Multi-pond wetland ecological hydrology regulation and control method and system

Abstract

The invention provides a multi-pond wetland ecological hydrology regulation and control method and system. The method comprises the steps that water quality data of multiple continuous water purification areas and meteorological data of geographic positions where the multiple continuous water purification areas are located are obtained; respectively constructing a water quality characteristic matrix and a meteorological characteristic matrix according to the water quality data and the meteorological data; according to the water quality characteristic matrix and the meteorological characteristic matrix, through a preset opening degree prediction strategy, the gate opening degree at the water inlet is predicted to serve as the predicted opening degree; obtaining constraint parameters representing hydraulic loads and pollution loads of the plurality of continuous water purification areas; according to the predicted opening degree and the constraint parameters, the gate control quantity at the water inlet is determined; and controlling the gate at the water inlet to operate according to the gate control quantity so as to regulate and control the water flow of the plurality of continuous water purification areas. The technical problems that water flow regulation and control are not flexible, and a sewage purification system is likely to be overloaded or the sewage purification effect is poor are solved.

Description

Technical Field

[0001] This invention relates to the field of hydrological regulation technology, specifically to a method and system for ecological hydrological regulation of Duotang Wetland. Background Technology

[0002] With the development of the times, environmental governance and the sustainable development of the ecological environment have become key issues of concern to all sectors. Among these issues, the discharge and purification of wastewater are particularly important.

[0003] In existing technologies, wastewater treatment (including discharge and purification) typically involves two methods. One method is to purify pollutants in wastewater at a wastewater treatment plant through sedimentation, chemical treatment, etc., and then use the purified water for irrigation, cleaning, or discharge into natural rivers. The second method is to directly discharge wastewater into artificially constructed purification wetlands (such as ponds, swamps, etc.), which not only purifies the water but also transforms the artificially constructed wetland purification system into everyday attractions such as wetland parks, beautifying the urban environment.

[0004] In existing wastewater treatment methods, wastewater discharge into wetland purification systems typically requires a stable preliminary testing process. After determining the purification capacity of the wetland purification system, wastewater is continuously discharged into it at a fixed flow rate. The wastewater is then purified by the wetland purification system (also known as a wastewater purification system, which refers to an aquatic ecosystem that purifies wastewater through a natural or artificially constructed wetland environment) before being discharged into natural watersheds, reservoirs, or for secondary use. However, in practical applications, the purification capacity of wetland purification systems is unstable due to variations in wastewater quality and real-time wetland flow rates. Continuous high-intensity wastewater discharge can easily overload the purification capacity of the wetland purification system or lead to poor purification results.

[0005] The existing invention patent application document CN120409960A, entitled "An Integrated Prevention and Control System for Targeted Interception and Ecological Restoration of Pollution Sources in Rivers Flowing into the Sea," describes a system comprising: a pollution source identification unit, a targeted interception unit, an ecological restoration unit, and a collaborative control platform. The identification unit utilizes a combination of multispectral UAVs and water quality sensors to locate pollution sources using artificial intelligence. The targeted interception unit implements physical-chemical combined interception based on pollution type. The ecological restoration unit purifies water quality and rebuilds the ecosystem through modular artificial wetlands. The collaborative control platform optimizes the strategies of each unit based on hydrological data. Formulas associated with each unit accurately calculate parameters, such as identification algorithms and dam opening control formulas. However, this existing solution primarily regulates water quality through pollution source identification, interception, and ecological restoration, neglecting the impact of meteorological factors on pond water quality.

[0006] In summary, existing wastewater treatment methods suffer from technical problems such as inflexible water flow control, which can easily lead to overload of the wastewater purification system or poor wastewater purification effect. Summary of the Invention

[0007] The technical problem to be solved by this invention is: how to solve the technical problem that the water flow regulation is inflexible in the prior art, which easily causes the sewage purification system to overload or the sewage purification effect to be poor.

[0008] The present invention solves the above-mentioned technical problems by adopting the following technical solution: The method for ecological and hydrological regulation of Duotang Wetland includes: S1. Obtain water quality data from no less than two continuous water purification areas, as well as meteorological data of the geographical locations of the continuous water purification areas; S2. Construct water quality characteristic matrices and meteorological characteristic matrices based on water quality data and meteorological data, respectively; S3. Based on the water quality characteristic matrix and meteorological characteristic matrix, predict the gate opening at the inlet of the continuous water purification area using a preset opening prediction strategy, and use the predicted opening as the predicted opening. S4. Obtain the constraint parameters of hydraulic load and pollution load to characterize no less than two continuous water purification zones; S5. Determine the gate control quantity at the inlet based on the predicted opening degree and constraint parameters; S6. Based on the gate control quantity, control the working status of the gate at the water inlet and regulate the water flow of no less than two continuous water purification zones.

