A sewage plant aeration decision method, computer equipment and readable storage medium

Abstract

This invention discloses a wastewater treatment plant aeration decision-making method and equipment, comprising: acquiring wastewater treatment data to obtain a predicted aeration volume X(t) at time t; and acquiring a blower frequency combination {K} to achieve the predicted aeration volume X(t). i (N)}, where K i (N) represents the i-th satisfying blower frequency combination {N}. 1i (t),N 2i (t),…,N ni (t)}, i = 1, 2, ..., N ji (t) represents the frequency of the j-th blower, and n represents the number of blowers; obtain the blower frequency combination {K} i (N)} Blower frequency combination K with minimum power loss min (N)={N 1min (t),N 2min (t),…,N nmin (t)}, and control each blower to combine K at the blower frequency. min The invention operates at the frequency corresponding to (N). It achieves efficient control of the aeration process in wastewater treatment plants through an intelligent optimization algorithm, characterized by rapid convergence and high efficiency.

Description

Technical Field

[0001] This invention relates to an optimization method for aeration decision-making in the field of wastewater treatment, and particularly to an aeration decision-making method, computer equipment, and readable storage medium for wastewater treatment plants, which utilizes reinforcement learning and genetic algorithms to achieve efficient control of the aeration process in wastewater treatment plants. Background Technology

[0002] Wastewater treatment plants are vital facilities for urban environmental protection. The aeration process is a crucial step in wastewater treatment, significantly impacting treatment effectiveness and energy consumption. Therefore, effectively optimizing decision-making during the aeration process to achieve both optimal wastewater treatment results and reduced energy consumption is a significant current research topic.

[0003] Traditional aeration decision optimization methods are usually based on empirical or mathematical models, but due to the complexity of wastewater treatment processes, these methods often fail to achieve ideal results. Therefore, researchers have begun to explore the introduction of advanced optimization algorithms for decision optimization, among which genetic algorithms have received widespread attention. For example, Hai Trung Do et al.'s paper, "A design of higher-level control based genetic algorithms for wastewater treatment plants," published in *Engineering Science and Technology, an International Journal* (Vol. 24, No. 4, 2021, pp. 872-878), illustrates this. Genetic algorithms are global optimization algorithms that simulate natural selection and genetic mechanisms, enabling them to find the global optimum in large-scale, nonlinear, multi-peak, and multi-constraint complex optimization problems. Genetic algorithms, by simulating crossover, mutation, and selection processes in natural selection, can maintain diversity during the search process and avoid getting trapped in local optima. See Rainier Hreiz et al.'s article, "Optimal design and operation of activated sludge processes: State-of-the-art," published in the *Chemical Engineering Journal* (Vol. 218, 2015, pp. 900-920). The application of genetic algorithms in wastewater treatment mainly focuses on two aspects: first, parameter optimization, optimizing various parameters in the aeration process, such as aeration time and intensity, to achieve optimal treatment results; second, structural optimization, optimizing the structure of wastewater treatment plants, such as the size, shape, and location of ponds, to achieve optimal energy consumption. In recent years, genetic algorithms have achieved some success in the optimization of wastewater treatment processes.By optimizing various parameters in the aeration process using genetic algorithms, the goal of both ensuring wastewater treatment efficiency and reducing energy consumption was achieved. This can be seen in the article "Aeration optimization of a wastewater treatment plant using genetic algorithm" by B. Holenda et al., published in *Optimal control applications and methods* (Vol. 28, No. 3, 2007, pp. 191-208). However, genetic algorithms still have some limitations in wastewater treatment process optimization, such as slow convergence speed and a tendency to get trapped in local optima. Therefore, how to improve genetic algorithms to adapt to the optimization of wastewater treatment processes is an important research direction. For reference, see "Optimizing Control of Wastewater Treatment Plant with Reinforcement Learning: Technical Evaluation of Twin-Delayed Deep Deterministic Policy Gradient Agent" by KLAWIKOWSKA, Zuzanna, and GROCHOWSKI, Micha, published in Transactions on Industrial Informatics (2024). Summary of the Invention

[0004] The purpose of this invention is to provide a wastewater treatment plant aeration decision-making method, computer equipment, and readable storage medium, which achieves efficient control of the aeration process in wastewater treatment plants through intelligent optimization algorithms, and features rapid convergence and high efficiency.

