Sewage process parameter regulation method and system based on artificial intelligence

Abstract

The application discloses a sewage process parameter regulation and control method and system based on artificial intelligence, relates to the technical field of data analysis, comprises the action relationship between analysis basic parameters and the action relationship of the basic parameters to the evaluation parameters, accurately analyzes the relationship between the basic parameters and the relationship between the basic parameters and the evaluation parameters, and provides a reliable basis for subsequent multi-objective optimization and regulation. The weight of each evaluation parameter is set in combination with the action relationship of the basic parameters to the evaluation parameters and the fluctuation of the evaluation parameters. The inertia weight in the MOPSO algorithm is set, and the optimal solution of the multi-objective optimization model is solved through the MOPSO algorithm, so that the search ability of the local and the global is balanced, and the quality of the optimal solution is improved. The sewage process parameters are cooperatively regulated and controlled based on the optimal solution of the self-adaptive control strategy and the multi-objective optimization model, and the regulation and control effect of the process parameters is comprehensively and effectively improved in combination with the self-adaptive control and the multi-objective optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, in particular to a sewage process parameter regulation method and system based on artificial intelligence. BACKGROUND

[0002] With the rapid development of industrialization and urbanization, sewage treatment is facing increasingly complex challenges. Traditional sewage process parameter regulation relies on manual experience and fixed process parameters, which is difficult to adapt to water quality fluctuations and working condition changes. The sewage process parameter regulation scheme based on artificial intelligence emerges as the times require. This scheme uses neural networks, machine learning and other advanced algorithms to analyze a large amount of data in the sewage treatment process in real time, accurately predict water quality changes, and dynamically adjust key parameters such as DO and MLSS. By building a closed-loop system of "perception-analysis-decision-execution", the sewage treatment has made a leap from "experience-based water control" to "algorithm-based water control", effectively improving the treatment efficiency, reducing energy consumption, ensuring the stability of the effluent water quality, and promoting the development of the sewage treatment industry towards intelligence and efficiency.

[0003] In the prior art, because there are multiple targets in sewage treatment, it is difficult to ensure the balance between multiple targets, resulting in poor accuracy and adaptability of sewage process parameter regulation, which cannot meet the changing sewage treatment conditions, and the effect of sewage process parameter regulation is low.

[0004] Therefore, how to improve the accuracy and adaptability of sewage process parameter regulation is a technical problem to be solved at present. SUMMARY

[0005] The purpose of the present application is to solve the problem of poor accuracy and adaptability of sewage process parameter regulation in the prior art, and to provide a sewage process parameter regulation method based on artificial intelligence, which comprises,

[0006] Collecting all sewage process parameters involved in the sewage treatment process, classifying the sewage process parameters, the sewage process parameters including basic parameters and evaluation parameters, analyzing the action relationship between the basic parameters and the action relationship of the basic parameters on the evaluation parameters;

[0007] Statistical analysis of the fluctuation of the basic parameters and the evaluation parameters, and setting the weight of each evaluation parameter according to the action relationship of the basic parameters on the evaluation parameters, and defining the treatment condition stability level according to the fluctuation of the basic parameters;

[0008] Adjusting the weight of the evaluation parameter by the treatment condition stability level, constructing a multi-objective optimization model, setting the inertia weight in the MOPSO algorithm, and solving the optimal solution of the multi-objective optimization model by the MOPSO algorithm;

[0009] An adaptive control strategy is set, and the sewage process parameters are cooperatively regulated based on the adaptive control strategy and the optimal solution of the multi-objective optimization model.

[0010] In some embodiments of the present application, the action relationship between the basic parameters and the action relationship of the basic parameters on the evaluation parameters are analyzed, including,

[0011] The action relationship between different basic parameters is analyzed by multiple regression analysis, and the significant action relationship is screened based on the P value;

[0012] According to the sewage treatment mechanism model, the hypothetical causal path between the basic parameters and the evaluation parameters is given, and a theoretical causal path diagram is constructed. The nodes in the theoretical causal path diagram are basic parameters or evaluation parameters, and the edges or paths represent the relationship between the basic parameters or the relationship between the basic parameters and the evaluation parameters;

[0013] The path coefficient of each path is calculated and marked at the corresponding position of the theoretical causal path diagram. According to the proximity relationship between the nodes on the theoretical causal path diagram, the direct effect action relationship and the indirect effect action relationship are divided, and the action relationship of the basic parameters on the evaluation parameters is determined based on the path coefficient, the direct effect action relationship and the indirect effect action relationship.

