] The dam is divided into several monitoring areas. The number of management personnel terminals (including but not limited to PCs or mobile devices) is denoted as Rn, where n = 1, 2, 3...N, where N is a positive integer, N represents the total number of management personnel terminals, and n represents the serial number of any given terminal. The personnel information of the management personnel stored on the server is retrieved, including name, length of service, contact information, number of warning responses, warning response value, and number of warnings awaiting response. The values ​​of the management personnel's length of service 1Rn, number of warning responses 2Rn, and warning response value 3Rn are substituted into a neural network model to calculate and output the management personnel's priority response value XRn. The neural network model includes an input layer, a hidden layer, and an output layer. The input layer receives the values ​​of the management personnel's length of service 1Rn, number of warning responses 2Rn, and warning response value 3Rn, and performs a weighted calculation through the hidden layer. ; f is the activation function, ωj is the connection weight, and finally, the priority response value XRn of the manager is output through the output layer. The specific analysis steps for the warning response value 3Rn are as follows: After receiving the warning signal through the manager's terminal, the manager needs to confirm receipt of the warning signal and respond to it. After completing the response, the manager sends a warning response completion signal to the server through the manager's terminal. The warning response time is the time it takes for the manager to receive the warning signal and confirm it on the manager's terminal. A warning response time threshold is set. When the manager's warning response time is less than the threshold, the manager's response time is considered a first-level response time; when the manager's warning response time is greater than or equal to the threshold, the manager's response time is considered a second-level response time. The manager's warning response value CRn1 is calculated using the set formula: ; in This indicates the preset fixed value for the current alert response, where β represents the preset weight factor, β1 represents the corresponding weight factor set for the first-level response time of the manager, and β2 represents the corresponding weight factor set for the second-level response time of the manager, with β1 > β2. The manager's current alert response value 3Rn1 is sent to the server for storage, and the current alert response value CRn1 is summed with the historical alert response values ​​and the average is taken to obtain the current alert response value 3Rn. The manager's personnel information is then updated.

[0039] The priority response value is compared and analyzed with the set priority response threshold: when the priority response value is greater than or equal to the set priority response threshold, the manager is classified as a Level 1 response, used for rapid response to drastic changes in dam monitoring characteristic values; when the priority response value is less than the set priority response threshold, the manager is classified as a Level 2 response, used for rapid response to stable changes in dam monitoring characteristic values; the managers at Level 1 and Level 2 response are counted; the dam monitoring characteristic values ​​are sorted from largest to smallest, and the monitoring area with the largest dam monitoring characteristic value is allocated, selecting the manager at Level 1 with the largest priority response value as the response personnel for that dam monitoring area, while increasing the number of pending warnings for that manager by one, and updating the number of pending warnings for that manager; information on areas with drastic changes in dam monitoring characteristic values ​​and information on areas with stable but not drastic changes in dam monitoring characteristic values ​​are sent to the terminals of the managers at Level 1 and Level 2 response; the dam area information includes the location of the dam monitoring area, various monitoring data, and monitoring values.

[0040] By prioritizing the responses of management personnel, it ensures that when dam monitoring characteristic values ​​change drastically, efficient management personnel are given priority to handle the situation, thus improving response speed. Dynamically displaying the changing trends of various monitoring parameters helps management personnel quickly grasp the dam's status, enhancing decision support. Furthermore, through real-time data analysis and rapid response mechanisms, potential dam risks are reduced, and safety management levels are improved. Recording the response history and performance of management personnel provides a basis for future personnel training and management, optimizes the composition and efficiency of the early warning response team, and ensures the efficiency and timeliness of dam safety management, which is of great significance for ensuring the safe operation of the dam.

