An intelligent dosing and regulation system for sewage treatment based on fuzzy logic

Abstract

The present invention relates to the technical field of data processing, and is a sewage treatment intelligent dosing adjustment system based on fuzzy logic, comprising: a data acquisition module, which collects water quality data of sewage and optimizes the data acquisition process, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow; a data analysis module, which is used to receive real-time data and perform small batch gradient descent analysis on the data to obtain analysis results; an optimization module, which is used to optimize the parameters and rules of the fuzzy logic controller through differential evolution according to the analysis results to obtain optimized parameters and rules; a fuzzy logic control module, which performs fuzzy logic calculation of dosing amount through simulated annealing according to the received water quality parameters and optimized rules, and converts it into a dosing control signal. The present invention realizes comprehensive intelligent management from data acquisition to dosing adjustment by integrating advanced data acquisition, analysis, optimization and control technologies.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to an intelligent dosing and regulating system for sewage treatment based on fuzzy logic. Background Art

[0002] In the sewage treatment process, the dosing adjustment link is crucial to ensure that the final effluent quality meets environmental protection standards. Traditional dosing adjustment methods mostly rely on the operator's experience and some fixed parameter settings. However, this method has obvious limitations: it cannot be flexibly and accurately adjusted according to the real-time water quality changes of sewage.

[0003] Specifically, the composition and concentration of sewage may fluctuate with time, weather, seasons and other external factors. Such fluctuations require the dosing system to respond quickly and accurately to maintain the stability and efficiency of the treatment effect. However, the traditional method that relies on manual experience and fixed parameters often cannot adjust the dosage in a timely and accurate manner, which may lead to substandard effluent quality, low treatment efficiency, and even waste of chemical agents, thereby increasing operating costs. What's more serious is that excessive dosage of chemicals may cause secondary pollution to the environment, thereby having a negative impact on the ecosystem.

[0004] In order to improve this situation, intelligent technologies have been widely used in the field of sewage treatment in recent years. These technologies make the dosing adjustment process more accurate and efficient by introducing advanced algorithms and control systems. Among them, fuzzy logic control technology has attracted much attention because of its unique advantages in dealing with uncertainty and fuzzy information. Fuzzy logic control can simulate the human reasoning process and deal with problems that are difficult to describe with precise mathematical models. It is particularly suitable for complex and changeable scenarios such as sewage treatment. However, although fuzzy logic control technology has great potential in theory, in practical applications, facing the changeable and complex water quality parameters, a single fuzzy logic control still seems to be powerless. The interaction of various chemical substances in sewage, changes in environmental factors, and uncertainties in the treatment process all increase the difficulty of control. Therefore, although fuzzy logic control technology can provide certain improvements in some cases, further technological innovation and optimization are still needed to achieve the ideal control effect. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide an intelligent dosing and adjustment system for sewage treatment based on fuzzy logic. By integrating advanced data acquisition, analysis, optimization and control technologies, comprehensive intelligent management from data acquisition to dosing adjustment is achieved.

[0006] In order to solve the above technical problems, the basic concept of the technical solution adopted by the present invention is:

[0007] In the first aspect, a sewage treatment intelligent dosing regulation system based on fuzzy logic comprises:

[0008] Data collection module, which collects sewage water quality data and optimizes the data collection process, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow;

[0009] The data analysis module is used to receive real-time data and perform small batch gradient descent analysis on the data to obtain analysis results;

[0010] An optimization module is used to optimize the parameters and rules of the fuzzy logic controller through differential evolution according to the analysis results to obtain optimized parameters and rules;

[0011] The fuzzy logic control module performs fuzzy logic calculation of the dosage through simulated annealing according to the received water quality parameters and the optimized rules, and converts it into a dosage control signal;

[0012] The dosing module, after receiving the dosing control signal, adjusts the constraint conditions of the dosing amount through the projected subgradient method, and combines the interior point method to perform integer programming in the dosing process to achieve automatic adjustment of the dosing amount;

[0013] The detection module detects the sewage water quality parameter data after drug addition through Kalman filtering, and feeds back the treated water quality parameters to the fuzzy logic controller, data analysis and optimization module to adjust the drug addition strategy.

[0014] Optimized, collect sewage water quality data and optimize the data collection process, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow, including:

[0015] Determine the frequency, cycle and strategy of data collection based on historical data and changing trends of water quality parameters;

[0016] According to the frequency, cycle and strategy of collection, water quality parameter data are collected and preprocessed to obtain preprocessed data. The water quality parameter data include chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow.

[0017] Optimized, receives real-time data, and performs small batch gradient descent analysis on the data to obtain analysis results, including:

[0018] Divide the preprocessed real-time data into multiple small batch data sets;

[0019] Initialize the learning rate and number of iterations; according to each mini-batch dataset, Calculate the average error between the predicted values ​​and the actual values;

[0020] According to the average error, Adjust the gradient of the weights by Adjust the gradient of the bias to minimize the difference between the predicted value and the actual value to obtain the final solution, where is the weight parameter, is the bias parameter, It is The sewage quality data values ​​of samples, It is The actual target value of samples, is the number of samples in the current mini-batch dataset, is the average error between the predicted value and the actual value, is the gradient of the weight, is the gradient of the bias;

[0021] According to the final solution, the optimized parameters are obtained;

[0022] The dosage is predicted according to the optimized parameters to obtain a dosage prediction result, wherein the prediction result is an analysis result.

[0023] Optimized, according to the analysis results, the parameters and rules of the fuzzy logic controller are optimized by differential evolution to obtain the optimized parameters and rules, including:

[0024] Determine the optimization target based on the dosage prediction results;

[0025] Initialize the population and generate a set of random fuzzy logic controller parameters and rules as initial candidate solutions;

[0026] The parameters and rules in the population are combined and applied to the fuzzy logic controller, and the fitness of the controller is evaluated based on the dosage prediction results;

[0027] Perform mutation, crossover and selection operations on the individuals in the population to generate new candidate solutions, and select the corresponding individuals to form a new population according to the fitness value until the preset number of iterations is reached, then stop the iteration. After the iteration is completed, obtain the final individuals in the current population as the optimized parameters and rules.

