Anti-clogging artificial wetland system based on intelligent algorithm

Abstract

The invention relates to an anti-clogging artificial wetland system based on an intelligent algorithm, and belongs to the technical field of ecological equipment, the system comprises a biological reaction tank, a sensing module, an automatic control module and a cleaning module; the sensing module is mounted in the biological reaction tank and is used for collecting state information of the biological reaction tank and sending the state information to the automatic control module; the state information comprises dissolved oxygen data, pH value, water flow, water level and water pressure; the automatic control module is used for predicting the water quality and the constructed wetland blockage degree according to the state information to obtain a blockage degree value, judging whether the blockage degree value is greater than a blockage threshold value or not, and sending a judgment result to the cleaning module if the blockage degree value is greater than the blockage threshold value; and the cleaning module is used for adjusting the water flow and the dissolved oxygen concentration of the biological reaction tank according to the judgment result to remove blockages. The problems that in the prior art, a constructed wetland substrate is prone to being blocked, so that the inflow water flow is unstable, and the effluent water quality is poor are solved.

Description

Technical Field

[0001] The invention relates to the technical field of ecological equipment, and in particular to an anti-clogging artificial wetland system based on an intelligent algorithm. Background Art

[0002] Artificial wetlands are artificially constructed facilities for sewage treatment that simulate the structure and function of natural wetlands. In recent years, they have become a commonly used sewage treatment method and have been highly praised for their high efficiency, economic applicability and eco-friendliness.

[0003] However, there are some drawbacks in the actual engineering application of artificial wetlands. The first is the area problem. Traditional artificial wetlands occupy a large area, have high requirements for engineering land, and have great restrictions. Secondly, the matrix is ​​prone to blockage, the influent flow is unstable, and the effluent water quality cannot be guaranteed for a long time. After completion, it is highly dependent on subsequent operations and has high operating costs.

[0004] Therefore, there is a need for an integrated artificial wetland equipment that occupies a small area and is highly automated and anti-clogging. Summary of the invention

[0005] In view of this, it is necessary to provide an anti-clogging artificial wetland system based on an intelligent algorithm to solve the problem that the artificial wetland matrix in the prior art is prone to clogging, resulting in unstable influent flow and poor effluent quality.

[0006] In order to solve the above problems, in a first aspect, the present invention provides an anti-clogging artificial wetland system based on an intelligent algorithm, comprising: a biological reaction tank, a sensor module, an automatic control module and a cleaning module; The sensor module is installed inside the bioreactor and is used to collect status information of the bioreactor and send the status information to the automation control module; the status information includes dissolved oxygen data, pH value, water flow, water level and water pressure; The automation control module is used to use a machine learning algorithm to predict the water quality and the blockage degree of the artificial wetland according to the state information to obtain a blockage degree value, and to determine whether the blockage degree value is greater than a blockage threshold value. If greater than, the determination result is sent to the cleaning module; The cleaning module is used to adjust the water flow and dissolved oxygen concentration of the biological reaction pool to remove blockages according to the judgment result.

[0007] In a possible implementation, the automation control module includes a data acquisition unit, a prediction unit, a judgment unit, and a transmission unit; The data acquisition unit is in communication with the sensor module, and is used to collect the state information and send the state information to the prediction unit; The prediction unit is used to use a machine learning algorithm to predict the water quality and the blockage degree of the artificial wetland according to the state information to obtain a blockage degree value; The judging unit is used to obtain the congestion degree value and judge whether the congestion degree value is greater than a congestion threshold value, and if so, send the judgment result to the transmission unit; The transmission unit is used to send the judgment result to the cleaning module.

[0008] In a possible implementation, the prediction unit includes an LSTM time series model subunit; The LSTM time series model is used to take the real-time dissolved oxygen data, pH value, water flow, water level and water pressure as input, and output a blockage degree value for real-time blockage trend prediction.

