A wetland pond water quantity and quality online monitoring method and system based on an internet of things

Abstract

The present application relates to water quantity and quality online monitoring technical field, especially to a kind of wetland pond water quantity and quality online monitoring method and system based on Internet of Things.The method comprises the following steps: obtaining liquid level monitoring data and flow monitoring data in wetland pond;According to liquid level monitoring data and flow monitoring data, the flow relationship between each pond is identified;Based on the flow relationship between the pond, the pond confluence link data is constructed;According to the pond confluence link data, the incoming water change data of each pond node is collected;According to the incoming water change data and the pond confluence link data, the water storage change state of each pond node is detected;According to the water storage change state, the water body residence state of each pond node is identified;The present application monitors the water quantity and quality of wetland pond to realize the continuous correlation monitoring of water quantity change, water quality change and discharge state, and improves the integrity of wetland pond operation state identification.

Description

Technical Field

[0001] This invention relates to the field of online water quantity and quality monitoring technology, and in particular to a method and system for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things. Background Technology

[0002] In the operation of existing wetland reservoirs, water bodies experience continuous confluence, diversion, and retention at multiple reservoir nodes. The water exchange relationship between different nodes directly affects the overall storage and discharge balance of the wetland system. During the retention process, pollutants are deposited, diffused, and degraded, creating a strong coupling relationship between water volume changes and water quality changes. Traditional wetland reservoir monitoring methods mainly rely on manual inspections or single-point timed sampling to independently record liquid level, flow rate, and water quality parameters. These methods can only reflect the status at a local moment and are difficult to form dynamic flow links between reservoir nodes. They cannot accurately identify the continuous impact of upstream water changes on downstream water storage status and pollution migration processes. In the series operation structure of wetland reservoirs, changes in the retention status of a single node often lead to changes in the pollutant accumulation status and purification capacity of subsequent nodes. Existing technologies lack overall monitoring methods for such chain-like transmission relationships. Current water quality monitoring mostly uses surface sampling methods, which cannot reflect the differences in pollutant concentration distribution at different water depths, making it difficult to identify bottom sedimentation pollution and stratified purification processes. Summary of the Invention

[0003] Therefore, it is necessary to provide an online monitoring method and system for wetland reservoir water quantity and quality based on the Internet of Things to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, an online monitoring method for water quantity and quality in wetlands and reservoirs based on the Internet of Things (IoT) includes the following steps: Step S1: Obtain water level monitoring data and flow monitoring data in wetland reservoirs; identify the water flow relationships between reservoirs based on the water level monitoring data and flow monitoring data; construct reservoir confluence link data based on the water flow relationships. Step S2: Collect inflow change data for each reservoir node based on reservoir confluence link data; detect water storage change status for each reservoir node based on inflow change data and reservoir confluence link data; identify water retention status for each reservoir node based on water storage change status. Step S3: Extract stratified water quality monitoring data corresponding to each reservoir node based on the water retention status; identify the pollutant accumulation status of each reservoir node based on the stratified water quality monitoring data and water retention status; assess the purification change status of each reservoir node based on the pollutant accumulation status. Step S4: Determine the water reuse status of the reservoir based on the purification change status of each reservoir node; detect the remaining discharge status of the reservoir based on the water reuse status; assess the reservoir operation risk data based on the remaining discharge status and purification change status, and upload the reservoir operation risk data to the cloud platform for early warning operation based on the Internet of Things.

[0005] This invention also provides an IoT-based online monitoring system for wetland and reservoir water quantity and quality, used to execute the IoT-based online monitoring method for wetland and reservoir water quantity and quality described above. The IoT-based online monitoring system for wetland and reservoir water quantity and quality includes: The reservoir confluence link construction module is used to acquire liquid level monitoring data and flow monitoring data in wetland reservoirs; identify the water flow relationship between reservoirs based on the liquid level monitoring data and flow monitoring data; and construct reservoir confluence link data based on the water flow relationship. The water retention status identification module is used to collect inflow change data of each reservoir node based on reservoir confluence link data; detect the water storage change status of each reservoir node based on the inflow change data and reservoir confluence link data; and identify the water retention status of each reservoir node based on the water storage change status. The purification change status assessment module is used to extract stratified water quality monitoring data corresponding to each reservoir node based on the water retention status of each reservoir node; identify the pollutant accumulation status of each reservoir node based on the stratified water quality monitoring data and water retention status; and assess the purification change status of each reservoir node based on the pollutant accumulation status. The risk warning module is used to determine the water reuse status of reservoirs based on the purification changes at each reservoir node; detect the remaining discharge status of reservoirs based on the water reuse status; assess the operational risk data of reservoirs based on the remaining discharge status and purification changes; and upload the operational risk data of reservoirs to the cloud platform for early warning operations based on the Internet of Things.

[0006] The beneficial effects of this invention are as follows: by jointly collecting liquid level monitoring data and flow monitoring data, the water flow relationship between wetland reservoirs can be continuously identified, and reservoir confluence link data can be further constructed, so that the originally discrete water exchange relationship between each reservoir node can form a complete chain transmission structure, which solves the problem of the disconnect between single-point monitoring and overall flow relationship in traditional wetland reservoir monitoring. By tracking the inflow changes at each reservoir node through the reservoir confluence link data, and combining the inflow change status to detect the water storage change status of the reservoir node, a continuous correlation is formed between the inflow process, the water storage accumulation process, and the internal water volume occupation process. This further identifies the water retention status and provides a quantifiable basis for the water retention process inside the reservoir, avoiding monitoring deviations caused by relying solely on static liquid levels to judge the reservoir operation status. Based on the water retention status, stratified water quality monitoring is carried out at different water depths to establish a direct correspondence between pollutant concentration changes and water retention locations. The pollutant accumulation status can be identified through the interlayer concentration change status, enabling continuous tracking of the deposition, diffusion, and aggregation processes of pollutants within the reservoir. This solves the problem that traditional whole sampling methods cannot identify differences in the spatial distribution of pollutants. By detecting changes in the purification status based on the accumulated state of pollutants, a closed-loop correlation is formed between the pollutant accumulation process and the purification attenuation process, thus providing a dynamic basis for identifying the self-purification capacity of the reservoir. The reuse status of the reservoir water is determined based on the changes in the purification status, and an operational risk assessment is conducted in conjunction with the remaining discharge status. This establishes a complete operational link between the reservoir and reuse, discharge, and risk warning. The entire process is achieved through the Internet of Things (IoT) to realize the synchronous transmission and centralized processing of data between level sensors, flow rate sensors, water quality probes, and cloud platforms. This enables a full-process online monitoring mechanism for wetland reservoirs from water inflow, storage, retention, purification to discharge, ensuring the continuity, correlation, and structural integrity of the monitoring process. At the same time, the coupling relationship between changes in water quantity and water quality within the reservoir is synchronously expressed, providing complete data support for wetland reservoir operation status identification, reuse scheduling, and risk warning. Attached Figure Description

[0007] Figure 1 A schematic diagram illustrating the steps of an online monitoring method for water quantity and quality in wetlands and reservoirs based on the Internet of Things (IoT). Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S2. Figure 3 for Figure 1 A detailed flowchart illustrating the implementation steps of step S3. Detailed Implementation

[0008] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0009] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0010] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0011] To achieve the above objectives, please refer to Figures 1 to 3 A method for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things includes the following steps: Step S1: Obtain water level monitoring data and flow monitoring data in wetland reservoirs; identify the water flow relationships between reservoirs based on the water level monitoring data and flow monitoring data; construct reservoir confluence link data based on the water flow relationships. Step S2: Collect inflow change data for each reservoir node based on reservoir confluence link data; detect water storage change status for each reservoir node based on inflow change data and reservoir confluence link data; identify water retention status for each reservoir node based on water storage change status. Step S3: Extract stratified water quality monitoring data corresponding to each reservoir node based on the water retention status; identify the pollutant accumulation status of each reservoir node based on the stratified water quality monitoring data and water retention status; assess the purification change status of each reservoir node based on the pollutant accumulation status. Step S4: Determine the water reuse status of the reservoir based on the purification change status of each reservoir node; detect the remaining discharge status of the reservoir based on the water reuse status; assess the reservoir operation risk data based on the remaining discharge status and purification change status, and upload the reservoir operation risk data to the cloud platform for early warning operation based on the Internet of Things.

