AI-based intelligent urban water purification sewage treatment process optimization system and method

Abstract

The invention relates to the technical field of water treatment intelligent control, in particular to an AI-based intelligent urban water purification and sewage treatment process optimization system and method, and the method comprises the steps: obtaining related sensor data information, and monitoring and recording the change of key data in real time; establishing a water quality change prediction model by using a multiple linear regression model, wherein the water quality change prediction model predicts the water quality change according to the key data change acquired in real time; the dynamic agent AI model adjusts the dosage, and the water treatment chemical / biological reaction kinetics AI model predicts the pollutant degradation condition; a local lightweight AI model optimizer and a multi-target genetic algorithm are combined with energy consumption, medicament cost, processing efficiency and carbon emission to obtain an optimal control model; the optimal control model optimizes the operation parameters of the sewage treatment process according to the pollutant degradation condition. According to the method, the operation parameters of the sewage treatment process are optimized through the optimal control model, the energy consumption, the agent cost and the carbon emission are reduced, and the treatment efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of water treatment, and particularly relates to an AI-based intelligent optimization system and method for urban water purification and sewage treatment processes. Background Art

[0002] In recent years, the water quality and quantity characteristics of municipal sewage have changed significantly, with the obvious increase in the nitrogen and phosphorus content in the sewage. Relevant departments have imposed stricter supervision on the operation of sewage treatment plants. However, there are still many intractable problems in the operation of sewage treatment plants, and the activated sludge process has become a difficult problem for research and improvement due to its own deficiencies. From the previous level B to level A and then to quasi-class IV, and from mainly removing carbon and phosphorus to taking nitrogen and phosphorus removal as the core, national standards and local standards have continuously raised the requirements for the effluent water quality and discharge standards of sewage treatment plants. Among them, the problem of stable compliance of effluent TN is particularly severe.

[0003] At the present stage, traditional water purification and sewage treatment all have various problems such as relying on experience to adjust parameters, lagging response to water quality fluctuations, and over-relying on chemical agents. This operation method will result in situations such as excessive energy consumption, excessive dosage of agents leading to high costs, or insufficient agents resulting in incomplete sewage treatment. Summary of the Invention

[0004] The present invention can adjust the operation parameters of the sewage treatment process in real time according to each model, balance energy consumption, agent cost, treatment efficiency, and carbon emissions, and obtain an optimal control method.

[0005] The technical solution proposed by the present invention is: an AI-based intelligent optimization method for urban water purification and sewage treatment processes, the method comprising: Obtain relevant sensor data information, and monitor and record the changes of key data in real time; Establish a water quality change prediction model using a multiple linear regression model, and the water quality change prediction model predicts the water quality change in the future period according to the changes of key data collected in real time; The dynamic agent AI model adjusts the dosage of agents in the future period according to the water quality change in the future period, and the water treatment chemical / biological reaction kinetics AI model predicts the degradation of pollutants in the future period; The local lightweight AI model optimizer combines with a multi-objective genetic algorithm to obtain an optimal control model considering energy consumption, agent cost, treatment efficiency, and carbon emissions; The optimal control model obtains the optimized operation parameters of the sewage treatment process according to the predicted degradation of pollutants in the future period; Switch to a traditional PID control system to monitor the actual operation parameters of the sewage treatment process in real time, calculate the deviation between the actual operation parameters of the sewage treatment process and the optimized operation parameters of the sewage treatment process, and perform feedback adjustment according to the deviation.

[0006] Preferably, the water quality change prediction model includes the following steps: Based on historical key data, the relevant regression coefficients of each key data are calculated using the least squares method. The sum of the products of the currently monitored key data and the corresponding relevant regression coefficients is obtained to determine the water quality change situation, and the changes in key data for a future period are analyzed based on the water quality change situation.

[0007] Preferably, the energy consumption is obtained from the aeration intensity and the equipment operation time. The coefficients corresponding to the aeration intensity and the equipment operation time are fitted based on the historical energy consumption data of the equipment, and the sum of the products of the preset aeration intensity and the preset equipment operation time and the corresponding coefficients is obtained to predict the energy consumption; the chemical agent cost is the product of the chemical agent dosage and the unit price of the chemical agent.