[0005] To achieve the above objectives, the solution of the present invention is:

[0006] A wastewater treatment plant aeration decision-making method includes,

[0007] Obtain wastewater treatment data to get the predicted aeration rate X(t) at time t;

[0008] Obtain the blower frequency combination {K} that achieves the predicted aeration volume X(t). i (N)}, where K i (N) is the i-th satisfying Blower frequency combination {N1i (t),N 2i (t),…,N ni (t)}, i = 1, 2, ..., N ji (t) represents the frequency of the j-th blower, and n represents the number of blowers;

[0009] Obtain the blower frequency combination {K i (N)} Blower frequency combination K with minimum power loss min (N)={N 1min (t),N 2min (t),…,N nmin (t)}, and control each blower to combine K at the blower frequency. min The frequency corresponding to (N) is used.

[0010] This includes acquiring wastewater treatment data and obtaining the predicted aeration rate X(t) at time t, including:

[0011] Construct a reinforcement learning model that includes a state model S(t), an action model A(t), and a reward mechanism R(t). Here, the state model S(t) is the environmental state at time t during the wastewater treatment process, the action model A(t) is the adjustment amount of aeration rate under state S(t), and the reward mechanism R(t) is the water quality performance at time t.

[0012] Construct an aeration volume prediction model.

[0013]

[0014] Where η is the learning rate and γ is the discount factor; The discount factor under the optimal action;

[0015] The aeration volume prediction model is optimized using a reinforcement learning algorithm to obtain the optimal aeration volume adjustment A(t).

[0016] Based on the aeration rate at time t-1 and the aeration rate adjustment A(t), the predicted aeration rate X(t) at time t is obtained.

[0017] X(t) = X(t-1) + A(t)

[0018] Where X(t) represents the predicted aeration volume at time t.

[0019] The expression for the state model S(t) in the reinforcement learning model is:

[0020]

[0021] Wherein, DO(t) represents the dissolved oxygen characteristic at time t during the wastewater treatment process. Let T(t) represent the ammonia nitrogen characteristics at time t during wastewater treatment, T(t) represent the temperature at time t during wastewater treatment, PH(t) represent the acidity / alkalinity at time t during wastewater treatment, and C represent the temperature at time t during wastewater treatment. TP (t) represents the total phosphorus characteristic of the effluent at time t during the wastewater treatment process, J TP (t) represents the influent total phosphorus characteristic at time t during the wastewater treatment process, C TN (t) represents the total nitrogen characteristic of the effluent at time t during the wastewater treatment process, J TN (t) represents the influent total nitrogen characteristic at time t during the wastewater treatment process, C SS (t) represents the characteristics of suspended solids in the effluent at time t during the wastewater treatment process, J SS (t) represents the characteristics of suspended solids in the influent at time t during the wastewater treatment process.

[0022] In the state model S(t), the method for obtaining each feature is as follows:

[0023] Acquire real-time data during the wastewater treatment process;

[0024] The real-time data is preprocessed, including filling in missing values ​​and correcting or deleting outliers;

[0025] Calculate the correlation coefficients between all features and the target variable, and select features whose correlation coefficients are higher than a set threshold.

[0026] Wherein, the blower frequency combination {K} is obtained. i (N)} Blower frequency combination K with minimum power loss min (N)={N 1min (t),N 2min (t),…,N nmin (t)}, including,

[0027] Construct the fitness function,

[0028] minE(t)=N1(t)+N2(t)+…+N n (t)

[0029] St.X(t)≤N1(t)+N2(t)+…+N n (t)

[0030] Wherein, minE(t) represents the minimum power loss;

[0031] A genetic algorithm was used to find the blower frequency combination that minimizes power loss.

