Sewage treatment plant automatic decision-making system based on AIoT driving

Abstract

The invention discloses a sewage treatment plant automatic decision-making system based on AIoT driving, and the system comprises a data collection layer which comprises a water quality sensor, a flow sensor and a liquid level sensor, and is used for monitoring various indexes of sewage in real time, including chemical oxygen demand, biochemical oxygen demand, ammonia nitrogen content and PH value; according to the data transmission layer, data collected by the sensor is transmitted to a data center based on a low-power-consumption wide area network and the 5G communication technology. According to the data processing and decision-making layer, a data center is provided with a server and a storage device and is responsible for receiving, storing and processing network data from the sensor. The AIoT technology is adopted, automatic operation of a sewage treatment plant is achieved, and the treatment efficiency and accuracy are greatly improved; various indexes of sewage are monitored in real time through a high-precision sensor, data analysis and prediction are performed by using a machine learning algorithm, potential problems can be found in time, and an optimization control strategy can be generated.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment, and in particular to an automated decision-making system for a sewage treatment plant driven by AIoT. Background Art

[0002] With the acceleration of urbanization, water pollution is becoming more and more serious, and sewage treatment plants play an important role in environmental protection. The traditional sewage treatment process relies on manual monitoring and control, which is inefficient and prone to errors. Therefore, the development of an efficient automated decision-making system is of great significance to improve the efficiency and quality of sewage treatment.

[0003] In the existing technology, sewage treatment plants usually adopt offline monitoring or semi-automatic control. Offline monitoring requires manual sampling and analysis, which has time delays and cannot respond to changes in the sewage treatment process in real time. Although semi-automatic control can achieve some basic process control, it still relies on the experience and judgment of the operator, making it difficult to achieve optimal control. In addition, traditional sewage treatment methods lack the ability to deeply analyze and predict data.

[0004] To this end, the present invention proposes an automated decision-making system for sewage treatment plants driven by AIoT. Summary of the invention

[0005] In view of the defects in the prior art, the present invention provides an automated decision-making system for a sewage treatment plant driven by AIoT, comprising:

[0006] Data collection layer: including water quality sensors, flow sensors, and liquid level sensors, which monitor various indicators of sewage in real time, including chemical oxygen demand, biochemical oxygen demand, ammonia nitrogen content, and pH value;

[0007] Data transmission layer: The data collected by the sensors is transmitted to the data center based on low-power wide area network and 5G communication technology;

[0008] Data processing and decision-making layer: The data center is equipped with servers and storage devices, responsible for receiving, storing, and processing network data from sensors; machine learning algorithms are used to analyze and predict data, identify potential problems in the sewage treatment process, and generate corresponding optimization control strategies;

[0009] Execution layer: including electric valves, pump stations, and dosing devices, which execute the control instructions issued by the data center to realize the automated operation of the sewage treatment process;

[0010] Aeration subsystem: The aeration subsystem provides oxygen by injecting air into the sewage, supporting the growth of microorganisms and promoting the decomposition of organic matter;

[0011] Biochemical subsystem: The biochemical subsystem removes organic pollutants and nitrogen and phosphorus substances in sewage through biochemical reactions;

[0012] Wherein, the aeration subsystem comprises:

[0013] Aeration device control: Automatically adjust aeration volume based on water quality monitoring data to optimize dissolved oxygen levels and improve treatment efficiency;

[0014] Aeration efficiency evaluation module: Use sensors to monitor dissolved oxygen concentration in real time, and combine data analysis and prediction modules to evaluate aeration effects;

[0015] Energy consumption management module: Analyze historical data through machine learning algorithms and optimize aeration strategies to reduce energy consumption.

[0016] Preferably: the data collection layer includes:

[0017] Water quality monitoring module: Based on chemical oxygen demand sensor, biochemical oxygen demand sensor, ammonia nitrogen content sensor, and pH value sensor, it monitors various indicators of sewage in real time;

[0018] Flow monitoring module: using high-precision flow meter to monitor the flow of sewage in real time;

[0019] Liquid level monitoring module: Use high-precision liquid level gauge to monitor the liquid level of process tanks and sedimentation tank nodes.

[0020] Preferably: the data processing and decision-making layer includes:

[0021] Data receiving and storage module: The data center is equipped with servers and storage devices, which are responsible for receiving and storing network data from sensors;

[0022] Data analysis module: Apply machine learning and deep learning to analyze and predict data and identify potential problems in the sewage treatment process;

[0023] Decision support module: Generate corresponding optimization control strategies based on the analysis results to achieve intelligent decision support.