[0008] Preferably, the carbon emissions are the product of the energy consumption and the carbon emission coefficient generated by certain chemical reactions in the treatment process; the treatment efficiency is the quotient of the water treatment volume per unit time and the coefficient of the complexity of the influent water quality. The sum of the products of chemical oxygen demand, ammonia nitrogen, total nitrogen, and total phosphorus and the corresponding coefficients is the coefficient of the complexity of the influent water quality.

[0009] Preferably, the specific formula of the optimal control model is as follows: ; Where: , , , are weight coefficients, and , and the weight coefficients are determined according to the multi-objective optimization algorithm; is the comprehensive objective function. The comprehensive objective function is calculated in real time. If the comprehensive objective function increases, the operating parameters of the sewage treatment process need to be adjusted.

[0010] Preferably, the multi-objective genetic algorithm uses the comprehensive objective function as the optimization objective orientation, and the specific content is as follows: By continuously iterating selection, crossover, and mutation, different combinations of control parameters are tried, and the values of the comprehensive objective function corresponding to these combinations are calculated. The roulette wheel selection method is used to select according to the fitness values of the individuals, and the selection probability of each group of individuals is calculated. A probability threshold is set to retain excellent combinations of control parameters to form a new population. Two individuals are randomly selected from the new population as parents to generate offspring individuals, and a certain mutation probability is used to mutate a certain control parameter in the offspring individuals. The iteration is repeated. When the termination condition is met, the combination of control parameters corresponding to the individual with the minimum comprehensive objective function is the optimal solution.

[0011] Preferably, the specific content of the AI model for water treatment chemical / biological reaction kinetics is as follows: According to the Arrhenius equation, considering the influence of pH value on the reaction and introducing a correction coefficient, the actual reaction rate constant is as follows: ; Where: is the correction coefficient; is the temperature; is the reaction rate constant; is a constant; is a constant.

[0012] The pollutant degradation formula is as follows: ; The integral form is: ; Where: is the initial pollutant concentration, and the correction coefficient is obtained by fitting historical data, , where is the fitting coefficient.

[0013] Preferably, the specific content of the dynamic chemical agent AI model is as follows: Calculate the deviation between the actual concentration and the predicted concentration of a certain pollutant in the current water quality. According to the chemical agent dosage calculation formula, establish a minimization objective function formula, take the derivative of the chemical agent dosage adjustment coefficient and set the derivative to 0 to solve for the specific value of the chemical agent dosage adjustment coefficient. The specific value of the chemical agent dosage is obtained by adding the product of the initial chemical agent dosage, the chemical agent dosage adjustment coefficient, and the deviation.

[0014] Preferably, when the switched traditional PID control system monitors that there is a deviation between the actual sewage treatment process operation parameters and the optimized sewage treatment process operation parameters, the control system is switched to the traditional PID control system and the operator is notified. The operator manually adjusts the sewage treatment process operation parameters according to the deviation value and checks whether the equipment fails.

[0015] The present invention also provides an AI-based intelligent urban water purification and sewage treatment process optimization system, and the system is used to execute the AI-based intelligent urban water purification and sewage treatment process optimization method described above.

[0016] Advantages of the present invention: Using a multiple linear regression model, based on a large amount of historical key data (13 parameters such as chemical oxygen demand, ammonia nitrogen, total nitrogen, total phosphorus, dissolved oxygen, etc.), the least squares method is used to calculate the relevant regression coefficients of each key data, so as to construct a water quality change prediction model. This model can predict the water quality change situation in the future period according to the change of real-time collected key data. The model comprehensively considers a variety of key parameters that affect water quality, and determines the influence weights (regression coefficients) of each parameter through historical data training. When the key parameters change are monitored in real time, the model can accurately calculate the predicted values of water quality change indicators based on the trained relationship. For example, if the regression coefficient corresponding to the ammonia nitrogen parameter is large and the real-time ammonia nitrogen value rises, the model can accurately reflect its impact on the overall water quality change (such as the increase of chemical oxygen demand, etc.). This enables the sewage treatment plant managers to plan in advance for equipment maintenance, chemical procurement, personnel allocation and other work. If the water quality is predicted to be stable, the equipment maintenance cycle can be extended. If the water quality is predicted to fluctuate, chemicals can be stocked up in advance and personnel monitoring can be strengthened, effectively improving the forward-looking and decision-making scientificity of the sewage treatment plant in dealing with water quality changes.