[0032] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor; when the processor executes the computer program, it implements the steps of the wastewater treatment plant aeration decision method as described above.

[0033] A computer-readable storage medium storing a computer program; when executed by a processor, the computer program implements the steps of the aforementioned wastewater treatment plant aeration decision method.

[0034] The beneficial effects of this invention, achieved by combining genetic algorithms with reinforcement learning, are that it enables intelligent control of the aeration process in wastewater treatment plants, optimizing the aeration decision-making process. Genetic algorithms quickly find the optimal solution by simulating natural selection, while reinforcement learning allows the system to continuously learn and optimize in actual operation, improving the efficiency of the control strategy. This method not only converges quickly, reducing computation time, but also significantly saves energy and reduces operating costs while ensuring water quality. Attached Figure Description

[0035] Figure 1 This is the training process of the aeration decision model;

[0036] Figure 2 It is a comparison between the predicted curves and the actual curves of aeration volume and water quality parameters based on reinforcement learning;

[0037] Figure 3 It is a comparison between the predicted curves and the actual curves of aeration volume and aeration strategy;

[0038] Figure 4 It is a comparison between the predicted curves and the actual curves of power consumption and aeration strategy.

[0039] Figure 5 This is the monthly power consumption curve per ton of water for a wastewater treatment plant without making an aeration decision.

[0040] Figure 6 This is the monthly power consumption curve per ton of water for a wastewater treatment plant under the aeration decision. Detailed Implementation

[0041] This invention provides a wastewater treatment plant aeration decision-making method, including,

[0042] Obtain wastewater treatment data to get the predicted aeration rate X(t) at time t;

[0043] Obtain the blower frequency combination {K} that achieves the predicted aeration volume X(t). i (N)}, where K i (N) is the i-th satisfying Blower frequency combination {N1i (t),N 2i (t),…,N ni (t)}, i = 1, 2, ..., N ji (t) represents the frequency of the j-th blower, and n represents the number of blowers;

[0044] Obtain the blower frequency combination {K i (N)} Blower frequency combination K with minimum power loss min (N)={N 1min (t),N 2min (t),…,N nmin (t)}, and control each blower to combine K at the blower frequency. min The frequency corresponding to (N) is used.

[0045] This includes acquiring wastewater treatment data and obtaining the predicted aeration rate X(t) at time t, including:

[0046] Construct a reinforcement learning model that includes a state model S(t), an action model A(t), and a reward mechanism R(t). Here, the state model S(t) is the environmental state at time t during the wastewater treatment process, the action model A(t) is the adjustment amount of aeration rate under state S(t), and the reward mechanism R(t) is the water quality performance at time t.

[0047] Construct an aeration volume prediction model.

[0048]

[0049] Where η is the learning rate and γ is the discount factor; The discount factor under the optimal action;

[0050] The aeration volume prediction model is optimized using a reinforcement learning algorithm to obtain the optimal aeration volume adjustment A(t).

[0051] Based on the aeration rate at time t-1 and the aeration rate adjustment A(t), the predicted aeration rate X(t) at time t is obtained.

[0052] X(t) = X(t-1) + A(t)

[0053] Where X(t) represents the predicted aeration volume at time t.