[0012] In this embodiment, IoT monitoring nodes are deployed within a wetland reservoir system. This system comprises multiple reservoir nodes arranged sequentially along the terrain elevation differences. These nodes are connected by diversion channels, connecting canals, and overflow weirs, forming a continuous water transport structure. A level sensor, employing a hydrostatic level gauge, is fixedly installed within each reservoir node. The sensor is mounted on a fixed support at the bottom of the reservoir and its contact with the water is maintained by an anti-siltation sleeve. Electromagnetic flow meters and ultrasonic flow velocity sensors are installed in the diversion channels between the reservoir nodes, positioned at the center of the channel cross-section, to continuously collect flow monitoring data within the channels.

[0013] The liquid level sensor acquires the current water level value at a sampling period of once every 10 seconds, and generates continuous liquid level monitoring data, denoted as... ,in Indicates the reservoir node number, Indicates the sampling time; the flow meter acquires the flow rate value of the guide channel cross-section according to the same sampling period, forming flow monitoring data. ,in and This indicates the node number of the adjacent reservoir.

[0014] After data acquisition, the liquid level monitoring data is sorted by time series to extract the liquid level change trend of each reservoir node over a continuous period. When the liquid level at an upstream node drops, and the liquid level at a downstream node rises within the corresponding delay time, and the flow rate in the connecting channel is greater than zero, the node pair is marked as having a water level transfer relationship. Further, combined with the flow direction, the upstream reservoir node is defined as the inflow node, and the downstream reservoir node is defined as the receiving node.

[0015] After all node relationships are identified, a node connection table is generated according to the liquid level transfer order between each node, and reservoir confluence link data is constructed based on the node connection table. The reservoir confluence link data records the inflow order, outflow order, and diversion connection relationship between each reservoir node, and serves as the basic input data for water inflow path identification and water storage status calculation in step S2.

[0016] Based on the reservoir confluence link data generated in step S1, the upstream inlet locations of each reservoir node are located. The upstream inlet is the connection point between the guide channel and the main body of the reservoir. A flow velocity sensor and a cross-sectional flow detection component are installed at each inlet. The flow velocity sensor is a Doppler flow meter, and the cross-sectional flow detection component includes three sets of flow sampling probes arranged along the width of the guide channel to synchronously acquire changes in the inflow at the inlet cross-section.

[0017] Real-time flow velocity and cross-sectional flow rate values ​​at each upstream inlet are extracted according to a preset sampling period and matched with the upstream node numbers in the confluence link to form inflow change data for the corresponding reservoir node. Inflow change data includes instantaneous inflow flow rate, inflow velocity, and flow rate change trend over a continuous time period. After the inflow change data is generated, continuous flow rate changes over a continuous time period are detected along the flow path, extracting the flow rate values ​​at the beginning, middle, and end of the flow path and calculating the flow rate difference at the three cross-sections. When the flow rate difference between adjacent cross-sections is continuously less than a preset threshold, the inflow process is determined to be a continuous flow state, and continuous inflow change data for the node is generated. Based on the continuous inflow change data for the node and the current reservoir level, the reservoir capacity increase during the current time period is calculated. The level change is converted into the corresponding reservoir capacity occupancy using a preset reservoir volume curve, forming water storage change status data. After the water storage change state is formed, the flow exchange value between the inlet and outlet of the reservoir is extracted, and the flow velocity decay of multiple monitoring points inside the reservoir is calculated. When the internal flow velocity is consistently lower than the inlet flow velocity and remains stable, the corresponding area is identified as a stagnation and delay area, and water body retention state data is generated based on the duration of the stagnation and delay area.

[0018] In this embodiment, stratified water quality monitoring is performed on each reservoir node based on the generated water retention state data. Each reservoir node is equipped with a vertically stratified monitoring assembly, which includes an upper monitoring probe, a middle monitoring probe, and a bottom monitoring probe, respectively fixed to a float support, a suspension support, and a bottom mounting frame, forming a continuous monitoring structure at different water depths. Each stratified monitoring probe collects data on total nitrogen concentration, total phosphorus concentration, ammonia nitrogen concentration, and chemical oxygen demand (COD). The sampling period is synchronized with the retention state to ensure consistency in monitoring timing. After data collection, stratified water quality monitoring data corresponding to each reservoir node is generated.

[0019] Inter-layer comparisons of pollutant concentrations at different levels are performed to calculate the concentration differences between the upper and middle layers, and between the middle and lower layers, thus forming pollutant concentration gradient data. When the pollutant concentration at a certain layer is continuously higher than that of adjacent layers, and the residence time at that layer is longer than a preset time threshold, that layer is marked as a pollutant accumulation layer. After the pollutant accumulation layer is identified, the concentration changes at the same layer over a continuous period are extracted. If the pollutant concentration continues to increase within a continuous monitoring period, it is recorded as pollutant accumulation state data. After the pollutant accumulation state is formed, the attenuation detection of the concentration change rate over a continuous time period continues. When the pollutant concentration changes from increasing to decreasing, and the decrease rate is continuously stable, the corresponding node is recorded as the pollutant attenuation state, and purification change state data is generated.

[0020] Based on the data on changes in purification status, the water reuse status of each reservoir node is determined. The concentration values ​​of total nitrogen, total phosphorus, and ammonia nitrogen for each node are read and compared with the preset farmland irrigation reuse standards. When all indicators are below the corresponding threshold, the node is marked as being in reuse status.

[0021] After the reuse status is established, reuse water is released through an electric gate installed at the outlet. The electric gate is connected to an Internet of Things (IoT) control module, which records the total amount of water released and updates the remaining reservoir capacity data of the current node synchronously.

[0022] After the reuse and discharge of water are completed, the remaining water in the reservoir is monitored for discharge status. Specifically, the current remaining discharge flow rate is collected by a flow meter at the outlet, and the discharge water is assessed in conjunction with purification change data to determine whether the discharge meets the discharge standards. When the discharge index exceeds the threshold, it is determined to be an abnormal discharge status. After completing the reuse and discharge status monitoring, the remaining discharge status, purification change status, and real-time liquid level status of the current node are jointly calculated to generate reservoir operation risk data. The operation risk data includes reuse abnormal risk, discharge exceeding standard risk, and excessive retention risk. The reservoir operation risk data is uploaded to the cloud management terminal via an IoT communication module. The communication module uses an NB-IoT communication unit, and the upload cycle is set to once every 60 seconds. After receiving the data, the cloud management terminal performs risk level classification and triggers corresponding early warning operations. This forms a complete closed loop for online monitoring of wetland reservoirs.

[0023] Preferably, step S1 includes the following steps: Step S11: Obtain liquid level monitoring data and flow monitoring data in the wetland reservoir, extract the liquid level changes at the reservoir nodes from the liquid level monitoring data, and obtain the node liquid level change data; Step S12: Detect the flow transmission direction between nodes of each reservoir based on the flow monitoring data, obtain node flow direction data, and identify the water level transmission relationship between each reservoir based on the node liquid level change data and node flow direction data. Step S13: Identify the flow sequence between nodes based on the water level transfer relationship between each reservoir, obtain node flow time series data, and construct the water flow relationship between reservoirs based on the node flow time series data. Step S14: Match upstream inflow nodes and downstream outflow nodes according to the water flow relationship between reservoirs and ponds to obtain node confluence connection data, and construct reservoir confluence link data based on the node confluence connection data.