[0024] Preferably: the data analysis module comprises:

[0025] Data preprocessing: Clean and normalize the collected data;

[0026] Feature extraction: extract key features from raw data, including trends, patterns, and mutation points;

[0027] Model training: Use convolutional neural networks and recurrent neural networks to train the extracted features to identify potential problems and anomalies, as follows:

[0028] y=f(W* x+b)

[0029] Where x is the input data, W is the weight of the convolution kernel, b is the bias term, f is the activation function, and * represents the convolution operation;

[0030] h t =tanh(W hh h t-1 +W xh x t +b h )

[0031] y t =W hy h t +b y

[0032] where h t is the hidden state at time t, x t is the input at time t, W hh , W xh and W hy They are the loop weight, the weight from input to hidden layer, and the weight from hidden layer to output layer, b h and b y are the bias terms of the hidden layer and the output layer respectively, and tanh is the hyperbolic tangent activation function;

[0033] Prediction and identification: Use the trained model to predict new data and identify abnormal conditions or potential problems in the sewage treatment process.

[0034] Preferably, the data analysis module uses a time series analysis method to predict sewage indicators, specifically:

[0035] φ(B)Δ d X t =c+θ(B)∈ t

[0036] Among them, X t is the time series data, φ(B) and θ(B) are lag operator polynomials, Δ d is a difference operation, ∈ t is the error term.

[0037] Preferably, the intelligent decision support of the decision support module comprises the following steps:

[0038] S1: Use linear programming, nonlinear programming or heuristic algorithms to develop optimal control strategies based on data analysis results;

[0039] S2: Generate specific control instructions based on the current working environment and historical data;

[0040] S3: Evaluate the effectiveness of control strategies in a simulated environment;

[0041] S4: Present the optimized control strategy in a visual way.

[0042] Preferably, the decision support module uses linear programming to determine the optimal dosage and pump station operation strategy, specifically:

[0043] minZ=c1x1+c2x2+…+c n x n

[0044]

[0045] a 21 x1+a 22 x2+…+a 2n x n ≤b2

[0046]

[0047] a m1 x1+a m2 x2+…+a mn x n ≤ b m

[0048] x1,x2,…,x n ≥0

[0049] Among them, x1,x2,…,x n are decision variables, c1, c2, …, c n is the cost coefficient, a aj and b i is the constraint coefficient.

[0050] Preferably: the execution layer includes:

[0051] Control equipment interface module: interface with electric valves, pump stations, and dosing devices to realize command transmission;

[0052] Automation control execution module: responsible for executing the control instructions issued by the data center to realize the automated operation of the sewage treatment process.

[0053] Preferably, the aeration subsystem combines the data analysis and prediction module to evaluate the aeration effect, specifically including the following steps:

[0054] S11: Data preprocessing: cleaning and normalizing the incoming dissolved oxygen data;

[0055] S12: Feature extraction: Extract key features from dissolved oxygen data, including average value, peak value, and valley value;

[0056] S13: Model training and prediction: Use historical data to train the machine learning model, input real-time data into the model to predict future dissolved oxygen levels, and evaluate whether the aeration effect meets expectations;

[0057] S14: Result feedback: Feedback the prediction results to the decision support module and adjust the aeration strategy according to the evaluation results.

[0058] Preferably: the biochemical subsystem comprises:

[0059] Biochemical reaction condition optimization module: adjust temperature and pH value based on data analysis results;

[0060] Index balance control module: monitors the concentrations of nitrogen, phosphorus and ammonia nitrogen, and automatically adjusts the dosage based on the dosing device to maintain index balance and prevent eutrophication.

[0061] The beneficial effects of the present invention are embodied in:

[0062] 1. The present invention adopts AIoT technology to realize the automated operation of sewage treatment plants, greatly improving treatment efficiency and accuracy; it monitors various indicators of sewage in real time through high-precision sensors, and uses machine learning algorithms for data analysis and prediction, which can timely discover potential problems and generate optimized control strategies.

[0063] 2. The present invention generates corresponding optimization control strategies based on the analysis results and provides them to decision makers as a reference to achieve intelligent decision support.

