A river channel regulation engineering monitoring method and system

Abstract

The application provides a river channel regulation engineering monitoring method and system, which belongs to the technical field of engineering monitoring. The method comprises the following steps: based on channel scene information, a matching monitoring parameter set, an abnormal diagnosis rule library and an abnormal threshold set are called from a pre-constructed knowledge base, a monitoring network device is cooperatively dispatched to collect multi-source monitoring data for abnormal identification, and potential abnormal problems and abnormal confidence are obtained; if the abnormal confidence is greater than the abnormal threshold, a drone is controlled to verify the potential abnormal problems on site, and a reliability evaluation result is generated; if the reliability evaluation result is greater than a reliability threshold, the multi-source monitoring data are input into a time series prediction model to obtain a trend prediction result, which is compared with a maintenance critical threshold to obtain a spatiotemporal position of future maintenance needs, a channel comprehensive monitoring index is obtained through a mechanism-data hybrid model, and if the comprehensive monitoring index is greater than a warning threshold, warning information is generated; finally, a cooperative scheduling strategy is generated through an optimization algorithm and is executed.

Description

Technical Field

[0001] This application relates to the field of engineering monitoring technology, and in particular to a monitoring method and system for river and waterway improvement projects. Background Technology

[0002] During the monitoring of river and waterway improvement projects, manual periodic inspections are used, with staff going to the site to check various facilities, consuming a significant amount of manpower, resources, and time. Another method involves using fixed monitoring equipment, such as sensors installed at key locations in the waterway, to collect data periodically. Historical experience and data analysis are also used to assess waterway conditions, and past cases and data from similar projects are used to predict potential problems. However, manual inspections cannot provide continuous monitoring, cannot promptly detect sudden anomalies, are greatly affected by human factors, and cannot guarantee the accuracy and reliability of monitoring results. Fixed monitoring equipment lacks flexibility, cannot be adjusted according to different waterway scenarios and actual conditions, and has limited data analysis and processing capabilities for complex and changing situations, failing to accurately identify potential anomalies and future maintenance needs.

[0003] Therefore, there is an urgent need for a monitoring method and system for river and waterway improvement projects. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a monitoring method and system for river channel improvement projects.

[0005] A first aspect of this application provides a method for monitoring river channel improvement projects, comprising: Based on the current waterway scenario information, a matching set of monitoring parameters, an anomaly diagnosis rule base, and an anomaly threshold set are generated by calling from a pre-built knowledge base; Based on the monitoring parameter set, the monitoring network equipment is coordinated and scheduled to collect multi-source monitoring data. Based on the multi-source monitoring data, the anomaly diagnosis rule base is invoked to identify anomalies and obtain potential anomalies and anomaly confidence levels. If the anomaly confidence level is greater than the preset anomaly threshold, then the drone in the monitoring network device is controlled to perform on-site verification of the potential anomaly and generate a reliability assessment result. If the reliability assessment result is greater than the preset reliability threshold, the multi-source monitoring data will be input into the preset time series prediction model to obtain the monitoring trend prediction result. The monitoring trend prediction results are compared with the maintenance critical threshold in the abnormal threshold set to obtain the spatiotemporal location of future maintenance needs. The comprehensive waterway monitoring index is calculated through the mechanism-data hybrid model. If the comprehensive monitoring index is greater than the preset warning threshold, the current warning information is generated. Based on the reliability assessment results, the spatiotemporal location of future maintenance needs, and current early warning information, a collaborative scheduling strategy is generated and executed through an optimization algorithm.

[0006] A second aspect of this application provides a monitoring system for river channel improvement projects, comprising: The data retrieval module is used to generate a matching set of monitoring parameters, an anomaly diagnosis rule base, and an anomaly threshold set based on the current waterway scenario information from a pre-built knowledge base. The data acquisition module is used to coordinate and schedule monitoring network devices based on the monitoring parameter set to acquire multi-source monitoring data. An anomaly identification module is used to identify anomalies based on the multi-source monitoring data and call the anomaly diagnosis rule base to obtain potential anomalies and anomaly confidence levels. The problem verification module is used to control the drone in the monitoring network device to perform on-site verification of the potential abnormal problem if the anomaly confidence level is greater than the preset anomaly threshold, and generate a reliability assessment result. The trend prediction module is used to input multi-source monitoring data into a preset time-series prediction model to obtain the monitoring trend prediction result if the reliability assessment result is greater than a preset reliability threshold. The early warning generation module is used to compare the monitoring trend prediction results with the maintenance critical threshold in the abnormal threshold set to obtain the spatiotemporal location of future maintenance needs, and calculate the waterway comprehensive monitoring index through the mechanism-data hybrid model. If the comprehensive monitoring index is greater than the preset early warning threshold, the current early warning information is generated. The strategy generation module is used to generate and execute a collaborative scheduling strategy based on the reliability assessment results, the spatiotemporal location of future maintenance needs, and current early warning information, through an optimization algorithm.

[0007] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described river channel regulation engineering monitoring method.

[0008] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described monitoring method for river channel improvement projects.

[0009] The beneficial effects of the river and waterway improvement project monitoring method and system provided in this application are as follows: This application generates a suitable monitoring parameter set, anomaly diagnosis rule base and anomaly threshold set by matching the current waterway scene information, collects multi-source monitoring data and identifies potential anomalies, conducts on-site verification and reliability assessment of anomalies, predicts monitoring trends, determines the spatiotemporal location of future maintenance needs and the comprehensive waterway monitoring index, generates early warning information when conditions are met, and finally generates and executes a collaborative scheduling strategy to achieve comprehensive, accurate monitoring and effective management of the river and waterway improvement project. Attached Figure Description

[0010] Figure 1 A flowchart illustrating a monitoring method for river channel improvement projects provided in an embodiment of this application; Figure 2 A structural block diagram of a river channel improvement engineering monitoring system provided in an embodiment of this application; Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0011] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0012] To make the purpose, technical solution, and advantages of this application clearer, the following will be described in conjunction with the appendix. Figure 1-3 The following is an explanation using specific examples.

[0013] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a monitoring method for river channel improvement projects according to an embodiment of this application. The method includes: S101: Based on the current waterway scenario information, generate a matching set of monitoring parameters, an anomaly diagnosis rule base, and an anomaly threshold set from the pre-built knowledge base.

[0014] In this embodiment, the current waterway scenario information is the operational and environmental characteristic data of the target monitoring area in the river waterway improvement project, including: construction stage characteristics: dredging, bank protection, silt removal, and other construction procedures; structural risk characteristics: waterway slope stability level, dam structure aging degree, etc.; ecological sensitivity characteristics: distribution of sensitive species, habitat importance, water body self-purification capacity, etc.; and hydrological environment characteristics: flow velocity, water level, sediment content, etc., which serve as the basis for calling knowledge base data. The pre-built knowledge base is a structured database built based on historical monitoring data of the waterway improvement project, industry standards, and expert experience, including three sub-bases: monitoring parameter sub-base: mapping relationship between monitoring indicators, frequencies, and thresholds under different scenarios; anomaly diagnosis rule sub-base: judgment rules for typical anomalies such as sudden siltation of the riverbed and abrupt structural displacement; and anomaly threshold sub-base: early warning thresholds and maintenance critical thresholds for monitoring indicators under different scenarios.

[0015] In this embodiment, the monitoring parameter set is a set of monitoring execution parameters matched to the current waterway scenario. This set includes monitoring indicators, monitoring frequency, and monitoring thresholds. Monitoring indicators include, for example, structural displacement and suspended solids concentration. Monitoring frequency includes, for example, data collected once every 2 hours or once daily. Monitoring thresholds are the critical values ​​that distinguish between normal and abnormal states. This monitoring parameter set serves as the basis for guiding the collaborative data collection of monitoring network devices. The anomaly diagnosis rule base is a set of judgment rules for typical anomalies in waterway improvement projects. This rule set is generated based on mechanism analysis and historical data mining. For example, the rule for sudden siltation of the riverbed is: a siltation rate greater than 5 cm / d and a duration greater than or equal to 3 days. The rule for abnormal water quality diffusion is: a suspended solids concentration increase of more than 50% within 24 hours, used for anomaly identification in subsequent multi-source monitoring data. The abnormal threshold set is a multi-level threshold set that matches the current waterway scenario. It includes the early warning threshold (the critical value that triggers an early warning), the maintenance critical threshold (the critical value that requires the initiation of maintenance operations), and the abnormal judgment threshold (the basic threshold that distinguishes between normal and abnormal). The threshold values ​​are different in different scenarios. For example, the suspended solids concentration threshold is higher in the dredging construction stage than in the bank protection construction stage.

[0016] S102: Based on the monitoring parameter set, coordinate and schedule monitoring network equipment to collect multi-source monitoring data.

[0017] In this embodiment, collaborative scheduling is based on the requirements of the monitoring parameter set. Through a unified scheduling management module, tasks are allocated, time is synchronized, and work is coordinated among various devices within the monitoring network. This ensures that different devices complete the collection of corresponding indicators at specified times and locations, avoiding data loss, duplicate collection, or time sequence misalignment, and guaranteeing the spatiotemporal consistency of multi-source data. The monitoring network equipment is a diversified cluster of equipment built in the waterway monitoring area, including fixed monitoring equipment, mobile monitoring equipment, and remote sensing equipment, including: structural monitoring equipment: total station, displacement gauge, settlement meter; hydrological and water quality monitoring equipment: Doppler current meter, water level gauge, water quality sensor; ecological monitoring equipment: underwater camera, biosensor; mobile patrol equipment: drone, monitoring vessel. The equipment achieves data communication and remote control through the Internet of Things.

[0018] S103: Based on multi-source monitoring data, call the anomaly diagnosis rule base to identify anomalies and obtain potential anomalies and anomaly confidence levels.

[0019] In this embodiment, multi-source monitoring data is a comprehensive data set covering multiple dimensions of the waterway improvement project, collected collaboratively by monitoring network devices. "Multi-source" refers to three levels: equipment sources, indicator dimensions, and data types. Equipment sources include total stations, water quality sensors, and drones; indicator dimensions include structural safety and slope settlement; hydrodynamics and flow velocity; and ecological environment (dissolved oxygen). Data types include: numerical data (settlement amount and concentration values); image data (slope appearance images); and time-series data (hourly flow velocity curves), which is the basic data source for anomaly identification.