[0009] In a possible implementation, the sensor module includes a flow meter, a water level sensor, a pressure sensor, a dissolved oxygen sensor, a turbidity sensor, and a pH sensor; The flow meter is used to detect the water flow of the bioreactor, and the flow meter is installed at the water inlet of the bioreactor; The water level sensor is used to detect the water level of the bioreactor tank, and the water level sensor is installed inside the bioreactor tank; The pressure sensor is used to detect the water pressure of the bioreactor, and the pressure sensor is installed inside the bioreactor; The dissolved oxygen sensor is used to detect the dissolved oxygen concentration of the bioreactor tank, the turbidity sensor is used to detect the water quality of the bioreactor tank, and the pH sensor is used to detect the pH value of the bioreactor tank; The water level sensor, the dissolved oxygen sensor, the turbidity sensor and the pH sensor are all installed at the water outlet of the biological reaction tank.

[0010] In a possible implementation, a display module is further included, and the automation control module further includes a blockage degree feedback unit; The blockage degree feedback unit is used to obtain water flow and water pressure, and send the water flow and water pressure to the display module; The display module is used to visually display the water flow and water pressure.

[0011] In a possible implementation, an aerator and a smart valve are also included; The air outlet end of the aerator is built into the biological reaction tank, and the aerator is used to provide oxygen to the biological reaction tank; The intelligent valve is installed at the water inlet of the biological reaction tank, and the intelligent valve is used to adjust the water inlet flow of the biological reaction tank.

[0012] In a possible implementation, the cleaning module includes a regulating unit and a feedback unit; The regulating unit is connected to the aerator and the smart valve, and is used to regulate the opening and closing degree of the aerator and the smart valve according to the judgment result; The feedback unit is used to obtain the real-time dissolved oxygen concentration and water flow rate based on the opening and closing degree of the adjusted aerator and the intelligent valve; and send the real-time dissolved oxygen concentration and water flow rate to the prediction unit.

[0013] In a possible implementation, a monitoring and early warning module is further included. The monitoring and early warning module is communicated with the automatic control module and is used to obtain a judgment result of comparing the congestion degree value with a threshold value and issue an alarm according to the judgment result.

[0014] In a possible implementation, a preprocessing module is also included; The pretreatment module comprises a grid and a sedimentation tank, wherein the grid and the sedimentation tank are sequentially distributed along the flow direction of the sewage treatment, and the grid, the sedimentation tank and the biological reaction tank are connected through a pipeline.

[0015] In a possible implementation, a solar power supply module is further included, and the solar power supply system is electrically connected to the sensor module, the automation control module and the cleaning module respectively, so as to supply power to the sensor module, the automation control module and the cleaning module.

[0016] The beneficial effects of the present invention are as follows: in order to solve the problem that the artificial wetland system is prone to clogging and the clogging control efficiency is low, the present invention installs a sensor module in the bioreactor, and based on the sensor module collects state information parameters that can reflect the clogging situation and water quality of the bioreactor, thereby ensuring the purpose of being able to grasp the water quality state of the bioreactor in real time, and adopts a machine learning algorithm to predict the water quality and clogging situation of the bioreactor based on the state information parameters, and when the clogging degree value is greater than a threshold value, a cleaning module is used to clean the bioreactor to evacuate the clogging situation, thereby realizing precise control of the clogging situation of the bioreactor, and improving the anti-clogging ability and processing efficiency in an automated manner, so that the present invention effectively improves the water quality management quality through efficient clogging processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A system architecture diagram of an embodiment of an anti-clogging artificial wetland system based on an intelligent algorithm provided by the present invention; Figure 2A schematic structural diagram of another embodiment of the anti-clogging artificial wetland system based on intelligent algorithm provided by the present invention. DETAILED DESCRIPTION

[0018] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0019] Before presenting the embodiments, the following terms are explained.

[0020] Artificial wetland: a sewage treatment technology that simulates the natural wetland ecosystem. It is artificially designed and constructed, and uses natural ecological processes to purify sewage.

[0021] Biological reactor: It is a facility specially used for sewage treatment. It removes organic pollutants, nitrogen, phosphorus and other nutrients in sewage through biological treatment technology. It usually combines the metabolism of microorganisms and physical and chemical processes, and is the carrier and important component of sewage treatment in artificial wetlands.

