An intelligent monitoring device for sewage pumping station

Abstract

The present invention discloses an intelligent monitoring device for a sewage pumping station, and relates to the fields of municipal engineering and intelligent control technology. The control system of the existing sewage pumping station mainly relies on fixed control strategies and manual operations, which is difficult to adapt to the complex and changeable operating environment, resulting in high energy consumption, increased equipment wear and reduced system reliability. With the development of artificial intelligence technology, it is possible to improve the operation and management level of sewage pumping stations by introducing intelligent control methods. The present invention proposes an efficient and intelligent sewage pumping station control method by combining hydraulic models and artificial intelligence models, which can significantly improve the operating efficiency and energy-saving effect of the pumping station, and has broad application prospects. Improve the operating efficiency and energy-saving effect of sewage pumping stations. Enhance the adaptability and robustness of the system, and be able to cope with complex and changeable operating environments. Reduce equipment wear and maintenance costs, and extend the service life of equipment. Improve the reliability and safety of sewage lifting.

Description

Technical Field

[0001] The invention relates to the field of municipal engineering and intelligent control technology, and in particular to an intelligent monitoring device for a sewage pump station. Background Art

[0002] The intelligent monitoring system of sewage pumping stations is a system that integrates advanced technologies and sensors to monitor, manage and optimize the operation of sewage pumping stations in real time.

[0003] In this regard, the patent document with publication number CN216310584U discloses an intelligent remote joint control system for sewage pumping stations, which involves the fields of intelligent control, industrial automation control, wireless communication technology, etc. It mainly includes a programmable controller, an ultrasonic liquid level sensor connected to the PLC, a submersible pump connected to the PLC, a touch screen connected to the PLC, a slave station wireless transmitting and receiving device connected to the PLC, an industrial computer, a master station wireless transmitting and receiving device connected to the industrial computer, and an audible and visual alarm connected to the industrial computer. The beneficial effect of the utility model is that the monitoring and management of the existing pumping station can be transformed through the utility model, and the operation of the sewage pumping station can be controlled by an advanced remote monitoring system, and the safe and stable operation of the pumping station can be ensured with the least human resources and the most effective management mode. The control performance that cannot be achieved by the traditional relay control system is achieved by combining the characteristics of PLC economy, flexibility, and reliability with the characteristics of remote transmission and remote control of wireless transmitting and receiving equipment.

[0004] In this regard, the patent document with publication number CN108803510A discloses an automatic monitoring system for a pump station, including several local control layers set up at the pump station site and a central control layer set up in the central control room. The central control room monitors and remotely controls the local control layer of the pump station in real time through the central control layer. The local control layer includes a sewage treatment module, a rainwater treatment module and an alarm module. It solves the problems of unattended and remote monitoring in pump station management, enables the municipal management department to monitor, control, manage and dispatch the rainwater and sewage pump stations within its jurisdiction in real time, and improves the management department's ability to respond quickly and deal with various environments of the pump station. In line with the trend of the Internet of Things era, it realizes centralized monitoring of multiple devices on the same platform, improves the level of automation, and reduces the maintenance and management intensity of staff; it can also reduce the safety hazards of urban waterlogging and facilitate remote control and early warning of faults.

[0005] The control systems of existing sewage pumping stations mainly rely on fixed control strategies and manual operations, which are difficult to adapt to complex and changing operating environments, resulting in high energy consumption, increased equipment wear and reduced system reliability. With the development of artificial intelligence technology, it is possible to improve the operation and management level of sewage pumping stations by introducing intelligent control methods.

[0006] In view of the above problems, an intelligent monitoring device for a sewage pumping station is proposed. Summary of the invention

[0007] The object of the present invention is to provide an intelligent monitoring device for a sewage pumping station, which solves the problem of high power consumption and poor monitoring in the background technology.