[0020] In this embodiment, the anomaly diagnosis rule base is a structured set of rules built based on the mechanism analysis of waterway regulation projects, historical anomaly cases, and industry standards. Each rule includes three elements: triggering conditions, anomaly type, and judgment logic. For example, the slope settlement anomaly rule is: the average daily settlement is greater than 0.2 cm and the duration is greater than or equal to 2 days; the water quality anomaly rule is: the suspended solids concentration increases by more than or equal to 50% within 24 hours. The rule base needs to be adjusted to adapt to different waterway scenarios. Anomaly identification is the process of matching and comparing multi-source monitoring data with the anomaly diagnosis rule base. This is achieved through three steps: spatiotemporal alignment, feature extraction, and rule matching. First, the multi-source data is spatiotemporally standardized. Then, the change characteristics of key monitoring indicators are extracted. Finally, the rule base is traversed to determine whether the anomaly triggering conditions are met. This is the step of mining potential problems from the monitoring data.

[0021] In this embodiment, potential anomalies are identified through rule matching as problems affecting the structural safety or ecological environment of the waterway. These correspond to typical anomalies in waterway improvement projects, such as structural anomalies like excessive slope settlement rate and sudden changes in slope displacement, hydrological anomalies like sudden siltation of the riverbed, and ecological anomalies like excessively low dissolved oxygen concentration. These problems have not yet caused substantial harm, but there is a risk of them developing into accidents. The anomaly confidence level is a probability value representing the authenticity of a potential anomaly, ranging from 0 to 1. A higher value indicates a more reliable anomaly determination. The confidence level calculation comprehensively considers three factors: the accuracy and completeness of the monitoring data, the fit of the rule matching, and the probability of similar anomalies occurring in the past. For example, if the slope settlement exceeds the preset anomaly threshold for two consecutive days, and the data accuracy meets the standard, the confidence level is 0.92.

[0022] S104: If the anomaly confidence level is greater than the preset anomaly threshold, control the drone in the monitoring network equipment to conduct on-site verification of potential anomalies and generate a reliability assessment result.

[0023] In this embodiment, the anomaly confidence score is a probability value representing the authenticity of a potential anomaly during the initial anomaly identification phase. Its value ranges from 0 to 1 and is calculated comprehensively based on the accuracy of monitoring data, rule matching fit, and the probability of similar anomalies occurring in the past. It serves as the basis for determining whether to initiate on-site verification; a higher value indicates a stronger credibility of the potential anomaly. The preset anomaly threshold is a confidence threshold value preset based on the risk level of the waterway improvement project, industry standards, and expert experience, ranging from 0.7 to 0.8. It is used to define whether a potential anomaly requires further on-site verification: when the anomaly confidence score is greater than the preset threshold, it indicates that the credibility of the potential anomaly meets the standard requiring verification; if it is less than the preset threshold, it is judged as a suspected misjudgment or a low-risk anomaly, and verification is not initiated temporarily.

[0024] In this embodiment, the drones in the monitoring network are drone devices integrated into the waterway monitoring network and equipped with monitoring functions. They are equipped with modules such as high-definition cameras, multispectral sensors, and lidar. Their advantages include flexibility, wide coverage, and the ability to quickly reach potential anomaly locations to obtain real-time images or data, such as slopes and the center of the waterway, thus compensating for the spatial coverage limitations of fixed monitoring equipment. On-site verification is the process of using drones to conduct on-site investigations and collect supplementary data at the locations of potential anomalies. The purpose is to verify the authenticity of the previous anomaly identification results, distinguish between genuine anomalies and data misjudgments, and obtain detailed information about the anomalies, such as the actual width of cracks and the extent of settlement, providing on-site data for subsequent risk assessment. The reliability assessment result is based on the on-site verification data from the drones, evaluating the authenticity and severity of potential anomalies. It is presented as a reliability score (ranging from 0 to 1, with higher scores indicating more genuine anomalies) or a verification conclusion (e.g., confirmed anomaly, ruled out anomaly, further investigation required). This serves as the basis for determining whether to initiate subsequent trend prediction and early warning scheduling.

[0025] S105: If the reliability assessment result is greater than the preset reliability threshold, the multi-source monitoring data will be input into the preset time series prediction model to obtain the monitoring trend prediction result.

[0026] In this embodiment, the reliability assessment result is a numerical result evaluating the authenticity and severity of potential anomalies after on-site verification by UAVs, presented as a reliability score (range 0-1). The score is calculated comprehensively based on the matching degree between the on-site verification data and the previous anomaly identification data, data integrity, and the risk level of the anomaly at the scene. It serves as the basis for determining whether to initiate trend prediction; a higher score indicates a genuine anomaly and a higher risk. The preset reliability threshold is a critical value for the reliability score preset based on the risk prevention and control needs of the waterway improvement project, historical engineering cases, and expert experience, ranging from 0.8 to 0.9. It is used to determine whether the anomaly requires further trend prediction: when the reliability assessment result is greater than the preset reliability threshold, it is determined to be a genuine high-risk anomaly, requiring the initiation of time-series prediction; when it is less than the preset reliability threshold, it is determined to be a low-risk anomaly, requiring only continuous tracking and monitoring.

[0027] In this embodiment, the multi-source monitoring data refers to the comprehensive historical and real-time data collected by the waterway monitoring network equipment, covering anomaly points and surrounding areas. This includes structural safety data (slope settlement, slope displacement), hydrodynamic data (flow velocity, water level), and ecological environment data (dissolved oxygen, suspended solids concentration). Data types include time-series numerical data and spatial image data, serving as the input basis for the time-series prediction model. The preset time-series prediction model is an algorithmic model pre-trained on the time-series evolution patterns of waterway monitoring indicators. This embodiment employs a CNN-BiGRU-Attention hybrid model, capable of capturing long-term trends and short-term abrupt changes in monitoring data. It can predict indicator changes within a preset future period based on historical data. The model parameters have been trained, optimized, and adapted to the current waterway scenario using historical monitoring data.

[0028] Specifically, the CNN-BiGRU-Attention hybrid model is a three-layer cascaded structure, with clear functions and connections between each layer: CNN feature extraction layer: adopts a cascaded structure of two one-dimensional convolutional layers and pooling layers. The first convolutional kernel size is set to 3×1 with 32 kernels, and the second convolutional kernel size is set to 5×1 with 64 kernels, both using the ReLU activation function; after the convolutional layers, a max pooling layer with a stride of 2 is connected to extract local spatiotemporal features from the time series of multi-source monitoring data, such as short-term abrupt changes in suspended matter concentration and gradient changes in slope settlement. The output of the pooling layer is converted into a one-dimensional feature vector through a flattening operation, which serves as the input to the next layer. BiGRU Temporal Modeling Layer: This layer consists of two BiGRU (Bidirectional Gated Recurrent Units) layers, each with 128 hidden nodes. The forward GRU and backward GRU capture the forward temporal dependencies of the monitoring data (e.g., the impact of increasing construction intensity on settlement) and the backward temporal dependencies (e.g., the inhibitory effect of water level decline on suspended solids concentration), respectively. The two BiGRU layers use residual connections to alleviate the gradient vanishing problem, outputting a 2×128 bidirectional temporal feature vector. Attention Mechanism Layer: This layer uses an additive attention mechanism to weight the temporal feature vector output by the BiGRU through a weight matrix. Key features in waterway monitoring (e.g., monitoring data near threshold values, abnormal mutation points) are given higher attention weights to weaken the interference of noisy data. The weighted feature vector output by the attention layer is mapped to the final predicted feature vector through a fully connected layer, completing the entire feature learning process from local feature extraction to temporal dependency modeling to key feature enhancement.

[0029] In this embodiment, the monitoring trend prediction result is the change trend data of the target monitoring indicator within a preset period (e.g., 7 days, 15 days) output by the time series prediction model, including: indicator value prediction curve, such as the daily slope settlement prediction value; trend characteristic determination, such as settlement rate acceleration / deceleration; indicator over-limit time prediction, such as predicting that the slope displacement will reach the maintenance critical threshold on the 5th day, which is the basis for subsequent maintenance demand analysis and early warning generation.

[0030] S106: Compare the monitoring trend prediction results with the maintenance critical threshold in the abnormal threshold set to obtain the spatiotemporal location of future maintenance needs, and calculate the waterway comprehensive monitoring index through the mechanism-data hybrid model. If the comprehensive monitoring index is greater than the preset warning threshold, the current warning information is generated.

[0031] In this embodiment, the maintenance critical threshold is a hierarchical threshold set of abnormal thresholds. Based on waterway engineering structural safety standards, ecological protection requirements, and equipment operation limits, it serves as the critical value for determining whether maintenance work needs to be initiated. When the predicted value of a monitoring indicator reaches or exceeds the maintenance critical threshold, it indicates that the corresponding area will face risks affecting waterway safety or function, requiring planned maintenance intervention. The spatiotemporal location of future maintenance needs is determined by combining the predicted time of exceeding limits by comprehensive indicators with the spatial location of anomalies, defining the target area and time window for maintenance operations. The spatial location corresponds to specific waterway chainages or monitoring points, such as the K12+300 slope section, while the temporal location corresponds to the predicted time of indicator exceeding limits, such as the 6th day in the future. This provides a precise basis for formulating maintenance plans. The mechanism-data hybrid model is a composite evaluation model that integrates a mechanism model and a data-driven model: the mechanism model is constructed based on the principles of waterway hydraulics, structural mechanics, and ecology, representing the physical / ecological relationships between indicators; the data-driven model is trained based on historical monitoring data to fit the statistical regularities of the indicators. The weighted combination of the two achieves a precise assessment of the overall waterway condition.