[0017] The local lightweight AI model optimizer is combined with the multi-objective genetic algorithm, comprehensively considering factors such as energy consumption, chemical cost, treatment efficiency and carbon emissions, to construct a comprehensive objective function. Through continuous iteration of selection, crossover and mutation of the multi-objective genetic algorithm, with the minimum of the comprehensive objective function as the guidance, the optimal control parameters are obtained. The water treatment chemical / biological reaction kinetics AI model can more accurately predict the degradation of pollutants compared with the traditional model that does not consider these factors because it takes into account various factors affecting the reaction rate such as temperature and pH value. For example, under different temperature and pH value conditions, it can accurately predict the pollutant degradation rate, providing a more accurate reference for the reaction process of the sewage treatment process. The dynamic chemical AI model accurately adjusts the chemical dosage based on the deviation between the actual concentration and the predicted concentration, avoiding the problem of excessive or insufficient chemicals caused by traditional experience-based chemical dosing. When the actual concentration is higher than the predicted concentration, the model calculates to increase the chemical dosage, and vice versa, so as to achieve accurate chemical dosing, improve the chemical utilization efficiency, reduce the chemical cost, and ensure the sewage treatment effect at the same time.

[0018] Adopt a switched traditional PID control system to monitor in real time the deviation between the actual operation parameters of the sewage treatment process and the optimized operation parameters. When a deviation occurs, switch to the traditional PID control system and notify the operator. The operator manually adjusts the operation parameters of the sewage treatment process according to the deviation value and checks whether the equipment fails. During the sewage treatment process, due to various unexpected factors (such as equipment failure, sudden significant change in water quality, etc.), the optimized operation parameters may not be effectively implemented. At this time, the switched traditional PID control system can respond quickly and maintain the basic operation of the sewage treatment system. The traditional PID control system ensures a certain stability of the system under abnormal conditions through feedback regulation of the deviation (such as adjusting the valve opening to control the influent volume, adjusting the power of the aeration equipment to control the aeration intensity, etc.). The operator's manual adjustment according to the deviation value and equipment failure detection can timely discover and solve problems, prevent the deterioration of the treatment effect caused by equipment failure or parameter deviation, and ensure the stability and reliability of the operation of the entire sewage treatment system. Description of the Drawings

[0019] Figure 1 It is a flowchart of an AI-based intelligent urban water purification and sewage treatment process optimization system and method of the present invention; Figure 2 It is a control system switching flowchart of an AI-based intelligent urban water purification and sewage treatment process optimization system and method of the present invention. Detailed Embodiments

[0020] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations. The basic principles defined in the following description can be applied to other implementation schemes, variant schemes, improvement schemes, equivalent schemes, and other technical schemes that do not depart from the spirit and scope of the present invention.

[0021] It can be understood that the term "a" should be understood as "at least one" or "one or more". That is, in one embodiment, the number of an element can be one, while in other embodiments, the number of the element can be multiple. The term "a" cannot be understood as a limitation on the number.

[0022] As Figure 1 shown, the present invention sets relevant sensors at positions such as the regulating tank, pretreatment system, biochemical system, sedimentation tank, ozone oxidation, biological filtration system, disinfection link, etc. to collect relevant data information, and monitors and records in real time the changes of key parameters such as chemical oxygen demand (COD), ammonia nitrogen, total nitrogen, total phosphorus, dissolved oxygen, sludge concentration meter, sludge interface meter, pH value, flow rate, turbidity, residual chlorine, conductivity, temperature, etc.

[0023] Let the water quality change index (such as pollutant concentration) be , and the key parameters collected by the sensor be independent variables, including chemical oxygen demand ( ), ammonia nitrogen ( ), total nitrogen ( ), total phosphorus ( ), dissolved oxygen ( ), sludge concentration meter ( ), sludge interface instrument ( ), pH value ( ), flow rate ( ), turbidity ( ), residual chlorine ( ), conductivity ( ), temperature ( ).