[0014] In some embodiments of the present application, the fluctuation of the basic parameters and the evaluation parameters is counted, including,

[0015] The standard deviation and the average value of the basic parameters and the evaluation parameters are counted, and the coefficient of variation of the basic parameters and the evaluation parameters is calculated according to the standard deviation and the average value. The fluctuation of the basic parameters and the evaluation parameters is described by the coefficient of variation.

[0016] In some embodiments of the present application, the weight of each evaluation parameter is set in combination with the action relationship of the basic parameters on the evaluation parameters, including,

[0017] The action relationship of the basic parameters on the evaluation parameters is divided into multiple groups according to the evaluation parameters, and the multiple path coefficients in the theoretical causal path diagram corresponding to the action relationship of the basic parameters on the evaluation parameters are integrated to obtain the action intensity.

[0018] For the same evaluation parameter, the weight of each evaluation parameter is determined according to the coefficient of variation and the action intensity.

[0019] In some embodiments of the present application, the weight of the evaluation parameter is dynamically adjusted by processing the process condition stability level, including,

[0020] Whether the current sewage treatment process condition belongs to a stable state is judged by the process condition stability level;

[0021] If the current sewage treatment working condition belongs to a stable state, the weight of the evaluation parameter does not need to be adjusted;

[0022] Otherwise, the weight of the evaluation parameter is adjusted according to the action relationship of the basic parameter to the evaluation parameter.

[0023] In some embodiments of the present application, a multi-objective optimization model is constructed, including,

[0024] A target function of each evaluation parameter is constructed based on the action relationship between the basic parameters and the action relationship of the basic parameter to the evaluation parameter, and a comprehensive target function model is generated by comprehensively combining the target functions of all evaluation parameters;

[0025] A plurality of constraint factors required by the sewage treatment process are collected, and the constraint conditions of the parameters in the multi-objective optimization model are determined according to the plurality of constraint factors.

[0026] In some embodiments of the present application, the inertia weight in the MOPSO algorithm is set, and the optimal solution of the multi-objective optimization model is solved by the MOPSO algorithm, including,

[0027] Based on the two action relationships of the action relationship between the basic parameters and the action relationship of the basic parameter to the evaluation parameter, local sensitivity analysis and global sensitivity analysis are respectively performed, and the sensitive index is determined by combining the results of the two sensitivity analyses;

[0028] The correlation between the evaluation parameters is analyzed to determine the conflict index of the evaluation parameters, the inertia weight is determined according to the sensitive index and the conflict index, and the MOPSO algorithm is implemented to calculate the optimal solution of the multi-objective optimization model.

[0029] In some embodiments of the present application, the sewage process parameters are cooperatively regulated based on the optimal solution of the multi-objective optimization model and the adaptive control strategy, including,

[0030] The first control target interval is output according to the adaptive control strategy, and the second control target interval is output according to the optimal solution of the multi-objective optimization model;

[0031] If the first control target interval and the second control target interval have an intersection, the control target interval intersection is used for sewage process parameter regulation;

[0032] Otherwise, the sewage process parameter regulation is performed according to the deviation of the first control target interval and the second control target interval.

[0033] Correspondingly, the present application also provides a sewage process parameter regulation system based on artificial intelligence, including,

[0034] The first module is used for collecting all sewage process parameters involved in the sewage treatment process, classifying the sewage process parameters, the sewage process parameters including basic parameters and evaluation parameters, analyzing the action relationship between the basic parameters and the action relationship of the basic parameters on the evaluation parameters;

[0035] The second module is used for counting the fluctuation of the basic parameters and the evaluation parameters, setting the weight of each evaluation parameter in combination with the action relationship of the basic parameters on the evaluation parameters, and defining the treatment working condition stability level according to the fluctuation of the basic parameters;

[0036] The third module is used for dynamically adjusting the weight of the evaluation parameters through the treatment working condition stability level, constructing a multi-objective optimization model, setting the inertia weight in the MOPSO algorithm, and solving the optimal solution of the multi-objective optimization model through the MOPSO algorithm.

[0037] The fourth module is used for setting an adaptive control strategy, and cooperatively regulating and controlling the sewage process parameters based on the optimal solution of the multi-objective optimization model and the adaptive control strategy.