[0032] Specifically, the mechanism-data hybrid model adopts a three-layer serial structure: mechanism constraint layer, data feature extraction layer, and fusion prediction layer. The functions and connections of each layer are as follows: Mechanism constraint layer: Based on the mechanisms of waterway engineering mechanics, hydrodynamics, and ecology, physical constraint equations are constructed, such as slope stability equilibrium equations, suspended solids diffusion and convection equations, and ecosystem service value assessment models. This transforms the inherent correlation between waterway structural safety, hydrodynamics, and the ecological environment into constraints, and provides a reasonable definition for the output results of the subsequent data-driven layer; Data feature extraction layer: Adopts a CNN-BiGRU combined structure. The CNN layer extracts data through two layers of one-dimensional convolution (3×1 kernel size, 64 kernels). The local abrupt changes in multi-source monitoring data, such as sudden changes in sedimentation rate and rapid increases in concentration, are captured by the BiGRU layer (2 layers, 128 hidden nodes), which captures long-term temporal dependencies in the data, such as the lagged effects of temporal changes in construction intensity on ecological indicators, and outputs high-dimensional data feature vectors. The fusion prediction layer uses a gated fusion unit to weightedly fuse the physical law parameters of the mechanism constraint layer with the high-dimensional feature vectors of the data feature extraction layer. The weight of influencing factors is strengthened through an attention mechanism, such as structural indicators under high-risk conditions and environmental indicators during ecologically sensitive periods. Finally, the comprehensive waterway monitoring index is output through a fully connected layer, achieving the dual goals of ensuring rationality through physical mechanisms and ensuring accuracy through data models.

[0033] In this embodiment, the comprehensive waterway monitoring index is an indicator representing the overall safety and functional status of the waterway, calculated by a mechanism-data hybrid model, with a value range of 0-10. A higher score indicates a higher overall risk to the waterway's structure, hydrology, and ecology; a lower score indicates a more stable waterway condition. The preset warning threshold is a critical value of the comprehensive monitoring index based on the waterway risk level (e.g., high-risk hub sections, low-risk ordinary sections), and serves as the standard for determining whether to trigger a warning. When the comprehensive monitoring index exceeds the preset warning threshold, a warning message needs to be generated immediately and pushed to the relevant management departments. The current warning message is a structured notification generated after the warning is triggered, including key information such as the warning level, risk points, core abnormal indicators, recommended handling measures, and response time limits, used to guide emergency response and maintenance scheduling.

[0034] S107: Based on the reliability assessment results, the spatiotemporal location of future maintenance needs, and current early warning information, generate and execute a collaborative scheduling strategy through an optimization algorithm.

[0035] In this embodiment, the reliability assessment result is a score derived from the on-site verification by the UAV regarding the authenticity and severity of potential anomalies in the waterway. The score ranges from 0 to 1 and serves as the basis for determining the priority of anomalies. A higher score indicates a higher degree of authenticity and risk level, directly determining the priority of resource allocation in collaborative scheduling. The spatiotemporal location of future maintenance needs is determined by comparing the monitoring trend prediction results with the maintenance critical threshold, defining the spatial range of maintenance operations (e.g., specific waterway chainage segments, slope / slope protection areas) and time window. For example, if a crack exceeds the maintenance critical threshold on the 5th day in the future, it needs to be repaired. This is an input parameter for task planning in the collaborative scheduling strategy. The current early warning information is a structured early warning notification triggered by the comprehensive waterway monitoring index, including key information such as the early warning level, core risk points, recommended handling measures, and response time limits. This defines the emergency handling boundaries for collaborative scheduling; for example, an orange warning requires a response within 24 hours. The optimization algorithm is selected based on the multi-objective optimization requirements of waterway maintenance scheduling. In this embodiment, the non-dominated sorting genetic algorithm NSGA-II is adopted, which can solve the Pareto optimal solution set under multiple constraints and achieve a multi-objective balance of minimizing maintenance costs, minimizing navigation interference, and minimizing operational risks.

[0036] Specifically, the NSGA-II algorithm adopts a four-layer progressive structure: encoding layer, evolutionary operation layer, non-dominated sorting layer, and selection layer. The functions and connections of each layer are as follows: Encoding layer: Using real number encoding, the core decision variables (task sequence, resource allocation, and time arrangement) of the channel coordinated scheduling are mapped to a one-dimensional chromosome vector. The chromosome length is determined by the dimension of the decision variables. For example, if there are 2 task sequences, 3 resource types, and 1 time window, the chromosome length is 9. Evolutionary operation layer: Includes two sub-modules: crossover and mutation. The crossover module uses a simulated binary crossover (SBX) strategy. The mutation module employs a multinomial mutation strategy to achieve gene recombination and maintain population diversity. The non-dominated sorting layer uses a fast non-dominated sorting algorithm to stratify each individual (scheduling scheme) within the population. Individuals within the same layer do not dominate each other (i.e., they cannot improve one optimization objective without harming others). Lower layers represent better schemes. The selection layer, based on crowding calculation and a roulette wheel selection strategy, selects individuals with high crowding (indicating a uniform distribution of schemes) within each layer to enter the next generation of the population, improving population diversity and convergence, and ultimately outputting a Pareto optimal solution set.

[0037] In this embodiment, the collaborative scheduling strategy is an integrated execution plan based on the output of an optimization algorithm, covering task allocation, resource configuration, and time scheduling. It includes a maintenance task list, equipment / personnel scheduling plan, work sequence table, and emergency plan, serving as the basis for guiding on-site maintenance operations. Execution involves distributing the collaborative scheduling strategy to the waterway maintenance management platform, work teams, and equipment terminals, completing maintenance tasks according to the plan, and simultaneously collecting status feedback data during execution, forming a closed-loop process of strategy generation, implementation, and data sharing.

[0038] As can be seen from the above, this application generates a suitable set of monitoring parameters, an anomaly diagnosis rule base, and an anomaly threshold set by matching the current waterway scene information, collects multi-source monitoring data and identifies potential anomalies, conducts on-site verification and reliability assessment of anomalies, predicts monitoring trends, determines the spatiotemporal location of future maintenance needs and the comprehensive waterway monitoring index, generates early warning information when conditions are met, and finally generates and executes a collaborative scheduling strategy to achieve comprehensive, accurate monitoring and effective management of river and waterway improvement projects.

[0039] In one embodiment of this application, based on current waterway scenario information, a matching set of monitoring parameters is generated by calling a pre-built knowledge base, including: Feature extraction is performed on the current waterway scene information to obtain the monitoring features of the target waterway area. The monitoring features include construction stage features, structural risk features, and ecological sensitivity features. Based on the characteristics of the construction stage and a predefined stage-threshold mapping table, the basic monitoring threshold is determined and then corrected according to the current hydrological conditions to obtain the first monitoring threshold. Based on the risk level in the structural risk characteristics, the basic monitoring frequency is determined and then adjusted according to the current construction activity intensity to obtain the first monitoring frequency; Based on ecological sensitivity characteristics, an ecological sensitivity index is calculated, and the monitoring focus is determined to be either structural safety-oriented or ecological environment-oriented based on the ecological sensitivity index. Based on the monitoring focus, preset parameter adjustment rules are invoked to adjust the first monitoring threshold and the first monitoring frequency, generating a monitoring parameter set that includes the second monitoring threshold and the second monitoring frequency.

[0040] In this embodiment, the monitoring features are those extracted from the current waterway scene information and play a decisive role in setting the monitoring parameters. They are divided into three categories: construction stage features: the current construction process of the waterway, such as dredging, still water revetment, and post-construction maintenance; structural risk features: the safety risk attributes of the waterway structure, such as slope stability level, dam crack risk level, and foundation settlement risk level; and ecological sensitivity features: the sensitivity attributes of the surrounding ecological environment of the waterway, such as the presence of sensitive species, habitat importance level, and water self-purification capacity level. The stage-threshold mapping table is a pre-existing knowledge base table that correlates different construction stages with the basic threshold values ​​of corresponding monitoring indicators. The table clearly defines the basic threshold value of a certain monitoring indicator under a certain construction stage, serving as the basis for determining the basic monitoring threshold.

[0041] In this embodiment, the basic monitoring threshold is an initial monitoring critical value directly matched based on construction stage characteristics and a stage-threshold mapping table, without considering on-site hydrological conditions. The first monitoring threshold is a threshold obtained by correcting the basic monitoring threshold according to current hydrological conditions (e.g., high flow velocity, high water level), which is closer to the actual on-site working conditions. The basic monitoring frequency is an initial monitoring time interval matched based on the risk level in the structural risk characteristics, without considering the intensity of construction activities; for example, the basic monitoring frequency for high-risk slopes is once every 2 hours. The first monitoring frequency is a frequency obtained by correcting the basic monitoring frequency according to the intensity of current construction activities (e.g., high-intensity work, routine work), which can avoid over-monitoring or under-monitoring.

[0042] In this embodiment, the ecological sensitivity index is a numerical value calculated based on ecological sensitivity characteristics to characterize the degree of ecological sensitivity of the waterway. The value ranges from 0 to 1, with higher scores indicating greater ecological sensitivity. The calculation dimensions include the presence of sensitive species, habitat importance, and water body self-purification capacity, serving as an indicator to determine the monitoring focus. The monitoring focus is determined based on the ecological sensitivity index and is divided into two categories: Structural safety orientation: When the ecological sensitivity index is low, the monitoring focus is on waterway structural safety indicators, such as slope settlement and revetment displacement; Ecological environment orientation: When the ecological sensitivity index is high, the monitoring focus is on ecological environment indicators, such as dissolved oxygen, suspended solids concentration, and benthic organism density.

[0043] In this embodiment, the second monitoring threshold and the second monitoring frequency are the final parameters obtained by further adjusting the first monitoring threshold and the first monitoring frequency according to the monitoring focus. They are components of the monitoring parameter set and directly guide the data acquisition work of the monitoring equipment. The monitoring parameter set is a standardized set of parameters including the final monitoring index, the second monitoring threshold, and the second monitoring frequency, and serves as the basis for the coordinated scheduling of monitoring network equipment.

[0044] As can be seen from the above, this embodiment extracts monitoring features, including construction stage features, structural risk features, and ecological sensitivity features, from the current waterway scene information, thus clarifying the key aspects of the target waterway that need to be monitored. Based on the construction stage features and a predefined stage-threshold mapping table, the basic monitoring threshold is determined and corrected according to hydrological conditions, resulting in a first monitoring threshold that better reflects the actual situation. Based on the risk level of the structural risk features, the basic monitoring frequency is determined and corrected according to the intensity of construction activities, resulting in a more reasonable first monitoring frequency. Based on the ecological sensitivity features, the ecological sensitivity index is calculated and the monitoring focus is determined, enabling targeted monitoring work. Based on the monitoring focus, parameter adjustment rules are invoked to adjust the first monitoring threshold and the first monitoring frequency to generate a monitoring parameter set, making the monitoring parameters more accurate and matching the current waterway scene, which is beneficial to improving the accuracy and effectiveness of waterway improvement project monitoring.