[0008] To achieve the above object, the present invention provides the following technical solution: an intelligent monitoring device for a sewage pumping station, comprising:

[0009] Data acquisition module, install sensors at key nodes of sewage pumps to collect water level, flow, pressure and pump power in real time;

[0010] The data preprocessing module cleans the data output from the sensor and unifies the data for accuracy and consistency;

[0011] Hydraulic model building module,

[0012] The hydraulic model building module includes: building a one-dimensional model, simulating the longitudinal profile of pipeline flow, collecting basic data related to the pipeline, determining the area and hydraulic radius of the section; setting the boundary conditions from the initial section to the end section of the pipeline; parameter sensitivity analysis, evaluating the impact of parameters on model output;

[0013] The hydraulic model building module also includes: building a three-dimensional model, building the model based on polygonal modeling, setting appropriate lighting effects to further enhance the realism of the model, adding animation to the static model through model interaction, testing the model in a virtual environment, and making adjustments and optimizations based on the test results;

[0014] The AI ​​model training module collects historical operation data and data output by the hydraulic model, processes missing values, outliers and standardized data, determines key features for model training, including water level, flow, pressure and pump power, divides the data into training set, validation set and test set, with a ratio of 70% training, 15% validation and 15% test, selects LSTM neural network, determines the hyperparameters of the model, uses the training set for model training, uses the optimization algorithm to minimize the loss function, and monitors the loss and accuracy of the training process;

[0015] LSTM neural network, the formula is as follows:

[0016] f t =σ(w f ·[h t-1 ,x t ]+b f );

[0017] enter:

[0018] i t =σ(w i ·[h t-1,x t ]+b i );

[0019] renew:

[0020]

[0021] Output:

[0022] o t =σ(w o ·[h t-1 ,x t ]+b o );h t =o t tanh(C t );

[0023] Where tan represents the activation function; w and b are the weight and bias matrices;

[0024] Use the validation set to evaluate the model performance, improve the generalization ability of the model by adjusting the hyperparameters, adjust the model architecture and hyperparameters based on the validation results, repeat the training and validation steps, and perform a final evaluation on the test set;

[0025] Hybrid model integration module: The hydraulic model and the artificial intelligence model are integrated to form a hybrid model. The hybrid model conducts comprehensive analysis and prediction of the operating status of the pumping station at different time scales and spatial scales;

[0026] Control algorithm, real-time monitoring data, designing algorithms for dynamically adjusting pump station operating parameters, including pump start and stop, frequency conversion regulation, and unit quantity regulation;

[0027] The sewage intelligent control module combines the intelligent control system with cloud computing, processes and analyzes data in real time, deploys the developed intelligent control system in the actual sewage pumping station, configures the on-site network connection, and ensures stable communication between sensors, control equipment and cloud servers;

[0028] The system testing and maintenance module conducts comprehensive testing on the deployed system to ensure that its functions and performance meet the expected requirements.

[0029] Preferably, based on the physical structure of the sewage pumping station and the principles of fluid mechanics, a hydraulic model describing the flow characteristics of sewage is constructed, and the hydraulic model and the artificial intelligence model are verified and optimized respectively to ensure the accuracy and reliability of the individual models. The data output by the hydraulic model is combined with the real-time monitoring data as the input of the artificial intelligence model, and a hybrid model is used for comprehensive analysis and prediction to analyze the operating status prediction at different time scales and spatial scales. The time scale includes short-term prediction and long-term prediction, and the spatial scale includes flow and water level at different locations.

[0030] Preferably, the prediction results of the hybrid model are verified, and the integration strategy and model parameters are adjusted;

[0031] Weighted average ensemble method:

[0032] Calculation formula: in and They are the prediction results of the hydraulic model and the artificial intelligence model respectively; w1 and w2 are weight coefficients, satisfying w1+w2=1.

[0033] Preferably, the field data is integrated into the cloud computing platform to achieve real-time data transmission, and deep learning algorithms are used for time series prediction to predict the operating status of the pumping station and possible problems;

[0034] Use PID control algorithm for real-time regulation to ensure the pump station operates in the optimal state;

[0035] PID control algorithm:

[0036]

[0037] Where e(t) is the error, K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient.

