Intelligent aeration control system and method for sewage treatment plant

Abstract

The application discloses a sewage treatment plant intelligent aeration control system and method, and relates to the technical field of sewage treatment control; comprising a data acquisition module, an aeration test module, a demand prediction module, a normal state analysis module and an aeration control module: the data acquisition module acquires real-time working data of a primary sedimentation tank, aeration tank test data, aeration tank historical working data and real-time working data, and performs preprocessing; the technical key points are that: the aeration tank is scientifically divided into multiple sections, professional monitoring equipment is arranged in each section, multi-dimensional data is collected, independent collection and test of data of each section can accurately master the operation condition and treatment effect of each section of aeration equipment, based on these accurate data, combined with an aeration efficiency evaluation model and a regulation and control strategy, fine adjustment can be carried out according to the actual demand of each section, resource waste or poor treatment effect caused by one-size-fits-all regulation and control is avoided, efficient and accurate operation of the aeration equipment is realized, and the application has a good application prospect.

Description

Technical Field

[0001] This invention relates to the field of wastewater treatment control technology, specifically to an intelligent aeration control system and method for wastewater treatment plants. Background Technology

[0002] Faced with the dual challenges of water scarcity and escalating water pollution, wastewater treatment plants have emerged as crucial facilities for protecting urban water environments. Like massive "urban kidneys," they purify domestic sewage and industrial wastewater layer by layer through physical treatments such as screen interception and sedimentation, biodegradation technologies such as activated sludge and biofilm processes, and advanced treatment methods such as chemical coagulation and disinfection. This not only effectively reduces pollutants but also enables the recycling of water resources, making them a core force in maintaining ecological balance and promoting sustainable development.

[0003] In the complex process of wastewater treatment, the biological treatment stage is the core link in removing pollutants. The efficient metabolism of microorganisms cannot be separated from sufficient oxygen. As a key device for supplying oxygen to the reaction tank, the control technology of the aeration system directly affects the efficiency and energy consumption of wastewater treatment. Aeration control, through intelligent adjustment of aeration volume, can not only accurately meet the dissolved oxygen requirements of microorganisms and ensure that organic matter in wastewater is fully degraded, but also avoid energy waste caused by excessive aeration. It has become an important technological breakthrough for improving the operational efficiency of wastewater treatment plants and achieving green and low-carbon treatment.

[0004] The existing patent, CN105152308B, entitled "Control Method and Control System for Aeration in MBR Aerobic Tank," describes the following steps: collecting ammonia nitrogen concentration signals from the membrane tank; obtaining a first residual value between the ammonia nitrogen concentration signal and a preset ammonia nitrogen concentration value; obtaining a preset dissolved oxygen concentration value in the aerobic tank using a first-order PI algorithm; collecting dissolved oxygen concentration signals from the aerobic tank; obtaining a second residual value between the dissolved oxygen concentration signal and the preset dissolved oxygen concentration value; obtaining the aeration rate value in the aerobic tank using a second-order PI algorithm; and controlling multiple blowers based on the aeration rate value in the aerobic tank and the blower performance curve. The control method of this invention further considers the biochemical effects of the membrane tank, improves control accuracy, better realizes automatic control of the MBR biochemical treatment aeration process, stabilizes effluent quality, reduces wastewater treatment energy consumption, and is simple and easy to implement.

[0005] The core idea of ​​the solution described in the aforementioned patent is to collect ammonia nitrogen concentration signals from the membrane tank and dissolved oxygen concentration signals from the aerobic tank, calculate the preset dissolved oxygen concentration value and aeration rate value using a two-stage PI algorithm, control the blower by combining the blower performance curve, and include signal validity verification, water volume compensation mechanism, and a control strategy for switching the working status of the influent flow meter, online ammonia nitrogen meter, and online dissolved oxygen meter, so as to achieve precise aeration control, stabilize water quality and reduce energy consumption.

[0006] However, the solution described in the aforementioned patent has the following drawbacks when used:

[0007] In the process of aeration control, it adopts a centralized data collection and maintenance method without considering the situation of local blockage of aeration equipment. In actual use, due to the influence of sludge, some areas of the aeration equipment are severely blocked, which greatly reduces the aeration effect in that part. This method cannot achieve stable oxygen supply and will greatly increase aeration energy consumption.

[0008] In addition, existing aeration control typically uses real-time accurate prediction. However, during use, the water flow will change due to environmental or other factors. At this time, the amount of data that needs to be processed is large, and the equipment parameters are usually adjusted frequently and significantly. This not only affects the life of the equipment, but also causes energy waste and unstable treatment effect, resulting in abnormally high costs and failing to meet the requirements of energy saving. Therefore, we have developed an intelligent aeration control system and method for sewage treatment plants. Summary of the Invention

[0009] (a) Technical problems to be solved

[0010] To address the shortcomings of existing technologies, this invention provides an intelligent aeration control system and method for wastewater treatment plants. The system scientifically divides the aeration tank into multiple functional sections, deploying various professional monitoring devices in each section to comprehensively collect multi-dimensional data such as flow rate, water quality, and dissolved oxygen. Through independent data collection and testing of each section, the operating status and treatment effect of the aeration equipment in each section can be accurately grasped. Based on this precise data, combined with an aeration efficiency evaluation model and control strategies, fine-tuning can be performed according to the actual needs of each section, avoiding resource waste or poor treatment effects caused by one-size-fits-all control. This achieves efficient and precise operation of the aeration equipment, has good application prospects, and effectively solves the problems raised in the background technology.

[0011] (II) Technical Solution

[0012] To achieve the above objectives, the present invention provides the following technical solution:

[0013] The intelligent aeration control system for wastewater treatment plants includes a data acquisition module, an aeration testing module, a demand forecasting module, a routine analysis module, and an aeration control module.

[0014] Data acquisition module: Collects real-time operating data of the primary sedimentation tank, test data of the aeration tank, historical operating data and real-time operating data of the aeration tank, and performs preprocessing.

[0015] Aeration test module: Extracts pre-treated aeration tank test data according to the segmentation standard of the aeration tank, analyzes the segmented aeration data, generates evaluation values, and compares the evaluation values ​​with the preset maintenance values;

[0016] If the assessed value is less than the maintenance value, no action is taken.

[0017] Conversely, delete the evaluation value, maintain the aeration equipment, and retest the aeration tank to obtain the evaluation value;

[0018] Demand forecasting module: Input the real-time operating data of the primary sedimentation tank and the aeration tank into the trained prediction model to predict the aeration demand data.

[0019] Routine Analysis Module: Analyzes historical operating data of aeration tanks, formulates aeration level indicators and corresponding control strategies, and uses evaluation values ​​to correct aeration demand data;

[0020] Aeration control module: Using the above set of corrected aeration demand data as a baseline, calculate the change value of the corrected aeration demand data, and compare the change value with the set standard range;

[0021] If the change value is within the standard range, the adjustment strategy corresponding to the aeration demand level corresponding to the corrected aeration demand data will be implemented.