[0028] Optimized, based on the received water quality parameters and optimized rules, the fuzzy logic calculation of the dosage is performed through simulated annealing and converted into a dosing control signal, including:

[0029] Obtain water quality parameter data and optimized rules;

[0030] Initialize simulated annealing algorithm parameters;

[0031] According to the optimized fuzzy logic rules, a fuzzy logic system is constructed. The system includes a fuzzification interface, a fuzzy inference engine and a defuzzification interface. The water quality parameter data is converted into a fuzzy set. According to the fuzzy logic rules, the fuzzy dosage decision is obtained and the fuzzy dosage decision is converted into a specific dosage value.

[0032] Initialize the simulated annealing algorithm parameters, execute the simulated annealing process, generate a new dosing decision based on the current temperature and state in each iteration, and use the fuzzy logic system to evaluate its advantages and disadvantages to obtain the final solution;

[0033] When the simulated annealing algorithm reaches the termination condition, the current final dosage is obtained;

[0034] The dosage calculated by simulated annealing and fuzzy logic is converted into a specific dosage control signal.

[0035] After receiving the dosing control signal, the dosing constraint is adjusted by the projected subgradient method, and integer programming is performed on the dosing process in combination with the interior point method to achieve automatic adjustment of the dosing amount, including:

[0036] Receive drug dosing control signals and preset drug dosage constraints;

[0037] According to the preset drug dosage constraints and the current control signal, the drug dosage is optimized by the projected subgradient method to obtain the dosage that meets all the constraints;

[0038] Under the premise of meeting all constraints, find the corresponding dosage plan and automatically adjust the dosage.

[0039] Optimized, based on the preset constraints and the current control signal, the dosage is optimized by the projected subgradient method to obtain the dosage that meets all constraints, including:

[0040] Set relevant parameters, including learning rate sequence, stopping conditions, and constraint sets;

[0041] According to the current dosage plan, the sub-gradient of the objective function is calculated. The sub-gradient is the sum of the absolute values ​​of the components in the dosage plan.

[0042] Use the projection operation to project the updated dosage scheme into the constraint set to obtain a new dosage scheme, set the new dosage scheme as the current dosage scheme, and prepare for the next iteration;

[0043] If the improvement of the objective function value is less than the preset threshold, the iteration is stopped and the dosage scheme is obtained.

[0044] Optimized, the sewage water quality parameter data after dosing is detected by Kalman filtering, and the treated water quality parameters are fed back to the fuzzy logic controller, data analysis and optimization module to adjust the dosing strategy, including:

[0045] Collect various water quality parameter data of sewage after adding drugs;

[0046] The collected sewage water quality parameter data is processed by Kalman filtering to obtain the processed water quality parameter data;

[0047] The water quality parameter data after Kalman filtering is transmitted to the fuzzy logic controller, data analysis and optimization module to identify the changing trend and periodic fluctuation data of water quality parameters and obtain the range of water quality parameters;

[0048] Compare the water quality parameter range with the expected range, and adjust the dosing strategy based on the comparison results, including changing the type, dosage and time of the agent.

[0049] In the second aspect, a sewage treatment intelligent dosing adjustment method based on fuzzy logic comprises the following steps:

[0050] Collect and optimize wastewater quality data, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content, and wastewater flow;

[0051] Perform a small batch gradient descent analysis on the data to obtain the analysis results;

[0052] According to the analysis results, the parameters and rules of the fuzzy logic controller are optimized by differential evolution to obtain the optimized parameters and rules;

[0053] According to the water quality parameters and the optimized rules, the fuzzy logic calculation of the dosage is performed through simulated annealing and converted into a dosing control signal;

[0054] According to the dosing control signal, the constraint conditions of the dosage are adjusted by the projection subgradient method, and integer programming is performed in the dosing process in combination with the interior point method to achieve automatic adjustment of the dosage.

[0055] The sewage water quality parameter data after dosing is detected by Kalman filtering, and the treated water quality parameters are fed back to the fuzzy logic controller, data analysis and optimization module to adjust the dosing strategy.

[0056] After adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0057] By optimizing the data collection process through dynamic programming algorithms, key water quality data such as chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow can be collected more efficiently; by using small batch gradient descent analysis methods, the system can quickly process a large amount of real-time data and obtain accurate analysis results. By optimizing the parameters and rules of the fuzzy logic controller through the differential evolution algorithm, the system can adaptively adjust the control strategy to cope with different water quality conditions and treatment requirements. Combining simulated annealing algorithm and fuzzy logic to calculate the dosage, the system can handle the uncertainty and ambiguity in water quality parameters while considering the global optimum; by automatically adjusting the dosage through the projected subgradient method and the interior point method, the system can automatically optimize the dosage according to the preset constraints and real-time control signals; by detecting the sewage water quality parameter data after the addition of drugs through Kalman filtering, and feeding back the treated water quality parameters to other modules, the system can identify the changing trend and periodic fluctuation of water quality parameters in real time. This enables the system to respond quickly and adjust the dosing strategy to maintain the stability and optimization of the treatment effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a schematic diagram of the intelligent dosing and regulating system for sewage treatment based on fuzzy logic of the present invention.

[0059] Figure 2 It is a flow chart of the intelligent dosing adjustment method for sewage treatment based on fuzzy logic of the present invention. DETAILED DESCRIPTION

[0060] The following embodiment of the present application takes the intelligent dosing and regulation system for sewage treatment based on fuzzy logic as an example to explain the scheme of the present application in detail, but this embodiment cannot limit the scope of protection of the present application.