[0022] A specific embodiment of the present invention discloses an anti-clogging artificial wetland system based on an intelligent algorithm, please refer to Figure 1 , including: a biological reaction pool 1, a sensor module 2, an automatic control module 3 and a cleaning module 4; The sensor module is installed inside the bioreactor and is used to collect status information of the bioreactor and send the status information to the automation control module; the status information includes dissolved oxygen data, pH value, water flow, water level and water pressure; It should be noted that various status information parameters that can reflect the blockage status and water quality of the biological reactor are collected through the sensor module, thereby achieving real-time monitoring of the water quality and blockage status of the biological reactor.

[0023] In order to achieve comprehensive monitoring, the sensing module has a plurality of different sensing bodies. In some embodiments, the sensing module includes a flow meter, a water level sensor, a pressure sensor, a dissolved oxygen sensor, a turbidity sensor and a pH sensor; The flow meter is used to detect the water flow of the bioreactor, and the flow meter is installed at the water inlet of the bioreactor; The water level sensor is used to detect the water level of the bioreactor tank, and the water level sensor is installed inside the bioreactor tank; The pressure sensor is used to detect the water pressure of the bioreactor, and the pressure sensor is installed inside the bioreactor; The dissolved oxygen sensor is used to detect the dissolved oxygen concentration of the bioreactor tank, the turbidity sensor is used to detect the water quality of the bioreactor tank, and the pH sensor is used to detect the pH value of the bioreactor tank; The water level sensor, the dissolved oxygen sensor, the turbidity sensor and the pH sensor are all installed at the water outlet of the biological reaction tank.

[0024] It should be noted that dissolved oxygen data is mainly used to reflect the concentration of dissolved oxygen in the bioreactor. The higher the dissolved oxygen concentration, the higher the aerobic metabolic activity of the microorganisms, and thus the higher the water degradation efficiency in the bioreactor, and the lower the probability of blockage in the bioreactor. Water flow, water level and water pressure data are mainly used to indirectly reflect the degree of blockage in the bioreactor. If the water flow and water pressure are normal, it means that there is no blockage or the blockage is slight. If the water flow is significantly reduced or the water pressure is increased, it means that there is a blockage. If the water level increases significantly, it means that there is a blockage. The pH value is used to reflect the quality of water.

[0025] Furthermore, the acquisition frequency was set according to the dynamic characteristics of the system, such as once per minute for dissolved oxygen and turbidity and 10 times per second for pressure and flow data, to accommodate the temporal resolution of the blockage formation process.

[0026] The automation control module is used to use a machine learning algorithm to predict the water quality and the blockage degree of the artificial wetland according to the state information to obtain a blockage degree value, and to determine whether the blockage degree value is greater than a blockage threshold value. If greater than, the determination result is sent to the cleaning module; It should be noted that the degree of congestion is predicted through the automatic control module, and the need for cleaning is determined based on the relationship between the degree of congestion and the threshold. Congestion prediction is performed through real-time collected data, which greatly improves the efficiency of anti-congestion.

[0027] In some embodiments of the present invention, see Figure 1 , the automation control module 3 includes a data acquisition unit 31, a prediction unit 32, a judgment unit 33 and a transmission unit 34; The data acquisition unit is in communication with the sensor module, and is used to collect the state information and send the state information to the prediction unit; The prediction unit is used to use a machine learning algorithm to predict the water quality and the blockage degree of the artificial wetland according to the state information to obtain a blockage degree value; The judging unit is used to obtain the congestion degree value and judge whether the congestion degree value is greater than a congestion threshold value, and if so, send the judgment result to the transmission unit; The transmission unit is used to send the judgment result to the cleaning module.

[0028] It should be noted that this embodiment realizes real-time monitoring, accurate prediction, rapid judgment and automatic transmission of the sewage treatment system by introducing an automated control module, which significantly improves the anti-blockage management efficiency and response speed.

[0029] In a specific embodiment, the prediction unit includes an LSTM time series model subunit; The LSTM time series model is used to take the real-time dissolved oxygen data, pH value, water flow, water level and water pressure as input, and output a blockage degree value for real-time blockage trend prediction.