[0045] In one embodiment of this application, a basic monitoring threshold is determined based on construction stage characteristics and a predefined stage-threshold mapping table, and then corrected according to the current hydrological conditions to obtain a first monitoring threshold, including: Based on the characteristics of each construction stage, the corresponding construction stage is determined. Based on the construction phase and a predefined phase-threshold mapping table, the corresponding basic monitoring thresholds are determined. The basic monitoring threshold is revised based on the current hydrological parameters and environmental risk characteristics to generate the first monitoring threshold.

[0046] In this embodiment, construction stage features are attributes extracted from the current waterway scenario information that characterize the current stage of the waterway improvement project, such as dredging construction stage, still water revetment construction stage, ecological slope protection restoration stage, and completion maintenance stage. These features serve as the basis for matching basic monitoring thresholds. The predefined stage-threshold mapping table is a structured lookup table pre-existing in a knowledge base, establishing a one-to-one correspondence between construction stages, monitoring indicators, and basic monitoring thresholds. The data in the table is based on waterway improvement industry standards and historical engineering experience; for example, the basic threshold for suspended solids concentration is 100 mg / L for the dredging stage and 50 mg / L for the revetment stage. The basic monitoring thresholds are initial monitoring critical values ​​directly matched from the stage-threshold mapping table, without considering on-site conditions. They serve as the benchmark for determining whether monitoring indicators are abnormal. The basic monitoring thresholds are only strongly correlated with the construction stage and are not adjusted according to real-time environmental conditions.

[0047] In this embodiment, the current hydrological parameters are the hydrological characteristic data of the target waterway area, including flow velocity, water level, sediment content, and flow direction. These are key field parameters for correcting the basic monitoring threshold. For example, high flow velocity accelerates the diffusion of suspended solids, requiring adjustment of the corresponding threshold. Environmental risk characteristics are environmental attributes related to hydrological conditions that affect the determination of monitoring indicators, such as the risk of high water levels during the flood season, the risk of sediment deposition during the dry season, and the risk of water pollution in sensitive ecological areas. These are used to assist in determining the direction and magnitude of threshold correction. The first monitoring threshold is a monitoring critical value that closely matches the actual field conditions, obtained by correcting the basic monitoring threshold based on the current hydrological parameters and environmental risk characteristics. Compared to the basic monitoring threshold, it is more targeted and accurate. As can be seen from the above, this embodiment determines the construction stage by identifying the characteristics of the construction stage, determines the basic monitoring threshold according to the stage-threshold mapping table, and corrects it according to the current hydrological parameters and environmental risk characteristics. This makes the monitoring threshold more in line with the actual situation of the current waterway, improves the accuracy and pertinence of monitoring, and provides a more reliable basis for the monitoring of subsequent waterway improvement projects.

[0048] In one embodiment of this application, a foundation monitoring frequency is determined based on the risk level in the structural risk characteristics, and then corrected according to the current construction activity intensity to obtain a first monitoring frequency, including: The basic monitoring frequency is determined based on the risk level coefficient in the structural risk characteristics; The basic monitoring frequency is adjusted based on the current intensity of construction activities and seasonal hydrological characteristics to generate the first monitoring frequency.

[0049] In this embodiment, structural risk characteristics are a set of attributes directly related to the safety of the waterway structure, extracted from waterway scenario information. These include slope soil type, dam aging degree, historical foundation settlement data, and crack development, serving as the basis for assessing the structural risk level. The risk level coefficient is a numerical value obtained by quantifying the structural risk characteristics, ranging from 0 to 1. A higher coefficient indicates a higher safety risk to the waterway structure; for example, the risk level coefficient for a silty clay slope is 0.7, and for a hard rock slope, it is 0.2. The basic monitoring frequency is an initial monitoring time interval based on the risk level coefficient, without considering the intensity of construction activities and seasonal hydrological characteristics. It is the benchmark value for setting the monitoring frequency and is determined by pre-stored risk level-monitoring frequency mapping rules in the knowledge base.

[0050] In this embodiment, the current construction activity intensity refers to the busyness and disturbance intensity of the waterway improvement project site, categorized into three levels: high intensity (e.g., large-scale dredging, blasting operations), medium intensity (e.g., conventional slope protection laying), and low intensity (e.g., routine inspections, minor repairs). Higher construction intensity results in greater disturbance to the structure, necessitating a higher monitoring frequency. Seasonal hydrological characteristics refer to the hydrological patterns of the waterway in different seasons, such as high water levels and high flow velocities during the flood season, low water levels and low sediment content during the dry season, and ice floes during the ice jam season. Seasonal hydrological changes directly affect structural stability and are a key environmental factor for adjusting the monitoring frequency. The first monitoring frequency is a monitoring time interval obtained by adjusting the base monitoring frequency based on the current construction activity intensity and seasonal hydrological characteristics. This frequency is more closely aligned with the site conditions than the base monitoring frequency, avoiding over-monitoring or under-monitoring.

[0051] As can be seen from the above, this embodiment determines the basic monitoring frequency by using the risk level coefficient of structural risk characteristics, and adjusts it according to the current construction activity intensity and seasonal hydrological characteristics, so that the generated first monitoring frequency can more accurately adapt to the actual situation of the waterway, improve monitoring efficiency and effectiveness, and provide more reliable data support for the monitoring of subsequent waterway improvement projects.

[0052] In one embodiment of this application, an ecological sensitivity index is calculated based on ecological sensitivity characteristics, and the monitoring focus is determined to be either structural safety-oriented or ecological environment-oriented based on the ecological sensitivity index, including: The ecological sensitivity index is calculated based on ecological sensitivity characteristics, which include the presence of sensitive species, the importance level of habitat, the value of ecosystem service functions, and the level of water body self-purification capacity. Based on the ecological sensitivity index, the monitoring focus is determined, including structural safety-oriented monitoring and ecological environment-oriented monitoring.

[0053] In this embodiment, ecological sensitivity features are a set of attributes extracted from waterway scene information that are directly related to the safety of the waterway and its surrounding ecological environment. They are the input parameters for calculating the ecological sensitivity index, including: Presence of sensitive species: the distribution density and population size of protected aquatic / terrestrial organisms in and around the waterway, such as the presence and size of fish spawning grounds and rare waterbird habitats; Habitat importance level: the level of support provided by the waterway ecosystem for the survival and reproduction of sensitive species, divided into core level, important level, and general level; Ecosystem service function value: the quantitative value of the functions provided by the waterway ecosystem, such as water conservation, water purification, and biodiversity maintenance; Water body self-purification capacity level: the level of the waterway water body's ability to dissolve pollutants through physical, chemical, and biological processes, divided into strong, medium, and weak levels.

[0054] In this embodiment, the ecological sensitivity index is a numerical value representing the sensitivity of the waterway's ecological environment, calculated using weighted summation and other methods based on four types of ecological sensitivity characteristics. The value ranges from 0 to 1. The closer the ecological sensitivity index is to 1, the more sensitive the ecological environment, and the higher the risk of being affected by construction or structural anomalies; the closer it is to 0, the lower the ecological sensitivity. The monitoring focus is based on the waterway monitoring guidelines defined by the ecological sensitivity index, serving as the basis for adjusting monitoring thresholds and frequencies. It is divided into two categories: Structural safety-oriented monitoring: When the ecological sensitivity index is low, the monitoring focus is on waterway structural safety indicators, such as slope settlement and slope displacement, prioritizing the stability of the engineering structure; Ecological environment-oriented monitoring: When the ecological sensitivity index is high, the monitoring focus is on ecological environment indicators, such as dissolved oxygen, suspended solids concentration, and benthic organism density, prioritizing the prevention and control of ecological risks.

[0055] As can be seen from the above, this embodiment calculates the ecological sensitivity index by considering the presence of sensitive species, the importance level of habitats, the value of ecosystem services, and the level of water body self-purification capacity, which can quantify the degree of ecological sensitivity. Based on the ecological sensitivity index, the monitoring focus can be determined as either structural safety-oriented or ecological environment-oriented, making the monitoring more targeted, improving monitoring efficiency and accuracy, better adapting to the ecological characteristics of different waterways, and ensuring that river waterway improvement projects can be effectively monitored in terms of both structural safety and ecological environment.

[0056] In one embodiment of this application, a first monitoring threshold and a first monitoring frequency are adjusted according to the monitoring focus to generate a second monitoring threshold and a second monitoring frequency, which serve as a monitoring parameter set, including: If the monitoring focus is on structural safety-oriented monitoring, then the first monitoring frequency of the structure displacement and foundation settlement parameters is increased based on the first adjustment step size to obtain the second monitoring frequency, and the second monitoring threshold is obtained by decreasing the allowable range of the first monitoring threshold of the corresponding parameter based on the second adjustment step size. If the monitoring focus is on ecological environment-oriented monitoring, then the second monitoring threshold is obtained by reducing the first monitoring threshold of suspended matter concentration and dissolved oxygen parameters based on the second adjustment step size, and the second monitoring frequency is obtained by increasing the first monitoring frequency of benthic biodiversity indicators based on the first adjustment step size.

[0057] In this embodiment, the monitoring focus is based on the monitoring orientation defined by the ecological sensitivity index, divided into two categories: structural safety-oriented monitoring and ecological environment-oriented monitoring, which serve as the basis for adjusting monitoring parameters. The first adjustment step size is used to adjust the monitoring frequency, and the second adjustment step size is used to adjust the monitoring threshold. The first adjustment step size is used to increase the monitoring frequency, for example, by increasing the frequency by 50% and shortening the interval to 2 / 3 of the original duration. The second adjustment step size is used to decrease the monitoring threshold (tightening the allowable range), for example, by lowering the threshold by 10% and reducing the allowable deviation to 0.9 times the original range.

[0058] Specifically, the calculation of the first adjustment step size (used for monitoring frequency adjustment) needs to be divided into two steps based on the importance weight of the indicators and the intensity of disturbances in the field conditions: First, determine the importance weight W of the indicators. Based on the monitoring focus, assign high weights to core indicators and a weight of 1 (no adjustment) to non-core indicators. Structural safety orientation: Structural indicators (settlement, displacement) W=1.5-2.0; Ecological indicators W=1.0. Ecological environment orientation: Ecological indicators (suspended matter, dissolved oxygen) W=1.5-2.0; Structural indicators W=1.0.