[0064] 3. The aeration subsystem of the present invention increases the dissolved oxygen concentration in the water through effective aeration, provides sufficient oxygen for microorganisms, promotes their growth and organic matter decomposition activities; and can combine the data analysis and prediction module to evaluate the aeration effect, so as to facilitate the adjustment of the strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0066] Figure 1 The figure is a flow chart of the intelligent decision support method of the decision support module of the present invention. DETAILED DESCRIPTION

[0067] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0068] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0069] Embodiment 1:

[0070] An automated decision-making system for sewage treatment plants driven by AIoT, including:

[0071] Data collection layer: Use a variety of high-precision sensors, including water quality sensors, flow sensors, liquid level sensors, etc., to monitor various indicators of sewage in real time, such as chemical oxygen demand (COD), biochemical oxygen demand (BOD), ammonia nitrogen content, pH value, etc.;

[0072] Data transmission layer: The data collected by the sensors is transmitted to the data center based on low-power wide area network (LPWAN) and 5G communication technology;

[0073] Data processing and decision-making layer: The data center is equipped with servers and storage devices, responsible for receiving, storing, and processing network data from sensors; machine learning algorithms are used to analyze and predict data, identify potential problems in the sewage treatment process, and generate corresponding optimization control strategies;

[0074] Execution layer: including electric valves, pump stations, and dosing devices, which execute the control instructions issued by the data center to realize the automated operation of the sewage treatment process.

[0075] Wherein, the data collection layer includes:

[0076] Water quality monitoring module: Use a variety of high-precision sensors, such as chemical oxygen demand (COD) sensor, biochemical oxygen demand (BOD) sensor, ammonia nitrogen content sensor, pH value sensor, etc. to monitor various indicators of sewage in real time;

[0077] Flow monitoring module: uses a high-precision flow meter to monitor the flow of sewage in real time and provide important parameters for data processing;

[0078] Liquid level monitoring module: Use high-precision liquid level gauges to monitor the liquid levels of key nodes such as process tanks and sedimentation tanks to ensure the stability of the treatment process.

[0079] Wherein, the data transmission layer includes:

[0080] Communication module: The data collected by the sensor is transmitted to the data center based on low-power wide area network and 5G communication technology;

[0081] Security module: Through data encryption and identity authentication technology, it ensures the security of data transmission and prevents data leakage and tampering.

[0082] The data processing and decision-making layer includes:

[0083] Data receiving and storage module: The data center is equipped with high-performance servers and storage devices, responsible for receiving and storing network data from sensors;

[0084] Data analysis module: Apply artificial intelligence algorithms, such as machine learning and deep learning, to analyze and predict data and identify potential problems in the sewage treatment process;

[0085] Decision support module: Generates corresponding optimization control strategies based on analysis results and provides them to decision makers as a reference to achieve intelligent decision support;

[0086] Aeration subsystem: The aeration subsystem provides oxygen by injecting air into the sewage, supporting the growth of microorganisms and promoting the decomposition of organic matter;

[0087] Wherein, the aeration subsystem comprises:

[0088] Aeration device control: Automatically adjust aeration volume based on water quality monitoring data to optimize dissolved oxygen levels and improve treatment efficiency;

[0089] Aeration efficiency evaluation module: Use sensors to monitor dissolved oxygen concentration in real time, and combine data analysis and prediction modules to evaluate aeration effects;

[0090] Energy consumption management module: Analyze historical data through machine learning algorithms and optimize aeration strategies to reduce energy consumption;

[0091] During implementation, on the basis of the above technical solution, the energy consumption management module can also optimize the dosage to save the use of medicine;

[0092] Biochemical subsystem: The biochemical subsystem removes organic pollutants and nitrogen and phosphorus substances in sewage through biochemical reactions;

[0093] Wherein, the biochemical subsystem includes:

[0094] Biochemical reaction condition optimization module: adjust temperature and pH value based on data analysis results;

[0095] Index balance control module: monitors the concentrations of nitrogen, phosphorus and ammonia nitrogen, automatically adjusts the dosage based on the dosing device, maintains index balance and prevents eutrophication.

[0096] Wherein, the data analysis module includes:

[0097] Data preprocessing: cleaning, normalizing and other preprocessing operations are performed on the collected data to prepare for subsequent analysis and training;

[0098] Feature extraction: extract key features from raw data, such as trends, patterns, mutation points, etc.;

[0099] Model training: Use convolutional neural networks and recurrent neural networks to train the extracted features to identify potential problems and anomalies, as follows:

[0100] y=f(W * x+b)

[0101] Where x is the input data, W is the weight of the convolution kernel, b is the bias term, f is the activation function, and * represents the convolution operation;

[0102] h t =tanh(W hh h t-1 +W xh x t +b h )

[0103] y t =W hy h t +b y

[0104] where h t is the hidden state at time t, x t is the input at time t, W hh , W xh and W hy They are the loop weight, the weight from input to hidden layer, and the weight from hidden layer to output layer, b h and b y are the bias terms of the hidden layer and the output layer respectively, and tanh is the hyperbolic tangent activation function;

[0105] Prediction and identification: Use the trained model to predict new data and identify abnormal conditions or potential problems in the sewage treatment process.