[0038] Compared with the prior art, the present application has the following beneficial effects:

[0039] 1. The action relationship between the basic parameters and the action relationship of the basic parameters on the evaluation parameters are analyzed, the relationship between the basic parameters and the relationship between the basic parameters and the evaluation parameters are accurately analyzed, and a reliable foundation is provided for subsequent multi-objective optimization and regulation. The weight of each evaluation parameter is set in combination with the action relationship of the basic parameters on the evaluation parameters and the fluctuation of the evaluation parameters, the adaptability and stability of the multi-objective optimization are improved, and the demands of multiple objectives are balanced as much as possible. The weight of the evaluation parameters is dynamically adjusted through the treatment working condition stability level, the sewage treatment working condition is monitored, and it is judged whether the weight of the multi-objective parameters is adjusted, so as to adjust in time, and the timeliness is ensured.

[0040] 2. The inertia weight in the MOPSO algorithm is set, and the optimal solution of the multi-objective optimization model is solved through the MOPSO algorithm, the local and global search ability is balanced, and the optimal solution quality is improved. The sewage process parameters are cooperatively regulated and controlled based on the optimal solution of the multi-objective optimization model and the adaptive control strategy, the process parameter regulation effect is comprehensively and effectively improved by combining adaptive control and multi-objective optimization, the accuracy and adaptability of the sewage process parameter regulation are improved, and the balance between multiple objectives of the sewage treatment and the sewage treatment effect are ensured. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 The flowchart of the sewage process parameter regulation method based on artificial intelligence provided by the present application is shown in the figure;

[0042] Figure 2A structure schematic diagram of the sewage process parameter regulation system based on artificial intelligence is provided for the present application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments.

[0044] Reference Figure 1 The sewage process parameter regulation method based on artificial intelligence comprises the following steps:

[0045] In step S101, all sewage process parameters involved in the sewage treatment process are collected, the sewage process parameters are classified, the sewage process parameters include basic parameters and evaluation parameters, and the action relationship between the basic parameters and the action relationship of the basic parameters on the evaluation parameters are analyzed.

[0046] In this embodiment, the basic parameters and the evaluation parameters include the following contents:

[0047] Physical parameters: flow (Q), temperature (T), dissolved oxygen (DO), sludge concentration (MLSS).

[0048] Chemical parameters: chemical oxygen demand (COD), ammonia nitrogen (NH3-N), total phosphorus (TP), pH value.

[0049] Operating parameters: aeration amount (A), reflux ratio (R), sludge age (SRT), and hydraulic retention time (HRT).

[0050] Evaluation parameters:

[0051] Water quality targets: effluent COD, effluent ammonia nitrogen, and effluent TP.

[0052] Energy consumption targets: aeration energy consumption and pumping energy consumption.

[0053] Cost targets: reagent cost and sludge disposal cost.

[0054] In this embodiment, the basic parameters may have mutual influences, the basic parameters may have influences on the evaluation parameters, and both of the two influences need to be considered to provide a basis for the subsequent.

[0055] In some embodiments of the present application, the action relationship between the basic parameters and the action relationship of the basic parameters on the evaluation parameters are analyzed, including,

[0056] The action relationship between different basic parameters is analyzed by multiple regression analysis, and the significant action relationship is screened out based on P value;

[0057] According to the sewage treatment mechanism model, an assumed causal path between the basic parameters and the evaluation parameters is given, a theoretical causal path diagram is constructed, the nodes in the theoretical causal path diagram are basic parameters or evaluation parameters, and the edges or paths represent the relationship between the basic parameters or the relationship between the basic parameters and the evaluation parameters;

[0058] The path coefficients of each path are calculated and marked at the corresponding positions of the theoretical causal path diagram, the direct effect relationship and the indirect effect relationship are divided according to the proximity relationship between the nodes on the theoretical causal path diagram, and the action relationship of the basic parameters to the evaluation parameters is determined based on the path coefficients, the direct effect relationship and the indirect effect relationship.

[0059] In this embodiment, the action relationship between different basic parameters is analyzed by multiple regression analysis, and whether it is significant is distinguished by a threshold. Different basic parameters are analyzed as independent variables and dependent variables respectively. Based on the sewage treatment mechanism, all possible assumed causal paths are proposed.