[0054] The expression for the state model S(t) in the reinforcement learning model is:

[0055]

[0056] Wherein, DO(t) represents the dissolved oxygen characteristic at time t during the wastewater treatment process. Let T(t) represent the ammonia nitrogen characteristics at time t during wastewater treatment, T(t) represent the temperature at time t during wastewater treatment, PH(t) represent the acidity / alkalinity at time t during wastewater treatment, and C represent the temperature at time t during wastewater treatment. TP (t) represents the total phosphorus characteristic of the effluent at time t during the wastewater treatment process, J TP (t) represents the influent total phosphorus characteristic at time t during the wastewater treatment process, C TN (t) represents the total nitrogen characteristic of the effluent at time t during the wastewater treatment process, J TN (t) represents the influent total nitrogen characteristic at time t during the wastewater treatment process, C SS (t) represents the characteristics of suspended solids in the effluent at time t during the wastewater treatment process, J SS (t) represents the characteristics of suspended solids in the influent at time t during the wastewater treatment process.

[0057] In the state model S(t), the method for obtaining each feature is as follows:

[0058] Acquire real-time data during the wastewater treatment process;

[0059] The real-time data is preprocessed, including filling in missing values ​​and correcting or deleting outliers;

[0060] Calculate the correlation coefficients between all features and the target variable, and select features whose correlation coefficients are higher than a set threshold.

[0061] Wherein, the blower frequency combination {K} is obtained. i (N)} Blower frequency combination K with minimum power loss min (N)={N 1min (t),N 2min (t),…,N nmin (t)}, including,

[0062] Construct the fitness function,

[0063] minE(t)=N1(t)+N2(t)+…+N n (t)

[0064] St.X(t)≤N1(t)+N2(t)+…+N n (t)

[0065] Wherein, minE(t) represents the minimum power loss;

[0066] A genetic algorithm was used to find the blower frequency combination that minimizes power loss.

[0067] The technical solution and beneficial effects of the present invention will be described in detail below with reference to the accompanying drawings.

[0068] The embodiments of the present invention include the following steps:

[0069] 1) Data acquisition and preprocessing: Obtaining multiple relevant real-time data from the wastewater treatment process, including but not limited to dissolved oxygen (DO) and ammonia nitrogen. Parameters such as temperature and pH value, as well as various indicators of effluent and influent, such as total phosphorus (C) in effluent. TP ), total phosphorus in the influent (J TP ), total nitrogen in effluent (C) TN ), total nitrogen in influent (J TN ), effluent suspended solids (C) SS ), influent suspended solids (J) SS The parameters include ammonia nitrogen in the aerobic zone of the biological treatment tank, nitrate nitrogen in the aerobic zone of the biological treatment tank, ORP in the anaerobic zone of the biological treatment tank, influent flow rate, effluent COD, influent COD and dissolved oxygen, as well as sludge return flow rate and chemical oxygen demand (COD) of the water. These data will be used as the basis for model input and optimization objectives, providing an accurate basis for aeration decisions.

[0070] 2) Feature engineering: The collected data undergoes feature engineering processing, including the following specific steps:

[0071] First, all the raw data is preprocessed, including filling missing values ​​and handling outliers, to obtain preprocessed features. The first step is to handle missing values ​​in the raw data, which can be done by using the mean, median, or front and back data to fill the missing values. Then, outliers are handled by identifying values ​​that are significantly different from the overall data distribution and correcting or deleting these values.

[0072] Then, the correlation coefficients between all features and the target variable (e.g., treatment effect or energy consumption) are calculated to select features with high correlation, and further filter out variables that have a greater impact on the model, thereby improving the accuracy and efficiency of the model.

[0073] By following the steps above, select dissolved oxygen (DO) and ammonia nitrogen. Total phosphorus in effluent (C TP ), total phosphorus in the influent (J TP ), total nitrogen in effluent (C) TN ), total nitrogen in influent (J TN ), effluent suspended solids (C) SS ), influent suspended solids (J) SS Features such as )

[0074] 3) Obtain the optimal aeration rate and predict the aeration rate using a reinforcement learning model; this includes the following specific steps:

[0075] A mathematical model for aeration control is established, and the model takes the following form:

[0076] X(t)=X(t-1)+△X(t) (1)

[0077] Where X(t) represents the predicted aeration volume at time t, X(t-1) represents the aeration volume at time t-1, and ΔX(t) is the aeration volume adjustment, which is obtained from the aeration volume adjustment strategy based on reinforcement learning.