[0061] like Figure 1 As shown, the present invention provides a sewage treatment intelligent dosing regulation system based on fuzzy logic, comprising:

[0062] The data collection module 11 collects the water quality data of the sewage and optimizes the data collection process, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow;

[0063] The data analysis module 12 is used to receive real-time data and perform small batch gradient descent analysis on the data to obtain analysis results;

[0064] An optimization module 13 is used to optimize the parameters and rules of the fuzzy logic controller by differential evolution according to the analysis results to obtain optimized parameters and rules;

[0065] The fuzzy logic control module 14 performs fuzzy logic calculation of the dosage through simulated annealing according to the received water quality parameters and the optimized rules, and converts it into a dosage control signal;

[0066] The dosing module 15, after receiving the dosing control signal, adjusts the constraint conditions of the dosing amount by using the projected subgradient method, and performs integer programming on the dosing process in combination with the interior point method to achieve automatic adjustment of the dosing amount;

[0067] The detection module 16 detects the sewage water quality parameter data after the addition of the drug through Kalman filtering, and feeds back the treated water quality parameters to the fuzzy logic controller, data analysis and optimization module to adjust the dosing strategy.

[0068] In an embodiment of the present invention, a dynamic programming algorithm is used to optimize the data collection process to ensure real-time and accurate acquisition of key water quality data. The application of the dynamic programming algorithm makes data collection more intelligent and efficient, avoiding unnecessary waste of resources. Real-time data analysis is performed through a small batch gradient descent algorithm, which can quickly respond to water quality changes. The analysis module can help predict the changing trend of water quality parameters, so as to make corresponding processing preparations in advance. The parameters and rules of the fuzzy logic controller are optimized by the differential evolution algorithm, so that the system can be adaptively adjusted according to actual conditions, improve processing efficiency and effect, and the optimization module can improve the intelligence level of the system, reduce human intervention, and reduce operating costs. The fuzzy logic calculation of the dosage is performed by the simulated annealing algorithm to ensure accurate control of the dosage and improve the sewage treatment effect. Fuzzy logic control enables the system to respond more flexibly when facing complex and changeable water quality conditions. The automatic adjustment of the dosage is achieved by the projection subgradient method and the interior point method to ensure the accuracy and efficiency of the dosing process. The water quality parameters after dosing are detected in real time by Kalman filtering, and the feedback mechanism of the detection module enables the entire system to form a closed-loop control, which can continuously optimize and adjust the treatment strategy to ensure the stability and reliability of the sewage treatment effect. At the same time, this closed-loop control mechanism also helps to detect and solve problems in a timely manner and improve the robustness of the system.

[0069] In the intelligent dosing and regulating system for sewage treatment based on fuzzy logic described in the embodiment of the present invention, the data acquisition module 11 collects water quality data of sewage and optimizes the data acquisition process, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow, including:

[0070] Determine the frequency, cycle and strategy of data collection based on historical data and changing trends of water quality parameters;

[0071] According to the frequency, cycle and strategy of collection, water quality parameter data are collected and preprocessed to obtain preprocessed data. The water quality parameter data include chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow.

[0072] In an embodiment of the present invention, by optimizing the data collection process, the frequency, cycle and strategy of data collection can be dynamically adjusted according to the historical data and the changing trend of water quality parameters; the data collection module can remove outliers, noise and interference by preprocessing the collected data to obtain more accurate and reliable water quality parameter data. Since the data collection module can monitor water quality parameters in real time, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow, it can respond quickly to changes in water quality and adjust the dosing strategy in time to ensure that the sewage treatment effect is optimal. The optimized data collection strategy reduces unnecessary sampling and data processing costs, while increasing the service life of the equipment, thereby reducing the operating costs of the sewage treatment plant.

[0073] In the specific implementation process of the present invention, the specific steps include:

[0074] Select sensors based on the water quality parameters that need to be collected, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow; install the selected sensors at key nodes of the sewage treatment process, configure the data transmission system, conduct in-depth analysis of the water quality data collected in the past period of time, identify the changing patterns and periodic characteristics of water quality parameters, and predict the possible changing trends of water quality parameters in the future based on the historical data analysis results, combined with current environmental factors and treatment conditions. According to the predicted changing trends, dynamically adjust the frequency and period of data collection. For example, increase the collection frequency during periods when water quality parameters change more dramatically, and appropriately reduce the collection frequency during periods when changes are stable, so as to achieve a balance between data collection efficiency and accuracy.

[0075] After determining the data collection strategy, the installed sensors are used to collect real-time data on water quality parameters such as chemical oxygen demand, ammonia nitrogen content, total phosphorus content, and sewage flow in the sewage. The collected raw data are then preprocessed, including steps such as data cleaning, denoising, and outlier detection and correction.

[0076] The data analysis module 12 receives real-time data and performs small batch gradient descent analysis on the data to obtain analysis results, including:

[0077] Divide the preprocessed real-time data into multiple small batch data sets;

[0078] Initialize the learning rate and number of iterations; according to each mini-batch dataset, Calculate the average error between the predicted values ​​and the actual values;

[0079] According to the average error, Adjust the gradient of the weights by Adjust the gradient of the bias to minimize the difference between the predicted value and the actual value to obtain the final solution, where is the weight parameter, is the bias parameter, It is The sewage quality data values ​​of samples, It is The actual target value of samples, is the number of samples in the current mini-batch dataset, is the average error between the predicted value and the actual value, is the gradient of the weight, is the gradient of the bias;

[0080] According to the final solution, the optimized parameters are obtained;

[0081] The dosage is predicted according to the optimized parameters to obtain a dosage prediction result, wherein the prediction result is an analysis result.