[0060] For example, DO→microbial activity→effluent COD (path 1);

[0061] MLSS→sludge load→effluent COD (path 2);

[0062] Aeration quantity→DO→microbial activity→effluent COD (path 3, indirect).

[0063] Theoretical causal path diagram:

[0064] Nodes: DO, MLSS, aeration quantity, microbial activity, sludge load, effluent COD.

[0065] Edges / paths:

[0066] DO→microbial activity (path coefficient 0.6);

[0067] Microbial activity→effluent COD (path coefficient-0.8);

[0068] MLSS→sludge load (path coefficient-0.4);

[0069] Sludge load→effluent COD (path coefficient 0.5);

[0070] Aeration quantity→DO (path coefficient 0.3).

[0071] The path coefficient is the degree of influence between two nodes, the direct effect relationship and the indirect effect relationship are divided according to the proximity relationship between the nodes on the theoretical causal path diagram, the adjacent nodes (between the basic parameters and the evaluation parameters) are the direct effect relationship, otherwise they are the indirect effect relationship. For example:

[0072] Direct effects:

[0073] DO → Microbial activity → Effluent COD (Path 1);

[0074] MLSS → Sludge Load → Effluent COD (Path 2);

[0075] Indirect effects:

[0076] Aeration rate → Discharge (DO) → Microbial activity → Effluent COD (Path 3).

[0077] Step S102: Statistically analyze the fluctuations of basic parameters and evaluation parameters, and set the weight of each evaluation parameter based on the relationship between the basic parameters and the evaluation parameters. Define the stability level of the processing condition based on the fluctuations of the basic parameters.

[0078] In this embodiment, the fluctuations in the basic parameters and evaluation parameters will affect the weights of subsequent multi-objective optimization and the stability of the operating conditions, so these factors are considered in the settings and monitoring.

[0079] In some embodiments of this application, the fluctuations of statistical basic parameters and evaluation parameters are analyzed, including:

[0080] The standard deviation and mean of the basic parameters and evaluation parameters are calculated, and the coefficient of variation of the basic parameters and evaluation parameters is calculated based on the standard deviation and mean. The coefficient of variation is used to describe the fluctuation of the basic parameters and evaluation parameters.

[0081] In this embodiment, the coefficient of variation for each parameter is calculated by the ratio of the standard deviation to the mean, which describes the fluctuation of the parameter.

[0082] In some embodiments of this application, the weight of each evaluation parameter is set in conjunction with the relationship between the basic parameters and the evaluation parameters, including:

[0083] The interaction between multiple sets of basic parameters and evaluation parameters is divided into units based on the evaluation parameters. Then, multiple path coefficients in the theoretical causal path graph corresponding to the interaction between basic parameters and evaluation parameters are integrated to obtain the intensity of the interaction.

[0084] For the same evaluation parameter, the weight of each evaluation parameter is determined based on the coefficient of variation and the intensity of the effect.

[0085] In this embodiment, each evaluation parameter corresponds to the influence of a basic parameter on the evaluation parameter. A calibration value is determined based on the coefficient of variation and the strength of the influence. Weights are assigned to each evaluation parameter by comparing the calibration values ​​of all evaluation parameters. The formula for calculating the calibration value is as follows:

[0086] ;

[0087] wherein, is the correction value of the first evaluation parameter, is the action intensity of the first evaluation parameter, is the coefficient of variation of the first evaluation parameter, is the first constant of the first evaluation parameter, represents the correction of the coefficient of variation to the action intensity, and the coefficient of variation describes the reliability and stability of the evaluation parameter. In this embodiment, the processing condition stability level is defined according to the fluctuation of the basic parameters, and the processing condition stability level is determined in combination with the coefficients of variation of all basic parameters. Here, the coefficients of variation of the basic parameters have two, one is the real-time coefficient of variation, and the other is the coefficient of variation determined by the historical data of the parameter, which are respectively denoted as the first coefficient of variation and the second coefficient of variation. A variation evaluation value (describing the stability compared with the previous variation) is determined by comparing the two, and the processing condition stability level is determined based on the variation evaluation value of each basic parameter. The calculation formula is as follows: ;

[0088] wherein,

[0089] is the processing condition stability level, is the number of basic parameters, is the combined weight of the first basic parameter,

[0090] is the variation evaluation value of the first basic parameter, is the maximum value in , , are the second constant and the third constant (used to balance the correction function size and the level) respectively, represents the correction of the maximum value to the class average value, and the class average value is slightly larger than the ordinary average value, and [] is the rounding symbol. Step S103, the weight of the evaluation parameter is dynamically adjusted through the processing condition stability level, a multi-objective optimization model is constructed, the inertia weight in the MOPSO algorithm is set, and the optimal solution of the multi-objective optimization model is solved through the MOPSO algorithm.