[0024] In this embodiment, a fixed monitoring unit is set up for each reservoir node within the wetland reservoir area. A hydrostatic level sensor is installed at the center of each reservoir node. The level sensor is fixed to the bottom of the reservoir by a metal bracket, with the sensing end vertically upward and submerged in the bottom water, for continuously collecting level monitoring data inside the reservoir node. An electromagnetic flow meter and an ultrasonic velocity sensor are installed in the middle of the guide channel between adjacent reservoir nodes. The electromagnetic flow meter is used to collect the cross-sectional flow rate value of the guide channel, and the ultrasonic velocity sensor is used to collect the real-time flow velocity value within the guide channel, forming flow monitoring data. The level monitoring data and flow monitoring data are synchronously transmitted to the edge processing unit through an IoT acquisition terminal. The edge processing unit arranges the level monitoring data into a time series according to a fixed sampling period and extracts the level difference between adjacent moments. The level difference is calculated using the level change calculation formula. ; in This represents the change in liquid level at the node. This indicates the liquid level value at the next sampling time. The liquid level value at the previous sampling time is represented; a liquid level change sequence is constructed by the liquid level change values ​​of multiple consecutive sampling periods, and node liquid level change data is formed; the node liquid level change data is used as the basic input data in step S12 for subsequent identification of the relationship between flow direction and water level transmission.

[0025] The edge processing unit receives the node liquid level change data generated in step S11 and simultaneously reads the real-time flow velocity value and cross-sectional flow rate value of the guide channel from the flow monitoring data; it compares the flow velocity values ​​at the beginning and end of the guide channel, and when the flow velocity value at the beginning is greater than that at the end and the cross-sectional flow rate remains continuous, it records the direction of flow velocity decay as the direction of water flow; it extracts the liquid level change data of adjacent reservoir nodes within the same time window, and performs time correspondence matching on the liquid level decline trend of the upstream node and the liquid level rise trend of the downstream node; when the liquid level decline time of the upstream node and the flow rate change time of the guide channel maintain a continuous correspondence, it records this direction as the node flow direction data; after the node flow direction data is generated, it associates the node liquid level change data with the node flow direction data to detect the liquid level transfer path between the liquid level decline node and the liquid level rise node; it records the water level transfer delay through the time difference, and marks the transfer sequence between each node according to the delay order to form the water level transfer relationship data between each reservoir.

[0026] The water level transfer relationship data generated in step S12 enters the sequence identification unit. The sequence identification unit sorts the nodes according to the transfer delay value between each reservoir node, defines the node with the shortest delay value as the priority flow node, and defines the nodes with increasing delay values ​​as the subsequent flow nodes. During the sorting process, the liquid level start change time and liquid level stabilization time corresponding to each node are extracted, and the duration of liquid level change at each node is calculated. A flow time sequence linked list between nodes is constructed based on the liquid level change duration and transfer delay time. The flow time sequence linked list between nodes records the flow sequence between each node in the order of "source node - transfer node - receiving node", forming node flow time sequence data. After the node flow time sequence data is established, the flow direction and liquid level transfer relationship between adjacent nodes are mapped one by one, and the nodes on the same time chain are connected to form a continuous flow path, generating the water flow relationship between reservoirs. The generated water flow relationship includes node number, flow direction, and time sequence, which is used for matching confluence nodes.

[0027] After establishing the water flow relationships between reservoirs and ponds, the confluence link construction unit reads the flow connection relationships between each node and performs node matching based on the starting and ending nodes in the flow relationships. When a node has only an output relationship and no input relationship, it is marked as an upstream inflow node; when a node has only an input relationship and no output relationship, it is marked as a downstream outflow node; for nodes that have both input and output relationships, they are marked as intermediate confluence nodes. After completing the node classification, the connection number, inflow direction number, and outflow direction number of each node are extracted to construct a node confluence connection data table. The node confluence connection data table records the inflow source, outflow target, and connection sequence of each node. Based on the node confluence connection data table, the data is chained together in chronological order to generate complete reservoir confluence link data.

[0028] Preferably, step S12 includes the following steps: Step S121: Extract instantaneous flow velocity change data between nodes of each reservoir based on flow monitoring data; Step S122: Calculate the flow fluctuation amplitude between each reservoir node based on the instantaneous flow velocity change data to obtain the node flow fluctuation data; Step S123: Detect the water level difference transmission between nodes of each reservoir based on node liquid level change data and node flow fluctuation data, and obtain node flow direction data; Step S124: Extract water level response delay data between adjacent reservoir nodes based on node flow direction data; Step S125: Determine the water level transfer relationship between each reservoir based on the node flow direction data and water level response delay data.

[0029] In this embodiment, after the liquid level monitoring data and flow monitoring data are collected in step S11, the edge acquisition terminal reads the flow monitoring data inside the guide channel in real time. Three sets of ultrasonic flow velocity sensors are fixedly installed inside each guide channel along the water flow direction, corresponding to the beginning, middle and end positions of the guide channel, respectively. Each set of sensors is fixed in the mounting groove embedded in the side wall of the guide channel, with the sensor probe facing the flow direction towards the center of the guide channel. After the flow monitoring data enters the edge processing unit, the instantaneous flow velocity value corresponding to each sampling moment is extracted to form a single-moment flow velocity sampling sequence. The flow velocity values ​​of multiple consecutive sampling moments are arranged in chronological order, and the flow velocity change value between adjacent moments is calculated to form an instantaneous flow velocity change sequence. The instantaneous flow velocity change sequence corresponding to each guide channel is archived according to the node number to form instantaneous flow velocity change data between nodes of each reservoir. The instantaneous flow velocity change data includes the instantaneous flow velocity at the beginning, the instantaneous flow velocity in the middle, the instantaneous flow velocity at the end, and the corresponding sampling timestamp.

[0030] The instantaneous velocity change data generated in step S121 is input into the flow fluctuation calculation unit; the flow fluctuation calculation unit calculates the flow change value at the corresponding moment based on the cross-sectional area of ​​the guide channel and the instantaneous velocity change value; the flow change value is calculated according to the formula: ; Perform calculations, where Indicates the instantaneous flow rate value. Indicates the instantaneous flow velocity value. The cross-sectional area of ​​the diversion channel is represented. After the flow rate value is calculated at each time point, the maximum and minimum flow rates within the continuous time window are extracted, and the difference between the two is calculated to form the flow fluctuation amplitude. The flow fluctuation amplitude is used to characterize the intensity of the change in flow rate inside the diversion channel within a continuous time period. Subsequently, the flow fluctuation amplitude of each diversion channel is bound to the corresponding node number according to the node connection relationship to generate node flow fluctuation data. The node flow fluctuation data records the flow change range of each node connection segment within the continuous monitoring period and is transmitted to the water level difference transmission detection.

[0031] Following the node flow fluctuation data, the edge processing unit synchronously retrieves the node liquid level change data generated in step S11; extracts the liquid level change values ​​of adjacent reservoir nodes within the same time window, and calculates the real-time liquid level difference between the two nodes; the liquid level difference values ​​are arranged in chronological order to form a liquid level height difference change sequence; the liquid level height difference change sequence is matched with the node flow fluctuation data, and when the liquid level difference of an upstream node continues to decrease and the flow fluctuation amplitude of the corresponding diversion channel continues to increase, the node is recorded as an active transport node; when the liquid level difference of the corresponding downstream node continues to increase, the node is recorded as a receiving node; the water level height difference transmission process between each node is detected by the synchronous change of the liquid level difference change direction and the flow fluctuation direction; after the height difference transmission relationship is established, the actual transmission offset direction of the water body is determined according to the direction of the height difference change, forming node flow direction data; the node flow direction data includes the starting node number, the ending node number, and the corresponding flow direction identifier.