[0009] This invention combines real-time water quality data from multiple continuous water purification zones with meteorological data of the geographical locations of these zones to determine their real-time wastewater purification capacity. This allows for the prediction of gate opening, and by adjusting the gate opening, dynamic control of wastewater flow is achieved. This avoids excessive water flow caused by wastewater or rainstorms, and improves upon the problems of inflexible water flow control in traditional wastewater treatment methods, which can easily lead to overload of the wastewater purification system or poor wastewater purification effect.

[0010] In a more specific technical solution, in S2, water quality data and meteorological data are preprocessed separately to obtain preprocessed water quality data and preprocessed meteorological data. Based on the pre-processed water quality data, a water quality feature matrix is ​​constructed. In the water quality feature matrix, one row of elements represents different types of pre-processed water quality data within the same time step, and one column of elements represents the element values ​​of the same type of pre-processed water quality data at different time steps. Based on the preprocessed meteorological data, a meteorological feature matrix is ​​constructed. Each row of the meteorological feature matrix represents different types of preprocessed meteorological data within the same time step, and each column of the meteorological feature matrix represents the element values ​​of the same type of preprocessed meteorological data at different time steps.

[0011] This invention improves the standardization of data structure by preprocessing water quality and meteorological data and constructing a time-series feature matrix (rows represent the values ​​of multiple types of data at the same time, and columns represent the values ​​of a single type of data at different times), providing high-dimensional feature support for subsequent accurate prediction.

[0012] In a more specific technical solution, S3 utilizes a dual-channel LSTM to extract features from the water quality feature matrix and the meteorological feature matrix, respectively, to obtain the first hidden state corresponding to the water quality feature matrix and the second hidden state corresponding to the meteorological feature matrix.

[0013] In the formula, This indicates the first hidden state. This indicates the primary channel LSTM of a dual-channel LSTM. Represents the water quality characteristic matrix. Let represent the trainable weight matrix and bias of the main channel LSTM, respectively. This indicates the second hidden state. This indicates the auxiliary channel LSTM of the dual-channel LSTM. Represents the meteorological feature matrix. , These represent the trainable weight matrix and bias of the auxiliary channel LSTM, respectively; Feature fusion is performed on the first hidden state and the second hidden state to obtain the fused state:

[0014] In the formula, Indicates the fusion state. This indicates element-wise multiplication. This represents the sigmoid activation function. , This represents the trainable gated weight matrix and gated bias; Based on the fusion status, the gate opening at the inlet of multiple continuous water purification zones is predicted as the predicted opening:

[0015] In the formula, Indicates the predicted opening degree. This represents the activation function. This represents the trainable prediction weight matrix and prediction bias.

[0016] This invention utilizes a dual-channel LSTM to extract temporal features of water quality and meteorological data separately, fuses the hidden states of the two channels through a gating mechanism, and then outputs the aperture prediction value through a fully connected layer. This effectively captures the spatiotemporal correlation of multi-source data and improves the accuracy of aperture prediction.

[0017] Compared with existing technologies, this invention, based on reasonable monitoring of water quality changes, takes into account the impact of meteorological factors (such as rainfall, sun exposure, etc.) on pond water quality through dual-channel LSTM, thereby enhancing the adaptability of multi-pond water quality control systems under complex natural conditions. This invention uses Long Short-Term Memory Neural Network (LSTM) as the core technology for predicting the opening of the inlet gate. LSTM has a more stable feature extraction capability in processing time-series data (changes in pond water quality and meteorological changes are necessarily time-series data).

[0018] In a more specific technical solution, S4 includes constraint parameters such as real-time influent pollutant concentration, real-time rainfall, design maximum flow rate, pollutant discharge limit, and time-varying maximum allowable flow rate. Obtain constraint parameters characterizing the hydraulic load and pollution load of multiple continuous water purification zones, including: The system obtains real-time influent pollutant concentration, real-time rainfall, design maximum flow rate, and pollutant discharge limits to determine the time-varying maximum allowable flow rate.

[0019] In a more specific technical solution, the following logic is used to determine the time-varying maximum allowable flow:

[0020] In the formula, Indicates the time-varying maximum allowable flow rate. This indicates the hydraulic safety factor and the pollution safety factor. Indicates the maximum design flow rate. Indicates real-time rainfall. Indicates pollutant weights, Indicates the first Emission limits for Class III pollutants Indicates the first Pollutants in Real-time concentration of pollutants in the incoming water.

[0021] This invention calculates the time-varying maximum allowable flow rate using dynamic parameters such as real-time rainfall and pollutant concentration, and dynamically controls the upper limit of water flow rate through dual constraints (hydraulic safety factor and pollution safety factor) to prevent the multi-pond wetland purification system from collapsing due to pollutant overload or excessive water flow caused by heavy rain or highly polluted inflow.