[0082] In an embodiment of the present invention, by dividing the pre-processed real-time data into multiple small batch data sets, the data analysis module can process these data in parallel, thereby significantly improving the computational efficiency; the small batch gradient descent algorithm combines the advantages of batch gradient descent and stochastic gradient descent, which can converge to the optimal solution quickly and avoid falling into the local optimum. By continuously adjusting the gradient of weights and biases, the algorithm can minimize the gap between the predicted value and the actual value, thereby obtaining a better parameter solution. The data analysis module can monitor the convergence of the algorithm in real time and predict the dosage according to the optimized parameters. This real-time monitoring mechanism enables the system to respond quickly to changes in water quality, adjust the dosage strategy in time, and ensure the sewage treatment effect. By predicting the dosage through the optimized parameters, the data analysis module can provide more accurate prediction results, which helps to reduce the waste of drugs and improve the economy and environmental protection of sewage treatment.

[0083] In the specific implementation process of the present invention, the specific steps include:

[0084] The data analysis module first receives real-time data from the data acquisition module or other data sources, and divides the received real-time data into multiple small batch data sets according to a certain size, such as 100 data points per batch.

[0085] Before starting the gradient descent analysis, the learning rate and number of iterations are initialized, where the learning rate is the update amplitude of the weights and biases in each iteration, and the number of iterations is the upper limit of the operation of the gradient descent algorithm.

[0086] For each small batch data set, the current weight and bias parameters are used to calculate the predicted value based on the input feature value, and compared with the actual target value to obtain the error. The input feature value is the sewage quality data value; the average error is used as the metric of the loss function; the gradient of the weight and bias is calculated based on the error, and the gradient indicates the direction and rate of change of the loss function with respect to the weight and bias; by adjusting these parameters, the gap between the predicted value and the actual value can be reduced. Using the calculated gradient and the set learning rate to update the weight and bias parameters is the core step of the gradient descent algorithm. Through multiple iterations, the model gradually approaches the optimal solution.

[0087] During the iteration process, after each iteration, the current loss value is recorded. If the loss value changes little or remains basically unchanged for multiple consecutive iterations, the algorithm can be considered to have converged. At this time, the iteration can be stopped and the current parameters are used as the optimized parameters. When the algorithm converges, the current weight and bias parameters are obtained as the optimized parameters. These parameters are automatically adjusted by the gradient descent algorithm, which can make the prediction performance of the model reach the optimal level.

[0088] According to the optimized parameters, new sewage quality data is received as the input of the model, and the predicted value of the dosage is calculated according to the input characteristic value using the optimized weight and bias parameters.

[0089] In the intelligent dosing and regulating system for sewage treatment based on fuzzy logic described in the embodiment of the present invention, the optimization module 13 optimizes the parameters and rules of the fuzzy logic controller by differential evolution according to the analysis results to obtain optimized parameters and rules, including:

[0090] Determine the optimization target based on the dosage prediction results;

[0091] Initialize the population and generate a set of random fuzzy logic controller parameters and rules as initial candidate solutions;

[0092] The parameters and rules in the population are combined and applied to the fuzzy logic controller, and the fitness of the controller is evaluated based on the dosage prediction results;

[0093] Perform mutation, crossover and selection operations on the individuals in the population to generate new candidate solutions, and select the corresponding individuals to form a new population according to the fitness value until the preset number of iterations is reached, then stop the iteration. After the iteration is completed, obtain the final individuals in the current population as the optimized parameters and rules.

[0094] In an embodiment of the present invention, the control accuracy of the controller can be significantly improved by optimizing the parameters and rules of the fuzzy logic controller through the differential evolution algorithm. The optimized parameters and rules are more in line with the needs of the actual sewage treatment process, making the control of the dosage more accurate, thereby improving the sewage treatment effect. The differential evolution algorithm has a strong global search capability and can find the optimal solution in a complex solution space, so that the optimization module can quickly adjust the parameters and rules of the fuzzy logic controller when facing water quality changes or system disturbances to adapt to the new environment and maintain the stability and efficiency of the system. The setting of traditional fuzzy logic controller parameters and rules often depends on the experience and knowledge of experts. The optimization module reduces the need for human intervention and the dependence on expert knowledge through an automated optimization process, thereby improving the intelligence level of the system. The differential evolution algorithm can quickly converge to the global optimal solution through the mutation, crossover and selection operations of individuals in the population. The efficient optimization process improves the computing efficiency, and the system can respond to water quality changes more quickly and adjust the dosing strategy in time.

[0095] In the specific implementation process of the present invention, the specific steps include:

[0096] According to the dosage prediction results and the actual sewage treatment needs, the optimization goals are determined, including reducing the deviation of dosage, improving treatment efficiency, and reducing treatment costs.

[0097] A set of parameters and rules of fuzzy logic controllers are randomly generated, which will be used as the initial candidate solutions of differential evolution algorithm. Each candidate solution represents a possible fuzzy logic controller configuration; the number of individuals in the population is determined, and each combination of parameters and rules in the population is applied to the fuzzy logic controller to simulate the operation of the controller.

[0098] According to the comparison between the predicted dosage and the actual dosage, the fitness of each controller is evaluated. The fitness function can reflect the optimization goal, and the fitness is the error between the predicted dosage and the actual dosage.

[0099] For each individual in the population, three different individuals are randomly selected, and mutant individuals are generated by differential vectors. The mutant individuals are cross-operated with the original individuals to generate new test individuals. The fitness of the test individuals and the original individuals is compared, and the individuals with higher fitness are selected to enter the next generation of population. The mutation, cross-over and selection operations are repeated until the maximum number of iterations is reached or other stop conditions are met. During the iteration process, the fitness changes of individuals in the population are monitored, and the optimal solution of each iteration is recorded. When the maximum number of iterations is reached or the optimal solution has no significant changes in multiple consecutive iterations, the iteration is stopped, and the individual with the highest fitness is selected from the current population as the optimized parameters and rules.