[0078] The following section will use a reinforcement learning algorithm to obtain the aeration rate adjustment.

[0079] First, a reinforcement learning model is established to control the aeration rate. The aeration rate X(t) is adaptively adjusted according to the environmental conditions and water quality targets. The model introduces a state model S(t), an action model A(t), and a reward mechanism R(t).

[0080] Establish a state mathematical model, the model form is as follows:

[0081]

[0082] S(t) represents the environmental state at time t during the wastewater treatment process, including dissolved oxygen (DO) and ammonia nitrogen. Total phosphorus in effluent (C TP ), that is, the features selected in step 2); PH(t) represents the acidity or alkalinity at time t, and T(t) represents the temperature at time t.

[0083] Establish a mathematical model of the action, in the form of:

[0084] A(t)=△X(t) (3)

[0085] A(t) represents the aeration rate adjustment strategy adopted by the system in state S(t), i.e., the aeration rate adjustment amount ΔX(t).

[0086] Establish a mathematical model for the reward mechanism. The model is in the following form:

[0087] R(t)=α·WaterQuality(t) (4)

[0088] The reward function R(t) is defined based on the system's water quality performance, α is a weighting coefficient, and WaterQuality(t) is the water quality score, calculated based on the effluent indicators. The mathematical model of WaterQuality(t) is shown below:

[0089]

[0090] Then, reinforcement learning uses a deep Q-network (DQN) to learn the optimal aeration strategy.

[0091] The Q-value update formula (gas demand forecasting model) is as follows:

[0092]

[0093] Where η is the learning rate, γ is the discount factor used to balance the weights of current rewards and future rewards, and is a constant with a value between [0,1]. The discount factor is the optimal action; the Q value reflects the long-term cumulative reward of taking action A(t) in state S(t). Through multiple iterations, reinforcement learning will continuously optimize the aeration rate, enabling the system to provide the optimal aeration rate prediction while meeting water quality requirements.

[0094] Through the above calculations, the predicted aeration rate X(t) at time t is obtained, which is the aeration rate that needs to be controlled.

[0095] 4) Obtain equipment scheduling schemes based on genetic algorithms;

[0096] 41) Construct an optimal equipment scheduling model based on a genetic algorithm. The genetic algorithm is used to optimize the equipment scheduling strategy under the aeration volume prediction value X(t) given by reinforcement learning, to ensure the minimum power loss. The objective function of the genetic algorithm is:

[0097] Fitness = minE(t) (7)

[0098] Where E(t) represents the power loss under the current aeration strategy, and the aeration strategy is the current frequency combination of each blower.

[0099] 42) Construct an energy consumption prediction model to describe the relationship between electricity consumption and blower frequency. The model form is as follows:

[0100] E(t)=f1(N1(t)+N2(t)+…+N n (t)) (8)

[0101] Where E represents power loss; N1(t) to N n (t) represents the frequency of the 1st, 2nd, ..., nth blower at time t, and f1 represents the relationship between power loss and blower frequency.

[0102] Construct an aeration volume prediction model to describe the relationship between aeration volume and aeration strategy. The model is in the form of:

[0103] X(t)=f2(N1(t)+N2(t)+…+N n (t)) (9)

[0104] Where f2 represents the relationship between aeration volume and blower frequency.

[0105] 43) Finding the optimal equipment scheduling strategy

[0106] Initial population generation: Based on a given aeration rate X(t), multiple equipment scheduling schemes (N1(t) + N2(t) + ... + N) are randomly generated. n (t)) constitutes the initial population.

[0107] Fitness assessment: The fitness value of each individual is evaluated using a fitness function.