[0032] After generating node flow direction data, the time delay extraction unit reads the liquid level change time series corresponding to adjacent nodes according to the flow direction; extracts the time point of the first drop in liquid level of upstream node and records it as the start time point; extracts the time point of the first rise in liquid level of downstream node and records it as the response time point; subtracts the start time point from the response time point to obtain the water level response time delay value between nodes; accumulates and records the time delay values ​​within multiple consecutive monitoring periods and calculates the average time delay value to form stable water level response time delay data; the water level response time delay data is used to reflect the actual transmission time difference between adjacent reservoir nodes; the generated water level response time delay data is synchronously bound to the node flow direction data.

[0033] After extracting the water level response delay data, the relationship determination unit reads the flow direction identifier between each node based on the node flow direction data, and extracts the response order between each node based on the water level response delay data. When the flow direction of a node is the output direction and the corresponding delay value is the smallest, the node is determined as the starting node of water level transmission. When the response delay of adjacent nodes increases sequentially, the water level transmission path is constructed in ascending order. Each water level transmission path is recorded by combining the node number, direction identifier, and delay value to form complete water level transmission relationship data. The water level transmission relationship data includes the upstream node number, downstream node number, transmission direction, and response delay.

[0034] Preferably, step S123 includes: Calculate the real-time liquid level difference between adjacent reservoir nodes based on the node liquid level change data to obtain node liquid level difference data; Based on the node liquid level difference data, the water level gradient change status between adjacent reservoir nodes is detected to obtain node water level gradient data. Identify the transmission of water level differences between adjacent reservoir nodes based on node water level gradient data and node flow fluctuation data; Based on the water level difference transmission, the direction of water transport offset between adjacent reservoir nodes is detected to obtain node flow direction data.

[0035] In this embodiment, after the node flow fluctuation data is generated in step S122, the edge acquisition terminal synchronously retrieves the node liquid level change data generated in step S11 and performs data pairing according to the relationship between adjacent nodes in the reservoir confluence link. The liquid level monitoring device of each reservoir node is fixedly installed at the calibration reference surface position on the side wall of the reservoir slope. The liquid level monitoring device adopts a pressure-type liquid level probe, and the bottom end of the probe is fixed at the elevation reference point at the bottom of the reservoir. The liquid level sampling period is set to 30 seconds. Between two adjacent reservoir nodes, the difference is calculated based on the liquid level value corresponding to the same sampling time. The liquid level value of the upstream node is subtracted from the liquid level value of the downstream node to form a real-time liquid level difference sequence, which is then arranged continuously in chronological order to obtain the node liquid level difference data. The node liquid level difference data includes the upstream node number, the downstream node number, the liquid level difference value, and the corresponding timestamp. After the node liquid level difference data is generated, it is input into the gradient detection unit for the next step of processing.

[0036] The gradient detection unit reads the node liquid level difference data from the continuous time series and compares the liquid level difference values ​​between two adjacent sampling times one by one. When the liquid level difference value at a later time is higher than that at a previous time, it is recorded as a gradient growth state; when the liquid level difference value at a later time is lower than that at a previous time, it is recorded as a gradient decay state. The gradient growth state and gradient decay state in multiple consecutive sampling periods are combined in chronological order to form a node water level gradient change sequence. Based on the node water level gradient change sequence, the duration and direction of gradient change are statistically analyzed to generate node water level gradient data. The node water level gradient data includes the gradient change direction, gradient duration, and gradient change interval, which serve as input data for subsequent water level difference transmission detection.

[0037] After the node water level gradient data is generated, it is synchronously matched with the node flow fluctuation data generated in step S122. During the matching, the node water level gradient change state and flow fluctuation amplitude are matched according to the same timestamp. When the node water level gradient is in a continuous increasing state and the node flow fluctuation amplitude increases synchronously, it is determined that there is a continuous height difference driven transmission process between the adjacent nodes. When the node water level gradient is continuously decreasing and the flow fluctuation amplitude decreases synchronously, it is determined that the height difference driven process has entered the decay stage. Through the synchronous change relationship in multiple consecutive monitoring cycles, the water level height difference transmission process between adjacent reservoir nodes is completely recorded to form water level height difference transmission data. This data includes the height difference start time, the height difference peak time, and the height difference decay end time.

[0038] After the water level difference transmission data is generated, the direction detection unit reads the liquid level change direction corresponding to the starting and ending nodes of the height difference transmission. If the liquid level at the starting node continues to decrease and the liquid level at the ending node continues to rise, the water transmission direction is marked as the direction from the starting node to the ending node. If the change trend is opposite, the direction is marked in the opposite direction. Then, cross-validation is performed by combining the flow change direction in the node flow fluctuation data. When the liquid level change direction is consistent with the flow change direction, node flow direction data is formed. The node flow direction data includes the starting node number, the ending node number, the direction identifier, and the corresponding time period. The generated node flow direction data is directly transmitted to step S124 to extract the water level response delay data between adjacent reservoir nodes.

[0039] Preferably, step S2 includes the following steps: Step S21: Extract the upstream inlet location data corresponding to each reservoir node based on the reservoir confluence link data, and collect the inflow change data of each reservoir node by using flow velocity sensors deployed at each upstream inlet location. Step S22: Detect the continuous flow status in the inlet diversion channel of each reservoir node based on the inflow change data, and obtain the continuous inflow change data of the node; Step S23: Detect the reservoir capacity occupancy change status of each reservoir node based on the continuous change data of water inflow at the node, and obtain the water storage change status of each reservoir node; Step S24: Identify the water retention and delay status within each reservoir node based on the water storage change status of each reservoir node, and obtain the water retention status of each reservoir node.

[0040] In this embodiment, after constructing the reservoir confluence link data, each upstream connection in the confluence link is mapped to the actual water body connection location in the wetland reservoir. The upstream inlet of each reservoir node is spatially located and marked, and the width, elevation, and slope parameters of the inlet cross-section and the diversion channel are recorded. A flow velocity sensor is fixedly installed at each upstream inlet location. The flow velocity sensor adopts an immersion installation method, with the sensor probe arranged along the main water flow direction of the diversion channel, ensuring that the sensitive surface of the probe is consistent with the water flow direction to reduce sampling errors caused by lateral disturbances. The flow velocity sensor continuously samples at a fixed sampling period through an IoT acquisition terminal, outputting the instantaneous flow velocity value at each sampling period, while also considering the inlet cross-sectional area. Flow conversion is performed; during conversion, pre-calibrated cross-sectional dimension data of the guide channel is called, and the instantaneous flow velocity is multiplied by the corresponding flow area to obtain the inflow flow value at the current moment; multiple consecutive sampling periods form a time series flow set, which is arranged in chronological order, and the flow increase or decrease value between adjacent sampling points is extracted to form an inflow change curve; the peak points, valley points and stable flow sections in the inflow change curve are marked to obtain the inflow change data of each reservoir node. The inflow change data includes instantaneous inflow flow, cumulative inflow, inflow duration and flow change gradient; the inflow change data is used as direct input data for continuous flow state detection and is bound one-to-one with the corresponding inlet location data.