[0022] In a more specific technical solution, in S5, the predicted flow rate corresponding to the predicted opening is determined based on the predicted opening degree and the design maximum flow rate in the constraint parameters:

[0023] In the formula, Indicates predicted flow. Indicates the predicted opening degree. Indicates the maximum design flow rate; Determine the upper limit of water flow based on the time-varying maximum allowable flow rate in the constraint parameters:

[0024] In the formula, Indicates the upper limit of water flow. Indicates the redundancy safety factor. Indicates the time-varying maximum allowable flow rate; Determine the target flow rate based on the predicted flow rate and the upper limit of water flow:

[0025] In the formula, Indicates the target traffic; Determine the target gate opening based on the target flow rate and the design maximum flow rate:

[0026] In the formula, Indicates the target opening degree. Indicates the target traffic. Indicates the maximum design flow rate; Determine the difference in opening degree based on the target opening degree and the current opening degree of the gate:

[0027] In the formula, Indicates differences in aperture. Indicates the target opening degree. Indicates the current opening degree; Determine the gate control quantity based on the opening difference:

[0028] In the formula, This indicates the gate control quantity.

[0029] This invention converts the predicted opening degree into a flow rate value, determines the target flow rate by combining it with the time-varying flow rate upper limit, then converts it into a target opening degree, and finally generates a stepped control command (±5% or maintain) based on the difference between the target and the current opening degree. This closed-loop control strategy ensures the safety of the multi-pond wetland purification system while achieving smooth regulation of water flow.

[0030] In a more specific technical solution, in S6, the gate is closed when the water quality data meets the preset conditions. The preset conditions include: The target dissolved oxygen level in the water quality data is less than or equal to the preset dissolved oxygen level. The target dissolved oxygen level represents the dissolved oxygen level at the outlet of the last water purification zone in multiple continuous water purification zones. Alternatively, the target pH value in the water quality data is within a preset pH range. The target pH value represents the pH value at the outlet of the last water purification zone in a series of continuous water purification zones.

[0031] In a more specific technical solution, the adjusted gate opening is determined based on the gate control quantity and the current opening degree, and is used as the new current opening degree.

[0032] In a more specific technical solution, S6 issues a warning message indicating excessive sewage in the continuous water purification area when the water quality data meets preset conditions.

[0033] This invention sets preset conditions for low dissolved oxygen or abnormal pH (outside the range), automatically triggering a gate closing mechanism to prevent ecological damage caused by water quality deterioration and improve the system's emergency protection capabilities.

[0034] This invention updates the gate opening status at the inlet in real time based on the gate control quantity, ensuring the continuity of control commands and providing accurate current status input for the next water flow regulation.

[0035] This invention proactively issues warning messages when water quality meets preset abnormal conditions, assisting maintenance personnel in responding quickly to sewage overload events and reducing environmental risks.

[0036] In a more specific technical solution, the Duotang Wetland Eco-hydrological Regulation System includes: The data acquisition module is used to acquire water quality data from no less than two continuous water purification areas, as well as meteorological data of the geographical location of the continuous water purification areas. The matrix construction module is used to construct water quality feature matrices and meteorological feature matrices based on water quality data and meteorological data, respectively. The matrix construction module is connected to the data acquisition module. The gate opening prediction module is used to predict the gate opening at the inlet of the continuous water purification area based on the water quality feature matrix and meteorological feature matrix and through a preset gate opening prediction strategy. The gate opening prediction module is connected to the matrix construction module. The constraint parameter acquisition module is used to obtain the constraint parameters of hydraulic load and pollution load for characterizing no less than two continuous water purification zones; The gate control quantity determination module is used to determine the gate control quantity at the inlet based on the predicted opening and constraint parameters. The gate control quantity determination module is connected to the constraint parameter acquisition module and the opening prediction module. The water flow control module is used to control the working status of the gate at the inlet according to the gate control quantity, and to regulate the water flow of no less than two continuous water purification zones. The water flow control module is connected to the gate control quantity determination module.

[0037] The present invention has the following advantages over the prior art: This invention combines real-time water quality data from multiple continuous water purification zones with meteorological data of the geographical locations of these zones to determine their real-time wastewater purification capacity. This allows for the prediction of gate opening, and by adjusting the gate opening, dynamic control of wastewater flow is achieved. This avoids excessive water flow caused by wastewater or rainstorms, and improves upon the problems of inflexible water flow control in traditional wastewater treatment methods, which can easily lead to overload of the wastewater purification system or poor wastewater purification effect.