[0108] In genetic algorithms, fitness evaluation is the process of determining the merits of each individual (i.e., each candidate solution). Each individual represents a potential equipment scheduling strategy or aeration rate adjustment scheme. The fitness value guides the genetic algorithm to select better individuals and use them to generate the next generation.

[0109] In this embodiment, the equipment scheduling scheme is set as N(t) = N1(t) + N2(t) + ... + N n (t), calculate its fitness according to the following steps:

[0110] Step a, calculate energy consumption E(t):

[0111] In this embodiment, N1(t) = 10kW, N2(t) = 12kW, and N3(t) = 15kW, then the total energy consumption is:

[0112] E(t) = 10 + 12 + 15 = 37 kWh

[0113] Step b, calculate the water quality score WaterQuality(t):

[0114] In this embodiment, the effluent quality is: C COD =45, C SS =40, C TP =0.4, C TN =8, which meets the water quality standards, therefore:

[0115] WaterQuality(t) = 1

[0116] Step c, calculate the fitness value:

[0117] In this embodiment, the weighting coefficient λ1 for energy consumption is set to 0.7, and the weighting coefficient λ2 for water quality score is set to 0.3. Therefore, the fitness value is:

[0118] Fitness(X(t))=0.7*37+0.3*1=26.2

[0119] Step d, compare the fitness of different individuals: After calculating the fitness value of all individuals, the genetic algorithm selects individuals with high fitness (i.e., equipment scheduling schemes with low energy consumption and qualified water quality) as the parents of the next generation for crossover and mutation operations.

[0120] Selection: Based on fitness values, select individuals with high fitness to be the parents of the next generation.

[0121] Crossover: The selected parents are crossovered to generate new individuals. The crossover formula is:

[0122] child i =parent i λ+child i (1-λ) (10)

[0123] Among them, parent i This represents the characteristic or parameter value of the i-th parent individual. In this embodiment, it represents the scheduling strategy of the i-th parent individual, which usually refers to the energy consumption or running time of each device in device scheduling; child i The crossover operation represents the feature or parameter value of the new individual generated by the crossover operation, which is generated from some information of the parent individual; λ represents the crossover coefficient (crossover rate), which is used to control the ratio of the two parent individuals in the crossover process. Its value is between 0 and 1. The higher the value, the more the new individual inherits the features of the corresponding parent individual.

[0124] Mutation: Mutation operations are performed on newly generated individuals to increase population diversity. The mutation formula is:

[0125] child > ′ =child > +Δ (11)

[0126] Among them, child > ′ Δ represents the individual characteristics or parameter values ​​after mutation; Δ is the amount of mutation, which represents the magnitude of the change in the individual's characteristics. Its magnitude determines the magnitude of the mutation. A larger amount of mutation will bring about a larger change, but may destroy the good solution that has been obtained. A smaller amount of mutation will help to explore a small range of solution space.

[0127] Iterative optimization: Repeat selection, crossover, and mutation operations until a preset termination condition is met. The termination condition may be reaching a preset maximum number of iterations, or the fitness value no longer significantly improves.

[0128] 44) Design a fitness function with energy loss and effluent quality as optimization objectives. The fitness function is defined as:

[0129] minE(t)=N1(t)+N2(t)+…+N n (t)

[0130] St.X(t-1)+△X(t)≤N1(t)+N2(t)+…+N n(t) (12)

[0131] Where minE(t) represents minimizing energy loss, and N1(t) + N2(t) + ... + N n The aeration rate under the aeration strategy (t) can satisfy the optimal aeration rate X(t) provided by reinforcement learning.

[0132] 5) Generating optimal aeration rate and equipment scheduling strategy: Through joint optimization using reinforcement learning and genetic algorithms, two key strategies are generated:

[0133] Optimal aeration rate X(t): provided by reinforcement learning to ensure water quality meets standards and minimize energy consumption.