[0022] If the change value is higher than the upper limit of the standard range, the control strategy of adjusting the aeration demand data to the next higher level of aeration demand will be implemented.

[0023] If the change value is lower than the lower limit of the standard range, maintain the current state and implement the abnormal monitoring strategy.

[0024] Furthermore, the real-time operating data of the primary sedimentation tank includes effluent indicators, instantaneous effluent flow rate, and sludge ratio. The historical and real-time operating data of the aeration tank both include dissolved oxygen, ammonia nitrogen concentration, sludge concentration, oxidation-reduction potential, effluent indicators, instantaneous effluent flow rate, and sludge ratio. The historical operating data of the aeration tank also includes actual oxygen demand data.

[0025] Furthermore, the collected data undergoes preprocessing, including cleaning and transformation, as detailed below:

[0026] Outlier handling: Outlier detection is performed using the Raida criterion to identify and remove data points that exceed the normal fluctuation range;

[0027] Missing value handling: Missing values ​​are estimated using data trends at adjacent time points, and then imputed using a time-series-based linear interpolation method.

[0028] Normalization: Normalize all data after outlier and missing value handling.

[0029] Furthermore, the aeration tank test involves operating each section of the aeration tank under stable operating conditions according to pre-set standard aeration parameters for a test duration of T. Test data is collected, and an evaluation value is calculated based on the test data. The formula for calculating the evaluation value is:

[0030] In the formula, Epg is the evaluated value, Ps is the power of the aeration equipment, and B is the pressure loss coefficient. This is the pressure loss data for the current aeration equipment. V represents the baseline pressure loss under initial conditions, and V is the water volume of each section of the aeration tank after segmentation. The increase in dissolved oxygen concentration in the water body over time T. Let A be the saturated dissolved oxygen concentration at the current wastewater temperature, and A be the water quality correction factor.

[0031] Furthermore, the prediction model employs an autoregressive integrated moving average model, and the training steps for the prediction model are as follows:

[0032] Data partitioning: The historical working data of each aeration tank section was divided into training and testing sets in a 7:3 ratio;

[0033] Stationarity processing: The unit root test is used to analyze the stationarity of the original oxygen demand time series. If the original oxygen demand time series does not meet the stationarity condition, it is transformed into a stationary series through difference operation, and the difference order is determined.

[0034] Model order determination: Plot the processed autocorrelation function and partial autocorrelation function. Combine the Akaike information criterion and Bayesian information criterion, and use the particle swarm optimization algorithm to determine the autoregression order and the moving average order to construct a preliminary model.

[0035] Parameter estimation: The parameters of the preliminary model are estimated using the training set data, and the maximum likelihood estimation method is used to solve for the model parameters;

[0036] Model testing: Input the test set data into the trained model for prediction, and use root mean square error and mean absolute error to evaluate the model's prediction accuracy.

[0037] Furthermore, the steps for analyzing historical operating data of the aeration tank and formulating aeration level indicators are as follows:

[0038] Data preparation: Retrieve the historical actual oxygen demand data for each aeration tank from the database for the most recent N times;

[0039] Statistical data: Calculate the mean and standard deviation of historical actual oxygen demand data for N periods;

[0040] Grading: Based on statistical principles and the calculated mean and standard deviation, the actual oxygen demand of each aeration tank section is divided into H levels;

[0041] Strategy formulation: Obtain the upper limit of the actual oxygen demand data in H levels, preset the parameters of the aeration equipment based on the upper limit, and summarize to generate control strategies, resulting in a total of H control strategies.

[0042] Furthermore, the steps for correcting the aeration demand data using the assessed values ​​are as follows:

[0043] The reduction rate of oxygen supply efficiency is calculated based on the assessed values, using the following formula: In the formula, G is the reduction ratio of oxygen supply efficiency, En is the energy consumption per unit oxygen supply under normal conditions, and Css is the pressure loss correction coefficient, 0 < Css < 1.

[0044] The aeration demand data is adjusted based on the percentage decrease in oxygen supply efficiency. The specific formula is as follows: In the formula, The adjusted aeration demand data, This is the predicted aeration demand data.

[0045] Furthermore, the formula for calculating the change in aeration demand is as follows: In the formula, This represents the change in aeration demand. This is the corrected, real-time predicted aeration demand data. This is the revised set of aeration requirement data from the previous set.

[0046] Furthermore, the abnormal monitoring strategy is to keep the baseline value unchanged, calculate the next set of change values, and if the change values ​​are lower than the lower limit of the standard range for three consecutive times, a detection warning is issued and feedback is awaited. If an abnormality is detected, maintenance is performed.

[0047] If the received feedback signal is correct, calculate the real-time corrected aeration demand data and execute the control strategy corresponding to the aeration demand level of the real-time corrected aeration demand data.

[0048] Furthermore, the intelligent aeration control method for wastewater treatment plants includes the following steps:

[0049] Collect real-time operating data of the primary sedimentation tank, test data of the aeration tank, historical operating data and real-time operating data of the aeration tank, and perform preprocessing.

[0050] According to the segmentation standard of the aeration tank, the test data of the pretreated aeration tank is extracted, the segmented aeration data is analyzed, the evaluation value is generated, and the evaluation value is compared with the preset maintenance value.

[0051] If the assessed value is less than the maintenance value, no action is taken.

[0052] Conversely, delete the evaluation value, maintain the aeration equipment, and retest the aeration tank to obtain the evaluation value;

[0053] Input the real-time operating data of the primary sedimentation tank and the aeration tank into the trained prediction model to predict the aeration demand data.

[0054] Analyze historical operating data of aeration tanks, formulate aeration level indicators and corresponding control strategies, and use evaluation values ​​to correct aeration demand data.

[0055] Using the above set of corrected aeration demand data as a baseline, calculate the change value of the corrected aeration demand data, and compare the change value with the set standard range;

[0056] If the change value is within the standard range, the adjustment strategy corresponding to the aeration demand level corresponding to the corrected aeration demand data will be implemented.

[0057] If the change value is higher than the upper limit of the standard range, the control strategy of adjusting the aeration demand data to the next higher level of aeration demand will be implemented.

[0058] If the change value is lower than the lower limit of the standard range, maintain the current state and implement the abnormal monitoring strategy.

[0059] (III) Beneficial Effects

[0060] This invention provides an intelligent aeration control system and method for wastewater treatment plants, which has the following beneficial effects:

[0061] 1. This invention describes an intelligent aeration control system and method for wastewater treatment plants. The aeration tank is scientifically divided into multiple functional sections, and specialized monitoring equipment is deployed in each section to collect multi-dimensional data such as flow rate, water quality, and dissolved oxygen. Through independent data collection and testing of each section, the operating status and treatment effect of the aeration equipment in each section can be accurately grasped. Based on this precise data, combined with an aeration efficiency evaluation model and control strategies, fine-tuning can be performed according to the actual needs of each section, avoiding resource waste or poor treatment effects caused by a one-size-fits-all approach. This achieves efficient and precise operation of the aeration equipment and has promising application prospects.