[0038] This invention improves the standardization of data structure by preprocessing water quality and meteorological data and constructing a time-series feature matrix (rows represent the values ​​of multiple types of data at the same time, and columns represent the values ​​of a single type of data at different times), providing high-dimensional feature support for subsequent accurate prediction.

[0039] This invention utilizes a dual-channel LSTM to extract temporal features of water quality and meteorological data separately, fuses the hidden states of the two channels through a gating mechanism, and then outputs the aperture prediction value through a fully connected layer. This effectively captures the spatiotemporal correlation of multi-source data and improves the accuracy of aperture prediction.

[0040] This invention calculates the time-varying maximum allowable flow rate using dynamic parameters such as real-time rainfall and pollutant concentration, and dynamically controls the upper limit of water flow rate through dual constraints (hydraulic safety factor and pollution safety factor) to prevent the multi-pond wetland purification system from collapsing due to pollutant overload or excessive water flow caused by heavy rain or highly polluted inflow.

[0041] This invention converts the predicted opening degree into a flow rate value, determines the target flow rate by combining it with the time-varying flow rate upper limit, then converts it into a target opening degree, and finally generates a stepped control command (±5% or maintain) based on the difference between the target and the current opening degree. This closed-loop control strategy ensures the safety of the multi-pond wetland purification system while achieving smooth regulation of water flow.

[0042] This invention sets preset conditions for low dissolved oxygen or abnormal pH (outside the range), automatically triggering a gate closing mechanism to prevent ecological damage caused by water quality deterioration and improve the system's emergency protection capabilities.

[0043] This invention updates the gate opening status at the inlet in real time based on the gate control quantity, ensuring the continuity of control commands and providing accurate current status input for the next water flow regulation.

[0044] This invention proactively issues warning messages when water quality meets preset abnormal conditions, assisting maintenance personnel in responding quickly to sewage overload events and reducing environmental risks.

[0045] This invention solves the technical problem in the prior art that the water flow control is inflexible, which can easily cause the sewage purification system to overload or the sewage purification effect to be poor. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the basic steps of the multi-pond wetland ecological hydrological regulation method in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the structure of the Duotang Wetland Purification System in Embodiment 1 of the present invention. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Example 1 like Figure 1 As shown, the method for ecological and hydrological regulation of multi-pond wetlands provided by the present invention includes the following basic steps: S1. Obtain water quality data from multiple continuous water purification areas, as well as meteorological data of the geographical locations of these areas. In this embodiment, for the convenience of implementing the above technical solution, please refer to... Figure 2 This application provides a multi-pond wetland purification system comprising multiple continuous water purification zones. Specifically, the system includes a stabilization pond for preliminary sedimentation treatment, subsurface flow wetlands and surface flow wetlands for parallel purification, and a submerged plant pond for deep purification, among other continuous water purification zones. Wastewater is discharged from the outlet into the stabilization pond via a gate. After sedimentation in the stabilization pond, the wastewater is diverted to the subsurface flow wetlands and surface flow wetlands for parallel purification, then flows into the submerged plant pond for deep purification. Finally, the purified water is discharged into rivers, reservoirs, or reused. The arrows in the figure indicate the direction of water flow.

[0049] In step S1, the water quality data includes, but is not limited to, COD (Chemical Oxygen Demand), BOD (Biochemical Oxygen Demand), TN (Total Nitrogen), TP (Total Phosphorus), DO (Dissolved Oxygen), and pH (acidity / alkalinity) at locations such as the inlet, subsurface flow wetland outlet, surface flow wetland outlet, and submerged plant pond outlet.

[0050] In this embodiment, meteorological data includes, but is not limited to: rainfall, temperature, wind speed, humidity, and solar radiation intensity.

[0051] In this embodiment, the acquisition of water quality and meteorological data can be achieved during the application phase of the technical solution proposed in this application. This can be done by real-time sensing and uploading of water quality and meteorological data through water quality sensors and meteorological sensors deployed in multiple continuous water purification areas and their geographical locations for subsequent processing. Alternatively, the acquisition of water quality and meteorological data can also be achieved during the testing phase of the technical solution proposed in this application. Users can pre-input and save multiple sets of data to simulate the water quality and meteorological conditions of multiple continuous water purification areas in the multi-pond wetland purification system, and then retrieve this data based on user-issued instructions during subsequent processing. The specific method for acquiring water quality and meteorological data is not limited here.