[0091] In this embodiment, the multi-objective optimization means as follows:

[0092] Solve conflicting goals: such as minimizing the effluent COD and energy consumption at the same time, and the two usually have a trade-off relationship.

[0093]

[0094] ​​​​​​Provide decision support: Generate Pareto front for decision makers to choose optimal solution based on preferences.

[0095] Enhance system robustness: Avoid suboptimal solutions caused by single objective through multi-objective optimization.

[0096] Significance of MOPSO algorithm solution:

[0097] Suitable for continuous / discrete problems: Flexible handling of mixed variables in wastewater treatment (such as continuous aeration quantity, discrete equipment switching).

[0098] Efficient search of Pareto front: Rapidly approach optimal solution set through particle swarm cooperation.

[0099] Dynamic adaptability: Combine inertia weight adjustment to adapt to operating condition changes. Inertia weight size in MOPSO algorithm (balance global and local search ability, ensure solution quality and solving speed).

[0100] Framework process:

[0101] Input: Real-time data of basic parameters (DO, MLSS, aeration quantity).

[0102] Step 1: Calculate the operating condition stability level (high / medium / low stability).

[0103] Step 2: Dynamically adjust the evaluation parameter weight according to the stability level.

[0104] Step 3: Construct multi-objective optimization model (such as effluent quality, energy consumption, operating cost).

[0105] Step 4: Set the inertia weight of MOPSO algorithm to solve the optimal solution.

[0106] Output: Pareto optimal solution set, support decision selection.

[0107] MOPSO algorithm process:

[0108] Initialize particle swarm: Randomly generate particle position (decision variable value) and velocity.

[0109] External archive maintenance: Store non-dominated solutions (Pareto front).

[0110] Iterative update:

[0111] Calculate the objective function value of each particle.

[0112] Update individual optimal solution (pbest) and global optimal solution (gbest, based on crowding distance selection). Update particle velocity and position;

[0113] Termination condition: maximum iteration number is reached or Pareto front converges.

[0114] In some embodiments of the present application, the weights of the evaluation parameters are dynamically adjusted by processing the stability level of the working condition, including,

[0115] Whether the current sewage treatment working condition belongs to a stable state is determined by processing the stability level of the working condition;

[0116] If the current sewage treatment working condition belongs to a stable state, the weights of the evaluation parameters do not need to be adjusted;

[0117] Otherwise, the weights of the evaluation parameters are adjusted according to the action relationship of the basic parameters to the evaluation parameters.

[0118] In the present embodiment, the stability level of the working condition is processed to describe the stability of the current sewage treatment, so as to determine whether adjustment is needed, thereby ensuring the reliability of multi-objective optimization. Precise control: only when the working condition is unstable, the weights are adjusted to avoid excessive intervention. Strong adaptability: combining real-time data and process mechanism, dynamically matching the operation demand.

[0119] In some embodiments of the present application, a multi-objective optimization model is constructed, including,

[0120] Taking each evaluation parameter as a unit, the objective function of each evaluation parameter is constructed based on the action relationship between the basic parameters and the action relationship of the basic parameters to the evaluation parameters, and the objective functions of all evaluation parameters are integrated to generate a comprehensive objective function model;

[0121] Multiple constraint factors required in the sewage treatment process are collected, and the constraint conditions of the parameters in the multi-objective optimization model are determined according to the multiple constraint factors.

[0122] In the present embodiment, the objective functions of all evaluation parameters are integrated and weightedly summed. The multiple constraint factors include process constraints, water quality constraints, equipment constraints, safety constraints, etc. The constraint conditions of the parameters in the multi-objective optimization model are controllable parameters.