[0041] The dynamic adjustment module identifies early warnings and executes corresponding data acquisition and dynamic adjustment strategies to achieve dynamic monitoring of the dam's safe operation; the specific steps are as follows: The data acquisition frequency determines the response speed of the monitoring system. To optimize resource utilization and response speed under different risk levels, three acquisition frequencies are set: P1 is low-frequency acquisition, suitable for daily monitoring when all monitoring parameters are within normal ranges and the environment is stable, with an acquisition cycle of 1 hour or longer; P2 is medium-frequency acquisition, where the acquisition frequency is increased to more closely track changes when certain monitoring parameters approach warning values, with an acquisition cycle of 10 to 30 minutes; P3 is high-frequency acquisition, which enters high-frequency acquisition mode when monitoring parameters exceed warning values ​​or show obvious anomalies, tracking data changes in real time, with an acquisition cycle of 1 to 5 minutes to ensure real-time response. It should be noted that the specific acquisition frequency cycle is set by industry professionals according to actual needs.

[0042] A warning threshold range is set, and the warning values ​​obtained from the server are compared and analyzed with the set warning threshold range. When the warning value is less than the minimum value of the set warning threshold range, it indicates that the dam is currently in a stable and risk-free state, and the dam's state is recorded as safe. The sensors in the data acquisition module are controlled to maintain low-frequency acquisition at P1. When the warning value is within the set warning threshold range, it indicates that the dam is currently experiencing significant but not drastic changes, and the sensors switch to medium-frequency acquisition at P2. When the warning value is greater than the maximum value of the set warning threshold range, it indicates that the dam's current state is changing drastically, and the sensors switch to high-frequency acquisition at P3. The system collects data at a frequency and immediately issues an early warning signal to the management terminal of relevant personnel. The early warning signal is sent via various methods, including SMS, email, and voice notification. Based on the real-time data collected, the system continuously monitors the early warning value and adjusts the collection frequency accordingly. When a significant change in the dam's condition is detected, the system immediately switches to a higher collection frequency (P3) for more intensive sampling and monitoring of the trend. If the condition stabilizes, the sensor switches to P1 or P2 to reduce the collection frequency and minimize system resource consumption. This process continues and repeats until the dam's monitoring characteristics return to a safe state. By continuously monitoring and providing real-time data feedback, the monitoring strategy is dynamically adjusted to ensure optimal resource utilization and guarantee that the dam monitoring system can respond quickly to risks at critical moments, avoiding delayed detection or inappropriate responses. By dynamically adjusting the acquisition frequency, energy consumption and storage costs are minimized while ensuring safe monitoring, thereby improving the overall operational efficiency of the system. The multi-level notification mechanism of the early warning execution module ensures timely information transmission, provides a more comprehensive understanding of the dam's operational status, allows for the advance development of emergency plans, and ensures the long-term safety of the dam. By achieving precise monitoring, timely response, resource conservation, real-time early warning, and continuous risk assessment, the system enhances dam safety, makes the entire monitoring process more efficient and accurate, and provides a strong guarantee for the long-term stable operation of the dam.

[0043] Example 3: Based on Example 2, such as Figure 2 As shown, the server also contains a central processing unit (CPU), a network card, and a solid-state drive (SSD); the CPU analyzes the dam and adjusts various sensors. The specific analysis process is as follows: The central processing unit (CPU) connects to the drone via a network interface card (NIC) and generates dam inspection commands. Upon receiving the commands, the drone performs a 3D scan and image acquisition of the dam according to a preset inspection path, obtaining 3D and image data. The drone then sends this data to a solid-state drive (SSD) for storage via the NIC. The CPU retrieves the 3D data from the SSD and uses BIM technology to create a 3D model of the dam. It then divides the 3D model into 3D meshes, obtaining several meshes corresponding to the dam. Finally, it retrieves the image data from the SSD, extracts the corresponding mesh images, and sends them to an image processor. The image processor identifies cracks in the mesh images. If a crack is found, the grid is marked as the initial grid. The previously acquired image of the initial grid is obtained. If the previous image did not contain a crack, the initial grid is marked as a first-level risk grid. If the previous image contained a crack, the outlines of the cracks in the two images are identified, and the center points of the two outlines are compared and overlapped. If the crack in the previous image is smaller than the crack in the current image, the initial grid is marked as a second-level risk grid. The area difference between the two crack outlines is calculated to obtain the crack propagation value. If the crack propagation value is greater than a set propagation threshold and the crack outline area is greater than a set area threshold, the second-level risk grid is marked as a third-level risk grid. If no crack exists, the grid is marked as a maintenance grid and sent to the central processing unit. Then, the central processing unit (CPU) acquires the dam monitoring feature value at the corresponding location of the primary risk grid. If the dam monitoring feature value is greater than the set monitoring threshold, sensor command one is generated; if the dam monitoring feature value is less than or equal to the set monitoring threshold, sensor command two is generated. Similarly, the CPU acquires the dam monitoring feature value at the corresponding location of the secondary and tertiary risk areas, makes a judgment, and generates the corresponding sensor command one or sensor command two. The CPU acquires the dam monitoring feature value corresponding to the maintenance grid. If the dam monitoring feature value is greater than the set monitoring threshold, it remains unchanged; if the dam monitoring feature value is less than or equal to the set monitoring threshold, the maintenance grid is marked as a transmission reduction grid, and sensor command three is generated. The CPU sends the primary, secondary, and tertiary risk grids, their corresponding location coordinates, the corresponding sensor command one or sensor command two, and the location coordinates and sensor command three of the transmission reduction grid to the early warning execution module.