[0024] The formula for the multiple linear regression model is: ; where: is the intercept; is the error term; is the regression coefficient. First, use a large amount of historical data that has been collected to train this model. By using methods such as the least squares method, the regression coefficient can be estimated, thus establishing a model for predicting water quality changes, predicting the water quality changes in real time, and adjusting relevant content according to the water quality changes. Find the influence degree of each key parameter on the water quality change index, that is, determine each coefficient in the model. After training the model, collect the current key parameter values in real time, substitute these values into the model, and a predicted value of the water quality change index can be calculated. This value can let us understand how the water quality will probably change in the future.

[0025] For example, if it is predicted that the value of the water quality change index (such as chemical oxygen demand COD) will increase, then it can be further analyzed which key parameters are playing a role. If it is found that the coefficient corresponding to the key parameter of ammonia nitrogen is relatively large, and the current real-time value of ammonia nitrogen is also rising, then it can be speculated that the change of the key parameter of ammonia nitrogen may be an important reason for the predicted water quality change (increase in COD). Similarly, if the coefficient of total phosphorus is not small and the real-time value of total phosphorus also changes, then the change of total phosphorus also has an impact on the water quality change. In this way, from the predicted water quality change situation, combined with the coefficients of the key parameters in the model and the real-time collected key parameter values, analyze the influence of each key parameter on the water quality change, and then roughly judge how the key parameters change and their relative importance to the overall water quality change.

[0026] The management staff of the sewage treatment plant can plan equipment maintenance plans, purchase chemicals, adjust personnel allocation, etc. according to the predicted water quality changes in the next period. When the model predicts that the water quality will be relatively stable in the next period, the equipment maintenance cycle can be appropriately extended. If it is predicted that the water quality will fluctuate greatly, sufficient chemicals need to be reserved in advance, and professional technical personnel need to be arranged to strengthen the monitoring and maintenance of the system.

[0027] Moreover, the water quality prediction model can accurately predict water quality changes, which helps to optimize the entire sewage treatment process. If the model predicts that a certain water quality index (such as ammonia nitrogen) is about to exceed the discharge standard, the aeration intensity, microbial culture conditions, etc. in the biochemical system can be adjusted in advance to enhance the removal ability of ammonia nitrogen. By continuously optimizing the treatment process according to the prediction results, the sewage treatment efficiency can be improved, the treatment cost can be reduced, and at the same time, the effluent water quality can be ensured to meet the standards stably. For example, when the model predicts that the total phosphorus content will increase, the dosage of phosphorus removal chemicals can be increased in a timely manner or the phosphorus removal process parameters can be adjusted to avoid excessive total phosphorus discharge. It can also provide key references for the system operation status. When abnormal fluctuations occur in water quality changes, the prediction results of the model can help the operation and maintenance personnel quickly judge the problem. For example, if the model predicts that the dissolved oxygen value decreases abnormally and other related water quality indicators also change abnormally, it may mean that the growth environment of microorganisms in the biochemical system is affected, which may in turn affect the stable operation of the entire sewage treatment system. The operation and maintenance personnel can check the equipment operation status and adjust the process parameters in a timely manner according to the model prediction results to ensure the stable operation of the system and reduce system failures and downtime caused by water quality changes.

[0028] Through the local lightweight AI model optimizer, combining the prediction module with the multi-objective genetic algorithm, optimal control parameters are generated to achieve multi-objective optimizations such as reducing energy consumption, improving treatment efficiency, reducing chemical costs, and balancing carbon emissions.

[0029] The specific formula steps for energy consumption are as follows: ; Where: is the energy consumption; is the aeration intensity; is the equipment operation time; , are constants, which can be obtained by fitting historical data according to the energy consumption characteristics of the equipment.

[0030] The formula for chemical cost is as follows: ; Where: is the chemical cost; is the chemical dosage; is the unit price of chemicals.

[0031] The calculation formula for the treatment efficiency is as follows: ; Where: is the treatment efficiency; is the water treatment volume per unit time; is the complexity coefficient of the influent water quality, which can be calculated through a certain algorithm based on key parameters such as chemical oxygen demand (COD), ammonia nitrogen, total nitrogen, and total phosphorus ; Where: is the weight coefficient, obtained through training with historical data; is the chemical oxygen demand; is the ammonia nitrogen; is the total nitrogen; is the total phosphorus.

[0032] The calculation formula for carbon emissions is as follows: ; Where: is the carbon emission; is the energy consumption; is the carbon emission coefficient generated by certain chemical reactions in the treatment process.