[0052] S2. Construct water quality characteristic matrices and meteorological characteristic matrices based on water quality data and meteorological data, respectively; In step S2, specific references for constructing the water quality feature matrix and meteorological feature matrix include, but are not limited to: Water quality data and meteorological data are preprocessed separately to obtain preprocessed water quality data and preprocessed meteorological data; Based on the preprocessed water quality data, a water quality feature matrix is ​​constructed. Each row of the water quality feature matrix represents different types of preprocessed water quality data within the same time step, and each column of the water quality feature matrix represents the element values ​​of the same type of preprocessed water quality data at different time steps. Based on the preprocessed meteorological data, a meteorological feature matrix is ​​constructed. Each row of the meteorological feature matrix represents different types of preprocessed meteorological data within the same time step, while each column of the meteorological feature matrix represents the element values ​​of the same type of preprocessed meteorological data at different time steps.

[0053] In this embodiment, the preprocessing of water quality data and meteorological data can be to convert the water quality data or meteorological data into dynamic features that characterize the relationship between the water quality data or meteorological data and the proportion of pollutants in the wastewater, and use them as preprocessed water quality data and preprocessed meteorological data.

[0054] For example, preprocessed water quality data may include:

[0055] In the formula, express Chemical oxygen demand removal rate at any given time. This indicates the chemical oxygen demand at the water inlet. This indicates the chemical oxygen demand (COD) at the outlet of the subsurface flow wetland. These represent the pollutant removal rates of subsurface flow wetlands and surface flow wetlands, respectively. This indicates the efficiency ratio between wetlands. Indicates dissolved oxygen level Rate of change within hours express Dissolved oxygen levels at any given time. express Dissolved oxygen at any given time (the rate of change of dissolved oxygen in each water purification zone can be calculated using formula (3), or the total rate of change of dissolved oxygen in multiple consecutive water purification zones). In this embodiment, the preprocessed meteorological data may include:

[0056] In the formula, express Temperature removal rate at any given time express The biochemical oxygen demand removal rate at any given time (the calculation logic is the same as that for chemical oxygen demand removal rate, the only difference being that the chemical oxygen demand in the water quality data is replaced with biochemical oxygen demand, which will not be elaborated here). express Temperature at any moment Indicates the pollutant load from rainfall. express Rainfall intensity at any given moment This indicates the total phosphorus at the water inlet. Indicates the characteristics of reoxygenation. express Wind speed at any moment express The dissolved oxygen saturation concentration in water at a given time represents the maximum amount of dissolved oxygen that a body of water can achieve under the current temperature and air pressure. express The actual dissolved oxygen level at any given time.

[0057] Thus, based on preprocessed water quality and meteorological data from multiple time points, and using a time window size of k, a water quality feature matrix and a meteorological feature matrix are constructed for each k time points. In the water quality feature matrix, each row represents different types of preprocessed water quality data within the same time step (i.e., the same moment), and each column represents the element values ​​of the same type of preprocessed water quality data at different time steps. Similarly, a meteorological feature matrix is ​​constructed based on the preprocessed meteorological data. In this matrix, each row represents different types of preprocessed meteorological data within the same time step, and each column represents the element values ​​of the same type of preprocessed meteorological data at different time steps. By preprocessing the water quality and meteorological data and constructing time-series feature matrices (rows representing values ​​of multiple data types at the same moment, and columns representing values ​​of a single data type at different moments), the standardization of the data structure is improved, providing high-dimensional feature support for subsequent accurate predictions.

[0058] S3. Based on the water quality characteristic matrix and meteorological characteristic matrix, predict the gate opening at the inlet using a preset opening prediction strategy, and use this as the predicted opening. In step S3, the specific operations for predicting the gate opening at the inlet of multiple continuous water purification zones include, but are not limited to: Using a dual-channel LSTM, features are extracted from the water quality feature matrix and the meteorological feature matrix respectively, resulting in the first hidden state corresponding to the water quality feature matrix and the second hidden state corresponding to the meteorological feature matrix.

[0059] In the formula, This indicates the first hidden state. This indicates the primary channel LSTM of a dual-channel LSTM. Represents the water quality characteristic matrix. Let represent the trainable weight matrix and bias of the main channel LSTM, respectively. This indicates the second hidden state. This indicates the auxiliary channel LSTM of the dual-channel LSTM. Represents the meteorological feature matrix. , These represent the trainable weight matrix and bias of the auxiliary channel LSTM, respectively; Feature fusion is performed on the first hidden state and the second hidden state to obtain the fused state:

[0060] In the formula, Indicates the fusion state. This indicates element-wise multiplication. This represents the sigmoid activation function. , This represents the trainable gated weight matrix and gated bias; Based on the fusion status, the gate opening at the inlet of multiple continuous water purification zones is predicted as the predicted opening:

[0061] In the formula, Indicates the predicted opening degree. This represents the activation function. This represents the trainable prediction weight matrix and prediction bias.