[0123] In some embodiments of the present application, the inertia weight in the MOPSO algorithm is set, and the optimal solution of the multi-objective optimization model is solved by the MOPSO algorithm, including,

[0124] Based on the two action relationships of the action relationship between the basic parameters and the action relationship of the basic parameters to the evaluation parameters, local sensitivity analysis and global sensitivity analysis are respectively performed, and the sensitive index is determined by combining the results of the two sensitivity analyses;

[0125] The correlation between the evaluation parameters is analyzed and evaluated to determine a conflict index of the evaluation parameters, the inertia weight is determined according to the sensitive index and the conflict index, the MOPSO algorithm is implemented, and the optimal solution of the multi-objective optimization model is calculated.

[0126] In this embodiment, the inertia weight is affected by both the data sensitivity degree and the conflict of the evaluation parameters, and a reasonable inertia weight is determined by considering both to balance the global and local search capabilities. Improve search efficiency: dynamically adjust the inertia weight according to the conflict degree, avoid falling into local optimum. Example: expand the search range when the conflict is high, fine optimization when the conflict is low. Enhance the quality of the solution: combine the basic parameter sensitivity, optimize the key parameters first. Example: when DO is sensitive to effluent COD, adjust the DO value more finely. Adapt to complex systems: suitable for multi-objective, strongly coupled complex systems (such as sewage treatment). Example: multi-objective optimization of effluent quality, energy consumption, and chemical cost. Automation and robustness: reduce the need for manual parameter adjustment, improve algorithm robustness. Example: automatically adapt to changes in conflict and sensitivity under different working conditions.

[0127] In this embodiment, the purpose of local sensitivity analysis is to quantify the impact of small changes in a single basic parameter on the evaluation parameters within a local range. It is realized by the partial derivative method, that is, the partial derivative of the objective function with respect to the basic parameter is calculated to evaluate the local sensitivity. Global sensitivity analysis evaluates the comprehensive influence of the basic parameters on the evaluation parameters within the entire feasible domain. Sobol index method: decomposes the variance of the objective function to calculate the main effect and interaction effect of each parameter. The sensitive index is determined by combining the results of the two sensitivity analyses (local sensitivity analysis and global sensitivity analysis), the Pearson correlation coefficient between the evaluation parameters is calculated, and the conflict index of the evaluation parameters is determined. The inertia weight is determined according to the sensitive index and the conflict index.

[0128] Step S104, set the adaptive control strategy, and cooperatively regulate the sewage process parameters based on the adaptive control strategy and the optimal solution of the multi-objective optimization model.

[0129] In this embodiment, the idea is as follows:

[0130] Dual-target interval generation:

[0131] Adaptive control strategy: dynamically output the first control target interval based on real-time working conditions.

[0132] Multi-objective optimization model: output the second control target interval based on the Pareto optimal solution set.

[0133] Interval coordination mechanism:

[0134] Intersection priority: if there is an intersection between the two intervals, directly use the intersection for parameter regulation.

[0135] Deviation compensation: if there is no intersection, dynamically adjust the control strategy according to the interval deviation.

[0136] In some embodiments of the present application, the wastewater process parameters are cooperatively regulated based on the optimal solution of the adaptive control strategy and the multi-objective optimization model, including,

[0137] The first control target interval is output according to the adaptive control strategy, and the second control target interval is output according to the optimal solution of the multi-objective optimization model;

[0138] If the first control target interval and the second control target interval have an intersection, the control target interval intersection is used for wastewater process parameter regulation;

[0139] Otherwise, the wastewater process parameters are regulated according to the deviation of the first control target interval and the second control target interval.

[0140] In this embodiment, fuzzy logic control is used as the adaptive control strategy. According to the deviation, dynamic compensation is carried out, small deviation uses weighted average value, and large deviation uses the lower limit of the second control target interval. The adaptive strategy real-time corrects the deviation, the model provides a fault tolerance interval, avoids single-point control failure, and enhances the anti-interference ability. Adaptive control strategy: real-time response: dynamically adjust the parameters according to the real-time working condition (such as influent load, equipment state), to ensure stable operation of the system. Local optimization: quickly respond to sudden working conditions (such as sudden increase of influent COD during rainstorm period), to avoid system collapse.

[0141] Correspondingly, the present application also provides a wastewater process parameter regulation system based on artificial intelligence, as shown in Figure 2 , which includes,

[0142] The first module is used for collecting all wastewater process parameters involved in the wastewater treatment process, classifying the wastewater process parameters, and analyzing the action relationship between the basic parameters and the evaluation parameters and the action relationship of the basic parameters on the evaluation parameters.