[0062] 2. This invention describes an intelligent aeration control system and method for wastewater treatment plants. Based on statistical principles, it analyzes the historical actual oxygen demand data of each aeration tank, calculates the mean and standard deviation, and divides it into multiple distinct demand levels. The system matches the predicted aeration demand to the corresponding level and adjusts according to preset rules. This level-based adjustment method significantly reduces the amount of data calculation compared to complex real-time calculations, improves system response speed, and effectively filters out short-term fluctuations and rapid changes in data through comprehensive consideration of historical data. This makes the aeration strategy adjustment more stable and reliable, ensuring the smooth operation of the aeration system and demonstrating promising application prospects.

[0063] 3. This invention describes an intelligent aeration control system and method for wastewater treatment plants. From data acquisition, testing of each section of the aeration tank, demand prediction, routine analysis to aeration control, a complete and systematic segmented operation management system for the aeration tank is formed. Through multi-dimensional data acquisition and scientific analysis, combined with accurate demand prediction and reasonable control strategies, the system achieves comprehensive perception, accurate assessment, and timely adjustment of the aeration tank's operating status, ensuring the efficient and stable operation of the aeration process in wastewater treatment. This not only improves wastewater treatment efficiency but also reduces operating costs, optimizes resource allocation, and provides strong support for the stable operation and sustainable development of wastewater treatment plants. It has good application results and promising prospects.

[0064] 4. This invention describes an intelligent aeration control system and method for wastewater treatment plants. It constructs a systematic architecture that includes data acquisition and processing, model prediction, hierarchical management, and intelligent control. It overcomes the shortcomings of traditional aeration control, such as single data, insufficient prediction, and extensive control. It achieves refined management by accurately predicting aeration demand and classifying aeration demand levels. It can flexibly adjust the system according to the degree of demand change, significantly improving the intelligence level and adaptability of the aeration system. It lays the foundation for the intelligent and digital development of the wastewater treatment field and has good application prospects. Attached Figure Description

[0065] Figure 1 This is a flowchart of the intelligent aeration control system for wastewater treatment plants according to the present invention.

[0066] Figure 2 This is a flowchart of the intelligent aeration control method for wastewater treatment plants according to the present invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and 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.

[0068] The core idea of ​​the solution described in the existing patent is to collect the ammonia nitrogen concentration signal of the membrane tank and the dissolved oxygen concentration signal of the aerobic tank, calculate the residuals respectively, and then obtain the preset dissolved oxygen concentration value and aeration rate value through a two-order PI algorithm. Then, combined with the blower performance curve, multiple blowers are controlled. It also has the functions of instrument signal verification, water volume compensation and automatic switching of control strategy in case of failure, which can improve the control accuracy, realize the automatic control of the aeration process, stabilize the effluent water quality and reduce the energy consumption of sewage treatment.

[0069] However, in practical use, this solution has the following limitations:

[0070] Centralized equipment management: In the process of aeration control, it adopts centralized collection of aeration tank data and analyzes the condition of aeration tank by combining the data collected periodically with existing technology, and centrally maintains aeration equipment. However, in actual use, the impact of sludge on aeration equipment varies, and it does not consider the situation of local blockage of aeration equipment. In actual use, due to the influence of sludge, some areas of aeration equipment are severely blocked, which greatly reduces the aeration effect in that part. This method cannot achieve stable oxygen supply and will greatly increase aeration energy consumption.

[0071] Too frequent adjustments: Existing aeration control usually adopts a real-time accurate prediction method, that is, real-time data collection, real-time analysis and real-time processing. When in use, the water flow will change due to environmental or other factors. At this time, the amount of data to be processed is large. Therefore, the equipment parameters are usually adjusted significantly and frequently during use, which not only affects the life of the equipment, but also causes energy waste and unstable treatment effect, resulting in abnormally high costs.

[0072] This invention forms a complete and systematic segmented operation and management system for aeration tanks, encompassing data acquisition, testing of each section of the aeration tank, demand prediction, routine analysis, and aeration control. It also constructs a novel technical framework for slope prestress monitoring and slope anomaly assessment. Its core lies not in the improvement of a single technology, but in overcoming the shortcomings of traditional aeration control, such as limited data, insufficient prediction, and crude regulation. Accurate prediction of aeration demand combined with the classification of aeration demand levels enables refined management, allowing for flexible adjustment based on the degree of demand change. This significantly improves the intelligence and adaptability of the aeration system, laying the foundation for the intelligent and digital development of the wastewater treatment field and leading the industry's technological upgrading and innovative development.

[0073] Example 1

[0074] Please see Figure 1 This embodiment provides an intelligent aeration control system for wastewater treatment plants, and the specific solution is as follows:

[0075] 1. System architecture;

[0076] The system architecture of the intelligent aeration control system for wastewater treatment plants consists of a data acquisition layer, a data processing layer, an analysis and prediction layer, a strategy decision-making layer, and an equipment control layer. Each layer works in concert to achieve efficient management and precise control of the aeration tank.

[0077] 1.1 Data Acquisition Layer;

[0078] The data acquisition layer is responsible for acquiring various key data during the operation of the aeration tank. The aeration tank is divided into multiple functional sections. Electromagnetic flow meters and multi-parameter online water quality analyzers are deployed at the inlet to collect real-time data on influent flow rate, chemical oxygen demand (COD), and other indicators. Dissolved oxygen sensors and automatic water quality samplers are installed at the outlets of each section to acquire dissolved oxygen concentration and effluent water quality data. Ultrasonic sludge concentration meters and other equipment are installed in the middle of the tank to monitor parameters such as sludge concentration and oxidation-reduction potential. Gas flow meters are installed on the air supply pipelines of the aeration equipment, and combined with the PLC control system, they collect operating parameters such as air supply and operating voltage of the aeration equipment, providing raw data support for subsequent treatment.

[0079] 1.2 Data Processing Layer;

[0080] The data processing layer receives the raw data transmitted from the data acquisition layer, processes the noise, missing values ​​and outliers in the data, uses the Laida criterion to identify and remove outliers, fills missing values ​​with linear interpolation, and uses normalization formulas to eliminate differences in data dimensions, thereby improving data quality and ensuring that the data input to subsequent layers is accurate and reliable, meeting the needs of analysis and modeling.