[0062] S4. Obtain constraint parameters characterizing the hydraulic load and pollution load of multiple continuous water purification zones; In step S4, the constraint parameters may include real-time influent pollutant concentration, real-time rainfall, design maximum flow rate, pollutant discharge limit, and time-varying maximum allowable flow rate; The specific operations for obtaining the constraint parameters of hydraulic load and pollution load include, but are not limited to: Obtain real-time influent pollutant concentration, real-time rainfall, maximum design flow rate, and pollutant discharge limits; The time-varying maximum allowable flow rate is determined based on real-time influent pollutant concentration, real-time rainfall, design maximum flow rate, and pollutant discharge limits: (12) In the formula, Indicates the time-varying maximum allowable flow rate. The hydraulic safety factor and pollution safety factor can be set to values ​​such as 0.8, 0.9, 0.95, etc., to further reduce water flow and pollutant load constraints, ensuring the stable operation of the multi-pond wetland purification system. This indicates the maximum design flow rate (the maximum design value for the Duotang Wetland Purification System). Indicates real-time rainfall. express Pollutant weights for different pollutant classes (which can be determined using the analytic hierarchy process). Indicates the first Emission limits for Class III pollutants Indicates the first Pollutants in Real-time concentration of pollutants in the incoming water.

[0063] S5. Determine the gate control quantity at the inlet based on the predicted opening degree and constraint parameters; In step S5, the specific operation for determining the gate control quantity at the inlet includes, but is not limited to: Based on the predicted opening size and the design maximum flow rate in the constraint parameters, determine the predicted flow rate corresponding to the predicted opening size:

[0064] In the formula, Indicates predicted flow. Indicates the predicted opening degree. Indicates the maximum design flow rate; Determine the upper limit of water flow based on the time-varying maximum allowable flow rate in the constraint parameters:

[0065] In the formula, Indicates the upper limit of water flow. Indicates the redundancy safety factor. Indicates the time-varying maximum allowable flow rate; Determine the target flow rate based on the predicted flow rate and the upper limit of water flow:

[0066] In the formula, Indicates the target traffic; Determine the target gate opening based on the target flow rate and the design maximum flow rate: (16) In the formula, Indicates the target opening degree. Indicates the target traffic. Indicates the maximum design flow rate; Determine the difference in opening degree based on the target opening degree and the current opening degree of the gate: (17) In the formula, Indicates differences in aperture. Indicates the target opening degree. Indicates the current opening degree; Determine the gate control quantity based on the opening difference: (18) In the formula, This indicates the gate control quantity.

[0067] S6. Control the gate at the water inlet to operate with gate control quantity to regulate the water flow of multiple continuous water purification zones.

[0068] In the practical application of step S6, after determining the predicted opening degree, the gate control quantity is determined based on the difference between the target opening degree and the current opening degree. Then, the gate is controlled to operate according to this gate control quantity. The target opening degree and the current opening degree are compared cyclically, and the gate at the inlet is controlled to operate with the new gate control quantity until the gate opening degree reaches the target opening degree, thus completing this round of regulation. In practical applications, cyclic regulation can be performed based on user-defined time steps (e.g., one hour, half an hour, one day, or any data in water quality / meteorological data exceeding a user-defined threshold).

[0069] As an optional implementation, the method may further include: When the water quality data meets the preset conditions, the gate is closed; The preset conditions are as follows: The target dissolved oxygen level in the water quality data is less than or equal to the preset dissolved oxygen level. The target dissolved oxygen level represents the dissolved oxygen level at the outlet of the last water purification zone in multiple continuous water purification zones. Alternatively, the target pH value in the water quality data is within a preset pH range. The target pH value represents the pH value at the outlet of the last water purification zone in multiple continuous water purification zones, that is, the pH value at the outlet of the submerged plant pond.

[0070] In this embodiment, the preset dissolved oxygen level and preset pH range can be flexibly set according to actual needs. For example, the preset dissolved oxygen level can be... Wait, the preset pH range can be... .

[0071] As an optional implementation, the method may further include: Based on the gate control quantity and the current opening degree, determine the adjusted gate opening degree, which will be used as the new current opening degree: like ;like ;like .in, This indicates the new current opening degree.

[0072] As an optional implementation, the method may further include: When the water quality data meets the preset conditions, a warning message indicating excessive sewage in the continuous water purification area is issued.

[0073] In this embodiment, when the water quality data meets the preset conditions, a warning message indicating that there is too much sewage in the continuous water purification area is issued through a prompting device (such as a buzzer, display, audible and visual alarm, etc.).

[0074] In summary, this invention combines real-time water quality data from multiple continuous water purification zones with meteorological data of the geographical locations of these zones to determine their real-time wastewater purification capacity. This allows for the prediction of gate opening, and by adjusting the gate opening, dynamic control of wastewater flow is achieved. This avoids excessive water flow caused by wastewater or rainstorms, and improves upon the problems of inflexible water flow control in traditional wastewater treatment methods, which can easily lead to overload of the wastewater purification system or poor wastewater purification effect.