[0059] Secondly, corrections are made based on the disturbance intensity K under the working conditions. The higher the disturbance intensity, such as during high-intensity construction or high water levels during the flood season, the larger the step size coefficient, and the more significant the frequency increase, as shown in Table 1. Table 1. Values ​​of Disturbance Intensity under Operating Conditions

[0060] Finally, the first adjustment step size coefficient = W×K, and the second monitoring frequency = the first monitoring frequency / the first adjustment step size.

[0061] For example, a certain waterway is ecologically oriented, with benthic biodiversity as the core indicator, and the current construction disturbance intensity is medium disturbance: the indicator importance weight W=2.0, the working condition disturbance correction coefficient K=1.2, the first adjustment step size=2.0×1.2=2.4, if the first monitoring frequency is once a week, then the second monitoring frequency=7 / 2.4≈3 days / time.

[0062] Specifically, the calculation of the second adjustment step size (used for monitoring threshold adjustment) needs to be divided into two steps based on the indicator's risk impact and ecological / structural safety standards: First, determine the indicator's risk impact weight R, as shown in Table 2. The higher the risk impact, for example, if suspended solids exceed the standard and directly affect fish spawning, the smaller the step size coefficient, and the more stringent the threshold tightening.

[0063] Table 2 Weighting of Indicator Risk Impact

[0064] Secondly, adjustments are made based on the threshold safety margin S. The safety margin is the ratio of the difference between the threshold and the mandatory industry standard. The smaller the difference, the smaller the coefficient, to prevent the threshold from exceeding the mandatory standard. S = 1 - (first monitoring threshold / mandatory industry standard threshold). If the first monitoring threshold is greater than or equal to the mandatory standard, then S = 0, adjustment is prohibited, and an immediate warning is required.

[0065] Finally, the second adjustment step size = R × (1-S), and the second monitoring threshold = the first monitoring threshold × the second adjustment step size.

[0066] For example, for a certain waterway with an ecological environment orientation, the concentration of suspended solids is the core indicator: the first monitoring threshold = 40 mg / L, the industry mandatory standard threshold = 50 mg / L, the risk impact weight R = 0.8, the safety margin S = 1 - (40 / 50) = 0.2, the second adjustment step size = 0.8 × (1 - 0.2) = 0.64, then the second monitoring threshold = 40 × 0.64 = 25.6 mg / L.

[0067] In this embodiment, the first monitoring threshold is an initial threshold based on the construction phase matching basic threshold, corrected for hydrological conditions, such as slope settlement of 0.3 cm / day and suspended solids concentration of 40 mg / L; the first monitoring frequency is an initial frequency based on the structural risk level matching basic frequency, corrected for construction intensity and seasonal hydrological characteristics, such as slope settlement once every 4 hours and suspended solids concentration once daily. The second monitoring threshold / second monitoring frequency is the final monitoring parameter obtained by optimizing the first monitoring threshold and first monitoring frequency with corresponding adjustment step sizes according to the monitoring focus: strengthening structural indicator monitoring under structural safety guidance (increasing frequency and tightening thresholds), and strengthening ecological indicator monitoring under ecological environment guidance (increasing frequency and tightening thresholds).

[0068] In this embodiment, the structural displacement / foundation settlement parameters are monitoring indicators representing the stability of the waterway structure: structural displacement includes horizontal displacement of the slope protection and lateral displacement of the dike; foundation settlement includes slope settlement and waterway bed settlement, and is an indicator for structural safety-oriented monitoring. The suspended solids concentration / dissolved oxygen / benthic biodiversity indicators are monitoring indicators representing the ecological environment quality of the waterway: suspended solids concentration and dissolved oxygen directly affect water quality and the survival of aquatic organisms; benthic biodiversity represents ecosystem integrity and is an indicator for ecological environment-oriented monitoring.

[0069] As can be seen from the above, this embodiment, by increasing the monitoring frequency of structural displacement and foundation settlement parameters and reducing the allowable range of monitoring thresholds for the corresponding parameters when the monitoring focus is on structural safety-oriented monitoring, can more accurately monitor the structural safety status and promptly detect potential structural safety problems; when the monitoring focus is on ecological environment-oriented monitoring, by reducing the monitoring thresholds for suspended solids concentration and dissolved oxygen parameters and increasing the monitoring frequency of benthic biodiversity indicators, it can better grasp the changes in the ecological environment and promptly detect ecological environment anomalies.

[0070] In one embodiment of this application, it further includes, If the construction stage characteristics are in the dredging construction stage, and the hydrodynamic characteristics indicate that the flow velocity is greater than the first threshold, then the second monitoring frequency of the suspended solids concentration index is increased based on the third adjustment step size to obtain the third monitoring frequency. If the construction stage is in the still water revetment construction stage, and the presence of sensitive species in the ecologically sensitive characteristics is less than the preset presence threshold, then the third monitoring frequency is obtained by reducing the second monitoring frequency of the comprehensive water quality index based on the fourth adjustment step size.

[0071] Specifically, the calculation of the third adjustment step size quantifies the risk of suspended solids diffusion under high flow velocities based on the magnitude of flow velocity exceeding the limit and the intensity of dredging operations. First, the flow velocity exceeding the limit coefficient L is determined. The greater the magnitude of the flow velocity exceeding the limit, the faster the suspended solids diffuse, and the higher the step size. L = 1 + (current channel actual flow velocity - first flow velocity threshold / first flow velocity threshold). Second, the dredging operation intensity coefficient Q is determined. The higher the dredging operation intensity, the more intense the sediment agitation, the greater the fluctuation in suspended solids concentration, and the higher the step size, as shown in Table 3. Table 3 Dredging Operation Intensity Table

[0072] The final calculation formula is: Third adjustment step size = L × Q, with the constraint that the final third adjustment step size is less than or equal to 3.0, thus avoiding excessive monitoring frequency that could increase equipment wear and labor costs.

[0073] For example, a waterway is in a high-intensity dredging phase (average daily dredging volume 2500 m³), ​​with a measured flow velocity of 0.8 m / s and a first velocity threshold of 0.6 m / s. The velocity exceedance coefficient is calculated as follows: L = 1 + (0.8 + 0.6 / 0.6) ≈ 1.33; the matching dredging intensity coefficient is: Q = 1.5 for high-intensity dredging; the third adjustment step is calculated as: 1.33 × 1.5 = 1.995 ≈ 2.0; the frequency adjustment example is: if the second monitoring frequency for suspended solids concentration is twice daily, then the third monitoring frequency = 2 × 2.0 = 4 times / day.

[0074] Specifically, the calculation of the fourth adjustment step size needs to quantify the reduction in monitoring frequency under low-risk conditions based on the degree of ecological sensitivity deficiency and the intensity of construction disturbance. First, determine the ecological sensitivity deficiency coefficient P. The lower the presence of sensitive species, the smaller the ecological risk, and the smaller the step size (the greater the frequency reduction). P = current quantitative value of sensitive species presence / preset presence threshold of sensitive species.

[0075] Secondly, the disturbance intensity coefficient Y during the construction of the still water revetment was determined. The smaller the construction disturbance, the weaker the impact on water quality, and the lower the step size, as shown in Table 4. Table 4. Disturbance Intensity During Still Water Bank Revetment Construction

[0076] Finally, the fourth adjustment step size = P × Y, with the following constraints: the final fourth adjustment step size is greater than or equal to 0.3, and at least 30% of the original monitoring frequency is retained.

[0077] For example, a waterway is in a low-disturbance phase of calm water revetment (ecological grid slope protection), with a measured value of 0.2 for the presence of sensitive species and a preset presence threshold of 0.5. The ecological sensitivity deficiency coefficient is calculated as: P = 0.2 / 0.5 = 0.4; the construction disturbance intensity coefficient is matched: Y = 0.8 for low disturbance; the fourth adjustment step is calculated as: 0.4 × 0.8 = 0.32. Frequency adjustment example: If the second monitoring frequency for comprehensive water quality indicators is twice daily, then the third monitoring frequency is approximately 2 × 0.32 ≈ 0.64 times / day, rounded down to once daily to ensure basic data collection.

[0078] In this embodiment, the dredging construction stage is a construction process in waterway improvement projects that uses dredgers and other equipment to remove silt and sediment from the waterway bed, widening or deepening the waterway. This stage significantly agitates the water, causing a sharp increase in suspended solids concentration, making it a critical stage for water quality monitoring. Hydrodynamic characteristics—flow velocity—are the speed at which water flows through the waterway and are a hydrological parameter affecting the spread of suspended solids: the higher the flow velocity, the faster and wider the spread of suspended solids; the lower the flow velocity, the easier it is for suspended solids to accumulate locally. The first threshold is a pre-existing critical value in the knowledge base used to determine whether the flow velocity needs to trigger an adjustment of the monitoring frequency. It is determined by the correlation analysis between the waterway hydrological conditions and the intensity of disturbance during dredging construction. For example, the first threshold for flow velocity during the dredging construction stage of inland waterways is often set to 0.6 m / s. The third adjustment step is an adjustment range set for the special working condition of increased flow velocity during dredging construction, used to increase the monitoring frequency of suspended solids concentration. Essentially, it is a frequency boosting coefficient, with a value greater than 1 and an amplitude greater than the first adjustment step, to cope with rapid water quality changes under high-risk conditions.

[0079] In this embodiment, the still-water revetment construction stage involves laying slope protection structures such as masonry and ecological grids in a waterway section with slow water flow. This stage minimizes construction disturbance and has a low impact on water quality. The presence of sensitive species refers to the distribution density and population size of sensitive species such as fish spawning grounds and rare aquatic organisms within the waterway. The preset presence threshold is a critical value for determining whether sensitive species are scarce; values ​​below the preset threshold indicate a low distribution of sensitive species and a low sensitivity of the ecological environment to construction disturbance. The fourth adjustment step is set for the special condition of still-water revetment with low sensitive species presence, used to reduce the frequency of comprehensive water quality indicator monitoring. Essentially, it is a frequency reduction coefficient, with a value less than 1, to reduce meaningless monitoring costs. The comprehensive water quality indicator is a composite evaluation index that summarizes multiple water quality parameters such as suspended solids concentration, dissolved oxygen, and pH value, used to comprehensively determine the waterway's environmental quality. The third monitoring frequency is the final execution monitoring frequency obtained by further optimizing the second monitoring frequency based on the special working conditions of the construction stage and hydrological / ecological characteristics through a third or fourth adjustment step size. It serves as the final basis for guiding the data collection of monitoring equipment.