[0106] The data analysis module uses a time series analysis method to predict sewage indicators, specifically:

[0107] φ(B)Δ d X t =c+θ(B)ε t

[0108] Among them, X t is the time series data, φ(B) and θ(B) are lag operator polynomials, Δ d is a difference operation, ∈t is the error term.

[0109] The intelligent decision support of the decision support module includes the following steps:

[0110] S1: Use linear programming, nonlinear programming or heuristic algorithms to develop optimal control strategies based on data analysis results;

[0111] S2: Combine the current working environment and historical data to generate specific control instructions, such as adjusting the dosage, adjusting the operating power of the pump station, etc.;

[0112] S3: Evaluate the effectiveness of control strategies in a simulated environment;

[0113] S4: Present the optimized control strategy in a visual way.

[0114] The decision support module uses linear programming to determine the optimal dosage and pump station operation strategy, specifically:

[0115] minZ=c1x1+c2x2+…+c n x n

[0116]

[0117] a 21 x1+a 22 x2+…+a 2n x n ≤b2

[0118]

[0119] a m1 x1+a m2 x2+…+a mn x n ≤ b m

[0120] x1,x2,…,x n ≥0

[0121] Among them, x1,x2,…,x n are decision variables, c1, c2, …, c n is the cost coefficient, a aj and b i is the constraint coefficient.

[0122] The execution layer includes:

[0123] Control equipment interface module: interface with various automation control equipment, such as electric valves, pump stations, dosing devices, etc., to realize command transmission;

[0124] Automation control execution module: responsible for executing the control instructions issued by the data center to realize the automated operation of the sewage treatment process.

[0125] The aeration subsystem combines the data analysis and prediction module to evaluate the aeration effect, specifically including the following steps:

[0126] S11: Data preprocessing: cleaning and normalizing the incoming dissolved oxygen data;

[0127] S12: Feature extraction: Extract key features from dissolved oxygen data, including average value, peak value, and valley value;

[0128] S13: Model training and prediction: Use historical data to train the machine learning model, input real-time data into the model to predict future dissolved oxygen levels, and evaluate whether the aeration effect meets expectations;

[0129] S14: Result feedback: Feedback the prediction results to the decision support module and adjust the aeration strategy according to the evaluation results.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. An automated decision-making system for sewage treatment plants driven by AIoT, characterized by: include: Data collection layer: including water quality sensors, flow sensors, and liquid level sensors, which monitor various indicators of sewage in real time, including chemical oxygen demand, biochemical oxygen demand, ammonia nitrogen content, and pH value; Data transmission layer: The data collected by the sensors is transmitted to the data center based on low-power wide area network and 5G communication technology; Data processing and decision-making layer: The data center is equipped with servers and storage devices, responsible for receiving, storing, and processing network data from sensors; machine learning algorithms are used to analyze and predict data, identify potential problems in the sewage treatment process, and generate corresponding optimization control strategies; Execution layer: including electric valves, pump stations, and dosing devices, which execute the control instructions issued by the data center to realize the automated operation of the sewage treatment process; Aeration subsystem: The aeration subsystem provides oxygen by injecting air into the sewage, supporting the growth of microorganisms and promoting the decomposition of organic matter; Biochemical subsystem: The biochemical subsystem removes organic pollutants and nitrogen and phosphorus substances in sewage through biochemical reactions; Wherein, the aeration subsystem comprises: Aeration device control: Automatically adjust aeration volume based on water quality monitoring data to optimize dissolved oxygen levels and improve treatment efficiency; Aeration efficiency evaluation module: Use sensors to monitor dissolved oxygen concentration in real time, and combine data analysis and prediction modules to evaluate aeration effects; Energy consumption management module: Analyze historical data through machine learning algorithms and optimize aeration strategies to reduce energy consumption.