[0081] 1.3 Analysis and prediction layer;

[0082] Based on the preprocessed data, under stable system conditions, each aeration tank section was tested according to standard aeration parameters. By calculating the evaluation values ​​of the aeration equipment and comparing the actual oxygen supply with the standard oxygen supply, the operating status of the equipment was assessed, and it was determined whether there were any problems such as blockage, providing a basis for strategy formulation.

[0083] The ARIMA model was used to predict aeration demand and oxygen supply from equipment. The difference order was determined by data partitioning and stationarity testing. The autoregression order and moving average order were determined by combining the autocorrelation function and partial autocorrelation function plots with the AIC and BIC criteria. After training the model, future oxygen demand and oxygen supply were predicted to anticipate demand trends.

[0084] 1.4 Strategic Decision-Making Level;

[0085] Based on historical aeration demand and equipment oxygen supply data, the strategic decision-making level calculates the mean and standard deviation, classifies aeration demand levels, and dynamically updates the levels in conjunction with the evaluation values ​​of aeration equipment, providing a basis for a graded strategy for aeration control.

[0086] The system calculates the degree of change in aeration demand in real time and compares it with the preset standard value. Based on different situations, such as small changes in demand, small increases or decreases in demand, and other factors, it generates corresponding adjustment strategies for the operating parameters of the aeration equipment.

[0087] 1.5 Equipment Control Layer;

[0088] The system receives control strategies generated by the strategy decision layer and uses actuators such as frequency converters and PLC control systems to precisely adjust operating parameters of aeration equipment, such as blower air volume and sludge return ratio, to achieve optimized control of the aeration process, ensuring efficient and stable operation of the aeration tank. At the same time, it triggers equipment inspection and maintenance processes based on equipment blockage judgment results.

[0089] 2. Hardware basics;

[0090] 2.1 Data acquisition hardware;

[0091] Electromagnetic flow meter: Select an electromagnetic flow meter such as the E+H Promag series, install it at the inlet of the aeration tank, and connect it securely to the pipeline through a flange connection. This flow meter is used to monitor the inlet flow in real time and provide basic flow data for the system.

[0092] Multi-parameter online water quality analyzer: The Hach Amtax NA8000 model is selected and installed near the water inlet. It is fixed by a special bracket and connected to the data processing equipment via an RS485 communication cable. It can simultaneously collect water quality indicators such as chemical oxygen demand, five-day biochemical oxygen demand, suspended solids and total phosphorus, providing data support for analyzing the influent water quality.

[0093] Dissolved oxygen sensor: The InPro 6050i model is selected. It is submerged in water at the outlet of each aeration tank through a special mounting bracket to ensure that the probe is in full contact with the water body and realize the real-time acquisition and transmission of dissolved oxygen concentration data.

[0094] Automatic water sampler: The HACH 3540 model is selected and installed near the sampling points at the outlets of each section. It is connected to the sampling location through pipelines and connected to the data processing equipment via RS232 or RS485 communication lines to realize remote control of sampling and data transmission. Water samples are collected at regular intervals for laboratory testing of water quality indicators.

[0095] Ultrasonic sludge concentration meter: The E+H FMX167 model is selected and installed on the pipeline in the middle of the tank through a pipe clamp to ensure that the measuring probe is aligned with the center of the fluid. The signal is connected to the data processing equipment through a dedicated cable to obtain sludge concentration data in real time, reflecting the sludge condition in the middle of the tank.

[0096] Oxidation-reduction potential (ORP) meter: Model PHS-3E is selected. It is wall-mounted in a suitable position in the middle of the tank. The electrode is immersed in the water. The electrode signal transmission line is connected to the data processing equipment through a dedicated signal line. It is used to monitor the oxidation-reduction potential and help judge the reaction status in the aeration tank.

[0097] Sludge concentration monitor: The WTW Oxi3310 model is selected. The installation method is similar to that of the ultrasonic sludge concentration meter. After being fixed in the middle of the tank, it is connected to the data processing equipment through the matching cable to measure the sludge concentration and provide multi-dimensional sludge concentration data for process control.

[0098] Gas flow meter: AXF series gas flow meters are selected and installed on the air supply pipeline of the aeration equipment, using flange connections. Real-time air supply data is collected, and the actual oxygen supply is calculated based on the equipment performance curve.

[0099] PLC control system: The S7-300 series PLC is selected and installed in the control cabinet in the control room. It receives analog signals from sensors through various input and output modules, such as analog input modules, and controls the switching on and off of aeration equipment, frequency converters and other equipment operations through digital input and output modules.

[0100] Frequency converter: Installed in the electrical control cabinet near the aeration equipment, it is connected to the motor of the aeration equipment. By changing the frequency of the motor power supply, it adjusts the operating speed of the aeration equipment, thereby controlling the aeration volume. At the same time, it is connected to the PLC control system through a communication line to receive control commands issued by the PLC.

[0101] 2.2 Control hardware;

[0102] Industrial control computer: Located in the control room, it serves as the core of the entire system's data processing and management. It connects to the PLC control system and data acquisition equipment (devices supporting network communication) via Ethernet, enabling centralized data storage, analysis, processing, and control command issuance. Equipped with a high-performance processor, large-capacity memory, and hard drive, it meets the demands of large-scale data processing and system operation.

[0103] Data acquisition card: Installed in an industrial control computer, it is used to acquire analog and digital signals from various sensors and convert them into digital quantities that the computer can recognize. Depending on the type and number of sensors, a data acquisition card with an appropriate number of channels and accuracy is selected to ensure the accuracy and stability of data acquisition.

[0104] 2.3 Software:

[0105] Data acquisition and transmission software: Running on an industrial control computer, it is responsible for communicating with various data acquisition hardware devices, acquiring data at set time intervals, and transmitting the data to a database for storage. It supports multiple communication protocols, such as RS232, RS485, and Ethernet, and is compatible with different types of devices, ensuring the timeliness and accuracy of data acquisition.

[0106] Database management software: MySQL database management system is selected to store massive amounts of data such as water quality, flow rate, and equipment operating parameters. A reasonable data table structure is established to classify and store the data, which facilitates subsequent data query, analysis and mining. It also has data backup and recovery functions to ensure data security and integrity.

[0107] Data analysis and modeling software: These software programs analyze historical data stored in databases to build aeration demand prediction models (ARIMA models), aeration equipment performance evaluation models, and more. Through data mining and machine learning algorithms, patterns and trends in the data are discovered, providing data support and model basis for strategic decision-making.

[0108] Control strategy execution software: Working in conjunction with the PLC control system, it generates control commands based on data analysis results and preset control strategies, and sends them to the PLC. The PLC then executes the control operations on the aeration equipment, such as adjusting the blower airflow and controlling the sludge return ratio, thereby achieving automated control of the aeration process. Simultaneously, it monitors the equipment's operating status in real time and provides alarm prompts for equipment malfunctions.