[0041] Based on the inflow change data generated in step S21, sampling points at the beginning, middle, and end are sequentially set along the flow direction inside the diversion channel at the inlet of each reservoir node. Each sampling point is equipped with a flow velocity sensor and a level probe. The flow velocity sensor is used to collect the flow velocity, and the level probe is used to synchronously record the cross-sectional water depth. Data collection is synchronized using a unified clock signal to ensure that each sampling point completes data output under the same time reference. The flow velocity value and corresponding water depth value at each sampling point are converted into cross-sectional flow rate, forming the initial cross-sectional flow rate, middle cross-sectional flow rate, and end cross-sectional flow rate. Subsequently, the cross-sectional flow rates of multiple consecutive sampling cycles are compared in chronological order to detect the flow rate. The flow rate attenuation along the flow channel is monitored. If the flow rate from the beginning to the end of the channel decreases continuously and the difference between adjacent sections is within a stable range, the channel is marked as being in a continuous flow state during the sampling period. If a sudden drop occurs at an intermediate section, the corresponding section position is recorded as the stagnation position. The continuous flow sections in multiple consecutive sampling periods are accumulated and statistically analyzed to form the continuous flow state data of the channel within the current monitoring window. The continuous transport start point, continuous transport end point, and continuous transport duration are further extracted based on the continuous flow state data to form the continuous change data of the node influent. This continuous change data of the node influent is directly transmitted to step S23 for reservoir capacity occupancy change detection.

[0042] The continuous water inflow data obtained in step S22 is time-aligned with the liquid level monitoring data deployed at each reservoir node. Using the continuous transport start and end points as time boundaries, liquid level change data within the corresponding intervals is extracted. The time-by-time difference of the liquid level change data is calculated to obtain a liquid level rise sequence. This sequence is then mapped to a pre-mapped reservoir capacity calibration table for each reservoir node. The calibration table records the effective water storage volume corresponding to different liquid levels. The liquid level rise is converted into reservoir capacity growth through a lookup table. Then, the cumulative increase in reservoir capacity over a continuous time period is summed to form the cumulative water storage volume within the current monitoring period. The cumulative water storage volume is then compared with the total effective reservoir capacity of the reservoir node to obtain the current reservoir capacity occupancy ratio. The changes in the reservoir capacity occupancy ratio are recorded for multiple consecutive monitoring periods to form a reservoir capacity occupancy change curve. The rate of increase, the length of the stable segment, and the peak occupancy ratio in the reservoir capacity occupancy change curve are extracted to obtain the water storage change status of each reservoir node. The water storage change status includes real-time water storage volume, cumulative occupancy ratio, and water storage growth rate.

[0043] Based on the water storage change status obtained in step S23, flow velocity monitoring points are set up at the inlet, middle, and outlet of each reservoir node, and real-time flow velocity data at the corresponding locations are acquired simultaneously. The difference between the flow velocity values ​​at the inlet and outlet is calculated to form the flow velocity difference data between the beginning and end of the node. The flow velocity value at the middle is then inserted between the flow velocity difference data between the beginning and end to construct a flow velocity decreasing sequence within the node. If the water storage change status continues to increase and the flow velocity decreasing sequence shows a continuous downward trend, it is determined that there is a stagnation and delay in the water flow transmission process within the reservoir node. Further analysis is performed on the flow velocity decreasing sequence for multiple consecutive sampling periods. The system performs overlay statistics to extract the duration and magnitude of velocity decay. It then matches the decay duration with the water storage growth rate to identify the coupling relationship between water storage growth and velocity slowdown. When the water storage growth rate consistently exceeds the outflow velocity change rate, the corresponding time period is recorded as the effective retention interval. All effective retention intervals are cumulatively statistically analyzed to form node water body retention time data. Combined with the average flow velocity value within the corresponding time period, the water body retention status of each reservoir node is output. The water body retention status includes the retention start time, retention end time, retention duration, and average flow velocity value during the retention period.

[0044] Preferably, step S22 includes the following steps: Step S221: Extract real-time flow rate change data of the diversion channel at the inlet of each reservoir node based on the inflow change data; Step S222: Detect the flow distribution at different cross-sectional positions within the guide channel based on real-time flow rate change data, and obtain the flow data of the guide channel cross-section; Step S223: Detect the velocity gradient change from the beginning to the end of the guide channel based on the cross-sectional flow data of the guide channel, and obtain the velocity gradient data of the guide channel; Step S224: Identify the continuous flow transmission state within the guide channel based on the flow velocity gradient data, and obtain the continuous change data of the incoming water at the node.

[0045] In this embodiment, the inflow variation data corresponding to the upstream inlet locations of each reservoir node is used. This inflow variation data includes the instantaneous flow velocity value and the corresponding sampling timestamp at each sampling moment. A cross-sectional geometric parameter table is pre-established at the inlet section of each upstream inlet guide channel. This table includes the guide channel width, channel bottom elevation, and the cross-sectional area corresponding to different liquid levels. During real-time monitoring, the current water depth value is synchronously collected by a liquid level sensor deployed at the inlet location of the guide channel, and the corresponding cross-sectional area is extracted from the cross-sectional geometric parameter table based on the current water depth value. Then, the instantaneous flow velocity value and the corresponding cross-sectional area are multiplied to form the real-time flow rate at the current sampling moment. The real-time flow rate calculation formula is as follows: ;

[0046] in: Indicates the first Real-time flow rate at any given moment, in cubic meters per second; Indicates the first The instantaneous flow velocity value collected by the flow velocity sensor at any time, in meters per second; Indicates the first The actual water flow area at the inlet section of the diversion channel at each moment is expressed in square meters. The real-time flow rate at each moment is arranged according to the continuous sampling period to form a real-time flow rate sequence. Further, the difference between the real-time flow rates at adjacent sampling moments is calculated to obtain the increase or decrease in flow rate at each moment. These increases or decreases are then combined in chronological order to form real-time flow rate change data. This real-time flow rate change data includes the initial flow rate, peak flow rate, terminal flow rate, and the amplitude of change between adjacent sampling points.

[0047] Based on the obtained real-time flow rate change data, multiple fixed cross-sectional monitoring positions are arranged along the water flow direction inside each guide channel. These monitoring positions include the initial, middle, and final cross-sections. Each cross-section is equipped with a flow velocity sensor and a liquid level sensor, all connected to an IoT data acquisition bus. During data acquisition, each cross-section synchronously outputs its flow velocity and liquid level values ​​according to the same sampling clock. The flow area of ​​each cross-section is calculated based on its liquid level and geometric dimensions. The flow velocity value is then converted to the flow area to generate the real-time flow rate value for each cross-section. Subsequently, the flow rate is determined according to the spatial dimensions of the guide channel. The real-time flow rate values ​​of each cross section are arranged in order to construct a cross-sectional flow distribution sequence. By comparing the flow rate differences between the first, middle, and last cross sections, the flow distribution status at different cross-sectional positions within the guide channel is detected. When the flow rate difference between adjacent cross sections continues to decrease, the flow distribution within the guide channel is recorded as stable. When the flow rate of a certain cross section suddenly drops, the position of that cross section is recorded as a local obstruction position. The flow rate values, cross-sectional position coordinates, and adjacent differences of each cross section are combined to form the guide channel cross-sectional flow data, which serves as the basic input for step S223.

[0048] Based on the flow data of the guide channel cross-section obtained in step S222, the flow rate values ​​of the first, middle, and last cross-sections of the guide channel are compared segment by segment, and the flow attenuation change per unit length is calculated in combination with the actual distance between each cross-section; the flow difference between the first and middle cross-sections and the flow difference between the middle and last cross-sections are extracted; the corresponding differences are divided by the corresponding cross-section distances to obtain the velocity gradient values ​​of the first and last segments; the velocity gradient values ​​of the first and last segments are then compared as a continuous time series. A velocity gradient decrease sequence is formed along the direction of the guide channel. If the velocity gradient decrease sequence changes in the same direction for multiple consecutive sampling periods, it is determined that the frictional resistance of the water flow inside the guide channel is stable. If the velocity gradient decrease sequence shows a sudden reversal at a certain position, the position is recorded as the point of abnormal velocity change. The gradient changes of all sections from the beginning to the end are further accumulated and integrated to form the velocity gradient data of the guide channel. The velocity gradient data of the guide channel includes the gradient value at the beginning, the gradient value in the middle section, the gradient value at the end, and the overall gradient decay rate.