2. According to claim 1, an automated decision-making system for sewage treatment plants driven by AIoT is characterized in that: The data collection layer includes: Water quality monitoring module: Based on chemical oxygen demand sensor, biochemical oxygen demand sensor, ammonia nitrogen content sensor, and pH value sensor, it monitors various indicators of sewage in real time; Flow monitoring module: using high-precision flow meter to monitor the flow of sewage in real time; Liquid level monitoring module: Use high-precision liquid level gauge to monitor the liquid level of process tanks and sedimentation tank nodes.

3. According to claim 2, an AIoT-driven automated decision-making system for sewage treatment plants is characterized by: The data processing and decision-making layer includes: Data receiving and storage module: The data center is equipped with servers and storage devices, which are responsible for receiving and storing network data from sensors; Data analysis module: Apply machine learning and deep learning to analyze and predict data and identify potential problems in the sewage treatment process; Decision support module: Generate corresponding optimization control strategies based on the analysis results to achieve intelligent decision support.

4. According to claim 3, an AIoT-driven automated decision-making system for sewage treatment plants is characterized by: The data analysis module comprises: Data preprocessing: Clean and normalize the collected data; Feature extraction: extract key features from raw data, including trends, patterns, and mutation points; Model training: Use convolutional neural networks and recurrent neural networks to train the extracted features to identify potential problems and anomalies, as follows: y=f(W * x+b) Where x is the input data, W is the weight of the convolution kernel, b is the bias term, f is the activation function, and * represents the convolution operation; h t =tanh(W hh h t-1 +W xh x t +b h ) y t =W hy h t +b y where h t is the hidden state at time t, x t is the input at time t, W hh , W xh and W hy They are the loop weight, the weight from input to hidden layer, and the weight from hidden layer to output layer, b h and b y are the bias terms of the hidden layer and the output layer respectively, and tanh is the hyperbolic tangent activation function; Prediction and identification: Use the trained model to predict new data and identify abnormal conditions or potential problems in the sewage treatment process.

5. According to claim 4, an AIoT-driven automated decision-making system for sewage treatment plants is characterized by: The data analysis module uses a time series analysis method to predict sewage indicators, specifically: φ(B)δ d X t =c+θ(B)∈ t Among them, X t is the time series data, φ(B) and θ(B) are lag operator polynomials, Δ d is a difference operation, ∈ t is the error term.

6. According to claim 5, an automated decision-making system for sewage treatment plants driven by AIoT is characterized in that: The intelligent decision support of the decision support module comprises the following steps: S1: Use linear programming, nonlinear programming or heuristic algorithms to develop optimal control strategies based on data analysis results; S2: Generate specific control instructions based on the current working environment and historical data; S3: Evaluate the effectiveness of control strategies in a simulated environment; S4: Present the optimized control strategy in a visual way.

7. The AIoT-driven automated decision-making system for sewage treatment plants according to claim 6 is characterized by: The decision support module uses linear programming to determine the optimal dosage and pump station operation strategy, specifically: minZ=c1x1+c2x2+…+c n x n a 21 x1+a 22 x2+…+a 2n x n ≤b2 a m1 x1+a m2 x2+…+a mn x n ≤b m x1,x2,…,x n ≥0 Among them, x1,x2,…,x n are decision variables, c1, c2, …, c n is the cost coefficient, a aj and b i is the constraint coefficient.

8. The AIoT-driven automated decision-making system for sewage treatment plants according to claim 7 is characterized in that: The execution layer includes: Control equipment interface module: interface with electric valves, pump stations, and dosing devices to realize command transmission; Automation control execution module: responsible for executing the control instructions issued by the data center to realize the automated operation of the sewage treatment process.

9. The AIoT-driven automated decision-making system for sewage treatment plants according to claim 1, characterized in that: The aeration subsystem combines the data analysis and prediction module to evaluate the aeration effect, specifically including the following steps: S11: Data preprocessing: cleaning and normalizing the incoming dissolved oxygen data; S12: Feature extraction: Extract key features from dissolved oxygen data, including average value, peak value, and valley value; S13: Model training and prediction: Use historical data to train the machine learning model, input real-time data into the model to predict future dissolved oxygen levels, and evaluate whether the aeration effect meets expectations; S14: Result feedback: Feedback the prediction results to the decision support module and adjust the aeration strategy according to the evaluation results.

10. The AIoT-driven automated decision-making system for sewage treatment plants according to claim 9, characterized in that: The biochemical subsystem includes: Biochemical reaction condition optimization module: adjust temperature and pH value based on data analysis results; Index balance control module: monitors the concentrations of nitrogen, phosphorus and ammonia nitrogen, automatically adjusts the dosage based on the dosing device, maintains index balance and prevents eutrophication.

Images