[0100] In the intelligent dosing regulation system for sewage treatment based on fuzzy logic described in the embodiment of the present invention, the fuzzy logic control module 14 performs fuzzy logic calculation of the dosing amount through simulated annealing according to the received water quality parameters and the optimized rules, and converts it into a dosing control signal, including:

[0101] Obtain water quality parameter data and optimized rules;

[0102] Initialize simulated annealing algorithm parameters;

[0103] According to the optimized fuzzy logic rules, a fuzzy logic system is constructed. The system includes a fuzzification interface, a fuzzy inference engine and a defuzzification interface. The water quality parameter data is converted into a fuzzy set. According to the fuzzy logic rules, the fuzzy dosage decision is obtained and the fuzzy dosage decision is converted into a specific dosage value.

[0104] Initialize the simulated annealing algorithm parameters, execute the simulated annealing process, generate a new dosing decision based on the current temperature and state in each iteration, and use the fuzzy logic system to evaluate its advantages and disadvantages to obtain the final solution;

[0105] When the simulated annealing algorithm reaches the termination condition, the current final dosage is obtained;

[0106] The dosage calculated by simulated annealing and fuzzy logic is converted into a specific dosage control signal.

[0107] In the embodiment of the present invention, the application of simulated annealing algorithm optimizes the decision-making process of dosing amount, can avoid falling into local optimum in the search process, and accepts poor solutions with a certain probability by simulating the physical annealing process, so that it is possible to jump out of the local optimum and find the global optimal solution, thereby ensuring the accuracy and rationality of dosing amount calculation. Fuzzy logic has significant advantages in dealing with uncertainty and ambiguity in water quality parameters. Since there are many factors that are difficult to accurately quantify in the sewage treatment process, fuzzy logic can provide a flexible and robust method to deal with these problems. Through fuzzy logic calculation, it can better adapt to the changes in water quality parameters and improve the accuracy and stability of dosing control. The fuzzy logic control module can receive water quality parameter data in real time, and respond and adjust quickly according to these data and optimized rules, ensuring the timeliness and accuracy of dosing control signals, and can quickly adapt to water quality changes and maintain sewage treatment effects. By combining simulated annealing and fuzzy logic, the fuzzy logic control module improves the intelligence level of the entire sewage treatment system. The system can automatically calculate a reasonable dosing amount according to the current water quality conditions and adjustment rules, and convert it into a specific dosing control signal, reducing the need for human intervention.

[0108] In the specific implementation process of the present invention, the specific steps include:

[0109] Obtain water quality parameter data from the data acquisition module, including key indicators such as chemical oxygen demand, ammonia nitrogen content, total phosphorus content, and sewage flow, and obtain optimized fuzzy logic rules;

[0110] After receiving the water quality parameters and optimization rules, the fuzzy logic control module will conduct an in-depth analysis of these data and determine the preliminary range of dosing according to the optimized rules. During the analysis, factors such as the changing trend of water quality parameters, historical data, and fuzzy sets and membership functions in the optimization rules will be considered.

[0111] The initial temperature, the temperature drop rate, the termination temperature and the number of iterations at each temperature or the number of inner loops are set, and the fuzzy logic rules optimized by the differential evolution algorithm are received from the optimization module.

[0112] The precise water quality parameters are converted into corresponding fuzzy sets, such as "low", "medium", and "high" through the fuzzification interface; the membership function is defined for each water quality parameter to determine the degree to which the parameter value belongs to each fuzzy set, and the fuzzy inference engine is constructed according to the optimized fuzzy logic rules, such as "if the pH value is low and the chemical oxygen demand is high, the dosage is large". The maximum membership method is used to convert the fuzzy dosage decision obtained by the fuzzy inference engine into a specific dosage value range.

[0113] Starting from the initial temperature, it enters the annealing cycle. In each iteration, a new dosage decision is generated according to the current temperature and the current dosage decision by making a small random perturbation to the current decision. The corresponding dosage value is obtained through the fuzzy logic system, and the degree of matching or error with the actual demand is calculated. If the new decision is better, it is accepted unconditionally and the current state is updated. If the new decision is worse, it is accepted with a certain probability. The probability of accepting a worse decision decreases as the temperature decreases, which helps the algorithm to jump out of the local optimum and converge to the global optimal solution.

[0114] After completing a certain number of iterations at each temperature, the temperature is reduced according to the temperature drop rate. When the temperature drops to the termination temperature or the preset maximum number of iterations is reached, the algorithm stops.

[0115] When the simulated annealing algorithm converges, the dosage is mapped to the actual control instruction, for example, a control signal is sent to the dosing equipment through a programmable logic controller or a distributed control system; the fuzzy logic control module outputs the converted dosing control signal to the dosing pump or valve to achieve precise control of the dosage in the sewage treatment process.

[0116] In the intelligent dosing regulation system for sewage treatment based on fuzzy logic described in the embodiment of the present invention, the dosing module 15, after receiving the dosing control signal, adjusts the constraint conditions of the dosage by the projection subgradient method, and performs integer programming in the dosing process in combination with the interior point method to realize automatic adjustment of the dosage, including:

[0117] Receive drug dosing control signals and preset drug dosage constraints;

[0118] According to the preset drug dosage constraints and the current control signal, the drug dosage is optimized by the projected subgradient method to obtain the dosage that meets all the constraints;

[0119] Under the premise of meeting all constraints, find the corresponding dosage plan and automatically adjust the dosage.

[0120] In an embodiment of the present invention, by optimizing the dosage through the projected subgradient method, the dosage can be controlled more accurately to ensure that it is within the preset constraints, avoiding the problem of too much or too little dosage, thereby improving the efficiency and effect of sewage treatment. In the sewage treatment process, the dosage is often subject to a variety of constraints, such as equipment capacity and treatment cost; it can ensure that the calculated dosage meets all preset constraints, ensuring the compliance and safety of the treatment process. On the premise of meeting the constraints, the dosing module finds a dosage scheme that is closer to the global optimum through the optimization algorithm, which helps to improve the performance of the entire sewage treatment system and achieve more efficient resource utilization. The dosing module can automatically receive control signals and automatically adjust the settings of the dosing equipment according to the algorithm, realizing the automation and intelligence of the dosing process.