[0109] 3. System solution;

[0110] like Figure 1 As shown, the intelligent aeration control system for wastewater treatment plants of the present invention includes a data acquisition module, an aeration testing module, a demand prediction module, a routine analysis module, and an aeration control module.

[0111] 3.1 Data Collection;

[0112] Data acquisition is the process of obtaining and analyzing the relevant data required for aeration control. It relies on a data acquisition module to collect and preprocess the data.

[0113] Data acquisition module: Collects real-time operating data of the primary sedimentation tank, test data of the aeration tank, historical operating data and real-time operating data of the aeration tank, and performs preprocessing.

[0114] The real-time operating data of the primary sedimentation tank includes effluent indicators, instantaneous effluent flow rate, and sludge ratio. The historical and real-time operating data of the aeration tank include dissolved oxygen, ammonia nitrogen concentration, sludge concentration, oxidation-reduction potential, effluent indicators, instantaneous effluent flow rate, and sludge ratio. The historical operating data of the aeration tank also includes actual oxygen demand data (real oxygen demand data).

[0115] Each section of the aeration tank has its own aeration equipment or aeration equipment that operates independently, without interfering with each other.

[0116] The test data for the aeration tank includes the power of the aeration equipment, the pressure loss data of the aeration equipment, and the water volume and dissolved oxygen concentration of each section of the aeration tank after it is divided into segments.

[0117] The collected data undergoes preprocessing, including cleaning and transformation, as detailed below:

[0118] Outlier handling: Outlier detection is performed using the Raida criterion. The mean and standard deviation of the data are calculated, and data points that exceed the normal fluctuation range are identified and removed.

[0119] For example, the aeration tank is divided into 5 sections. The chemical oxygen demand (COD) of the first aeration tank at a certain time period is 80, 85, 200, 90 and 95. The calculated mean is 110 and the standard deviation is 49.5. All data are within the range and there are no outliers.

[0120] Missing value handling: Missing values ​​are estimated using data trends at adjacent time points, and then imputed using a time-series-based linear interpolation method.

[0121] For example, the data sequence of dissolved oxygen content in one aeration tank is 2.2, 2.4, empty, 2.8, 3, 0. The missing value is calculated to be (2.4+2.8) / 2=2.6mg / L by linear interpolation.

[0122] Normalization: Normalization is performed on all data after outlier and missing value handling. Normalization is a common technique in existing data processing and is common knowledge, so it will not be described in detail.

[0123] This invention describes an intelligent aeration control system and method for wastewater treatment plants. The system scientifically divides the aeration tank into multiple functional sections, deploying various professional monitoring devices in each section to comprehensively collect multi-dimensional data such as flow rate, water quality, and dissolved oxygen. Through independent data collection and testing of each section, the system can accurately grasp the operating status and treatment effect of the aeration equipment in each section. Based on this precise data, combined with an aeration efficiency evaluation model and control strategies, the system can make refined adjustments according to the actual needs of each section, avoiding resource waste or poor treatment results caused by a one-size-fits-all approach. This achieves efficient and precise operation of the aeration equipment and has promising application prospects.

[0124] 3.2 Aeration Analysis;

[0125] To achieve precise control of aeration, it is necessary to understand the condition of the aeration equipment. Therefore, it is necessary to analyze the aeration equipment accordingly, and this step is based on the aeration test module.

[0126] The aeration test module extracts the pre-treated aeration tank test data according to the segmentation standard of the aeration tank, analyzes the segmented aeration data, generates evaluation values, and compares the evaluation values ​​with the preset maintenance values.

[0127] If the assessed value is less than the maintenance value, no action is taken.

[0128] Conversely, delete the evaluation value, maintain the aeration equipment, and retest the aeration tank to obtain the evaluation value;

[0129] The aeration tank test involves operating each section of the aeration tank under stable operating conditions according to pre-set standard aeration parameters. The test duration is T, and test data is collected. An evaluation value is calculated based on the test data. The formula for calculating the evaluation value is:

[0130] In the formula, Epg is the evaluated value, Ps is the power of the aeration equipment, and B is the pressure loss coefficient. This is the pressure loss data for the current aeration equipment. V represents the baseline pressure loss under initial conditions, and V is the water volume of each section of the aeration tank after segmentation. The increase in dissolved oxygen concentration in the water body over time time T. Here, A represents the saturated dissolved oxygen concentration at the current wastewater temperature, and A is the water quality correction factor. A < 1, and is normally taken as 0.7-0.9.

[0131] For example: The aeration equipment uses a blower. Under normal conditions, the voltage is 380V and the current is 25A, with a calculated input power of 13.9. The actual oxygen supply increases from 2 mg / L to 8 mg / L in 15 minutes. Each section of the aeration tank has a volume of 500 m³. 3 The pressure loss is 4.2 kPa, while the normal estimated value is 2.5 kW·h.

[0132] After a period of use, the current status monitoring showed that the voltage remained unchanged, the power meter displayed a power of 15.6, the actual oxygen supply increased from 2 mg / L to 6 mg / L in 15 minutes, the pressure loss was 6.5 kPa, and the calculated evaluation value was 2.78 kW·h.

[0133] At this point, if the set maintenance value is 2.75 kWh, it does not meet the requirements; if the set maintenance value is 2.85 kWh, it meets the requirements. The maintenance value is set based on the energy consumption requirements of the wastewater treatment plant.

[0134] 3.3 Demand Forecasting;

[0135] Demand forecasting involves processing and analyzing existing and historical data to predict the current aeration demand. This step is based on the demand forecasting module.

[0136] Demand forecasting module: Input the real-time operating data of the primary sedimentation tank and the aeration tank into the trained prediction model to predict the aeration demand data.

[0137] The prediction model uses an autoregressive integrated moving average (ARIMA) model, and the training steps for the prediction model are as follows:

[0138] Data partitioning: The historical working data of each aeration tank section was divided into training and testing sets in a 7:3 ratio;

[0139] For example, a certain aeration tank has 100 historical oxygen demand data points. The first 70 data points are used as the training set, and the last 30 data points are used as the test set.

[0140] Historical data is the most recent data, and the recent data environment is the same, so it is of reference value. In addition, it is necessary to use data for the same wastewater treatment. For new wastewater treatment, this solution is not applicable if there is insufficient data.

[0141] Stationarity processing: The unit root test (ADF test) was used to analyze the stationarity of the original oxygen demand time series. If the original oxygen demand time series did not meet the stationarity conditions, it was transformed into a stationary series through differencing and the difference order was determined.

[0142] For example, the original oxygen demand sequence of an aeration tank, after plotting the sequence and calculating the statistic, shows a clear upward trend, initially indicating a non-stationary sequence. An ADF test is performed, yielding a test statistic of -1.5. Compared to the critical values ​​at different significance levels (e.g., -3.5 at the 1% significance level), this statistic is greater than the critical value, so the null hypothesis cannot be rejected, meaning the sequence is non-stationary. After first differencing, another ADF test is performed, and the test statistic becomes -3.8, which is less than the critical value at the 1% significance level. At this point, the sequence can be considered to have transformed into a stationary sequence, thus determining d=1.