[0134] Optimal equipment scheduling strategy: provided by a genetic algorithm, ensuring the optimal combination of equipment and operation mode under a given aeration rate, further reducing energy consumption.

[0135] 10) System Implementation: Through the established control system, the optimized strategies are applied to the actual wastewater treatment plant to achieve intelligent and precise aeration control, ensuring that the effluent quality meets standards and reducing operating costs. The control system includes a data acquisition module, a predictive model module, an optimization algorithm module, and an execution control module. The overall architecture is as follows:

[0136] 1. Data acquisition module: Real-time acquisition of various parameters during the wastewater treatment process.

[0137] 2. Prediction Model Module: Based on the collected data, a prediction model is built, and the optimal aeration rate is obtained by combining reinforcement learning methods.

[0138] 3. Optimization Algorithm Module: Uses a genetic algorithm to continuously optimize aeration decisions.

[0139] 4. Execution control module: Adjusts the frequency parameters of the blower in real time based on the optimization results.

[0140] The method of this invention optimizes the aeration decision of wastewater treatment plants through genetic algorithms, effectively improving the intelligence level of the aeration process, significantly reducing energy consumption, optimizing operating costs, and providing an innovative solution for energy conservation and emission reduction in the wastewater treatment industry.

[0141] Water quality parameters and aeration volume data were collected from January to September 2024. Data from January to July were used as training data to train the model, and data from July to September were used for testing. Figure 2 The test results are shown, predicting the required gas volume based on water quality parameters. The predicted gas volume is compared with the actual gas volume, with an R2 score as high as 0.96 and a mean square error of 8.6, which is within 10.

[0142] Data on aeration volume and blower frequency were collected from January to September 2024. Data from January to July was used as training data to train the model, while data from July to September was used for testing. The model was then used to predict the aeration volume that could be provided at the current blower frequency. Figure 3 The test results are shown, comparing the predicted air volume with the actual air volume. The R2 score is as high as 0.96, and the mean square error is 7.3, which is within 10.

[0143] Data on blower power consumption and frequency parameters were collected from January to September 2024. Data from January to July was used for training the model, while data from July to September was used for testing. The model was then used to predict power consumption at the current blower frequency. Figure 4 The test results are shown, with 300 values ​​used as a reference. The predicted air volume is compared with the actual air volume, and the R2 score is as high as 0.99 with a mean square error of 2.1.

[0144] See again Figure 5 and Figure 6 , Figure 5 The daily energy consumption curve for September 2024, when this invention was not used, shows an average daily energy consumption of 0.265 kilowatt-hours of water; while... Figure 6 The daily energy loss curve for October 2024, based on the decision-making recommendations of this invention, shows an average daily energy consumption of 0.238 kilowatt-hours per ton of water, resulting in a 10.2% saving. Theoretically, this could save 20% of the energy loss.

[0145] This invention also provides another computer device, including a processor and a memory configured to store a computer program capable of running on the processor; wherein, when the processor is configured to run the computer program, it performs the method steps described in the foregoing embodiments.

[0146] In practical applications, the aforementioned processor includes a Field-Programmable Gate Array (FPGA), and the processor can be a Central Processing Unit (CPU) or a Digital Signal Processor (DSP). It is understood that for different devices, the electronic devices used to implement the functions of the aforementioned processor can also be other types, and this embodiment of the invention does not impose specific limitations.

[0147] The aforementioned memory can be volatile memory, such as random-access memory (RAM); or non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD); or a combination of the above types of memory, and provides instructions and data to the processor.

[0148] In an exemplary embodiment, the present invention also provides a computer-readable storage medium for storing a computer program.

[0149] Optionally, the computer-readable storage medium can be applied to any of the methods in the embodiments of the present invention, and the computer program causes the computer to execute the corresponding processes implemented by the processor in the various methods of the embodiments of the present invention. For the sake of brevity, these will not be described in detail here.