[0075] This invention improves the standardization of data structure by preprocessing water quality and meteorological data and constructing a time-series feature matrix (rows represent the values ​​of multiple types of data at the same time, and columns represent the values ​​of a single type of data at different times), providing high-dimensional feature support for subsequent accurate prediction.

[0076] This invention utilizes a dual-channel LSTM to extract temporal features of water quality and meteorological data separately, fuses the hidden states of the two channels through a gating mechanism, and then outputs the aperture prediction value through a fully connected layer. This effectively captures the spatiotemporal correlation of multi-source data and improves the accuracy of aperture prediction.

[0077] This invention calculates the time-varying maximum allowable flow rate using dynamic parameters such as real-time rainfall and pollutant concentration, and dynamically controls the upper limit of water flow rate through dual constraints (hydraulic safety factor and pollution safety factor) to prevent the multi-pond wetland purification system from collapsing due to pollutant overload or excessive water flow caused by heavy rain or highly polluted inflow.

[0078] This invention converts the predicted opening degree into a flow rate value, determines the target flow rate by combining it with the time-varying flow rate upper limit, then converts it into a target opening degree, and finally generates a stepped control command (±5% or maintain) based on the difference between the target and the current opening degree. This closed-loop control strategy ensures the safety of the multi-pond wetland purification system while achieving smooth regulation of water flow.

[0079] This invention sets preset conditions for low dissolved oxygen or abnormal pH (outside the range), automatically triggering a gate closing mechanism to prevent ecological damage caused by water quality deterioration and improve the system's emergency protection capabilities.

[0080] This invention updates the gate opening status at the inlet in real time based on the gate control quantity, ensuring the continuity of control commands and providing accurate current status input for the next water flow regulation.

[0081] This invention proactively issues warning messages when water quality meets preset abnormal conditions, assisting maintenance personnel in responding quickly to sewage overload events and reducing environmental risks.

[0082] This invention solves the technical problem in the prior art that the water flow control is inflexible, which can easily cause the sewage purification system to overload or the sewage purification effect to be poor.

[0083] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for ecological and hydrological regulation of multi-pond wetlands, characterized in that, The method includes: S1. Obtain water quality data from no less than two continuous water purification areas, as well as meteorological data of the geographical locations of the continuous water purification areas; S2. Construct a water quality feature matrix and a meteorological feature matrix based on the water quality data and the meteorological data, respectively; S3. Based on the water quality feature matrix and the meteorological feature matrix, predict the gate opening at the inlet of the continuous water purification area using a preset opening prediction strategy, and use the predicted opening as the predicted opening. S4. Obtain the constraint parameters of hydraulic load and pollution load to characterize no less than two continuous water purification zones; S5. Determine the gate control quantity at the inlet based on the predicted opening degree and the constraint parameters; S6. According to the gate control quantity, control the working state of the gate at the water inlet and regulate the water flow of no less than two of the continuous water purification zones.

2. The method for ecological and hydrological regulation of multi-pond wetlands according to claim 1, characterized in that, In step S2, the water quality data and the meteorological data are preprocessed to obtain preprocessed water quality data and preprocessed meteorological data. Based on the preprocessed water quality data, a water quality feature matrix is ​​constructed, wherein one row of the water quality feature matrix represents different types of preprocessed water quality data within the same time step, and one column of the water quality feature matrix represents the element values ​​of the same type of preprocessed water quality data at different time steps. Based on the preprocessed meteorological data, a meteorological feature matrix is ​​constructed, wherein one row of the meteorological feature matrix represents different types of preprocessed meteorological data within the same time step, and one column of the meteorological feature matrix represents the element values ​​of the same type of preprocessed meteorological data at different time steps.

3. The method for ecological and hydrological regulation of multi-pond wetlands according to claim 1, characterized in that, In S3, Using a dual-channel LSTM, features are extracted from the water quality feature matrix and the meteorological feature matrix respectively, resulting in a first hidden state corresponding to the water quality feature matrix and a second hidden state corresponding to the meteorological feature matrix. In the formula, This indicates the first hidden state. This indicates the primary channel LSTM of a dual-channel LSTM. Represents the water quality characteristic matrix. Let represent the trainable weight matrix and bias of the main channel LSTM, respectively. This indicates the second hidden state. This indicates the auxiliary channel LSTM of the dual-channel LSTM. Represents the meteorological feature matrix. , These represent the trainable weight matrix and bias of the auxiliary channel LSTM, respectively; Feature fusion is performed on the first hidden state and the second hidden state to obtain the fused state: In the formula, Indicates a fusion state. This indicates element-wise multiplication. This represents the sigmoid activation function. , This represents the trainable gated weight matrix and gated bias; Based on the fusion state, the gate opening at the inlet of the multiple continuous water purification zones is predicted as the predicted opening: In the formula, Indicates the predicted opening degree. This represents the activation function. This represents the trainable prediction weight matrix and prediction bias.