[0143] Model order determination: Plot the processed autocorrelation function and partial autocorrelation function. Combine the Akaike information criterion and Bayesian information criterion, and use the particle swarm optimization algorithm to determine the autoregression order and the moving average order to construct a preliminary model.

[0144] The autocorrelation function (ACF) and partial autocorrelation function (PACF) plots show the degree of autocorrelation of a series at different lag orders. The autocorrelation function plot, after removing the influence of intermediate lag terms, shows the degree of partial autocorrelation of the series at different lag orders. By observing the truncation and tailing characteristics of these two plots, the range of values ​​for p and q can be preliminarily determined.

[0145] The Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) are metrics used to evaluate the goodness of fit and complexity of a model. The smaller the value, the better the model fit and the lower the complexity. When determining p and q, different combinations of values ​​are tried, the corresponding AIC and BIC values ​​are calculated, and the combination that minimizes the AIC and BIC values ​​is selected as the final p and q values.

[0146] For example, after plotting the ACF and PACF diagrams of oxygen supply data from a certain aeration tank, it was found that the ACF diagram rapidly approached 0 after a lag of 2, exhibiting a truncated characteristic; the PACF diagram approached 0 after a lag of 3, also exhibiting a truncated characteristic. Then, starting from p=0, q=0, different combinations of p and q were gradually tried to calculate the AIC and BIC values. When p=3, q=2, both the AIC and BIC values ​​reached their minimum, so p=3 and q=2 were finally determined.

[0147] Parameter estimation: The parameters of the preliminary model are estimated using the training set data, and the maximum likelihood estimation method is used to solve for the model parameters;

[0148] The basic idea of ​​maximum likelihood estimation is to find a set of parameter values ​​that maximize the probability of the training data, thereby enabling the model to best fit the features and patterns of the training data.

[0149] For example, using 70 training data points of oxygen demand in the second aeration tank, the parameters were estimated using the maximum likelihood estimation method. After iterative calculation, the autoregressive coefficients and moving average coefficients were finally obtained, thus determining the ARIMA model for predicting the oxygen demand in this aeration tank.

[0150] The calculation formula for the ARIMA model is as follows:

[0151]

[0152] In the formula: To perform d-order difference operations on a sequence, Let be the time series of oxygen demand or oxygen supply for the i-th segment of the aeration tank at time t. These are autoregressive coefficients, used to measure the weight of the influence of the series' historical values ​​on the current value. Let p be the time series of oxygen demand or oxygen supply in the i-th aeration tank at time tj, where p is the autoregressive order and q is the moving average order. The moving average coefficient is... Let i be the white noise sequence of the i-th segment of the aeration tank at time t. Let be the white noise sequence of the i-th aeration tank at time to.

[0153] Model testing: Input the test set data into the trained model for prediction, and use root mean square error and mean absolute error to evaluate the model's prediction accuracy.

[0154] For example, in a certain aeration tank oxygen demand prediction model, the test set contains 30 data points. The 30 actual oxygen demand data points in the test set and the oxygen demand data predicted by the model are substituted into the root mean square error and mean absolute error formulas for calculation. First, the error at each time point is calculated, and then the root mean square error and mean absolute error are calculated separately according to the root mean square error and mean absolute error formulas. The calculated root mean square error and mean absolute error are 4 and 2.5, respectively. If they are preset to 5 and 3, the model meets the threshold requirements.

[0155] This invention describes an intelligent aeration control system and method for wastewater treatment plants. Based on statistical principles, it analyzes the historical actual oxygen demand data of each aeration tank, calculates the mean and standard deviation, and divides it into multiple distinct demand levels. The system matches the predicted aeration demand to the corresponding level and adjusts according to preset rules. This level-based adjustment method significantly reduces the amount of data computation compared to complex real-time calculations, improves system response speed, and effectively filters out short-term fluctuations and rapid changes in data through comprehensive consideration of historical data. This makes the aeration strategy adjustment more stable and reliable, ensuring the smooth operation of the aeration system and demonstrating promising application prospects.

[0156] 3.4 Detailed Analysis;

[0157] Analyzing historical data, summarizing and generalizing the historical data, and formulating appropriate control strategies based on the historical data are steps based on the normal analysis module.

[0158] Routine Analysis Module: Analyzes historical operating data of aeration tanks, formulates aeration level indicators and corresponding control strategies, and uses evaluation values ​​to correct aeration demand data;

[0159] The steps for analyzing historical operating data of the aeration tank and formulating aeration level indicators are as follows:

[0160] Data preparation: Retrieve the historical actual oxygen demand data for each aeration tank from the database for the most recent N times;

[0161] The most recent N historical actual oxygen demand data need to be reviewed and evaluated to ensure they are accurate.

[0162] Statistical data: Calculate the mean and standard deviation of historical actual oxygen demand data for N periods;

[0163] The formulas for calculating the mean and standard deviation are commonly used techniques, so they will not be described in detail.

[0164] Grading: Based on statistical principles and the calculated mean and standard deviation, the actual oxygen demand of each aeration tank is divided into H levels to clarify the current operating status of the aeration tank and provide a reference for subsequent regulation;

[0165] For example, 2 standard deviations can be used as the grade difference value, that is, the difference between the upper limit and the lower limit of the grade is 2 standard deviations. Combined with the mean data, the historical actual oxygen demand data corresponding to H grades can be obtained.

[0166] Strategy formulation: Obtain the upper limit of the actual oxygen demand data in H levels, preset the parameters of the aeration equipment based on the upper limit, and summarize to generate control strategies, resulting in a total of H control strategies.

[0167] For example, the upper limit of the second level is Bmax and the lower limit is Bmin. The parameters of the aeration equipment are preset based on the actual oxygen demand data Bmax. The corresponding settings of the historical scheme can be directly found, including setting the aeration volume, aeration time, blower speed, etc.

[0168] Setting the upper limit ensures stable aeration performance.

[0169] The steps for correcting aeration demand data using assessed values ​​are as follows:

[0170] The reduction rate of oxygen supply efficiency is calculated based on the assessed values, using the following formula: In the formula, G is the reduction ratio of oxygen supply efficiency, En is the energy consumption per unit oxygen supply under normal conditions, and Css is the pressure loss correction coefficient, 0 < Css < 1.

[0171] For example, under normal circumstances, the energy consumption per unit of oxygen supply is 2.5 kWh, while under the current circumstances, the energy consumption per unit of oxygen supply is 3 kWh, and the calculated reduction in oxygen supply efficiency is 16.7%.