[0038] Preferably, the reinforcement learning algorithm adaptively adjusts the control strategy:

[0039] Define the state space S: including the liquid level, flow rate, and pressure of the pump station;

[0040] Define action space A: pump start and stop, valve switch;

[0041] Set the reward function R(S,A) to encourage the system to maintain the best motion state:

[0042] Formula: R(S,A) = -(deviation cost + energy cost); Use the updated Q value function to select the strategy:

[0043] Update Q value function:

[0044]

[0045] in, is the learning rate and γ is the discount factor.

[0046] Preferably, a neural network is used to approximate the Q-value function:

[0047]

[0048] Select an action based on the current state S During strategy optimization, data is continuously collected and the Q-value function or strategy network is optimized to adapt to different working conditions.

[0049] Preferably, an isolation forest anomaly detection algorithm is used to monitor data anomalies, set up an alarm mechanism, develop a visual interface on the cloud platform, monitor the operating status of the pump station in real time, generate reports and analysis charts, and assist in decision-making and optimization.

[0050] An application method of an intelligent monitoring device for a sewage pumping station, wherein sensors are arranged at key nodes of the sewage pumping station to collect water level, flow, pressure, and pump power data in real time, and the data is cleaned and preprocessed. A physical model of the sewage pumping station is constructed based on the hydraulic principle, and a one-dimensional and three-dimensional hydraulic model is established based on the physical structure of the pumping station and the hydraulic principle. The parameters of the model are calibrated through historical data and experimental data.

[0051] Preferably, historical operating data and data output by the hydraulic model are used to train machine learning in the LSTM neural network to predict the operating status and energy consumption of the pumping station under different working conditions. The hydraulic model and the artificial intelligence model are integrated to form a hybrid model to comprehensively analyze and predict the operating status of the pumping station at different time scales and spatial scales.

[0052] Preferably, real-time monitoring data is used to dynamically adjust the operating parameters of the pump station, including pump start and stop, frequency conversion regulation, and unit quantity adjustment, to optimize the operation strategy of the pump station and reduce energy consumption. This is deployed in an actual sewage pump station in combination with cloud computing.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] 1. The intelligent monitoring device for sewage pumping stations provided by the present invention combines hydraulic models and artificial intelligence models to propose an efficient and intelligent sewage pumping station control method, which can significantly improve the operating efficiency and energy-saving effect of the pumping station and has broad application prospects. Improve the operating efficiency and energy-saving effect of sewage pumping stations. Enhance the adaptability and robustness of the system to cope with complex and changeable operating environments. Reduce equipment wear and maintenance costs and extend equipment service life. Improve the reliability and safety of sewage lifting. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic diagram of the process of the present invention;

[0056] Figure 2 A logical schematic diagram for the integration of the model of the present invention;

[0057] Figure 3 It is a schematic diagram of the process of real-time monitoring and regulation of the present invention. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings.

[0060] Combination Figure 1-Figure 3 , a sewage pump station intelligent monitoring device of the present invention comprises:

[0061] Data acquisition module, install sensors at key nodes of sewage pumps to collect water level, flow, pressure and pump power in real time;

[0062] The data preprocessing module cleans the data output from the sensor and unifies the data for accuracy and consistency;

[0063] In the process of data standardization,

[0064]

[0065] Where x is the original data, u is the mean, and σ is the standard deviation.

[0066] After standardized data preprocessing, the data is normalized:

[0067]

[0068] Where min(x) and max(x) are the minimum and maximum values ​​of the data respectively;

[0069] Hydraulic model building module,

[0070] The hydraulic model building module includes: building a one-dimensional model, simulating the longitudinal profile of pipeline flow, collecting basic data related to the pipeline, determining the area and hydraulic radius of the section; setting the boundary conditions from the initial section to the end section of the pipeline; parameter sensitivity analysis, evaluating the impact of parameters on model output;

[0071] The hydraulic model building module also includes: building a three-dimensional model, building the model based on polygonal modeling, setting appropriate lighting effects to further enhance the realism of the model, adding animation to the static model through model interaction, testing the model in a virtual environment, and making adjustments and optimizations based on the test results;