4. The method for ecological and hydrological regulation of multi-pond wetlands according to claim 1, characterized in that, In S4, the constraint parameters include real-time influent pollutant concentration, real-time rainfall, design maximum flow rate, pollutant discharge limit, and time-varying maximum allowable flow rate; Obtaining constraint parameters characterizing the hydraulic load and pollution load of the multiple continuous water purification zones, including: The real-time influent pollutant concentration, the real-time rainfall, the design maximum flow rate, and the pollutant discharge limit are obtained to determine the time-varying maximum allowable flow rate.

5. The method for ecological and hydrological regulation of multi-pond wetlands according to claim 4, characterized in that, The time-varying maximum allowable flow rate is determined using the following logic: In the formula, Indicates the time-varying maximum allowable flow rate. This indicates the hydraulic safety factor and the pollution safety factor. Indicates the maximum design flow rate. Indicates real-time rainfall. Indicates pollutant weights, Indicates the first Emission limits for Class III pollutants Indicates the first Pollutants in Real-time concentration of pollutants in the incoming water.

6. The method for ecological and hydrological regulation of multi-pond wetlands according to claim 1, characterized in that, In step S5, the predicted flow rate corresponding to the predicted opening is determined based on the predicted opening and the design maximum flow rate in the constraint parameters: In the formula, Indicates predicted flow. Indicates the predicted opening degree. Indicates the maximum design flow rate; Based on the time-varying maximum allowable flow rate in the aforementioned constraint parameters, determine the upper limit of water flow rate: In the formula, Indicates the upper limit of water flow. Indicates the redundancy safety factor. Indicates the time-varying maximum allowable flow rate; Based on the predicted flow rate and the upper limit of water flow, determine the target flow rate: In the formula, Indicates the target traffic; Based on the target flow rate and the design maximum flow rate, determine the target opening degree of the gate: In the formula, Indicates the target opening degree. Indicates the target traffic. Indicates the maximum design flow rate; Based on the target opening degree and the current opening degree of the gate, determine the opening degree difference: In the formula, Indicates differences in aperture. Indicates the target opening degree. Indicates the current opening degree; Based on the opening difference, determine the gate control quantity: In the formula, This indicates the gate control quantity.

7. The method for ecological and hydrological regulation of multi-pond wetlands according to claim 1, characterized in that, In step S6, when the water quality data meets the preset conditions, the gate is closed; The preset conditions include: The target dissolved oxygen content in the water quality data is less than or equal to the preset dissolved oxygen content, and the target dissolved oxygen content represents the dissolved oxygen content at the outlet of the last water purification zone in the plurality of continuous water purification zones. Alternatively, the target pH value in the water quality data is within a preset pH range, and the target pH value represents the pH value at the outlet of the last water purification zone among the multiple continuous water purification zones.

8. The method for ecological and hydrological regulation of multi-pond wetlands according to claim 6, characterized in that, Based on the gate control quantity and the current opening degree, the adjusted gate opening degree is determined and used as the new current opening degree.

9. The method for ecological and hydrological regulation of multi-pond wetlands according to claim 1, characterized in that, In step S6, when the water quality data meets the preset conditions, a warning message indicating excessive sewage in the continuous water purification area is issued.

10. The Duotang Wetland Eco-hydrological Regulation System, characterized in that, The system includes: The data acquisition module is used to acquire water quality data from no less than two continuous water purification areas, as well as meteorological data of the geographical locations of the continuous water purification areas. A matrix construction module is used to construct a water quality feature matrix and a meteorological feature matrix based on the water quality data and the meteorological data, respectively. The matrix construction module is connected to the data acquisition module. The gate opening prediction module is used to predict the gate opening at the inlet of the continuous water purification area based on the water quality feature matrix and the meteorological feature matrix, and through a preset gate opening prediction strategy. The predicted gate opening is used as the predicted gate opening. The gate opening prediction module is connected to the matrix construction module. The constraint parameter acquisition module is used to obtain the constraint parameters of hydraulic load and pollution load for characterizing no less than two continuous water purification zones; A gate control quantity determination module is used to determine the gate control quantity at the inlet based on the predicted opening degree and the constraint parameters. The gate control quantity determination module is connected to the constraint parameter acquisition module and the opening degree prediction module. The water flow control module is used to control the working state of the gate at the inlet according to the gate control quantity, and to regulate the water flow of no less than two of the continuous water purification zones. The water flow control module is connected to the gate control quantity determination module.

Images