[0044] When the early warning execution module receives the first-level risk grid, the corresponding location coordinates, and sensor instruction one, it matches the sensor addition quantity one corresponding to sensor instruction one and sends the corresponding location coordinates and sensor addition quantity one to the management terminal. The management personnel then deploy new sensors at the corresponding location coordinates according to sensor addition quantity one, increasing sensor density and improving safety monitoring of that location. Sensor addition quantity one refers to the number of each type of sensor added, such as two of each type. When the module receives the first-level risk grid, the corresponding location coordinates, and sensor instruction two, it generates a sensor calibration instruction and sends it along with the sensor's location coordinates to the management terminal. The management personnel calibrate the sensor, and upon completion, upload a calibration completion command to improve sensor accuracy and achieve precise safety monitoring of the location.

[0045] When the early warning execution module receives the secondary risk grid, its corresponding location coordinates, and sensor instruction one, it matches the corresponding sensor addition quantity two and sends the corresponding location coordinates and sensor addition quantity two to the management terminal. The management personnel then deploy the new sensor at the corresponding location coordinates according to the sensor addition quantity two. When the secondary risk grid, its corresponding location coordinates, and sensor instruction two are received, all sensors corresponding to the secondary risk grid are obtained, and the installation time of the sensors is obtained. Sensors installed before the target time one are marked as replacement sensors. The location coordinates of the replacement sensors and the corresponding location coordinates of the unmarked sensors are sent to the management terminal. The management personnel then replace the sensors marked as replacement sensors and calibrate the unmarked sensors.

[0046] When the early warning execution module receives the Level 3 risk grid, its corresponding location coordinates, and sensor instruction one, it matches the corresponding sensor addition quantity three and sends the corresponding location coordinates and sensor addition quantity three to the management terminal. The management personnel then deploy the new sensor at the corresponding location coordinates based on the sensor addition quantity three. When it receives the Level 3 risk grid, its corresponding location coordinates, and sensor instruction two, it retrieves all sensors corresponding to the Level 2 risk grid, obtains the sensor installation time, marks sensors installed before the target time two as replacement sensors, and sends the location coordinates of the replacement sensors and the corresponding unmarked sensor locations to the management terminal. The management personnel then replace the marked sensors, calibrate the unmarked sensors, and perform maintenance on the corresponding location in the grid.

[0047] When the early warning execution module receives the location coordinates of the reduction grid and sensor command three, it matches the corresponding number of sensors to be reduced and sends the number of sensors to be reduced along with the corresponding location coordinates to the management terminal. The management personnel then reduce the number of sensors corresponding to that location based on the number of sensors to be reduced. The number of sensors to be reduced refers to the number of each type of sensor to be reduced, such as reducing one of each type of sensor.