[0121] In the specific implementation process of the present invention, the specific steps include:

[0122] The dosing module first receives the dosing control signal converted from the fuzzy logic control module. Before adjusting the dosing, some dosing constraints need to be preset, including the maximum dosing amount, the minimum dosing amount, and the dosing speed limit.

[0123] According to the received dosing control signal and the preset constraints, an initial dosing scheme is set. In each iteration, the objective function gradient of the current dosing scheme is calculated, and a small step update is performed in the opposite direction of the gradient to try to reduce the value of the objective function. The updated dosing scheme is projected into the feasible domain of the constraints using a projection operation, and the above iterative process is repeated until the stop condition is met.

[0124] The optimized dosing scheme obtained by the projected subgradient method is used as the initial point, and the parameters of the interior point method, such as the parameters of the barrier function and the step size, are set. The interior point method is used to further search for a dosing scheme that is closer to the global optimum while satisfying all constraints. Since the actual dosing equipment may only accept integer dosing settings, it is necessary to convert the floating-point dosing scheme into an integer scheme.

[0125] The dosing amount plan obtained after integer programming is output to the dosing equipment. After receiving the new dosing amount setting, the dosing equipment automatically adjusts to achieve automatic control and adjustment of the dosing amount, including adjusting the speed of the dosing pump, the opening of the valve and other parameters to ensure that the actual dosing amount is consistent with the optimized dosing amount plan.

[0126] In the intelligent dosing regulation system for sewage treatment based on fuzzy logic described in the embodiment of the present invention, the dosing module 15 optimizes the dosage by the projection subgradient method according to the preset constraints and the current control signal to obtain the dosage that satisfies all the constraints, including:

[0127] Set relevant parameters, including learning rate sequence, stopping conditions, and constraint sets;

[0128] According to the current dosage plan, the sub-gradient of the objective function is calculated. The sub-gradient is the sum of the absolute values ​​of the components in the dosage plan.

[0129] Use the projection operation to project the updated dosage scheme into the constraint set to obtain a new dosage scheme, set the new dosage scheme as the current dosage scheme, and prepare for the next iteration;

[0130] If the improvement of the objective function value is less than the preset threshold, the iteration is stopped and the dosage scheme is obtained, which should satisfy all constraints.

[0131] In an embodiment of the present invention, the projected subgradient method can effectively handle optimization problems involving multiple complex constraints. In the process of dosing in sewage treatment, multiple restrictions may be involved, such as the upper and lower limits of the dosage and the cost budget of the agent. Through this method, it can be ensured that the obtained dosing scheme strictly meets these preset constraints. By calculating the subgradient of the objective function and iteratively updating it, the optimal dosing scheme can be gradually approached. The projected subgradient method can achieve a faster convergence speed through reasonable iteration steps and learning rate adjustment, and can quickly find the optimal or approximately optimal dosing scheme that meets all constraints, thereby improving the response speed and efficiency of sewage treatment. When the requirements or conditions in the sewage treatment process change, it is only necessary to adjust the corresponding constraint set and objective function to re-perform the optimization calculation without making major changes to the algorithm itself.

[0132] In the specific implementation process of the present invention, the specific steps include:

[0133] Set relevant parameters and learning rate sequence to control the step size of dosage adjustment during the iteration process; the stopping condition is a sufficiently small positive number as a threshold to determine whether the improvement of the objective function value is small enough to decide whether to stop the iteration; the constraint condition set specifies all constraints that need to be met during the dosage adjustment process, such as the maximum dosage, minimum dosage, proportional relationship between dosages, etc.

[0134] Select an initial dosing scheme as the starting point of the iteration to ensure that the scheme is feasible, that is, it meets all the preset constraints. According to the current dosing scheme, calculate the subgradient of the objective function, which is the sum of the absolute values ​​of the components in the dosing scheme. This subgradient reflects the gap between the current dosing scheme and the optimal solution of the objective function.

[0135] The dosage scheme is updated by multiplying the learning rate by the subgradient. The updated dosage scheme may not satisfy the constraints, so it needs to be projected into the constraint set using a projection operation. The projection operation ensures that the new dosage scheme satisfies all preset constraints. If the updated dosage exceeds the range of a constraint, it is adjusted to the boundary of the range.

[0136] The new dosage scheme after the projection operation is set as the current dosage scheme, and the next iteration is prepared. After each iteration, check whether the improvement of the objective function value is less than the preset threshold. If so, stop the iteration; otherwise, continue to the next iteration. When the iteration stops, the current dosage scheme is obtained as the final dosage scheme. This scheme not only meets all the preset constraints, but also is the best solution obtained after optimization.

[0137] In the intelligent dosing regulation system for sewage treatment based on fuzzy logic described in the embodiment of the present invention, the detection module 16 detects the sewage water quality parameter data after dosing through Kalman filtering, and feeds back the treated water quality parameters to the fuzzy logic controller, the data analysis and optimization module to adjust the dosing strategy, including:

[0138] Collect various water quality parameter data of sewage after adding drugs;

[0139] The collected sewage water quality parameter data is processed by Kalman filtering to obtain the processed water quality parameter data;

[0140] The water quality parameter data after Kalman filtering is transmitted to the fuzzy logic controller, data analysis and optimization module to identify the changing trend and periodic fluctuation data of water quality parameters and obtain the range of water quality parameters;

[0141] Compare the water quality parameter range with the expected range, and adjust the dosing strategy based on the comparison results, including changing the type, dosage and time of the agent.