[0049] Based on the velocity gradient data of the formed guide channel, the gradient decay rate change within multiple consecutive sampling periods is extracted in chronological order, and the continuity of the gradient decay rate change between adjacent sampling periods is detected. When the gradient decay rate change amplitude within multiple consecutive sampling periods remains within a fixed threshold range, the water flow transmission state inside the guide channel is marked as continuous within the corresponding time period. When the gradient decay rate suddenly increases or decreases, the corresponding time point is marked as the flow interruption point. Subsequently, each sampling period is merged according to the continuous transmission interval, and the start time, end time, and duration of each continuous interval are calculated. Then, combined with the real-time flow rate change data in step S221, the average flow rate value within the continuous interval is calculated to form the continuous flow intensity parameter. The time information of the continuous interval and the continuous flow intensity parameter are combined to obtain the continuous change data of the node inflow. The continuous change data of the node inflow includes the continuous flow start time, continuous flow end time, continuous flow duration, and continuous flow average intensity.

[0050] Preferably, step S223 includes: Extract the flow rate data of the first, middle and last sections of the guide channel based on the cross-sectional flow data; Based on the cross-sectional flow rate data at the beginning, middle and end of the guide channel, the friction resistance change along the guide channel is detected to obtain the friction resistance change data of the guide channel. Based on the data on the change in flow resistance along the flow channel, the decreasing state of the flow velocity gradient from the beginning to the end of the flow channel is identified, and the flow velocity gradient data of the flow channel is obtained.

[0051] In this embodiment, the flow data of the guide channel cross-section is formed. The flow data of the guide channel cross-section includes the real-time flow rate values ​​of the first, middle, and last cross-sections of the guide channel, and the flow rate values ​​of each cross-section correspond to a unified sampling time. Based on the flow data of the guide channel cross-section, the positions of each cross-section are spatially sorted according to the length of the guide channel. The first cross-section is defined as the inflow start position of the guide channel, the middle cross-section is defined as the flow position in the middle of the guide channel, and the last cross-section is defined as the outflow end position of the guide channel. Then, the flow rate of the first, middle, and last cross-sections within the same sampling period is extracted and the flow rate of the cross-sections is arranged in sequence. This extraction operation is repeated in multiple consecutive sampling periods so that each sampling period corresponds to a set of flow rate data of the first, middle, and last cross-sections. The flow rate data of the cross-sections within consecutive sampling periods are stored in time sequence to form a flow rate change sequence of the guide channel cross-section. The flow rate change sequence of the guide channel cross-section serves as the basic input data for detecting friction loss changes.

[0052] After extracting the cross-sectional flow rate, the flow rate data at the beginning, middle, and end of the guide channel are used to detect the change in flow resistance along the channel. Specifically, the difference between the flow rate at the beginning and middle sections is calculated to obtain the flow attenuation value at the beginning section, and the difference between the flow rate at the middle and end sections is calculated to obtain the flow attenuation value at the end section. If both the flow attenuation value at the beginning and the flow at the end section continuously increase, it indicates that there is a flow resistance accumulation phenomenon inside the guide channel. Further analysis is performed in conjunction with the flow... A roughness testing scale is pre-laid on the inner wall of the channel to synchronously record the actual adhesion state of the channel wall between each section, and to establish a resistance correspondence based on the adhesion thickness and the corresponding cross-sectional flow attenuation value; when the adhesion thickness increases and the cross-sectional flow attenuation value increases synchronously, the section is recorded as a resistance enhancement section; the resistance enhancement degree of each section is arranged sequentially according to the spatial order of the guide channel to form the resistance change data along the guide channel; the resistance change data along the guide channel includes the resistance value of the first section, the resistance value of the middle section, and the resistance value of the last section.

[0053] After obtaining the resistance variation data along the guide channel, the velocity gradient decreasing from the beginning to the end of the guide channel is identified based on the resistance variation data. Specifically, the resistance values ​​of the first and middle sections are compared sequentially, and then the resistance values ​​of the middle and final sections are compared sequentially to determine whether the resistance variation increases progressively along the length of the guide channel. When the resistance variation value shows a continuous increasing state, it is determined that the velocity is continuously decreasing along the length of the guide channel. Furthermore, the duration of the resistance increase within each sampling period is extracted and recorded. Matching the flow rate change sequence with the corresponding cross-section to form a flow velocity reduction process record; then superimposing the flow velocity reduction process records in multiple sampling periods to calculate the average reduction amplitude and reduction duration from the beginning to the end; forming the flow velocity gradient data of the guide channel, which includes the flow velocity gradient value at the beginning, the flow velocity gradient value in the middle section, the flow velocity gradient value at the end, and the overall reduction trend; this flow velocity gradient data of the guide channel serves as the direct input data for identifying the continuous flow transmission state in the guide channel in step S224.

[0054] Preferably, step S24 includes the following steps: Step S241: Extract the real-time water storage growth data corresponding to each reservoir node based on the water storage change status of each reservoir node; Step S242: Detect the water exchange change status between the inlet and outlet of each reservoir node based on real-time water storage growth data, and obtain node water exchange data; Step S243: Detect the velocity decay change at different locations within each reservoir node based on the node water exchange data, and obtain node velocity decay data. Step S244: Identify the water retention and delay status within each reservoir node based on the node flow velocity decay data, and obtain the water retention status of each reservoir node.

[0055] In this embodiment, a set of liquid level monitoring rods is pre-deployed inside the wetland reservoir node. The liquid level monitoring rods are distributed longitudinally along the reservoir at the inlet, middle area, and outlet. Each set of liquid level monitoring rods is fixedly equipped with a pressure-type liquid level probe. The sampling period of the liquid level probe is set to 30 seconds, and each collection forms a set of liquid level height values. The water storage change status data obtained in step S23 is called synchronously, and the liquid level increment change value of each liquid level probe at the current moment is matched. The liquid level increments within six consecutive sampling periods are sorted by time to form a liquid level growth sequence. The water storage growth volume per unit time is calculated based on the liquid level growth sequence, and converted into real-time water storage growth data by combining the measured planar area data of the corresponding reservoir node. The real-time water storage growth data is written into the corresponding reservoir buffer area according to the node number to form the real-time water storage growth record of the current reservoir node, which can be called in the next step when performing the exchange change detection at the inlet and outlet.

[0056] The real-time water storage growth data generated in step S241 is called, and the data of the electromagnetic flowmeters installed at the inlet and outlet of the corresponding reservoir node are extracted. The electromagnetic flowmeters are fixed at the end of the inlet guide channel and the outlet overflow position, respectively, and the inlet flow rate and outlet flow rate are recorded in the same time period. The inlet flow rate and outlet flow rate are time-synchronized and corrected so that the sampling time difference is controlled within 3 seconds. Then, the difference between inlet and outlet flow rates is verified according to the real-time water storage growth data. If the flow rate difference is consistent with the direction of water storage growth volume change, it is determined that there is effective water exchange at the current node. The exchange change amplitude in multiple consecutive sampling periods is further accumulated and calculated to form node water exchange data. The node water exchange data records the inlet input, outlet discharge, and net retention in the node, which serves as the data basis for subsequent internal flow velocity decay detection.