[0143] The second module is used for counting the fluctuation of the basic parameters and the evaluation parameters, setting the weight of each evaluation parameter in combination with the action relationship of the basic parameters on the evaluation parameters, and defining the treatment working condition stability level according to the fluctuation of the basic parameters;

[0144] The third module is used for dynamically adjusting the weight of the evaluation parameters through the treatment working condition stability level, constructing a multi-objective optimization model, setting the inertia weight in the MOPSO algorithm, and solving the optimal solution of the multi-objective optimization model through the MOPSO algorithm;

[0145] The fourth module is used for setting an adaptive control strategy, and cooperatively regulating the wastewater process parameters based on the adaptive control strategy and the optimal solution of the multi-objective optimization model.

[0146] Compared with the prior art, the present application has the following beneficial effects:

[0147] 1. The action relationship between the basic parameters and the action relationship between the basic parameters and the evaluation parameters are analyzed, the relationship between the basic parameters and the relationship between the basic parameters and the evaluation parameters are accurately analyzed, and a reliable foundation is provided for subsequent multi-objective optimization and control. The weight of each evaluation parameter is set in combination with the action relationship of the basic parameters on the evaluation parameters and the fluctuation of the evaluation parameters, the adaptability and stability of the multi-objective optimization are improved, and the demand of multiple objectives is balanced as much as possible. The weight of the evaluation parameter is dynamically adjusted by processing the working condition stability level, the sewage treatment working condition is monitored, and whether the weight of the multi-objective parameter is adjusted is judged, so that timely adjustment is ensured, and the timeliness is ensured.

[0148] 2. The inertia weight in the MOPSO algorithm is set, and the optimal solution of the multi-objective optimization model is solved by the MOPSO algorithm, the local and global search ability is balanced, and the optimal solution quality is improved. Based on the adaptive control strategy and the optimal solution of the multi-objective optimization model, the sewage process parameters are cooperatively controlled, the adaptive control and multi-objective optimization are combined, the process parameter control effect is comprehensively and effectively improved, the accuracy and adaptability of the sewage process parameter control are improved, and the balance between multiple objectives of sewage treatment and the sewage treatment effect are ensured.

[0149] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by hardware, or can be implemented by means of software and necessary general hardware platform. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.), and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present application.

[0150] Those skilled in the art can understand that the drawings are only a schematic diagram of a preferred embodiment, and the modules or flows in the drawings are not necessarily required for implementing the present application.

[0151] Those skilled in the art can understand that the modules in the system in the embodiments can be distributed in the system in the embodiments, or can be changed and located in one or more systems different from the embodiments. The modules in the above embodiments can be combined into one module, or can be further split into multiple sub-modules.

[0152] The above merely describes preferred specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art, according to the technical solution and inventive concept of the present application, makes equivalent replacement or change within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A wastewater treatment process parameter control method based on artificial intelligence, characterized in that, include, All wastewater process parameters involved in the wastewater treatment process are collected, classified, including basic parameters and evaluation parameters. The interaction relationships between basic parameters and their effects on evaluation parameters are analyzed. The fluctuations of basic parameters and evaluation parameters are statistically analyzed, and the weight of each evaluation parameter is set based on the relationship between the basic parameters and the evaluation parameters. The stability level of the processing condition is defined according to the fluctuation of the basic parameters. By dynamically adjusting the weights of evaluation parameters based on the stability level of the operating conditions, a multi-objective optimization model is constructed. Inertial weights are set in the MOPSO algorithm, and the optimal solution for the multi-objective optimization model is obtained through the MOPSO algorithm. An adaptive control strategy is set up, and the wastewater process parameters are coordinated and regulated based on the optimal solution of the adaptive control strategy and the multi-objective optimization model. in, Analyze the interactions between the basic parameters and their effects on the evaluation parameters, including: Multiple regression analysis was used to analyze the interaction between different basic parameters, and significant interactions were screened based on p-values. Based on the wastewater treatment mechanism model, the hypothetical causal path between basic parameters and evaluation parameters is given, and a theoretical causal path diagram is constructed. The nodes in the theoretical causal path diagram are basic parameters or evaluation parameters, and the edges or paths represent the relationship between basic parameters or the relationship between basic parameters and evaluation parameters. Calculate the path coefficient for each path and mark it at the corresponding position on the theoretical causal path diagram. Divide the direct effect relationship and the indirect effect relationship according to the proximity relationship between nodes on the theoretical causal path diagram. Determine the relationship between the basic parameters and the evaluation parameters based on the path coefficient, the direct effect relationship and the indirect effect relationship. The fluctuations of statistical basic parameters and evaluation parameters, including, The standard deviation and mean of the basic parameters and evaluation parameters are calculated, and the coefficient of variation of the basic parameters and evaluation parameters is calculated based on the standard deviation and mean. The coefficient of variation is used to describe the fluctuation of the basic parameters and evaluation parameters. And based on the relationship between the basic parameters and the evaluation parameters, the weight of each evaluation parameter is set, including, The interaction between multiple sets of basic parameters and evaluation parameters is divided into units based on the evaluation parameters. Then, multiple path coefficients in the theoretical causal path graph corresponding to the interaction between basic parameters and evaluation parameters are integrated to obtain the intensity of the interaction. For the same evaluation parameter, the weight of each evaluation parameter is determined based on the coefficient of variation and the intensity of the effect.