[0072] The AI ​​model training module collects historical operation data and data output by the hydraulic model, processes missing values, outliers and standardized data, determines key features for model training, including water level, flow, pressure and pump power, divides the data into training set, validation set and test set, with a ratio of 70% training, 15% validation and 15% test, selects LSTM neural network, determines the hyperparameters of the model, uses the training set for model training, uses the optimization algorithm to minimize the loss function, and monitors the loss and accuracy of the training process;

[0073] f t =σ(w f ·[h t-1 ,x t ]+b f );

[0074] enter:

[0075] i t =σ(w i ·[h t-1 ,x t ]+b i );

[0076] renew:

[0077]

[0078] Output:

[0079] o t =σ(w o ·[h t-1 ,x t ]+b o );h t =ot tanh(C t );

[0080] Where tan represents the activation function; w and b are the weight and bias matrices;

[0081] Use the validation set to evaluate the model performance, and improve the generalization ability of the model by adjusting the hyperparameters. According to the validation results, adjust the model architecture and hyperparameters, repeat the training and validation steps, and perform a final evaluation on the test set. In the evaluation and selection of the model, select the cross-validation algorithm. The algorithm formula is:

[0082]

[0083] where J is the performance index, The model is trained with data other than the i-th data; compared with the traditional time series analysis method, it can more accurately predict the operating status of the pumping station.

[0084] Hybrid model integration module: The hydraulic model and the artificial intelligence model are integrated to form a hybrid model. The hybrid model conducts comprehensive analysis and prediction of the operating status of the pumping station at different time scales and spatial scales.

[0085] Based on the physical structure and fluid mechanics principles of the sewage pumping station, a hydraulic model that describes the flow characteristics of sewage is constructed. The hydraulic model and artificial intelligence model are verified and optimized respectively to ensure the accuracy and reliability of the individual models. The data output by the hydraulic model is combined with the real-time monitoring data as the input of the artificial intelligence model. The hybrid model is used for comprehensive analysis and prediction. The operating status prediction at different time scales and spatial scales is analyzed. The time scale includes short-term prediction and long-term prediction, and the spatial scale includes flow and water level at different locations. The prediction results of the hybrid model are verified, and the integration strategy and model parameters are adjusted. The reinforcement learning algorithm enables the system to adaptively adjust the control strategy under different working conditions to optimize the operating efficiency and energy consumption of the pumping station. By real-time monitoring of the working conditions and dynamically adjusting the weights of the hydraulic model and the artificial intelligence model, it is ensured that the optimal prediction results can be obtained under different conditions.

[0086] Weighted average ensemble method:

[0087] Calculation formula: in and are the prediction results of the hydraulic model and the artificial intelligence model respectively; w1 and w2 are weight coefficients, satisfying w1+w2=1;

[0088] Real-time monitoring data X = [1.54, 4.0]; hydraulic model output Predictive AI Models Weight coefficient w1 = 0.4; w2 = 0.6

[0089]

[0090] Adjust the weights of the hydraulic model and artificial intelligence model in real time according to different working conditions to adapt to different operating conditions, use online learning mechanism to update the model so that the system can adapt to environmental changes and new data, introduce machine learning model interpretation tools and techniques to improve the transparency and trust of decision-making.

[0091] Control algorithm, real-time monitoring data, designing algorithms for dynamically adjusting pump station operating parameters, including pump start and stop, frequency conversion regulation, and unit quantity regulation;

[0092] The sewage intelligent control module combines the intelligent control system with cloud computing, processes and analyzes data in real time, deploys the developed intelligent control system in the actual sewage pumping station, configures the on-site network connection, and ensures stable communication between sensors, control equipment and cloud servers;

[0093] Integrate field data into the cloud computing platform to achieve real-time data transmission, and use deep learning algorithms to perform time series forecasting to predict the operating status of the pump station and possible problems;

[0094] Use PID control algorithm for real-time regulation to ensure the pump station operates in the optimal state;

[0095] PID control algorithm:

[0096]

[0097] Where e(t) is the error, K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient;

[0098] Reinforcement learning algorithm adaptively adjusts the control strategy:

[0099] Define the state space S: including the liquid level, flow rate, and pressure of the pump station;

[0100] Define action space A: pump start and stop, valve switch;

[0101] Set the reward function R(S,A) to encourage the system to maintain the best motion state:

[0102] Formula: R(S,A) = -(deviation cost + energy cost); Use the updated Q value function to select the strategy:

[0103] Update Q value function:

[0104]

[0105] in, is the learning rate, γ is the discount factor;

[0106] Use neural network to approximate Q value function:

[0107]

[0108] Select an action based on the current state S In strategy optimization, continuously collect data and optimize the Q-value function or strategy network to adapt to different tasks;

[0109] Adopt the Isolate Forest Anomaly Detection Algorithm to monitor data anomalies, set up alarm mechanisms, and develop a visual interface on the cloud platform to monitor the operating status of the pump station in real time, generate reports and analysis charts, and help decision-making and optimization;

[0110] The system testing and maintenance module conducts comprehensive testing on the deployed system to ensure that its functions and performance meet the expected requirements.

[0111] An application method of an intelligent monitoring device for a sewage pumping station, wherein sensors are arranged at key nodes of the sewage pumping station, water level, flow, pressure, and pump power data are collected in real time, and the data are cleaned and preprocessed. Based on the hydraulic principle, a physical model of the sewage pumping station is constructed. Based on the physical structure of the pumping station and the hydraulic principle, a one-dimensional and three-dimensional hydraulic model is established, and the parameters of the model are calibrated by historical data and experimental data.

[0112] Using historical operating data and data output by the hydraulic model, we train machine learning in the LSTM neural network to predict the operating status and energy consumption of the pump station under different working conditions. We integrate the hydraulic model and the artificial intelligence model to form a hybrid model to comprehensively analyze and predict the operating status of the pump station at different time scales and spatial scales.

[0113] Real-time monitoring of data, dynamic adjustment of pump station operating parameters, including pump start and stop, frequency conversion adjustment, and unit quantity adjustment, optimize the pump station operation strategy, reduce energy consumption, and deploy it in actual sewage pump stations in combination with cloud computing.

[0114] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0115] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent monitoring device for a sewage pumping station, characterized in that: include: Data acquisition module, install sensors at key nodes of sewage pumps to collect water level, flow, pressure and pump power in real time; The data preprocessing module cleans the data output from the sensor and unifies the data for accuracy and consistency; Hydraulic model building module, The hydraulic model building module includes: building a one-dimensional model, simulating the longitudinal profile of pipeline flow, collecting basic data related to the pipeline, determining the area and hydraulic radius of the section; setting the boundary conditions from the initial section to the end section of the pipeline; parameter sensitivity analysis, evaluating the impact of parameters on model output; The hydraulic model building module also includes: building a three-dimensional model, building the model based on polygonal modeling, setting appropriate lighting effects to further enhance the realism of the model, adding animation to the static model through model interaction, testing the model in a virtual environment, and making adjustments and optimizations based on the test results; The AI ​​model training module collects historical operation data and data output by the hydraulic model, processes missing values, outliers and standardized data, determines key features for model training, including water level, flow, pressure and pump power, divides the data into training set, validation set and test set, with a ratio of 70% training, 15% validation and 15% test, selects LSTM neural network, determines the hyperparameters of the model, uses the training set for model training, uses the optimization algorithm to minimize the loss function, and monitors the loss and accuracy of the training process; LSTM neural network, the formula is as follows: f t =σ(w f ·[h t-1 ,x t ]+b f ); enter: renew: Output: o t =σ(w o ·[h t-1 ,x t ]+b o );h t =o t ·tanh(C t ); Where tan represents the activation function; w and b are the weight and bias matrices; Use the validation set to evaluate the model performance, improve the generalization ability of the model by adjusting the hyperparameters, adjust the model architecture and hyperparameters based on the validation results, repeat the training and validation steps, and perform a final evaluation on the test set; Hybrid model integration module: The hydraulic model and the artificial intelligence model are integrated to form a hybrid model. The hybrid model conducts comprehensive analysis and prediction of the operating status of the pumping station at different time scales and spatial scales. Control algorithm, real-time monitoring data, designing algorithms for dynamically adjusting pump station operating parameters, including pump start and stop, frequency conversion regulation, and unit quantity regulation; The sewage intelligent control module combines the intelligent control system with cloud computing, processes and analyzes data in real time, deploys the developed intelligent control system in the actual sewage pumping station, configures the on-site network connection, and ensures stable communication between sensors, control equipment and cloud servers; System testing and maintenance module, which conducts comprehensive testing on the deployed system to ensure that its functions and performance meet the expected requirements; Integrate field data into the cloud computing platform to achieve real-time data transmission, and use deep learning algorithms to perform time series forecasting to predict the operating status of the pump station and possible problems; Use PID control algorithm for real-time regulation to ensure the pump station operates in the optimal state; PID control algorithm: Where e(t) is the error, K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient; Reinforcement learning algorithm adaptively adjusts the control strategy: Define the state space S: including the liquid level, flow rate, and pressure of the pump station; Define action space A: start and stop of pump, opening and closing of valve; Set the reward function R(S,A) to encourage the system to maintain the best motion state: Formula: R(S,A) = -(deviation cost + energy cost); Use the updated Q value function to select the strategy: Update Q value function: in, is the learning rate, γ is the discount factor; Use neural network to approximate Q value function: Select an action based on the current state S In strategy optimization, data is continuously collected and the Q-value function or strategy network is optimized to adapt to different working conditions; The Isolation Forest Anomaly Detection Algorithm is used to monitor data anomalies, set up alarm mechanisms, and develop a visual interface on the cloud platform to monitor the operating status of the pump station in real time, generate reports and analysis charts, and assist in decision-making and optimization.