[0080] As can be seen from the above, this embodiment can capture changes in suspended solids concentration more promptly and accurately by increasing the monitoring frequency of suspended solids concentration indicators during the dredging construction stage and when the flow velocity is greater than the first threshold, thus avoiding the failure to detect abnormal suspended solids concentration due to high flow velocity. During the still water revetment construction stage and when the presence of sensitive species is less than the preset presence threshold, the monitoring frequency of comprehensive water quality indicators can be reduced, ensuring necessary monitoring while reducing unnecessary monitoring work, improving monitoring efficiency, and reducing monitoring costs.

[0081] In one embodiment of this application, based on multi-source monitoring data and by calling an anomaly diagnosis rule base for anomaly identification, potential anomalies and anomaly confidence levels are obtained, including: Spatiotemporal alignment and feature extraction are performed on multi-source monitoring data to construct a spatiotemporal state matrix for the waterway; The channel spatiotemporal state matrix is ​​input into the anomaly diagnosis rule base, which contains predefined rules including the riverbed siltation pattern, the structural displacement mutation pattern, and the water quality anomaly diffusion pattern. By combining pattern matching and confidence calculation, the system outputs the matched anomaly pattern type, its location, and the corresponding anomaly confidence level.

[0082] In this embodiment, spatiotemporal alignment and feature extraction are two preprocessing operations performed on multi-source monitoring data: spatiotemporal alignment matches and associates data collected from different devices, locations, and times according to a unified timestamp and spatial coordinates, eliminating spatiotemporal misalignment; feature extraction extracts key features characterizing the channel state from the aligned data, such as settling rate, suspended solids concentration increase, and flow velocity gradient, providing feature input for subsequent anomaly identification. The channel spatiotemporal state matrix is ​​a structured data matrix constructed based on the spatiotemporal alignment and feature extraction results. The row dimension of the matrix is ​​the monitoring point / channel segment, the column dimension is the time series plus core monitoring features, and the matrix elements are the feature quantification values ​​of the corresponding point-time, serving as a data carrier describing the state distribution of the channel in the spatial and temporal dimensions. The anomaly diagnosis rule base is a pre-constructed set of judgment rules for typical anomalies in channel improvement projects. The base includes three elements: anomaly mode type, triggering condition, and judgment logic. This embodiment explicitly includes three types of anomaly modes: sudden siltation mode, structural displacement mutation mode, and water quality anomaly diffusion mode.

[0083] In this embodiment, the sudden siltation mode is an abnormal state where the siltation rate in the riverbed is far greater than the normal level within a short period of time, and the determination rule is related to features such as siltation rate and duration. The abrupt structural displacement mode is an abnormal state where the displacement of structures such as channel slopes and revetments experiences a jump in a short period of time, and the determination rule is related to features such as single displacement increment and daily average displacement rate. The abnormal water quality diffusion mode is an abnormal state where water quality indicators such as suspended solids and pollutants rapidly diffuse from the construction area to the surrounding waters, and the determination rule is related to features such as concentration increase, diffusion range, and flow velocity.

[0084] In this embodiment, pattern matching involves comparing the feature values ​​in the spatiotemporal state matrix of the airway with the triggering conditions of various patterns in the anomaly diagnosis rule base one by one to filter out anomaly patterns that meet the conditions. Confidence calculation is based on the degree of fit between the feature values ​​and the rule triggering conditions, the accuracy of the monitoring data, and the probability of occurrence of similar anomalies in the past. It quantifies the probability of the anomaly pattern's authenticity, taking a value from 0 to 1, with higher values ​​indicating a more credible anomaly. Potential anomalies are the combination of the anomaly pattern type and its spatiotemporal location output after pattern matching, such as the sudden displacement pattern of segment K12+300. Anomaly confidence is the quantified credibility value corresponding to the potential anomaly, and it serves as the basis for determining whether to initiate UAV verification later.

[0085] As can be seen from the above, this embodiment can summarize multi-source data for subsequent analysis by performing spatiotemporal alignment and feature extraction on multi-source monitoring data and constructing a spatiotemporal state matrix. By inputting the spatiotemporal state matrix of the waterway into a predefined anomaly diagnosis rule base with rules for sudden siltation patterns, structural displacement mutation patterns, and water quality anomaly diffusion patterns, anomaly identification can be performed using the rules. By outputting the matched anomaly pattern type, occurrence location, and anomaly confidence level through pattern matching and confidence level calculation, potential anomalies and their confidence levels can be accurately determined, providing a basis for subsequent on-site verification and decision-making.

[0086] In one embodiment of this application, based on reliability assessment results, the spatiotemporal location of future maintenance needs, and current early warning information, a collaborative scheduling strategy is generated and executed through an optimization algorithm, including: The optimization objectives are to minimize total maintenance cost, minimize air traffic disruption time, and minimize operational safety risks, while taking into account the availability of maintenance resources, time window limitations, and environmental constraints. The Pareto optimal solution set for task sequence, resource allocation, and time arrangement is generated by using a non-dominated sorting genetic algorithm. Based on the Pareto optimal solution set, the final scheduling scheme is selected as the cooperative scheduling strategy and executed.

[0087] In this embodiment, the multiple optimization objectives are three optimization directions that need to be satisfied simultaneously in the generation of this collaborative scheduling strategy. These three objectives are mutually restrictive and need to be balanced: Minimizing total maintenance cost: controlling the total cost of manpower, equipment, materials, and other resources to avoid cost waste; Minimizing navigation interference time: reducing the duration of waterway occupation during maintenance operations to reduce the impact on ship passage efficiency; Minimizing operational safety risks: reducing safety hazards to construction personnel, equipment, and waterway structures to avoid safety accidents. Constraints are the limiting factors that must be followed when formulating the scheduling strategy, and are the boundary conditions for the optimization algorithm to solve: Maintenance resource availability: the quantity and available time periods of available work teams, equipment (e.g., grouting machines, drones), and materials; Time window limitation: maintenance operations need to be completed within a time range where the future maintenance demand is determined in time and space to avoid risks caused by exceeding the limits; Environmental constraints: operations need to avoid severe weather (e.g., heavy rain, strong winds), ecologically sensitive periods (e.g., fish spawning season), and also comply with navigation rules, such as avoiding peak navigation periods. Non-dominated sorting genetic algorithm (NSGA-II) is an intelligent algorithm for solving multi-objective optimization problems. By simulating the selection, crossover and mutation process of biological evolution, it filters a large number of scheduling schemes and finally obtains a set of optimal schemes that are not dominated by each other (Pareto optimal solution set). It is suitable for complex scheduling scenarios with multiple objectives that are mutually constrained.

[0088] In this embodiment, the Pareto optimal solution set is the set of solutions in multi-objective optimization where no single solution can improve one objective without harming others. Each solution in the solution set is a feasible and high-quality scheduling strategy, and the final solution needs to be selected based on actual engineering needs (e.g., prioritizing navigation). The collaborative scheduling strategy is an integrated execution plan selected from the Pareto optimal solution set, including task sequence, resource allocation, and time arrangement, and serves as the basis for guiding on-site maintenance operations. Execution is the entire process of distributing the collaborative scheduling strategy to various execution units, such as maintenance teams, maritime departments, and equipment management centers, completing maintenance tasks according to the plan, and simultaneously providing feedback on the execution status.

[0089] As can be seen from the above, this embodiment takes minimizing the total maintenance cost, minimizing the navigation interference time, and minimizing the operational safety risk as multiple optimization objectives. By considering the availability of maintenance resources, time window limitations, and environmental constraints, and using a non-dominated sorting genetic algorithm to obtain the Pareto optimal solution set, the final scheduling scheme is selected as the collaborative scheduling strategy for execution. This can generate a scientific and reasonable collaborative scheduling strategy, effectively reducing maintenance costs, navigation interference, and operational safety risks.

[0090] In one embodiment of this application, it further includes: Execute the coordinated scheduling strategy and collect channel status feedback data after execution; The actual monitored values ​​in the channel status feedback data are compared with the predicted values ​​of the time series prediction model to obtain the comparison deviation value; When the comparison deviation value is greater than the preset deviation threshold, the time series prediction model is optimized and updated using the particle swarm optimization algorithm; Analyze the difference between the execution effect and the expected goal of the collaborative scheduling strategy, and based on the difference, revise the mapping relationship or adjustment rule in the knowledge base used to generate the monitoring parameter set, anomaly diagnosis rule and anomaly threshold set.

[0091] In this embodiment, the collaborative scheduling strategy execution involves distributing the optimized maintenance task sequence, resource allocation plan, and time schedule to each execution unit (maintenance team, maritime department, monitoring center), and completing the entire waterway maintenance operation according to the plan. This is the key step in strategy implementation. Waterway status feedback data, collected through monitoring network equipment after strategy execution, includes: structural safety indicators (slope settlement, slope crack width); ecological environment indicators (suspended solids concentration, dissolved oxygen); and navigation impact indicators (waterway traffic efficiency). This data serves as the basis for evaluating the effectiveness of strategy execution. The comparison deviation value is the absolute or relative difference between the actual monitored value in the waterway status feedback data and the predicted value output by the time-series prediction model. It represents the accuracy of the prediction model and is calculated using the formula: Comparison Deviation Value = |Actual Monitored Value - Predicted Value / Predicted Value| × 100%.

[0092] In this embodiment, the preset deviation threshold is a critical value pre-existing in the knowledge base to determine whether the prediction model needs optimization. It is determined by the model accuracy requirements and actual engineering needs; for example, it is set to 15%-20% for inland waterways. A deviation value greater than the preset deviation threshold indicates insufficient model prediction accuracy. Particle Swarm Optimization (PSO) is a swarm intelligence-based optimization algorithm that iteratively updates the position and velocity of particles by simulating the foraging behavior of bird flocks, searching for the optimal parameter combination for the time-series prediction model.