[0033] Combining energy consumption, chemical agent cost, treatment efficiency, and carbon emissions to obtain the comprehensive objective function. The specific formula for the comprehensive objective function is as follows: ; Where: , , , are weight coefficients, and , and these weight coefficients can be determined through expert evaluation or multi-objective optimization algorithms according to actual requirements and the importance of each objective.

[0034] Comprehensive objective function The value of can be used as a quantitative indicator to evaluate the performance of the entire sewage treatment system. During the operation of the system, calculate the value in real time. According to the value change situation, it is possible to intuitively understand the performance of the system in terms of multi-objective optimization. If the value gradually decreases, it indicates that the system is operating in a better direction, and the comprehensive optimization effect of each objective is getting better; conversely, if the value increases, it indicates that there may be problems in some aspects of the system, and it is necessary to adjust the operating parameters or treatment strategies to ensure that the system continues to operate efficiently, energy-saving, and environmentally friendly.

[0035] In the optimization process such as multi-objective genetic algorithm, the comprehensive objective function As the goal of optimization, we try different combinations of control parameters (such as aeration intensity) through continuous iteration of selection, crossover, mutation and other operations. , Equipment running time , dosage , Water volume treated per unit time etc.), calculate the comprehensive objective function values corresponding to these combinations .by For example, the smaller the better (when the fitness value is set to 1 / ), the algorithm moves towards The system searches in the direction with decreasing values, thereby guiding the system to find the optimal control parameter combination that can simultaneously reduce energy consumption, improve processing efficiency, reduce reagent costs, and balance carbon emissions, providing a clear direction for system operation optimization.

[0036] Calculate each individual according to the following calculation formula of value: ; The roulette wheel selection method is used to select individuals based on their fitness values. The probability of each individual being selected is calculated as: ; in: Is an individual The fitness value of is the population size. Through the random number generator, according to the probability Select individuals to form a new population. Individuals with smaller values have a greater probability of being selected, thereby retaining excellent control parameter combinations.

[0037] Randomly select two individuals from the new population as parents, e.g. and , and Represent different control parameters. Randomly select a crossover point to generate offspring individuals and The crossover operation helps to generate new control parameter combinations and explore a better solution space.

[0038] For individuals after crossover, with a certain mutation probability Mutate a gene (i.e. a control parameter) in an individual. ,if Mutation may become , is the variation quantity, which can be determined according to certain rules, such as randomly taking values within a certain range. The mutation operation increases the diversity of the population and prevents the algorithm from falling into a local optimal solution.

[0039] Repeat the above steps, continuously iteratively calculate the fitness values of the individuals in the new population, and perform selection, crossover, and mutation operations. Each time an iteration is performed, the individuals in the population gradually evolve towards a better direction, that is, the comprehensive objective function value gradually decreases. Set termination conditions, such as reaching a certain number of iterations, or the value changes very little (less than the set threshold) in several iterations. When the termination condition is met, at this time, among the individuals in the population the combination of control parameters corresponding to the individual with the smallest value is the optimal solution obtained by solving under the current model formula and conditions, and the corresponding value is the optimal comprehensive objective function value.

[0040] Take the water treatment chemical / biological reaction kinetics AI model and the dynamic reagent AI model as part of the dynamic strategy execution module to improve the optimization rationality of prediction. The specific content of the water treatment chemical / biological reaction kinetics AI model is as follows: According to the Arrhenius equation, while considering the influence of pH value on the reaction, a correction coefficient is introduced, and the actual reaction rate constant is as follows: ; Among them: is the correction coefficient; is the temperature; is the reaction rate constant; is a constant; is a constant.

[0041] The pollutant degradation formula is as follows: ; The integral form is , where is the initial pollutant concentration. The correction coefficient is obtained by fitting experimental data, , where is the fitting coefficient.

[0042] The specific content of the dynamic reagent AI model is as follows: Let the actual concentration of a certain pollutant in the current water quality be , the predicted concentration be , the predicted concentration is related to the predicted water quality change index , and can be regarded as , the deviation . The chemical dosage adjustment coefficient is , and the initial chemical dosage is , then the adjusted chemical dosage . The chemical dosage adjustment coefficient can be obtained by machine learning algorithms, such as least squares fitting of historical data. Suppose there are groups of data in the historical data, and each group of data contains the current water quality pollutant concentration , predicted concentration and actual chemical dosage . By minimizing the objective function , taking the derivative of and setting the derivative to 0, the value of can be solved.