2. The intelligent monitoring device for a sewage pumping station according to claim 1, characterized in that: Based on the physical structure of the sewage pumping station and the principles of fluid mechanics, a hydraulic model that describes the flow characteristics of sewage is constructed. The hydraulic model and artificial intelligence model are verified and optimized respectively to ensure the accuracy and reliability of the individual models. The data output by the hydraulic model is combined with the real-time monitoring data as the input of the artificial intelligence model. The hybrid model is used for comprehensive analysis and prediction, and the operation status prediction at different time scales and spatial scales is analyzed. The time scale includes short-term prediction and long-term prediction, and the spatial scale includes flow and water level at different locations.

3. The intelligent monitoring device for a sewage pumping station according to claim 2 is characterized in that: Verify the prediction results of the hybrid model and adjust the integration strategy and model parameters; Weighted average ensemble method: Calculation formula: in and They are the prediction results of the hydraulic model and the artificial intelligence model respectively; w1 and w2 are weight coefficients, satisfying w1+w2=1.

4. The application method of the intelligent monitoring device for a sewage pumping station according to any one of claims 1 to 3, characterized in that: Sensors are arranged at key nodes of the sewage pumping station to collect water level, flow, pressure, and pump power data in real time, and the data is cleaned and preprocessed. Based on the principles of hydraulics, a physical model of the sewage pumping station is constructed. Based on the physical structure of the pumping station and the principles of hydraulics, one-dimensional and three-dimensional hydraulic models are established, and the parameters of the model are calibrated using historical data and experimental data.

5. The application method of the intelligent monitoring device for sewage pumping station according to claim 4 is characterized by: Using historical operating data and data output by the hydraulic model, we train machine learning in the LSTM neural network to predict the operating status and energy consumption of the pumping station under different working conditions. We integrate the hydraulic model and the artificial intelligence model to form a hybrid model to comprehensively analyze and predict the operating status of the pumping station at different time scales and spatial scales.

6. The application method of the intelligent monitoring device for a sewage pumping station according to claim 5 is characterized by: Real-time monitoring of data, dynamic adjustment of pump station operating parameters, including pump start and stop, frequency conversion adjustment, and unit quantity adjustment, optimize the pump station operation strategy, reduce energy consumption, and deploy it in actual sewage pump stations in combination with cloud computing.

Images