[0142] In an embodiment of the present invention, Kalman filtering is an efficient recursive filter that can estimate the state of a dynamic system from a series of incomplete and noisy measurements. In sewage treatment, the sewage water quality parameter data collected by Kalman filtering can effectively remove noise and outliers, and improve the accuracy and reliability of the data. The water quality parameter data processed by Kalman filtering is transmitted to the fuzzy logic controller, data analysis and optimization module in real time, and can quickly identify the changing trend and periodic fluctuation of water quality parameters. By comparing the range of water quality parameters after treatment with the expected range, the effectiveness of the current dosing strategy can be accurately judged, and the dosing strategy can be adjusted according to the comparison results, including changing the type, dosage and time of the agent; so as to achieve more accurate sewage treatment and agent use efficiency. The close cooperation between the detection module and other modules makes the entire sewage treatment system more intelligent, and can automatically adjust the dosing strategy according to the changes in water quality parameters, reducing the need for human intervention and improving work efficiency and accuracy.

[0143] In the specific implementation process of the present invention, the specific steps include:

[0144] Various sensors are deployed at key nodes in the sewage treatment process to monitor and collect sewage water quality parameter data in real time. Data is continuously collected through the sensor network, and necessary preprocessing is performed on the collected raw data, such as removing outliers and smoothing, to reduce the impact of noise and interference on subsequent analysis.

[0145] The Kalman filtering algorithm is used to further process the preprocessed data. In the Kalman filtering process, the current state is first predicted based on the system's dynamic model and the previous state estimate. Then, the current actual measurement value is used to update the prediction, and continuous iterations are performed to obtain a more accurate state estimate.

[0146] The water quality parameter data after Kalman filtering is fed back to the fuzzy logic controller and data analysis and optimization module in real time. In the data analysis and optimization module, statistical analysis and machine learning techniques are used to identify the changing trends and periodic fluctuations of water quality parameters and predict future water quality conditions; based on historical data and current data, the normal range or expected range of water quality parameters is determined.

[0147] Compare the range of treated water quality parameters with the expected range to evaluate the current water quality and dosing effect. Based on the comparison results, if the water quality parameters exceed the expected range or show an unfavorable trend, the dosing strategy adjustment will be triggered. The adjustment may include changing the type of agent, adjusting the dosage, optimizing the dosing time, etc. The adjusted dosing strategy will be sent to the dosing module for execution, and the changes in water quality parameters will be continuously monitored to verify the effectiveness of the strategy. If necessary, further adjustments and optimizations can be made based on the actual results.

[0148] like Figure 2 As shown, a control method for a sewage treatment intelligent dosing regulation system based on fuzzy logic includes the following steps:

[0149] Collect and optimize wastewater quality data, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content, and wastewater flow;

[0150] Perform a small batch gradient descent analysis on the data to obtain the analysis results;

[0151] According to the analysis results, the parameters and rules of the fuzzy logic controller are optimized by differential evolution to obtain the optimized parameters and rules;

[0152] According to the water quality parameters and the optimized rules, the fuzzy logic calculation of the dosage is performed through simulated annealing and converted into a dosing control signal;

[0153] According to the dosing control signal, the constraint conditions of the dosage are adjusted by the projection subgradient method, and integer programming is performed in the dosing process in combination with the interior point method to achieve automatic adjustment of the dosage.

[0154] The sewage water quality parameter data after dosing is detected by Kalman filtering, and the treated water quality parameters are fed back to the fuzzy logic controller, data analysis and optimization module to adjust the dosing strategy.

[0155] In the embodiment of the present invention, by collecting key water quality data in sewage, the water quality status can be monitored in real time. The dynamic programming algorithm is used to optimize the data collection process, which can ensure that the most critical data is collected in the most efficient way under limited resources. The system can quickly identify trends and patterns in the data by analyzing water quality data through a small batch gradient descent algorithm. The application of the differential evolution algorithm optimizes the parameters and rules of the fuzzy logic controller, thereby improving the performance and accuracy of the controller when dealing with complex and uncertain problems. Through the simulated annealing algorithm, the system can calculate the optimal dosage under the consideration of multiple factors, and convert it into an actual control signal through fuzzy logic, thereby realizing the intelligence and precision of the dosing process. Combined with the projected subgradient method and the interior point method, the system can realize automatic adjustment and integer programming of the dosage while meeting various constraints, further improving the efficiency and reliability of the dosing process. The application of Kalman filtering technology helps to reduce noise and interference in water quality parameter detection and improve detection accuracy. By feeding back the treated water quality parameters to the fuzzy logic controller and the data analysis module, the system can continuously adjust and optimize the dosing strategy according to the actual situation, thereby realizing continuous improvement and adaptive control of the sewage treatment process.

[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A sewage treatment intelligent dosing regulation system based on fuzzy logic, characterized in that: include: Data collection module, which collects sewage water quality data and optimizes the data collection process, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow; The data analysis module is used to receive real-time data and perform small batch gradient descent analysis on the data to obtain analysis results; An optimization module is used to optimize the parameters and rules of the fuzzy logic controller through differential evolution according to the analysis results to obtain optimized parameters and rules; The fuzzy logic control module performs fuzzy logic calculation of the dosage through simulated annealing according to the received water quality parameters and the optimized rules, and converts it into a dosage control signal; The dosing module, after receiving the dosing control signal, adjusts the constraint conditions of the dosing amount through the projected subgradient method, and combines the interior point method to perform integer programming in the dosing process to achieve automatic adjustment of the dosing amount; The detection module detects the sewage water quality parameter data after drug addition through Kalman filtering, and feeds back the treated water quality parameters to the fuzzy logic controller, data analysis and optimization module to adjust the drug addition strategy; Among them, receiving real-time data and performing small batch gradient descent analysis on the data to obtain analysis results, including: Divide the preprocessed real-time data into multiple small batch data sets; Initialize the learning rate and number of iterations; according to each mini-batch dataset, Calculate the average error between the predicted values ​​and the actual values; According to the average error, Adjust the gradient of the weights by Adjust the gradient of the bias to minimize the gap between the predicted value and the actual value to obtain the final solution, where is the weight parameter, is the bias parameter, It is The sewage quality data values ​​of samples, It is The actual target value of samples, is the number of samples in the current mini-batch dataset, is the average error between the predicted value and the actual value, is the gradient of the weight, is the gradient of the bias; According to the final solution, the optimized parameters are obtained; The dosage is predicted according to the optimized parameters to obtain a dosage prediction result, wherein the prediction result is an analysis result.