[0093] Specifically, the Particle Swarm Optimization (PSO) algorithm adopts a four-layer closed-loop structure: particle encoding layer, population initialization layer, iterative update layer, and fitness evaluation layer. The functions of each layer are closely interconnected: Particle Encoding Layer: Maps the parameters to be optimized in the time-series prediction model (e.g., learning rate, number of hidden layer nodes, number of iterations) to particle position vectors. Each particle corresponds to a complete set of model parameters. The particle dimension is consistent with the number of parameters to be optimized; for example, if there are 3 parameters to be optimized, the particle dimension is 3. Population Initialization Layer: Randomly generates a preset number of particles based on the parameter value range to form the initial population, ensuring that the particle distribution covers the entire parameter search space, providing a diverse foundation for subsequent iterative optimization. Iterative Update Layer: Includes two core sub-modules: velocity update and position update. By adjusting the particle's flight speed and position through the individual particle's optimal position (its own historical optimal parameter combination) and the population's global optimal position (the entire population's historical optimal parameter combination), it achieves a balance between accuracy and diversity in parameter search. Fitness Evaluation Layer: Using the minimization of the time-series prediction model's contrast deviation as the fitness function, it calculates the fitness value corresponding to each particle, serving as the basis for judging the quality of parameter combinations and driving the population to converge towards the optimal parameter region.

[0094] In this embodiment, the time-series prediction model optimization update involves adjusting the model's parameters using a particle swarm optimization algorithm when the comparison deviation exceeds a preset deviation threshold. The model is then retrained and its accuracy verified, resulting in an updated model that outputs predictions that better reflect reality. The difference between the execution effect and the expected target is the gap between the actual execution result of the collaborative scheduling strategy (e.g., maintenance cost, air traffic interference time, risk control effectiveness) and the preset optimization target. For example, the actual maintenance cost is 280,000 yuan compared to the expected target of 250,000 yuan, and the actual air traffic interference is 40 hours compared to the expected target of 36 hours.

[0095] In this embodiment, knowledge base modification is based on the difference between the execution effect and the expected goal. The content in the knowledge base that supports the generation of monitoring parameters, anomaly diagnosis, and threshold setting is adjusted, such as the stage-threshold mapping table, risk-frequency mapping rules, and anomaly diagnosis mode triggering conditions, to achieve closed-loop optimization of algorithms, models, and rules.

[0096] As can be seen from the above, this embodiment executes a collaborative scheduling strategy and collects channel status feedback data. By comparing the actual monitored values ​​with the predicted values ​​to obtain the deviation value, and optimizing and updating the time-series prediction model when the deviation value is greater than a preset threshold, the model prediction can be made more accurate. Analyzing the difference between the execution effect of the collaborative scheduling strategy and the expected goal, and correcting the knowledge base mapping relationship or adjusting the rules, can make the knowledge base more adaptable to the monitoring needs of the channel improvement project, and improve the accuracy and reliability of the monitoring method.

[0097] In one embodiment of this application, it further includes: Construction activity data and ecological environment indicator data from historical and current multi-source monitoring data are subjected to spatiotemporal standardization to construct a spatiotemporal database of ecological impacts; ecological environment indicator data include water quality parameters, benthic animal community data, and fish resource data. Based on the spatiotemporal database of ecological impacts, spatial distribution maps and time series curves of key ecological and environmental indicators are generated at different stages before, during and after construction. Based on spatial distribution maps and time series curves, statistical testing methods were used to analyze the differences in ecological and environmental indicators between the construction area and the control area, as well as between different stages of construction, to obtain the indicators with the greatest impact on the ecological and environmental environment. The spatial distribution and time series of the indicators with the greatest impact on the ecological environment are spatially overlaid and coupled with the construction activity data of the same period to identify the dominant driving factors that lead to changes in the ecological environment. The dominant driving factors include construction procedures, construction intensity and construction location. Based on the driving factors of advantages, we identify the key construction links and ecologically sensitive areas that have the greatest impact on the ecological environment, and output decision-making recommendations for optimizing construction technology, adjusting construction sequence and implementing ecological protection measures.

[0098] In this embodiment, spatiotemporal standardization is a preprocessing operation that unifies the format and dimensions of historical and current construction activity data and ecological environment indicator data. Spatial standardization unifies the coordinates of different monitoring points to the same geographic coordinate system, and temporal standardization unifies data from different collection frequencies to the same time granularity, such as daily or weekly, eliminating inconsistencies in the spatiotemporal dimensions of the data. The ecological impact spatiotemporal database is a structured database built based on the spatiotemporally standardized data. It includes a construction activity dataset (the spatiotemporal distribution of construction procedures, intensity, and location) and an ecological environment indicator dataset (spatiotemporal monitoring values ​​of water quality, benthic animals, and fish resources), serving as the data carrier for analyzing the ecological impact of construction.

[0099] In this embodiment, the spatial distribution map uses a map as a base to visually display the spatial distribution characteristics of a certain ecological indicator in the construction area and surrounding areas, such as the gradient of suspended solids concentration and the density distribution of benthic animals. The time series curve uses time as the horizontal axis and indicator value as the vertical axis to show the changing trends of key ecological indicators at different stages before, during, and after construction, such as the fluctuation curve of dissolved oxygen concentration. Statistical testing methods are mathematical methods used to analyze the significance of data differences, such as independent samples t-tests and analysis of variance (ANOVA). In this embodiment, they are used to compare the differences in ecological indicators between the construction area and the control area, and at different stages of construction, to determine whether the differences are caused by construction activities or random factors. The ecologically impactful indicator is the ecologically impactful indicator that, after statistical testing, shows the differences between the construction area and the control area, and between different stages of construction, such as suspended solids concentration and the number of benthic animal species. These are characterizing indicators of the ecological impact of construction activities.

[0100] In this embodiment, spatial overlay analysis overlays the spatial distribution map of the most impactful ecological indicators with the spatial distribution map of concurrent construction activities to determine their spatial correlation, such as whether areas with high suspended solids concentrations overlap with dredging construction areas. Temporal series coupling analysis correlates the temporal variation curves of the most impactful ecological indicators with the temporal variation curves of concurrent construction intensity to determine their temporal response relationship, such as whether the peak of construction intensity is synchronized with the abnormal peak of ecological indicators. The dominant driving factors are the construction-related factors that play a leading role in ecological and environmental changes after coupling analysis; in this embodiment, these are specifically defined as construction procedures (dredging / bank protection), construction intensity (high / medium / low), and construction location (main channel / tributary estuary). Decision recommendations are targeted optimization schemes proposed based on the identification results of dominant driving factors, key construction links, and ecologically sensitive areas. These include: optimizing construction techniques, such as using environmentally friendly dredging vessels; adjusting the construction sequence, such as avoiding fish spawning seasons; and implementing ecological protection measures, such as setting up suspended solids isolation barriers.

[0101] As can be seen from the above, this embodiment constructs an ecological impact spatiotemporal database by performing spatiotemporal standardization processing on construction activity data and ecological environment indicator data, which can summarize historical and current multi-source monitoring data. Based on this database, spatial distribution maps and time series curves of key ecological environment indicators in different periods are generated, which is conducive to intuitively displaying changes in ecological environment indicators. By using statistical testing methods to analyze differences, the ecological environment indicators with the greatest impact can be obtained, which can identify the ecological environment indicators with the greatest impact. Analyzing the ecological environment indicators with the greatest impact and construction activity data to identify dominant driving factors can identify the factors that lead to changes in the ecological environment. Based on the dominant driving factors, key construction links and ecologically sensitive areas are identified and decision-making suggestions are output, which helps to optimize construction technology, adjust construction sequence, and implement ecological protection measures to reduce the impact on the ecological environment.

[0102] In one embodiment of this application, the spatiotemporal location of future maintenance needs is obtained by analyzing the monitoring trend prediction results and preset maintenance thresholds, including: The target waterway area is divided into regular or irregular spatiotemporal prediction units, and each unit is associated with spatial coordinates and a time slice of a future preset period. For each spatiotemporal prediction unit, the predicted value of the monitoring indicator in the corresponding time slice in the monitoring trend prediction result is compared with the corresponding indicator threshold in the preset maintenance critical threshold. If, within a spatiotemporal prediction unit, the predicted value of at least one monitoring indicator will be greater than or equal to its preset maintenance critical threshold within the corresponding time slice, then the monitoring indicator has a maintenance requirement. The overall urgency score of maintenance needs is calculated based on the number of indicators that are greater than or equal to the preset maintenance critical threshold, the importance level of the indicators, and the magnitude by which the predicted value is greater than or equal to the preset maintenance critical threshold. Cluster maintenance requirement units that are spatially and temporally adjacent and have similar comprehensive urgency scores to obtain a maintenance requirement set. Based on the spatial boundaries and suggested maintenance time windows in the set of maintenance needs, the spatiotemporal location of future maintenance needs can be obtained.

[0103] In this embodiment, the spatiotemporal prediction unit is the smallest analysis unit formed by dividing the target waterway area according to spatial division rules and time slice granularity. Each unit is simultaneously bound to a spatial coordinate range (e.g., waterway chainage segment K12+200-K12+300, slope area) and a future time slice, such as the 3rd day in the future, or the 5th-7th day in the future, serving as the basic carrier for maintenance demand analysis. The division method can flexibly select regular grids (e.g., 50m×50m spatial grid + 1-day time slice) or irregular units, such as division according to structural risk zones or construction sections. A time slice is a segment of the time dimension of a future preset period, with granularity determined according to the urgency of maintenance needs, such as short-term (1 day / slice), medium-term (3 days / slice), and long-term (7 days / slice), used to locate the time nodes of maintenance needs.

[0104] In this embodiment, maintenance requirement is determined when the predicted value of at least one monitoring indicator within a spatiotemporal prediction unit is greater than or equal to a preset maintenance critical threshold, thus establishing the necessity of maintenance for that unit and triggering the condition of indicator exceeding limits. The comprehensive urgency score is calculated based on three dimensions: the number of indicators exceeding limits within the unit, the importance level of the indicators, and the magnitude of the exceedance. It is used to represent the urgency of the maintenance requirement; a higher score indicates a higher priority for handling. Maintenance requirement unit clustering is the process of merging spatially adjacent spatiotemporal prediction units with similar comprehensive urgency scores. The purpose is to eliminate isolated units, delineate concentrated maintenance operation areas, and reduce the cost and navigation interference of decentralized operations.