[0043] According to the feedback of the above model, the water inflow, reflux ratio, chemical dosage and dosing frequency, aeration intensity and time, air-water backwashing intensity and air-water backwashing period of the membrane tank and filter tank are adjusted in real time to optimize the system, reduce costs and energy consumption.

[0044] For example Figure 2 as shown, the intelligent actuator linkage system switches the functions of the traditional PID control system, switches between the automatic control system and the traditional PID control system, and further ensures the stability of the overall project system operation.

[0045] Control output , where is the deviation between the system set value and the actual measured value. Taking the control of the water inflow as an example, suppose the set value of the water inflow is , and the actually measured water inflow is , then . is the proportional coefficient, is the integral coefficient, is the differential coefficient. These coefficients can be tuned by methods such as the empirical method and the Ziegler-Nichols method to achieve the best control effect. For example, adjusting the valve opening to control the water inflow and adjusting the power of the aeration equipment to control the aeration intensity to ensure the stable operation of the system.

[0046] The control output formula can comprehensively consider the relationships between multiple factors, combine with AI models such as water treatment chemical / biological reaction kinetics, and dynamically adjust the control parameters according to different water quality conditions and treatment requirements to optimize the entire sewage treatment process, reduce costs such as energy consumption and chemical consumption, and at the same time ensure that the effluent water quality meets the standards. Feedback regulation is carried out according to the real-time state of the system. When there are disturbances or deviations in the system, the corresponding control signals are calculated and output in a timely manner to make the system return to a stable operating state, avoiding large fluctuations in the treatment effect or system out-of-control caused by external factors. The deviation is corrected manually by the control personnel, and at the same time, the system equipment is repaired.

[0047] Embodiments disclosed by the present invention. The processes described above with reference to the flowcharts can be implemented as computer software programs. Embodiments disclosed by the present invention include a computer program product, which includes a computer program carried on a computer-readable medium. The computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), the above-mentioned functions defined in the methods of the present application are executed. It should be noted that the above-mentioned computer-readable medium in the present application can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium can include, but are not limited to: an electrical connection having one or more wire segments, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or combined with an instruction execution system, apparatus, or device. In the present application, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program codes. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, and this computer-readable medium can send, propagate, or transmit a program for use by or combined with an instruction execution system, apparatus, or device. The program codes contained on the computer-readable medium can be transmitted by any appropriate medium, including but not limited to: wireless segments, wire segments, optical cables, RF, etc., or any suitable combination of the above.

[0048] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks may occur in a different order than that noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0049] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and illustrated in the embodiments. Without departing from the said principles, the embodiments of the present invention may have any variations or modifications.

Claims

1. An AI-based intelligent optimization method for urban water purification and sewage treatment process, characterized in that, The method includes: Obtain relevant sensor data information, and monitor and record the changes of key data in real time; Use a multiple linear regression model to establish a water quality change prediction model, which predicts the water quality change in the future period according to the changes of key data collected in real time; The dynamic chemical dosing AI model adjusts the chemical dosing amount in the future period according to the water quality change in the future period, and the water treatment chemical / biological reaction kinetics AI model predicts the pollutant degradation situation in the future period; The local lightweight AI model optimizer combines with the multi-objective genetic algorithm to obtain the optimal control model considering energy consumption, chemical dosing cost, treatment efficiency and carbon emissions; The optimal control model obtains the optimized operation parameters of the sewage treatment process according to the predicted pollutant degradation situation in the future period; Switch to the traditional PID control system to monitor the actual operation parameters of the sewage treatment process in real time, calculate the deviation between the actual operation parameters of the sewage treatment process and the optimized operation parameters of the sewage treatment process, and perform feedback adjustment according to the deviation.

2. The optimization method for the intelligent urban water purification and sewage treatment process based on AI according to claim 1, characterized in that The water quality change prediction model includes the following steps: Through historical key data, use the least squares method to calculate the relevant regression coefficients of each key data, add the product of the currently monitored key data and the corresponding relevant regression coefficients to obtain the water quality change situation, and analyze the changes of key data in the future period according to the water quality change situation.