[0172] The aeration demand data is adjusted based on the percentage decrease in oxygen supply efficiency. The specific formula is as follows: In the formula, The adjusted aeration demand data, This is the predicted aeration demand data.

[0173] For example, the calculated aeration demand data here is 116.7% of the predicted aeration demand data.

[0174] This invention describes an intelligent aeration control system and method for wastewater treatment plants. From data acquisition, testing of each section of the aeration tank, demand prediction, routine analysis to aeration control, it forms a complete and systematic segmented operation management system for the aeration tank. Through multi-dimensional data acquisition and scientific analysis, combined with accurate demand prediction and reasonable control strategies, it achieves comprehensive perception, accurate assessment, and timely adjustment of the aeration tank's operating status, ensuring the efficient and stable operation of the aeration process in wastewater treatment. This not only improves wastewater treatment efficiency but also reduces operating costs, optimizes resource allocation, and provides strong support for the stable operation and sustainable development of wastewater treatment plants. It has good application results and promising prospects.

[0175] 3.5 Aeration control;

[0176] In use, data needs to be collected, compared, and adjusted in real time to achieve precise aeration control, and this process is based on the aeration control module.

[0177] Aeration control module: Using the above set of corrected aeration demand data as a baseline, calculate the change value of the corrected aeration demand data, and compare the change value with the set standard range;

[0178] The formula for calculating the change in aeration demand is: In the formula, This represents the change in aeration demand. This is the corrected, real-time predicted aeration demand data. This is the revised set of aeration requirement data from the previous set.

[0179] If the change value is within the standard range, the adjustment strategy corresponding to the aeration demand level corresponding to the corrected aeration demand data will be implemented.

[0180] For example, the previous set of aeration demand data was 25kg, and the real-time predicted aeration demand data after correction was 30kg, corresponding to an aeration demand level of 3. The calculated aeration demand change value was 0.2. If the standard range is (-0.1, 0.1), then a level 4 control strategy will be implemented.

[0181] If the change value is higher than the upper limit of the standard range, the control strategy of adjusting the aeration demand data to the next higher level of aeration demand will be implemented.

[0182] For example, the previous set of aeration demand data was 25kg, and the real-time predicted aeration demand data after correction was 27kg, corresponding to an aeration demand level of 3. The calculated aeration demand change value was 0.08. If the standard range is (-0.1, 0.1), then the level 3 control strategy will be implemented.

[0183] If the change value is lower than the lower limit of the standard range, maintain the current state and implement the abnormal monitoring strategy.

[0184] For example, if the previous set of aeration demand data was 25 kg, and the revised real-time predicted aeration demand data is 20 kg, corresponding to an aeration demand level of 3, the calculated aeration demand change value is -0.2. If the standard range is (-0.1, 0.1), the abnormal monitoring strategy is to keep the baseline value of 25 kg unchanged and calculate the next set of change values. If the change value is lower than the lower limit of the standard range for three consecutive times (3 collections and 3 tests, i.e., after 3 testing cycles), a detection warning is issued (usually an abnormality, such as reduced sewage volume or damaged detection instrument). Feedback is awaited (manual rapid detection and analysis to determine if it is a reduction in sewage volume). If an abnormality is detected, maintenance is performed (if it is a reduction in sewage volume, parameters need to be adjusted; if it is a damaged detection instrument, testing and maintenance are required).

[0185] If the received feedback signal is correct, calculate the real-time corrected aeration demand data and execute the control strategy corresponding to the aeration demand level of the real-time corrected aeration demand data.

[0186] This method primarily provides a range, within which no adjustments are made, thus avoiding frequent equipment adjustments. Additionally, it allows for advance adjustments when aeration demand data increases sharply to ensure effective aeration, and timely feedback when aeration demand data decreases sharply to ensure stable aeration by the aeration equipment.

[0187] This invention describes an intelligent aeration control system and method for wastewater treatment plants. It constructs a systematic architecture encompassing data acquisition and processing, model prediction, hierarchical management, and intelligent control. This system overcomes the shortcomings of traditional aeration control, such as limited data, insufficient prediction, and crude control. It achieves refined management by accurately predicting aeration demand and classifying aeration adjustment levels, allowing for flexible control based on changes in demand. This significantly improves the intelligence level and adaptability of the aeration system, laying the foundation for the intelligent and digital development of wastewater treatment and demonstrating promising application prospects.

[0188] Example 2

[0189] Based on Example 1, such as Figure 2 As shown, the intelligent aeration control method for wastewater treatment plants includes the following steps:

[0190] Collect real-time operating data of the primary sedimentation tank, test data of the aeration tank, historical operating data and real-time operating data of the aeration tank, and perform preprocessing.

[0191] According to the segmentation standard of the aeration tank, the test data of the pretreated aeration tank is extracted, the segmented aeration data is analyzed, the evaluation value is generated, and the evaluation value is compared with the preset maintenance value.

[0192] If the assessed value is less than the maintenance value, no action is taken.

[0193] Conversely, delete the evaluation value, maintain the aeration equipment, and retest the aeration tank to obtain the evaluation value;

[0194] Input the real-time operating data of the primary sedimentation tank and the aeration tank into the trained prediction model to predict the aeration demand data.

[0195] Analyze historical operating data of aeration tanks, formulate aeration level indicators and corresponding control strategies, and use evaluation values ​​to correct aeration demand data.

[0196] Using the above set of corrected aeration demand data as a baseline, calculate the change value of the corrected aeration demand data, and compare the change value with the set standard range;

[0197] If the change value is within the standard range, the adjustment strategy corresponding to the aeration demand level corresponding to the corrected aeration demand data will be implemented.

[0198] If the change value is higher than the upper limit of the standard range, the control strategy of adjusting the aeration demand data to the next higher level of aeration demand will be implemented.

[0199] If the change value is lower than the lower limit of the standard range, maintain the current state and implement the abnormal monitoring strategy.

[0200] In this application, the various formulas mentioned are all calculated by removing dimensions and taking their numerical values. The formulas are established by collecting a large amount of data and simulating the most recent real situation. Some coefficients or weights in the formulas are set by those skilled in the art according to the actual situation, so they will not be elaborated here.