[0048] By combining UAV 3D scanning with BIM modeling, high-precision 3D meshing of the entire dam area is achieved, eliminating blind spots in traditional manual inspections. Crack propagation analysis based on historical data comparison enables dynamic risk level classification. Sensor addition, calibration, or replacement instructions are automatically generated according to the risk level, dynamically adjusting monitoring resource allocation. Reduced-transmission mesh marking reduces the number of sensors in low-risk areas, lowering energy consumption and maintenance costs. This addresses the problem of traditional point sensors not being able to cover the entire area and avoids over-monitoring in low-risk areas and under-monitoring in high-risk areas.

Claims

1. An automated monitoring system for dam safety, characterized in that, include: Data acquisition module, server, data analysis module, monitoring and management module, and dynamic adjustment module; The data acquisition module is used to collect physical and environmental data of the dam in real time and send them to the server for storage; The data analysis module analyzes and evaluates physical and environmental data to obtain early warning values ​​and sends them to the monitoring and management module; After receiving the warning value, the monitoring and management module notifies the management personnel of the priority response value obtained from data analysis and performs a real-time warning response. The dynamic adjustment module executes a dynamic adjustment strategy for data acquisition by identifying early warning values.

2. The dam safety automated monitoring system according to claim 1, characterized in that, The data analysis module analyzes and evaluates physical and environmental data to obtain early warning values ​​and sends them to the monitoring and management module as follows: Physical quantity data and environmental quantity data are analyzed separately to obtain physical quantity characteristic values ​​and environmental quantity characteristic values. The physical quantity characteristic values ​​and environmental quantity characteristic values ​​are then processed to obtain dam monitoring characteristic values. A two-dimensional rectangular coordinate system is constructed with the acquisition time as the horizontal axis and the dam monitoring characteristic value as the vertical axis. The position of the dam monitoring characteristic value in the coordinate system is recorded as the monitoring point. A smooth curve is used to connect each monitoring point in sequence according to the acquisition time to obtain the curve of the dam monitoring characteristic value changing with time. Substitute the two adjacent warning coordinate points The calculated rate of change value ,in This represents the change in the dam monitoring characteristic value, where Gi represents the time corresponding to the i-th data collection moment. This represents the time interval between two adjacent data collection moments. The change rate value and the dam monitoring characteristic value are multiplied by their respective weighting coefficients, and then summed to obtain the warning value, which is then sent to the monitoring management module.

3. The dam safety automated monitoring system according to claim 1, characterized in that, The data acquisition module obtains monitoring data on the structural safety of the dam by installing various sensors on the dam, including physical data and environmental data. The sensors include water level sensors, pressure sensors, displacement sensors, seepage sensors, strain sensors, and meteorological monitoring instruments. The physical data includes water level values, pressure values, displacement values, seepage values, and strain values. The environmental data includes temperature values, rainfall, and wind speed values.

4. The dam safety automated monitoring system according to claim 1, characterized in that, The priority response values ​​obtained from data analysis are used to notify management personnel and issue real-time early warning responses. The specific steps are as follows: The dam is divided into several monitoring areas; personnel information of management personnel stored on the server is obtained, including name, length of service, contact information, number of early warning responses, early warning response value, and number of early warnings pending response; Analyze personnel information and output priority response values ​​through a neural network model; When the priority response value is greater than or equal to the set priority response threshold, the manager is classified as Level 1; when the priority response value is less than the set priority response threshold, the manager is classified as Level 2. The system involves: identifying management personnel at Level 1 and Level 2 response levels; sorting dam monitoring characteristic values ​​from largest to smallest; allocating monitoring areas with the highest dam monitoring characteristic values; selecting management personnel at Level 1 response level with the highest response priority value as the response personnel for that dam monitoring area; and sending information on areas with rapidly changing dam monitoring characteristic values ​​and areas with stable but not drastic changes to the terminals of management personnel at Level 1 and Level 2 response levels.