[0150] In the several embodiments provided by this invention, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0151] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0152] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0153] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0154] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0155] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0156] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A wastewater treatment plant aeration decision-making method, characterized in that: include, Obtain wastewater treatment data to get the predicted aeration rate X(t) at time t; Obtain the blower frequency combination {K} that achieves the predicted aeration volume X(t). i (N)}, where K i (N) is the i-th satisfying Blower frequency combination {N 1i (t),N 2i (t),…,N ni (t)}, i=1,2,…,N ji (t) represents the frequency of the j-th blower, and n represents the number of blowers; Obtain the blower frequency combination {K i (N)} Blower frequency combination K with minimum power loss min (N)={N 1min (t),N 2min (t),…,N nmin (t)}, and control each blower to combine K at the blower frequency. min Operating at the frequency corresponding to (N); This includes acquiring wastewater treatment data and obtaining the predicted aeration rate X(t) at time t, including: Construct a reinforcement learning model that includes a state model S(t), an action model A(t), and a reward mechanism R(t). Here, the state model S(t) is the environmental state at time t during the wastewater treatment process, the action model A(t) is the adjustment amount of aeration rate under state S(t), and the reward mechanism R(t) is the water quality performance at time t. Construct an aeration volume prediction model. ， in, For learning rate, Discount factor; The discount factor under the optimal action; The aeration volume prediction model is optimized using a reinforcement learning algorithm to obtain the optimal aeration volume adjustment A(t). Based on the aeration rate at time t-1 and the aeration rate adjustment A(t), the predicted aeration rate X(t) at time t is obtained. X(t) = X(t-1) + A(t), Where X(t) represents the predicted aeration volume at time t; Wherein, the blower frequency combination {K} is obtained. i (N)} Blower frequency combination K with minimum power loss min (N)={N 1min (t),N 2min (t),…,N nmin (t)}, including, Construct the fitness function, minE(t)=N1(t)+N2(t)+…+N n (t) St.X(t)≤N1(t)+N2(t)+…+N n (t) Wherein, minE(t) represents the minimum power loss; A genetic algorithm was used to find the blower frequency combination that minimizes power loss.

2. The wastewater treatment plant aeration decision-making method as described in claim 1, characterized in that: The expression for the state model S(t) in the reinforcement learning model is: ， Wherein, DO(t) represents the dissolved oxygen characteristic at time t during the wastewater treatment process. Let T(t) represent the ammonia nitrogen characteristics at time t during wastewater treatment, T(t) represent the temperature at time t during wastewater treatment, PH(t) represent the acidity / alkalinity at time t during wastewater treatment, and C represent the temperature at time t during wastewater treatment. TP (t) represents the total phosphorus characteristic of the effluent at time t during the wastewater treatment process, J TP (t) represents the influent total phosphorus characteristic at time t during the wastewater treatment process, C TN (t) represents the total nitrogen characteristic of the effluent at time t during the wastewater treatment process, J TN (t) represents the influent total nitrogen characteristic at time t during the wastewater treatment process, C SS (t) represents the characteristics of suspended solids in the effluent at time t during the wastewater treatment process, J SS (t) represents the characteristics of suspended solids in the influent at time t during the wastewater treatment process.

3. The wastewater treatment plant aeration decision-making method as described in claim 2, characterized in that: In the state model S(t), the method for obtaining each feature is as follows: Acquire real-time data during the wastewater treatment process; The real-time data is preprocessed, including filling in missing values ​​and correcting or deleting outliers; Calculate the correlation coefficients between all features and the target variable, and select features whose correlation coefficients are higher than a set threshold.

4. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor; characterized in that: When the processor executes the computer program, it implements the steps of the wastewater treatment plant aeration decision method as described in any one of claims 1 to 3.

5. A computer-readable storage medium storing a computer program; characterized in that: When the computer program is executed by the processor, it implements the steps of the wastewater treatment plant aeration decision method as described in any one of claims 1 to 3.

Images