2. The wastewater treatment process parameter control method based on artificial intelligence according to claim 1, characterized in that, The weights of the evaluation parameters are dynamically adjusted by processing the stability level of the operating conditions. include, The stability level of the treatment conditions determines whether the current wastewater treatment operation is in a stable state. If the current wastewater treatment operation is in a stable state, then there is no need to adjust the weights of the evaluation parameters; Otherwise, the weights of the evaluation parameters are adjusted based on the relationship between the basic parameters and the evaluation parameters.

3. The wastewater treatment process parameter control method based on artificial intelligence according to claim 1, characterized in that, Construct a multi-objective optimization model, including, Taking each evaluation parameter as a unit, an objective function is constructed based on the interaction relationship between the basic parameters and the interaction relationship between the basic parameters and the evaluation parameters. The objective functions of all evaluation parameters are then combined to generate a comprehensive objective function model. Collect various constraints required for the wastewater treatment process, and determine the constraints on the parameters in the multi-objective optimization model based on these constraints.

4. The wastewater treatment process parameter control method based on artificial intelligence according to claim 1, characterized in that, We define the inertia weights in the MOPSO algorithm and use the MOPSO algorithm to find the optimal solution for a multi-objective optimization model. include, Based on the interaction between the basic parameters and the interaction between the basic parameters and the evaluation parameters, local sensitivity analysis and global sensitivity analysis are performed respectively, and the sensitivity index is determined by combining the results of the two sensitivity analyses. The correlation between evaluation parameters is analyzed to determine the conflict index of the evaluation parameters. Based on the sensitivity index and the conflict index, the inertia weight is determined, and then the MOPSO algorithm is implemented to calculate the optimal solution of the multi-objective optimization model.

5. The wastewater treatment process parameter control method based on artificial intelligence according to claim 1, characterized in that, The optimal solution of the adaptive control strategy and multi-objective optimization model is used to coordinately regulate the parameters of wastewater treatment processes, including: The first control target interval is output based on the adaptive control strategy, and the second control target interval is output based on the optimal solution of the multi-objective optimization model. If the first control target interval and the second control target interval intersect, the intersection of the control target intervals will be used to regulate the wastewater process parameters. Otherwise, the wastewater treatment process parameters are adjusted based on the deviation between the first and second control target intervals.

6. A wastewater treatment process parameter control system based on artificial intelligence, characterized in that, The system is used to implement the artificial intelligence-based wastewater treatment process parameter control method as described in any one of claims 1-5, the system comprising: The first module is used to collect all wastewater process parameters involved in the wastewater treatment process, classify the wastewater process parameters, including basic parameters and evaluation parameters, and analyze the interaction between basic parameters and the interaction between basic parameters and evaluation parameters. The second module is used to statistically analyze the fluctuations of basic parameters and evaluation parameters, and to set the weight of each evaluation parameter based on the relationship between the basic parameters and the evaluation parameters. It also defines the stability level of the processing condition based on the fluctuations of the basic parameters. The third module is used to dynamically adjust the weights of evaluation parameters by processing the stability level of the operating condition, construct a multi-objective optimization model, set the inertia weights in the MOPSO algorithm, and solve the optimal solution of the multi-objective optimization model through the MOPSO algorithm. The fourth module is used to set the adaptive control strategy, and to coordinate the regulation of wastewater process parameters based on the optimal solution of the adaptive control strategy and the multi-objective optimization model.

Images