3. The optimization method for the intelligent urban water purification and sewage treatment process based on AI according to claim 2, wherein, The energy consumption is obtained through the aeration intensity and the equipment operation time. The coefficients corresponding to the aeration intensity and the equipment operation time are fitted according to the historical energy consumption data of the equipment, and the product of the preset aeration intensity and the preset equipment operation time and the corresponding coefficients is added to obtain the predicted energy consumption; the chemical dosing cost is the product of the chemical dosing amount and the chemical unit price.

4. An AI-based intelligent urban water purification and sewage treatment process optimization method according to claim 3, characterized in that The carbon emissions are the product of the energy consumption and the carbon emission coefficient generated by some chemical reactions in the treatment process; the treatment efficiency is the quotient of the water treatment volume per unit time and the coefficient of the complexity of the influent water quality. The sum of the products of chemical oxygen demand, ammonia nitrogen, total nitrogen and total phosphorus and the corresponding coefficients is the coefficient of the complexity of the influent water quality.

5. The optimization method for the intelligent urban water purification and sewage treatment process based on AI according to claim 1, characterized in that, The specific formula of the optimal control model is as follows: ； Wherein: , , , are weight coefficients, and , and the weight coefficients are determined according to the multi-objective optimization algorithm; is the comprehensive objective function, the comprehensive objective function is calculated in real time, and if the comprehensive objective function becomes larger, the operating parameters of the sewage treatment process need to be adjusted.

6. The method for optimizing the intelligent urban water purification and sewage treatment process based on AI according to claim 5, characterized in that, The multi-objective genetic algorithm takes the comprehensive objective function as the optimization objective orientation, and the specific content is as follows: By continuously iterating selection, crossover and mutation, try different combinations of control parameters, and calculate the comprehensive objective function values corresponding to these combinations; Adopt the roulette wheel selection method, select according to the fitness value of the individual, calculate the probability of each group of individuals being selected, set the probability threshold to retain excellent combinations of control parameters to form a new population, and randomly select two groups of individuals from the new population as parents to generate offspring individuals; Mutate a certain control parameter in the offspring individuals with a certain mutation probability, repeat the iteration, and when the termination condition is met, the combination of control parameters corresponding to the individual with the minimum comprehensive objective function is the optimal solution.

7. An AI-based intelligent urban water purification and sewage treatment process optimization method according to claim 1, characterized in that, The specific content of the water treatment chemical / biological reaction kinetics AI model is as follows: According to the Arrhenius equation, considering the influence of pH value on the reaction at the same time, introducing a correction coefficient, the actual reaction rate constant is as follows: ； Wherein: is the correction coefficient; is the temperature; is the reaction rate constant; is a constant; is a constant; The pollutant degradation formula is as follows: ； The integral form is: ； Wherein: is the initial pollutant concentration, and the correction coefficient is obtained by fitting historical data, , where is the fitting coefficient.

8. The optimized method for the intelligent urban water purification and sewage treatment process based on AI according to claim 7, characterized in that, The specific content of the dynamic reagent AI model is as follows: Calculate the deviation between the actual concentration and the predicted concentration of a certain pollutant in the current water quality. Establish a minimization objective function formula according to the chemical dosing amount calculation formula. Take the derivative of the chemical dosing amount adjustment coefficient and set the derivative to 0 to solve for the specific value of the chemical dosing amount adjustment coefficient. Add the initial chemical dosing amount to the product of the chemical dosing amount adjustment coefficient and the deviation to obtain the specific value of the chemical dosing amount.

9. An AI-based intelligent urban water purification and sewage treatment process optimization method according to claim 1, characterized in that, When the switched traditional PID control system monitors that there is a deviation between the actual sewage treatment process operation parameters and the optimized sewage treatment process operation parameters, switch the control system to the traditional PID control system and notify the operator. The operator manually adjusts the sewage treatment process operation parameters according to the deviation value and checks whether the equipment fails.

10. An AI-based intelligent optimization system for urban water purification and sewage treatment processes, characterized in that, The system is used to execute an AI-based intelligent urban water purification and sewage treatment process optimization method according to any one of claims 1-9.

Images