[0057] The node water exchange data generated in step S242 is used to fix multiple sets of flow velocity monitoring probes along the water flow path inside the reservoir node. The flow velocity monitoring probes adopt an ultrasonic velocity measurement structure and are arranged in sequence according to the inlet end, intermediate buffer zone and outlet end. The local flow velocity value at the corresponding position is collected synchronously by each probe and the position is mapped with the node water exchange data. The flow velocity at the inlet end is used as the reference flow velocity, and the flow velocities in the middle area and outlet end are compared with the reference flow velocity in sequence. If the flow velocity value decreases continuously along the flow path, the corresponding attenuation amplitude is extracted. The attenuation amplitudes in multiple consecutive sampling periods are superimposed to form node flow velocity attenuation data. The node flow velocity attenuation data includes the flow velocity difference at each measuring point, the attenuation position and the attenuation duration, which is used to reflect the kinetic energy loss process of water inside the reservoir.

[0058] The node velocity decay data generated in step S243 is used to perform time-related detection on the velocity decay positions of different measuring points within each reservoir node. When the velocity difference between adjacent measuring points increases continuously and the decay duration exceeds the preset monitoring period, it is determined that there is a stagnant section in the corresponding area. Multiple stagnant sections are connected in series according to their spatial locations to form complete stagnant area distribution data. Further, combined with the real-time water storage growth data in step S241, the water accumulation rate in the stagnant area is checked. When the water storage growth per unit time in the stagnant area is higher than the discharge at the outlet, it is identified as a water stagnant and delayed state. The water retention status data of each reservoir node is output according to the node number. The water retention status data includes the stagnant location, stagnant duration, and stagnant water volume value. This data is transmitted to step S3 as the positioning basis for the extraction of stratified water quality monitoring data.

[0059] Preferably, step S3 includes the following steps: Step S31: Extract stratified water quality monitoring data at different water depths within each reservoir node based on the water retention status of each reservoir node; Step S32: Detect the interlayer variation of pollutant concentration within each reservoir node based on the stratified water quality monitoring data to obtain node pollutant concentration gradient data; Step S33: Identify the pollutant accumulation status of each reservoir node based on the pollutant concentration gradient data; Step S34: Detect the purification change status of each reservoir node based on the pollutant accumulation status of each reservoir node, and obtain the purification change status of each reservoir node.

[0060] In this embodiment, after outputting the water retention status of each reservoir node in step S24, the identified retention areas within each reservoir node are designated as key areas for water quality monitoring. Layered monitoring rods are vertically fixedly installed within each reservoir node, extending from the water surface downwards to the reservoir bottom. Multiple sets of water quality probes are installed on the rods at fixed intervals, with installation depths corresponding to the surface, middle, and bottom layers, respectively. Each set of water quality probes includes a dissolved oxygen sensor, a conductivity sensor, a turbidity sensor, and an ammonia nitrogen detection electrode. Based on the water retention status data generated in step S24, the retention duration is determined. For longer areas, location matching is performed, and water quality sampling records of the corresponding monitoring rods within the same time period are retrieved. The water quality values ​​collected at each depth are uniformly corrected according to the timestamp to ensure that the data of each layer corresponds to the same water body state within the same sampling period. After correction, the dissolved oxygen concentration, conductivity value, turbidity value, and ammonia nitrogen concentration of each layer are archived separately, and a hierarchical correspondence is established according to the water depth position to form stratified water quality monitoring data at different water depth positions within each reservoir node. This stratified water quality monitoring data serves as the input basis for step S32 to perform interlayer pollutant concentration change detection.

[0061] Using the stratified water quality monitoring data generated in step S31, the pollution indicators corresponding to the surface, middle, and bottom layers within the same reservoir node are compared layer by layer. Ammonia nitrogen concentration and turbidity values ​​are extracted from each layer, and concentration difference records are established between the surface and middle layers, and between the middle and bottom layers. With the probe positions remaining fixed at each layer, the concentration changes over multiple sampling periods are continuously recorded to form an interlayer concentration change sequence. If the ammonia nitrogen concentration at the bottom layer is consistently higher than that at the middle layer, and the middle layer is consistently higher than that at the surface layer, it is determined that there is a vertical accumulation trend of pollutants. Simultaneously, turbidity changes are checked; if the direction of turbidity change is consistent with the direction of ammonia nitrogen concentration change, the interlayer transfer path of pollutants is confirmed to be stable. Then, the concentration differences between each layer are normalized according to the water depth distance to obtain the pollution change amplitude per unit water depth. This pollution change amplitude is stored according to the node number and monitoring depth to form node pollutant concentration gradient data. The node pollutant concentration gradient data records the direction of pollutant change, the amplitude of change, and the corresponding depth position at each layer, providing a basis for pollutant accumulation status identification in step S33.

[0062] The node pollutant concentration gradient data generated in step S32 is called and combined with the water body residence state data output in step S24 for joint matching; the spatial overlap detection is performed on areas with a residence time longer than the preset monitoring period and areas with a large concentration gradient; if high residence time and high concentration gradient appear simultaneously in the same area, the area is extracted as the key accumulation area of ​​pollutants; the bottom concentration change records of the area in the continuous sampling period are retrieved to detect the cumulative change of bottom pollutant concentration; if the bottom concentration value increases continuously and the increase rate is higher than the surface change rate, it is determined that pollutants have accumulated at the bottom of the reservoir; then, the accumulation area area, accumulation duration and accumulation concentration increment are comprehensively recorded to form pollutant accumulation state data; the pollutant accumulation state data includes the location of the accumulation area, the length of the accumulation time and the intensity of pollutant accumulation, and this data is output to step S34 as the input parameter for purification change state detection.

[0063] Based on the pollutant accumulation status data, the pollutant accumulation areas within each reservoir node are continuously monitored. At the corresponding locations of the pollutant accumulation areas, the stratified water quality monitoring data from step S31 is retrieved to track changes in ammonia nitrogen concentration, turbidity, and dissolved oxygen concentration over continuous time periods. If the ammonia nitrogen concentration and turbidity gradually decrease over multiple consecutive sampling cycles, while the dissolved oxygen concentration gradually increases, it is determined that a pollutant attenuation process exists in the current area. The initial pollutant accumulation value is further compared with the current detection value to extract the pollution attenuation amount per unit time. Then, the pollution attenuation amount is correlated with the water retention time in step S24 to determine whether the pollution attenuation process corresponds to the retention time. When the pollution attenuation amount continuously increases and the retention time extends synchronously, the purification change status of the current reservoir node is recorded. The purification change status data for each reservoir node is output, including the location of the purification area, the pollution attenuation rate, the purification duration, and the current residual pollution concentration. This data is directly transmitted to step S4 as the basis for judging the reservoir water reuse status.

[0064] This invention also provides an IoT-based online monitoring system for wetland and reservoir water quantity and quality, used to execute the IoT-based online monitoring method for wetland and reservoir water quantity and quality described above. The IoT-based online monitoring system for wetland and reservoir water quantity and quality includes: The reservoir confluence link construction module is used to acquire liquid level monitoring data and flow monitoring data in wetland reservoirs; identify the water flow relationship between reservoirs based on the liquid level monitoring data and flow monitoring data; and construct reservoir confluence link data based on the water flow relationship. The water retention status identification module is used to collect inflow change data of each reservoir node based on reservoir confluence link data; detect the water storage change status of each reservoir node based on the inflow change data and reservoir confluence link data; and identify the water retention status of each reservoir node based on the water storage change status. The purification change status assessment module is used to extract stratified water quality monitoring data corresponding to each reservoir node based on the water retention status of each reservoir node; identify the pollutant accumulation status of each reservoir node based on the stratified water quality monitoring data and water retention status; and assess the purification change status of each reservoir node based on the pollutant accumulation status. The risk warning module is used to determine the water reuse status of reservoirs based on the purification changes at each reservoir node; detect the remaining discharge status of reservoirs based on the water reuse status; assess the operational risk data of reservoirs based on the remaining discharge status and purification changes; and upload the operational risk data of reservoirs to the cloud platform for early warning operations based on the Internet of Things.