5. The dam safety automated monitoring system according to claim 1, characterized in that, By identifying early warnings and implementing corresponding dynamic adjustment management strategies for data collection, the specific steps are as follows: A warning threshold range is set. Received warning values ​​are compared and analyzed against this range. When the warning value is less than the minimum value of the range, the dam's current state is recorded as safe, and low-frequency data acquisition (P1) continues. When the warning value is within the range, the acquisition frequency is switched to medium-frequency (P2). When the warning value exceeds the maximum value, the acquisition frequency is switched to high-frequency (P3), and a warning signal is immediately issued to notify relevant management personnel. Based on the real-time data, the warning values ​​are continuously monitored, and the acquisition frequency is adjusted accordingly. When a significant change in the dam's state is detected, the acquisition frequency is immediately switched to a higher frequency (P3) for more intensive sampling and monitoring of the trend. If the state is stable, the acquisition frequency is switched back to P1 or P2 to reduce the acquisition frequency. This process continues, repeating the above steps until the dam's monitoring characteristics return to a safe state.

6. The dam safety automated monitoring system according to claim 5, characterized in that, The specific method for setting the sampling frequency is as follows: Three acquisition frequencies are set: P1 for low-frequency acquisition, P2 for medium-frequency acquisition, and P3 for high-frequency acquisition. When the monitored parameters exceed the warning value or show obvious abnormalities, the high-frequency acquisition mode is entered to track data changes in real time.

7. The dam safety automated monitoring system according to claim 1, characterized in that, The server includes a central processing unit (CPU), a network interface card (NIC), and a solid-state drive (SSD); the CPU is connected to both the NIC and the SSD, and is used to analyze the dam and adjust various sensors.

8. The dam safety automated monitoring system according to claim 7, characterized in that, The central processing unit executes: The drone is controlled by a network card to perform a 3D scan of the dam, generating 3D data and image data of the dam; A 3D model of the dam was created based on 3D data and meshed. Cracks are identified by comparing grid images and marked as level 1, level 2, or level 3 risk grids according to their expansion. Instructions for adding, calibrating, or replacing sensors are generated based on the risk level.

9. The dam safety automated monitoring system according to claim 4, characterized in that, The analysis of personnel information and the output of priority response values ​​through a neural network model specifically include: Let the management terminal be denoted as Rn, where n = 1, 2, 3...N, N is a positive integer, N represents the total number of management terminals, and n represents the serial number of any one of the management terminals; The neural network model calculates and outputs the manager's priority response value XRn by substituting the manager's length of service 1Rn, number of warning responses 2Rn, and warning response value 3Rn into the model. The neural network model includes an input layer, a hidden layer, and an output layer. The input layer receives the manager's length of service 1Rn, number of warning responses 2Rn, and warning response value 3Rn, and performs a weighted calculation through the hidden layer. ； Where f is the activation function and ωj is the connection weight; Finally, the priority response value XRn of the manager is output through the output layer.

10. The dam safety automated monitoring system according to claim 9, characterized in that, The specific analysis steps for the warning response value 3Rn are as follows: After receiving an early warning signal through the management terminal, the administrator needs to confirm receipt of the early warning signal and respond to it. After responding to the early warning signal, the administrator sends an early warning response completion signal to the server through the management terminal. The early warning response time is the time it takes for the administrator to receive the early warning signal and confirm the early warning on the management terminal. An early warning response time threshold is set. When the administrator's early warning response time is less than the early warning response time threshold, the administrator's response time is classified as a level 1 response time. When a manager's early warning response time is greater than or equal to the early warning response time threshold, the manager's response speed is classified as a level two response time. The manager's early warning response value CRn1 is calculated using the established formula: ； in This indicates the preset fixed value for the current alert response, where β represents the preset weight factor, β1 represents the corresponding weight factor set for the first-level response time of the manager, and β2 represents the corresponding weight factor set for the second-level response time of the manager, with β1 > β2. The manager's current alert response value 3Rn1 is sent to the server for storage, and the current alert response value CRn1 is summed with the historical alert response values ​​and the average is taken to obtain the current alert response value 3Rn. The manager's personnel information is then updated.

Images