[0030] It should be noted that LSTM is a time series neural network that can store context information through memory cells when processing time series data, effectively capture long-range dependencies, and dynamically adjust gradient flows through gating mechanisms. LSTM avoids the gradient vanishing problem of traditional RNNs during back propagation. LSTM can process dynamic time series and static input data at the same time, and has high applicability in scenarios such as congestion trend prediction.

[0031] Congestion trend prediction formula: W is the LSTM model parameter, which needs to be optimized through the optimizer.

[0032] Independent variables: X = {xt, xt-1, ..., xt-T+1}; xi = {Qwater, DO2, Turbidity, Pfilter} Target variable: Y=Pclog.

[0033] In the application process of the LSTM model, in order to ensure the accuracy of sensor data and the effectiveness of model training, the following data preprocessing steps are adopted: 1) Data cleaning ① Missing value processing: If the sensor generates missing data due to failure or communication interruption, interpolation methods are used to fill the data, such as time series interpolation or sliding window average method.

[0034] ②Data time alignment The sampling frequencies of sensors are different, and the time stamps are aligned to achieve a unified step size. Resampling technology is used to unify all data to a set time step size (for example, once every 10 seconds).

[0035] ③Data standardization In order to eliminate the influence of different sensor units and dimensions, the following method is used for standardization: ④Time series windowing To adapt the time series model (such as LSTM), the data is processed by sliding window: The sliding window length T is 10 time steps.

[0036] Each window corresponds to a congestion prediction label, and the label value is the value of Pclog at time t11.

[0037] The input matrix formed contains time series samples and their corresponding labels for subsequent machine learning model training.

[0038] 2) Data Labeling The tag Pclog for congestion trend prediction is generated by analyzing historical sensor data and system status: a continuous value in the range of [0,1] is generated according to the degree of congestion.

[0039] 3) Data storage and partitioning ①Data storage: The dataset is stored in the form of time series, including raw data and preprocessed data, to support subsequent model training and verification.

[0040] ②Data division: Partition the data in chronological order to ensure time series consistency of model evaluation: The training set accounts for 70% of the total data The test set each accounts for 30% 2. Model training and verification methods (1) Model training 1) Training data preparation ① Input features: Sensor data, including water volume (Qwater), dissolved oxygen concentration (DO2), turbidity (Turbidity) and filter layer pressure (Pfilter) The data structure of the windowed time series is in the form of X={xt,xt-1,…,xt-T+1}, where T is the window length.

[0041]

[0042]

[0043] ④Regularization measures: Dropout (0.2-0.5) was used to prevent overfitting.

[0044] Add L2 regularization to constrain model weights.

[0045] (2) Model validation 1) Verification indicators Returning Mission: Mean Squared Error (MSE): Measures the deviation of the predicted value from the true value.

[0046] Coefficient of determination (R2): Evaluates the goodness of fit of the model.

[0047] 2) Verification method Cross Validation: Use K-fold cross validation (usually K=5) to divide the dataset into K subsets, use one of the subsets for validation in turn, and use the remaining subsets for training.

[0048] Record the performance indicators of each verification and calculate the mean and variance.

[0049] Therefore, based on the above-mentioned constructed and trained model, an accurate congestion prediction value can be obtained, so as to judge the congestion situation based on the congestion prediction value, and perform clearing operations in time if there is congestion.

[0050] The cleaning module is used to adjust the water flow and dissolved oxygen concentration of the biological reaction pool to remove blockages according to the judgment result.

[0051] It should be noted that the bioreactor is cleaned by adjusting the water flow and dissolved oxygen concentration through the cleaning module, thereby alleviating the blockage and greatly improving the anti-blockage efficiency.

[0052] In some embodiments of the present invention, see Figure 2 ,in Figure 2 The equipment includes: a water pipeline (1), sewage passes through a grille (2) and a lifting pump (3) and enters a sedimentation tank (5) through an intelligent valve (4) for sewage pretreatment, and the sewage enters a biological unit (11) through a water pipeline after pretreatment, first enters a valve well through an intelligent valve (4), enters a biological unit through a water distribution pipe (10) for treatment, and flows out through a drainage pipe (14). It also includes an aerator and an intelligent valve; The air outlet end of the aerator is built into the biological reaction tank, and the aerator is used to provide oxygen to the biological reaction tank; The intelligent valve is installed at the water inlet of the biological reaction tank, and the intelligent valve is used to adjust the water inlet flow of the biological reaction tank.