[0065] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things, characterized in that, Includes the following steps: Step S1: Obtain water level monitoring data and flow monitoring data in wetland reservoirs; identify the water flow relationships between reservoirs based on the water level monitoring data and flow monitoring data; construct reservoir confluence link data based on the water flow relationships. Step S2: Collect inflow change data for each reservoir node based on reservoir confluence link data; detect water storage change status for each reservoir node based on inflow change data and reservoir confluence link data; identify water retention status for each reservoir node based on water storage change status. Step S3: Extract stratified water quality monitoring data corresponding to each reservoir node based on the water retention status; identify the pollutant accumulation status of each reservoir node based on the stratified water quality monitoring data and water retention status; assess the purification change status of each reservoir node based on the pollutant accumulation status. Step S4: Determine the water reuse status of the reservoir based on the purification change status of each reservoir node; detect the remaining discharge status of the reservoir based on the water reuse status; assess the reservoir operation risk data based on the remaining discharge status and purification change status, and upload the reservoir operation risk data to the cloud platform for early warning operation based on the Internet of Things.

2. The method for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things as described in claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain liquid level monitoring data and flow monitoring data in the wetland reservoir, extract the liquid level changes at the reservoir nodes from the liquid level monitoring data, and obtain the node liquid level change data; Step S12: Detect the flow transmission direction between nodes of each reservoir based on the flow monitoring data, obtain node flow direction data, and identify the water level transmission relationship between each reservoir based on the node liquid level change data and node flow direction data. Step S13: Identify the flow sequence between nodes based on the water level transfer relationship between each reservoir, obtain node flow time series data, and construct the water flow relationship between reservoirs based on the node flow time series data. Step S14: Match upstream inflow nodes and downstream outflow nodes according to the water flow relationship between reservoirs and ponds to obtain node confluence connection data, and construct reservoir confluence link data based on the node confluence connection data.

3. The method for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things as described in claim 2, characterized in that, Step S12 includes the following steps: Step S121: Extract instantaneous flow velocity change data between nodes of each reservoir based on flow monitoring data; Step S122: Calculate the flow fluctuation amplitude between each reservoir node based on the instantaneous flow velocity change data to obtain the node flow fluctuation data; Step S123: Detect the water level difference transmission between nodes of each reservoir based on node liquid level change data and node flow fluctuation data, and obtain node flow direction data; Step S124: Extract water level response delay data between adjacent reservoir nodes based on node flow direction data; Step S125: Determine the water level transfer relationship between each reservoir based on the node flow direction data and water level response delay data.

4. The method for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things as described in claim 3, characterized in that, Step S123 includes: Calculate the real-time liquid level difference between adjacent reservoir nodes based on the node liquid level change data to obtain node liquid level difference data; Based on the node liquid level difference data, the water level gradient change status between adjacent reservoir nodes is detected to obtain node water level gradient data. Identify the transmission of water level differences between adjacent reservoir nodes based on node water level gradient data and node flow fluctuation data; Based on the water level difference transmission, the direction of water transport offset between adjacent reservoir nodes is detected to obtain node flow direction data.

5. The method for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things as described in claim 1, characterized in that, Step S2 includes the following steps: Step S21: Extract the upstream inlet location data corresponding to each reservoir node based on the reservoir confluence link data, and collect the inflow change data of each reservoir node by using flow velocity sensors deployed at each upstream inlet location. Step S22: Detect the continuous flow status in the inlet diversion channel of each reservoir node based on the inflow change data, and obtain the continuous inflow change data of the node; Step S23: Detect the reservoir capacity occupancy change status of each reservoir node based on the continuous change data of water inflow at the node, and obtain the water storage change status of each reservoir node; Step S24: Identify the water retention and delay status within each reservoir node based on the water storage change status of each reservoir node, and obtain the water retention status of each reservoir node.

6. The method for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things according to claim 5, characterized in that, Step S22 includes the following steps: Step S221: Extract real-time flow rate change data of the diversion channel at the inlet of each reservoir node based on the inflow change data; Step S222: Detect the flow distribution at different cross-sectional positions within the guide channel based on real-time flow rate change data, and obtain the flow data of the guide channel cross-section; Step S223: Detect the velocity gradient change from the beginning to the end of the guide channel based on the cross-sectional flow data of the guide channel, and obtain the velocity gradient data of the guide channel; Step S224: Identify the continuous flow transmission state within the guide channel based on the flow velocity gradient data, and obtain the continuous change data of the incoming water at the node.

7. The method for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things according to claim 6, characterized in that, Step S223 includes: Extract the flow rate data of the first, middle and last sections of the guide channel based on the cross-sectional flow data; Based on the cross-sectional flow rate data at the beginning, middle and end of the guide channel, the friction resistance change along the guide channel is detected to obtain the friction resistance change data of the guide channel. Based on the data on the change in flow resistance along the flow channel, the decreasing state of the flow velocity gradient from the beginning to the end of the flow channel is identified, and the flow velocity gradient data of the flow channel is obtained.

8. The method for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things according to claim 5, characterized in that, Step S24 includes the following steps: Step S241: Extract the real-time water storage growth data corresponding to each reservoir node based on the water storage change status of each reservoir node; Step S242: Detect the water exchange change status between the inlet and outlet of each reservoir node based on real-time water storage growth data, and obtain node water exchange data; Step S243: Detect the velocity decay change at different locations within each reservoir node based on the node water exchange data, and obtain node velocity decay data. Step S244: Identify the water retention and delay status within each reservoir node based on the node flow velocity decay data, and obtain the water retention status of each reservoir node.

9. The method for online monitoring of water quantity and quality in wetlands and reservoirs based on the Internet of Things according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Extract stratified water quality monitoring data at different water depths within each reservoir node based on the water retention status of each reservoir node; Step S32: Detect the interlayer variation of pollutant concentration within each reservoir node based on the stratified water quality monitoring data to obtain node pollutant concentration gradient data; Step S33: Identify the pollutant accumulation status of each reservoir node based on the pollutant concentration gradient data; Step S34: Detect the purification change status of each reservoir node based on the pollutant accumulation status of each reservoir node, and obtain the purification change status of each reservoir node.

10. An online monitoring system for water quantity and quality in wetlands and reservoirs based on the Internet of Things, characterized in that, For executing the IoT-based online monitoring method for wetland reservoir water quantity and quality as described in claim 1, the IoT-based online monitoring system for wetland reservoir water quantity and quality includes: The reservoir confluence link construction module is used to acquire liquid level monitoring data and flow monitoring data in wetland reservoirs; identify the water flow relationship between reservoirs based on the liquid level monitoring data and flow monitoring data; and construct reservoir confluence link data based on the water flow relationship. The water retention status identification module is used to collect inflow change data of each reservoir node based on reservoir confluence link data; detect the water storage change status of each reservoir node based on the inflow change data and reservoir confluence link data; and identify the water retention status of each reservoir node based on the water storage change status. The purification change status assessment module is used to extract stratified water quality monitoring data corresponding to each reservoir node based on the water retention status of each reservoir node; identify the pollutant accumulation status of each reservoir node based on the stratified water quality monitoring data and water retention status; and assess the purification change status of each reservoir node based on the pollutant accumulation status. The risk warning module is used to determine the water reuse status of reservoirs based on the purification changes at each reservoir node; detect the remaining discharge status of reservoirs based on the water reuse status; assess the operational risk data of reservoirs based on the remaining discharge status and purification changes; and upload the operational risk data of reservoirs to the cloud platform for early warning operations based on the Internet of Things.

Images