[0201] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0202] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0203] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. An intelligent aeration control system for wastewater treatment plants, characterized in that: include: Data acquisition module: Collects real-time operating data of the primary sedimentation tank, test data of the aeration tank, historical operating data and real-time operating data of the aeration tank, and performs preprocessing. Aeration test module: Extracts pre-treated aeration tank test data according to the segmentation standard of the aeration tank, analyzes the segmented aeration data, generates evaluation values, and compares the evaluation values ​​with the preset maintenance values; If the assessed value is less than the maintenance value, no action is taken. Conversely, delete the evaluation value, maintain the aeration equipment, and retest the aeration tank to obtain the evaluation value; Demand forecasting module: Input the real-time operating data of the primary sedimentation tank and the aeration tank into the trained prediction model to predict the aeration demand data. Routine Analysis Module: Analyzes historical operating data of aeration tanks, formulates aeration level indicators and corresponding control strategies, and uses evaluation values ​​to correct aeration demand data; Aeration control module: Using the above set of corrected aeration demand data as a baseline, calculate the change value of the corrected aeration demand data, and compare the change value with the set standard range; If the change value is within the standard range, the adjustment strategy corresponding to the aeration demand level corresponding to the corrected aeration demand data will be implemented. If the change value is higher than the upper limit of the standard range, the control strategy of adjusting the aeration demand data to the next higher level of aeration demand will be implemented. If the change value is lower than the lower limit of the standard range, maintain the current state and implement the abnormal monitoring strategy; The aeration tank test involves operating each section of the aeration tank under stable operating conditions according to pre-set standard aeration parameters. The test duration is T, and test data is collected. An evaluation value is calculated based on the test data. The formula for calculating the evaluation value is: In the formula, Epg is the evaluated value, Ps is the power of the aeration equipment, and B is the pressure loss coefficient. This is the pressure loss data for the current aeration equipment. V represents the baseline pressure loss under initial conditions, and V is the water volume of each section of the aeration tank after segmentation. The increase in dissolved oxygen concentration in the water body over time time T. Let A be the saturated dissolved oxygen concentration at the current wastewater temperature, and A be the water quality correction factor.

2. The intelligent aeration control system for wastewater treatment plants according to claim 1, characterized in that: The real-time operating data of the primary sedimentation tank includes effluent indicators, instantaneous effluent flow rate, and sludge ratio. The historical and real-time operating data of the aeration tank both include dissolved oxygen, ammonia nitrogen concentration, sludge concentration, oxidation-reduction potential, effluent indicators, instantaneous effluent flow rate, and sludge ratio. The historical operating data of the aeration tank also includes actual oxygen demand data.

3. The intelligent aeration control system for wastewater treatment plants according to claim 2, characterized in that: The collected data undergoes preprocessing, including cleaning and transformation, as detailed below: Outlier handling: Outlier detection is performed using the Raida criterion to identify and remove data points that exceed the normal fluctuation range; Missing value handling: Missing values ​​are estimated using data trends at adjacent time points, and then imputed using a time-series-based linear interpolation method. Normalization: Normalize all data after outlier and missing value handling.

4. The intelligent aeration control system for wastewater treatment plants according to claim 3, characterized in that: The prediction model uses an autoregressive integrated moving average model, and the training steps for the prediction model are as follows: Data partitioning: The historical working data of each aeration tank section was divided into training and testing sets in a 7:3 ratio; Stationarity processing: The unit root test is used to analyze the stationarity of the original oxygen demand time series. If the original oxygen demand time series does not meet the stationarity condition, it is transformed into a stationary series through difference operation, and the difference order is determined. Model order determination: Plot the processed autocorrelation function and partial autocorrelation function. Combine the Akaike information criterion and Bayesian information criterion, and use the particle swarm optimization algorithm to determine the autoregression order and the moving average order to construct a preliminary model. Parameter estimation: The parameters of the preliminary model are estimated using the training set data, and the maximum likelihood estimation method is used to solve for the model parameters; Model testing: Input the test set data into the trained model for prediction, and use root mean square error and mean absolute error to evaluate the model's prediction accuracy.

5. The intelligent aeration control system for wastewater treatment plants according to claim 4, characterized in that: The steps for analyzing historical operating data of the aeration tank and formulating aeration level indicators are as follows: Data preparation: Retrieve the historical actual oxygen demand data for each aeration tank from the database for the most recent N times; Statistical data: Calculate the mean and standard deviation of historical actual oxygen demand data for N periods; Grading: Based on statistical principles and the calculated mean and standard deviation, the actual oxygen demand of each aeration tank section is divided into H levels; Strategy formulation: Obtain the upper limit of the actual oxygen demand data in H levels, preset the parameters of the aeration equipment based on the upper limit, and summarize to generate control strategies, resulting in a total of H control strategies.

6. The intelligent aeration control system for wastewater treatment plants according to claim 5, characterized in that: The steps for correcting aeration demand data using assessed values ​​are as follows: The reduction rate of oxygen supply efficiency is calculated based on the assessed values, using the following formula: In the formula, G is the reduction ratio of oxygen supply efficiency, En is the energy consumption per unit oxygen supply under normal conditions, and Css is the pressure loss correction coefficient, 0 < Css < 1. The aeration demand data is adjusted based on the percentage decrease in oxygen supply efficiency. The specific formula is as follows: In the formula, The adjusted aeration demand data, This is the predicted aeration demand data.

7. The intelligent aeration control system for wastewater treatment plants according to claim 6, characterized in that: The formula for calculating the change in aeration demand is: In the formula, This represents the change in aeration demand. This is the corrected, real-time predicted aeration demand data. This is the revised set of aeration requirement data from the previous set.

8. The intelligent aeration control system for wastewater treatment plants according to claim 7, characterized in that: The abnormal monitoring strategy is to keep the baseline value unchanged, calculate the next set of change values, and if the change values ​​are lower than the lower limit of the standard range for three consecutive times, a detection warning is issued and feedback is awaited. If the feedback is abnormal, maintenance is performed. If the received feedback signal is correct, calculate the real-time corrected aeration demand data and execute the control strategy corresponding to the aeration demand level of the real-time corrected aeration demand data.

9. A method for intelligent aeration control in a wastewater treatment plant, using the system described in any one of claims 1 to 8, characterized in that: Includes the following steps: Collect real-time operating data of the primary sedimentation tank, test data of the aeration tank, historical operating data and real-time operating data of the aeration tank, and perform preprocessing. According to the segmentation standard of the aeration tank, the test data of the pretreated aeration tank is extracted, the segmented aeration data is analyzed, the evaluation value is generated, and the evaluation value is compared with the preset maintenance value. If the assessed value is less than the maintenance value, no action is taken. Conversely, delete the evaluation value, maintain the aeration equipment, and retest the aeration tank to obtain the evaluation value; Input the real-time operating data of the primary sedimentation tank and the aeration tank into the trained prediction model to predict the aeration demand data. Analyze historical operating data of aeration tanks, formulate aeration level indicators and corresponding control strategies, and use evaluation values ​​to correct aeration demand data. Using the above set of corrected aeration demand data as a baseline, calculate the change value of the corrected aeration demand data, and compare the change value with the set standard range; If the change value is within the standard range, the adjustment strategy corresponding to the aeration demand level corresponding to the corrected aeration demand data will be implemented. If the change value is higher than the upper limit of the standard range, the control strategy of adjusting the aeration demand data to the next higher level of aeration demand will be implemented. If the change value is lower than the lower limit of the standard range, maintain the current state and implement the abnormal monitoring strategy.

Images