[0053] In this embodiment, an aerator is installed to adjust the oxygen concentration of the biological reaction pool, thereby adjusting the activity of microorganisms, and decomposing granular or blocky substances, further reducing the risk of blockage. In addition, by installing an intelligent valve and adjusting the water inlet flow based on the intelligent valve, when blockage occurs, the water inlet is accelerated through the intelligent valve, thereby flushing the biological reaction pool and reducing the degree of blockage.

[0054] It should be noted that the structure or type of the smart valve is not limited here.

[0055] Furthermore, for higher cleaning efficiency, please refer to Figure 1, the cleaning module includes a regulating unit 41 and a feedback unit 42; The regulating unit is connected to the aerator and the smart valve, and is used to regulate the opening and closing degree of the aerator and the smart valve according to the judgment result; The feedback unit is used to obtain the real-time dissolved oxygen concentration and water flow rate based on the opening and closing degree of the adjusted aerator and the intelligent valve; and send the real-time dissolved oxygen concentration and water flow rate to the prediction unit.

[0056] In this embodiment, the opening and closing degree of the aerator and the intelligent valve is first adjusted by the regulating unit. When blockage occurs, the dissolved oxygen content and water flow rate of the biological reaction tank can be increased to treat the blockage. At the same time, the dissolved oxygen concentration and water flow rate are collected in real time by the feedback unit and sent to the prediction unit to predict the blockage degree value in real time. When the blockage degree value is lower than the blockage threshold, the automatic control module is also used to control the cleaning module to stop cleaning work, thereby realizing cycle control, avoiding energy waste, and improving the efficiency of blockage treatment.

[0057] In order to achieve the purpose of visualization and real-time monitoring of the blockage situation, in some embodiments of the present invention, please refer to Figure 2 , further comprising a display module, and the automation control module further comprising a blockage degree feedback unit; The blockage degree feedback unit is used to obtain water flow and water pressure, and send the water flow and water pressure to the display module; The display module is used to visually display the water flow and water pressure.

[0058] In this embodiment, whether the water flow is stable, whether the size is appropriate, and whether the water pressure is stable can indirectly reflect the blockage condition of the biological reactor. By obtaining the water flow and water pressure through the blockage feedback unit, the backstage personnel can be qualitatively reminded and an early warning can be given to a certain extent.

[0059] Further, in order to have a more efficient processing action, in some embodiments of the present invention, please refer to Figure 2 , also includes a monitoring and early warning module, which is communicated with the automatic control module and is used to obtain a judgment result of comparing the congestion degree value with a threshold value, and issue an alarm based on the judgment result.

[0060] In order to make the embodiments of the present invention more energy-saving and environmentally friendly, some embodiments of the present invention also include a solar power supply module, and the solar power supply system is electrically connected to the sensor module, the automation control module and the cleaning module, respectively, for supplying power to the sensor module, the automation control module and the cleaning module.

[0061] Furthermore, in order to improve the anti-clogging efficiency, in some embodiments of the present invention, a pre-processing module is also included; The pretreatment module comprises a grid and a sedimentation tank, wherein the grid and the sedimentation tank are sequentially distributed along the flow direction of the sewage treatment, and the grid, the sedimentation tank and the biological reaction tank are connected through a pipeline.

[0062] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. An anti-clogging artificial wetland system based on intelligent algorithm, characterized in that: include: Biological reaction tank, sensor module, automatic control module and cleaning module; The sensor module is installed inside the bioreactor and is used to collect status information of the bioreactor and send the status information to the automation control module; the status information includes dissolved oxygen data, pH value, water flow, water level and water pressure; The automation control module is used to use a machine learning algorithm to predict the water quality and the blockage degree of the artificial wetland according to the state information to obtain a blockage degree value, and to determine whether the blockage degree value is greater than a blockage threshold value. If greater than, the determination result is sent to the cleaning module; The cleaning module is used to adjust the water flow and dissolved oxygen concentration of the biological reaction pool to remove blockages according to the judgment result.