[0105] In this embodiment, the maintenance requirement set is a clustered group of maintenance requirements that includes a centralized spatial boundary and a unified time window. It serves as the direct basis for outputting the spatiotemporal location of future maintenance requirements. The spatiotemporal location of future maintenance requirements refers to the target interval for maintenance operations determined based on the maintenance requirement set, which combines spatial range (e.g., waterway chainage interval, specific structural part) and time window (e.g., the 5th to 7th day in the future). It is the input that guides the generation of subsequent scheduling strategies.

[0106] As can be seen from the above, this embodiment can accurately locate the area to be monitored by dividing the target waterway area into spatiotemporal prediction units; it can accurately identify the monitoring indicators with maintenance needs by comparing the predicted values ​​of the monitoring indicators with the preset maintenance critical thresholds; it can calculate the comprehensive urgency score of the maintenance needs to represent the urgency of the maintenance needs; it can cluster the maintenance need units to obtain a set of maintenance needs, which facilitates the overall planning of maintenance work; and it can finally obtain the spatiotemporal location of future maintenance needs, which helps to arrange maintenance work in advance, rationally allocate resources, and improve the maintenance efficiency and pertinence of river waterway improvement projects.

[0107] Corresponding to the river channel improvement project monitoring method in the above embodiment, Figure 2 This is a structural block diagram of a river channel improvement engineering monitoring system provided according to an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The monitoring system 20 for the river channel improvement project includes: a data retrieval module 21, a data acquisition module 22, an anomaly identification module 23, a problem verification module 24, a trend prediction module 25, an early warning generation module 26, and a generation strategy module 27.

[0108] Among them, the data retrieval module 21 is used to retrieve and generate a matching set of monitoring parameters, an anomaly diagnosis rule base and an anomaly threshold set from a pre-built knowledge base based on the current waterway scenario information; Data acquisition module 22 is used to coordinate and schedule monitoring network devices based on the monitoring parameter set to acquire multi-source monitoring data; Anomaly identification module 23 is used to identify anomalies based on multi-source monitoring data and call the anomaly diagnosis rule base to obtain potential anomalies and anomaly confidence levels. Problem verification module 24 is used to control the drone in the monitoring network equipment to conduct on-site verification of potential abnormal problems and generate reliability assessment results if the anomaly confidence level is greater than the preset anomaly threshold. The trend prediction module 25 is used to input multi-source monitoring data into a preset time series prediction model to obtain the monitoring trend prediction result if the reliability assessment result is greater than the preset reliability threshold. The early warning generation module 26 is used to compare the monitoring trend prediction results with the maintenance critical threshold in the abnormal threshold set to obtain the spatiotemporal location of future maintenance needs, and calculate the waterway comprehensive monitoring index through the mechanism-data hybrid model. If the comprehensive monitoring index is greater than the preset early warning threshold, the current early warning information is generated. The strategy generation module 27 is used to generate and execute a collaborative scheduling strategy based on the reliability assessment results, the spatiotemporal location of future maintenance needs, and current early warning information through an optimization algorithm.

[0109] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the modules in the aforementioned device embodiments, for example... Figure 2 The functions of the data retrieval module 21, data acquisition module 22, anomaly identification module 23, problem verification module 24, trend prediction module 25, early warning generation module 26, and generation strategy module 27 are shown.

[0110] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0111] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0112] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.

[0113] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation methods described in any embodiment of the river channel regulation engineering monitoring method provided in the embodiments of this application, or they can execute the implementation methods of the electronic devices described in the embodiments of this application, which will not be repeated here.

[0114] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0115] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0116] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0117] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0118] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or units, or it may be an electrical, mechanical, or other form of connection.

[0119] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0120] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0121] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered 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 monitoring method for river channel improvement projects, characterized in that, include: Based on the current waterway scenario information, a matching set of monitoring parameters, an anomaly diagnosis rule base, and an anomaly threshold set are generated by calling from a pre-built knowledge base; Based on the monitoring parameter set, the monitoring network equipment is coordinated and scheduled to collect multi-source monitoring data. Based on the multi-source monitoring data, the anomaly diagnosis rule base is invoked to identify anomalies and obtain potential anomalies and anomaly confidence levels. If the anomaly confidence level is greater than the preset anomaly threshold, then the drone in the monitoring network device is controlled to perform on-site verification of the potential anomaly and generate a reliability assessment result. If the reliability assessment result is greater than the preset reliability threshold, the multi-source monitoring data will be input into the preset time series prediction model to obtain the monitoring trend prediction result. The monitoring trend prediction results are compared with the maintenance critical threshold in the abnormal threshold set to obtain the spatiotemporal location of future maintenance needs. The comprehensive waterway monitoring index is calculated through the mechanism-data hybrid model. If the comprehensive monitoring index is greater than the preset warning threshold, the current warning information is generated. Based on the reliability assessment results, the spatiotemporal location of future maintenance needs, and current early warning information, a collaborative scheduling strategy is generated and executed through an optimization algorithm.

2. The monitoring method for river channel improvement projects according to claim 1, characterized in that, The step of generating a matching set of monitoring parameters from a pre-built knowledge base based on the current waterway scenario information includes: Feature extraction is performed on the current waterway scene information to obtain the monitoring features of the target waterway area. The monitoring features include construction stage features, structural risk features, and ecological sensitivity features. Based on the construction stage characteristics and the predefined stage-threshold mapping table, the basic monitoring threshold is determined and corrected according to the current hydrological conditions to obtain the first monitoring threshold. Based on the risk level in the structural risk characteristics, the basic monitoring frequency is determined and then corrected according to the current construction activity intensity to obtain the first monitoring frequency; Based on the aforementioned ecological sensitivity characteristics, an ecological sensitivity index is calculated, and the monitoring focus is determined to be either structural safety-oriented or ecological environment-oriented based on the ecological sensitivity index. Based on the monitoring focus, preset parameter adjustment rules are invoked to adjust the first monitoring threshold and the first monitoring frequency, generating a monitoring parameter set including the second monitoring threshold and the second monitoring frequency.

3. The monitoring method for river channel improvement projects according to claim 2, characterized in that, Based on the construction stage characteristics and a predefined stage-threshold mapping table, the basic monitoring threshold is determined and corrected according to the current hydrological conditions to obtain the first monitoring threshold, including: Based on the characteristics of the construction stages, the corresponding construction stages are determined; Based on the construction stage and the predefined stage-threshold mapping table, the corresponding basic monitoring thresholds are determined; The basic monitoring threshold is corrected based on the current hydrological parameters and environmental risk characteristics to generate the first monitoring threshold.

4. The monitoring method for river channel improvement projects according to claim 2, characterized in that, The process of determining the foundation monitoring frequency based on the risk level in the structural risk characteristics, and then adjusting it according to the current construction activity intensity to obtain a first monitoring frequency, includes: The basic monitoring frequency is determined based on the risk level coefficient in the structural risk characteristics. The basic monitoring frequency is adjusted based on the current construction activity intensity and seasonal hydrological characteristics to generate the first monitoring frequency.

5. A monitoring method for river channel improvement projects according to claim 2, characterized in that, The process of calculating an ecological sensitivity index based on the aforementioned ecological sensitivity characteristics, and determining whether the monitoring focus is structural safety-oriented or ecological environment-oriented based on the ecological sensitivity index, includes: An ecological sensitivity index is calculated based on the aforementioned ecological sensitivity characteristics, which include the presence of sensitive species, the importance level of habitats, the value of ecosystem services, and the level of water body self-purification capacity. Based on the ecological sensitivity index, the monitoring focus is determined, including structural safety-oriented monitoring and ecological environment-oriented monitoring.

6. A monitoring method for river channel improvement projects according to claim 5, characterized in that, The step of adjusting the first monitoring threshold and the first monitoring frequency according to the monitoring focus direction to generate a second monitoring threshold and a second monitoring frequency, as the monitoring parameter set, includes: If the monitoring focus is on structural safety-oriented monitoring, then the first monitoring frequency of the structure displacement and foundation settlement parameters is increased based on the first adjustment step size to obtain the second monitoring frequency, and the second monitoring threshold is obtained by decreasing the allowable range of the first monitoring threshold of the corresponding parameter based on the second adjustment step size. If the monitoring focus is on ecological environment-oriented monitoring, then a second monitoring threshold is obtained by reducing the first monitoring threshold of suspended matter concentration and dissolved oxygen parameters based on the second adjustment step size, and a second monitoring frequency is obtained by increasing the first monitoring frequency of benthic biodiversity indicators based on the first adjustment step size.

7. A monitoring method for river channel improvement projects according to claim 6, characterized in that, It also includes, If the construction stage characteristics are in the dredging construction stage, and the hydrodynamic characteristics indicate that the flow velocity is greater than the first threshold, then the second monitoring frequency of the suspended solids concentration index is increased based on the third adjustment step size to obtain the third monitoring frequency. If the construction stage characteristics are in the still water revetment construction stage, and the presence of sensitive species in the ecologically sensitive characteristics is less than the preset presence threshold, then the third monitoring frequency is obtained by reducing the second monitoring frequency of the comprehensive water quality index based on the fourth adjustment step size.

8. A monitoring system for river channel improvement projects, characterized in that, include: The data retrieval module is used to generate a matching set of monitoring parameters, an anomaly diagnosis rule base, and an anomaly threshold set based on the current waterway scenario information from a pre-built knowledge base. The data acquisition module is used to coordinate and schedule monitoring network devices based on the monitoring parameter set to acquire multi-source monitoring data. An anomaly identification module is used to identify anomalies based on the multi-source monitoring data and call the anomaly diagnosis rule base to obtain potential anomalies and anomaly confidence levels. The problem verification module is used to control the drone in the monitoring network device to perform on-site verification of the potential abnormal problem if the anomaly confidence level is greater than the preset anomaly threshold, and generate a reliability assessment result. The trend prediction module is used to input multi-source monitoring data into a preset time-series prediction model to obtain the monitoring trend prediction result if the reliability assessment result is greater than a preset reliability threshold. The early warning generation module is used to compare the monitoring trend prediction results with the maintenance critical threshold in the abnormal threshold set to obtain the spatiotemporal location of future maintenance needs, and calculate the waterway comprehensive monitoring index through the mechanism-data hybrid model. If the comprehensive monitoring index is greater than the preset early warning threshold, the current early warning information is generated. The strategy generation module is used to generate and execute a collaborative scheduling strategy based on the reliability assessment results, the spatiotemporal location of future maintenance needs, and current early warning information, through an optimization algorithm.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images