2. The intelligent dosing and regulating system for sewage treatment based on fuzzy logic according to claim 1 is characterized in that: Collect sewage water quality data and optimize the data collection process, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow, including: Determine the frequency, cycle and strategy of data collection based on historical data and changing trends of water quality parameters; According to the frequency, cycle and strategy of collection, water quality parameter data are collected and preprocessed to obtain preprocessed data. The water quality parameter data include chemical oxygen demand, ammonia nitrogen content, total phosphorus content and sewage flow.

3. The intelligent dosing and regulating system for sewage treatment based on fuzzy logic according to claim 1 is characterized in that: According to the analysis results, the parameters and rules of the fuzzy logic controller are optimized by differential evolution to obtain the optimized parameters and rules, including: Determine the optimization target based on the dosage prediction results; Initialize the population and generate a set of random fuzzy logic controller parameters and rules as initial candidate solutions; The parameters and rules in the population are combined and applied to the fuzzy logic controller, and the fitness of the controller is evaluated based on the dosage prediction results; Perform mutation, crossover and selection operations on the individuals in the population to generate new candidate solutions, and select the corresponding individuals to form a new population according to the fitness value until the preset number of iterations is reached, then stop the iteration. After the iteration is completed, obtain the final individuals in the current population as the optimized parameters and rules.

4. The intelligent dosing and regulating system for sewage treatment based on fuzzy logic according to claim 1 is characterized in that: According to the received water quality parameters and optimized rules, the fuzzy logic calculation of the dosage is performed through simulated annealing and converted into a dosing control signal, including: Obtain water quality parameter data and optimized rules; Initialize simulated annealing algorithm parameters; According to the optimized fuzzy logic rules, a fuzzy logic system is constructed. The system includes a fuzzification interface, a fuzzy inference engine and a defuzzification interface. The water quality parameter data is converted into a fuzzy set. According to the fuzzy logic rules, the fuzzy dosage decision is obtained and the fuzzy dosage decision is converted into a specific dosage value. Initialize the simulated annealing algorithm parameters, execute the simulated annealing process, generate a new dosing decision based on the current temperature and state in each iteration, and use the fuzzy logic system to evaluate its advantages and disadvantages to obtain the final solution; When the simulated annealing algorithm reaches the termination condition, the current final dosage is obtained; The dosage calculated by simulated annealing and fuzzy logic is converted into a specific dosage control signal.

5. The intelligent dosing and regulating system for sewage treatment based on fuzzy logic according to claim 1 is characterized in that: After receiving the dosing control signal, the constraint conditions of the dosing amount are adjusted by the projected subgradient method, and integer programming is performed in the dosing process in combination with the interior point method to achieve automatic adjustment of the dosing amount, including: Receive drug dosing control signals and preset drug dosage constraints; According to the preset drug dosage constraints and the current control signal, the drug dosage is optimized by the projected subgradient method to obtain the dosage that meets all the constraints; Under the premise of meeting all constraints, find the corresponding dosage plan and automatically adjust the dosage.

6. The intelligent dosing and regulating system for sewage treatment based on fuzzy logic according to claim 1 is characterized in that: According to the preset constraints and the current control signal, the dosage is optimized by the projected subgradient method to obtain the dosage that satisfies all constraints, including: Set relevant parameters, including learning rate sequence, stopping conditions, and constraint sets; According to the current dosage plan, the sub-gradient of the objective function is calculated. The sub-gradient is the sum of the absolute values ​​of the components in the dosage plan. Use the projection operation to project the updated dosage scheme into the constraint set to obtain a new dosage scheme, set the new dosage scheme as the current dosage scheme, and prepare for the next iteration; If the improvement of the objective function value is less than the preset threshold, the iteration is stopped and the dosage scheme is obtained.

7. The intelligent dosing and regulating system for sewage treatment based on fuzzy logic according to claim 1 is characterized in that: The Kalman filter is used to detect the sewage water quality parameter data after dosing, and the treated water quality parameters are fed back to the fuzzy logic controller, data analysis and optimization module to adjust the dosing strategy, including: Collect various water quality parameter data of sewage after adding drugs; The collected sewage water quality parameter data is processed by Kalman filtering to obtain the processed water quality parameter data; The water quality parameter data after Kalman filtering is transmitted to the fuzzy logic controller, data analysis and optimization module to identify the changing trend and periodic fluctuation data of water quality parameters and obtain the range of water quality parameters; Compare the water quality parameter range with the expected range, and adjust the dosing strategy based on the comparison results, including changing the type, dosage and time of the agent.

8. A sewage treatment intelligent dosing adjustment method based on fuzzy logic, characterized in that: Applied to the system according to any one of claims 1 to 7, the method comprises: Collect and optimize wastewater quality data, including chemical oxygen demand, ammonia nitrogen content, total phosphorus content, and wastewater flow; Perform a small batch gradient descent analysis on the data to obtain the analysis results; According to the analysis results, the parameters and rules of the fuzzy logic controller are optimized by differential evolution to obtain the optimized parameters and rules; According to the water quality parameters and the optimized rules, the fuzzy logic calculation of the dosage is performed through simulated annealing and converted into a dosing control signal; According to the dosing control signal, the constraint conditions of the dosage are adjusted by the projection subgradient method, and integer programming is performed in the dosing process in combination with the interior point method to achieve automatic adjustment of the dosage. The sewage water quality parameter data after dosing is detected by Kalman filtering, and the treated water quality parameters are fed back to the fuzzy logic controller, data analysis and optimization module to adjust the dosing strategy.

Images