2. The anti-clogging artificial wetland system based on intelligent algorithm according to claim 1 is characterized in that: The automation control module includes a data acquisition unit, a prediction unit, a judgment unit and a transmission unit; The data acquisition unit is in communication with the sensor module, and is used to collect the state information and send the state information to the prediction unit; The prediction unit is used to use a machine learning algorithm to predict the water quality and the blockage degree of the artificial wetland according to the state information to obtain a blockage degree value; The judging unit is used to obtain the congestion degree value and judge whether the congestion degree value is greater than a congestion threshold value, and if so, send the judgment result to the transmission unit; The transmission unit is used to send the judgment result to the cleaning module.

3. The anti-clogging artificial wetland system based on intelligent algorithm according to claim 2 is characterized in that: The prediction unit includes an LSTM time series model subunit; The LSTM time series model is used to take the real-time dissolved oxygen data, pH value, water flow, water level and water pressure as input, and output a blockage degree value for real-time blockage trend prediction.

4. The anti-clogging artificial wetland system based on intelligent algorithm according to claim 1 is characterized in that: The sensor module includes a flow meter, a water level sensor, a pressure sensor, a dissolved oxygen sensor, a turbidity sensor and a pH sensor; The flow meter is used to detect the water flow of the bioreactor, and the flow meter is installed at the water inlet of the bioreactor; The water level sensor is used to detect the water level of the bioreactor tank, and the water level sensor is installed inside the bioreactor tank; The pressure sensor is used to detect the water pressure of the bioreactor, and the pressure sensor is installed inside the bioreactor; The dissolved oxygen sensor is used to detect the dissolved oxygen concentration of the bioreactor tank, the turbidity sensor is used to detect the water quality of the bioreactor tank, and the pH sensor is used to detect the pH value of the bioreactor tank; The water level sensor, the dissolved oxygen sensor, the turbidity sensor and the pH sensor are all installed at the water outlet of the biological reaction tank.

5. The anti-clogging artificial wetland system based on intelligent algorithm according to claim 4 is characterized in that: It also includes a display module, and the automation control module also includes a blockage degree feedback unit; The blockage degree feedback unit is used to obtain water flow and water pressure, and send the water flow and water pressure to the display module; The display module is used to visually display the water flow and water pressure.

6. The anti-clogging artificial wetland system based on intelligent algorithm according to claim 1 is characterized in that: Also included are aerators and smart valves; The air outlet end of the aerator is built into the biological reaction tank, and the aerator is used to provide oxygen to the biological reaction tank; The intelligent valve is installed at the water inlet of the biological reaction tank, and the intelligent valve is used to adjust the water inlet flow of the biological reaction tank.

7. The anti-clogging artificial wetland system based on intelligent algorithm according to claim 6 is characterized in that: The cleaning module includes a regulating unit and a feedback unit; The regulating unit is connected to the aerator and the smart valve, and is used to regulate the opening and closing degree of the aerator and the smart valve according to the judgment result; The feedback unit is used to obtain the real-time dissolved oxygen concentration and water flow rate based on the opening and closing degree of the adjusted aerator and the intelligent valve; and send the real-time dissolved oxygen concentration and water flow rate to the prediction unit.

8. The anti-clogging artificial wetland system based on intelligent algorithm according to claim 1 is characterized in that: It also includes a monitoring and early warning module, which is in communication with the automation control module and is used to obtain a judgment result of comparing the congestion degree value with a threshold value and issue an alarm based on the judgment result.

9. The anti-clogging artificial wetland system based on intelligent algorithm according to claim 1 is characterized in that: It also includes a pre-processing module; The pretreatment module comprises a grid and a sedimentation tank, wherein the grid and the sedimentation tank are sequentially distributed along the flow direction of the sewage treatment, and the grid, the sedimentation tank and the biological reaction tank are connected through a pipeline.

10. The anti-clogging artificial wetland system based on intelligent algorithm according to claim 1 is characterized in that: It also includes a solar power supply module. The solar power supply system is electrically connected to the sensor module, the automation control module and the cleaning module respectively, and is used to supply power to the sensor module, the automation control module and the cleaning module.

Images