A multi-objective optimal control method for sewage treatment process with over-standard suppression strategy

Abstract

The present application relates to a multi-objective optimization control method for sewage treatment process with over-standard suppression strategy, including the following steps: using the time series modeling ability of long short-term memory network to establish prediction models for effluent ammonia nitrogen concentration and total nitrogen concentration, and optimizing the set values of nitrate nitrogen concentration and dissolved oxygen concentration in the prediction models; judging the over-standard situation of water quality parameters according to the output of the prediction models. If the water quality parameters are over-standard, controlling the external reflux and the additional carbon source, and performing secondary optimization on the set values of dissolved oxygen concentration and nitrate nitrogen concentration to improve the prediction accuracy of effluent ammonia nitrogen concentration and total nitrogen concentration. When selecting the set values, comprehensively considering water quality and energy consumption, initially achieving the goal of reducing energy consumption on the basis of improving water quality; effectively suppressing the over-standard phenomenon of the predicted peaks of effluent ammonia nitrogen concentration and total nitrogen concentration while reducing energy consumption and improving water quality.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment, and particularly relates to a multi-objective optimization control method for sewage treatment process with an over-standard inhibition strategy. Background Art

[0002] Water is an important resource for human survival and development. With the rapid advancement of industrialization and urbanization, the urban sewage discharge is continuously increasing. It is particularly important to implement optimal control for the sewage treatment process. However, the sewage treatment process is time-varying, non-linear, and the variables are severely coupled. Therefore, how to reduce energy consumption and improve water quality as much as possible while ensuring that the effluent parameters meet the standards has become one of the important research directions in the field of sewage treatment. The effluent water quality in the sewage treatment process is affected by the nitrate nitrogen concentration (S NO2 ) and the dissolved oxygen concentration (S O5 ). The two can be controlled respectively through the reflux pump and the blower to achieve the effect of improving water quality, but the process equipment will generate high energy consumption during operation.

[0003] Therefore, it is necessary to perform multi-objective optimization control on water quality and energy consumption under certain process condition constraints. Although the multi-objective optimization control method has achieved the average water quality discharge standard and also obtained relatively satisfactory results in terms of energy consumption, there are still over-standard phenomena of the effluent ammonia nitrogen concentration (S Nh，e ) and the total nitrogen concentration (S Ntot，e ) during the optimization control process. If the sewage is directly discharged without taking any measures in the face of this phenomenon, it will still have an impact on the environment and human health.

[0004] In summary, in the sewage treatment process, the following main problems exist: the sewage treatment process has characteristics such as time-varying non-linearity and severe coupling between variables, and it is difficult to establish a prediction model for the parameter variables therein; for the over-standard situations of the effluent ammonia nitrogen concentration and the total nitrogen concentration during the multi-objective optimization control process, how to adopt an inhibition control strategy to constrain them within the control limits and achieve the goal of improving water quality and reducing energy consumption. Summary of the Invention

[0005] An embodiment of the present invention provides a multi-objective optimal control method for the sewage treatment process with an over-standard suppression strategy, aiming to solve the problems in the prior art that the sewage treatment process has characteristics such as time-varying nonlinearity and severe coupling between variables, making it difficult to establish a prediction model for parameter variables, and to address the situation of exceeding the standard of the effluent ammonia nitrogen concentration and total nitrogen concentration during the multi-objective optimal control process. The method is to adopt a suppression control strategy to constrain them within the control limits, so as to achieve the purpose of improving water quality and reducing energy consumption. To have a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary part is not a general review, nor is it intended to identify key / important constituent elements or delineate the protection scope of these embodiments. Its sole purpose is to present some concepts in a simple form as a prelude to the subsequent detailed description.

[0006] The present application provides a multi-objective optimal control method for the sewage treatment process with an over-standard suppression strategy, including the following steps:

[0007] Utilize the time series modeling ability of the long short-term memory network to establish a prediction model for the effluent ammonia nitrogen concentration and total nitrogen concentration, and optimize the set values of the nitrate nitrogen concentration and dissolved oxygen concentration in the prediction model;

[0008] Judge the over-standard situation of water quality parameters according to the output of the prediction model. If the water quality parameters exceed the standard, control the external reflux and the added carbon source, and perform secondary optimization on the set values of the dissolved oxygen concentration and nitrate nitrogen concentration.

[0009] As a preferred implementation manner, the process of optimizing the set values of the nitrate nitrogen concentration and dissolved oxygen concentration in the prediction model is carried out using a multi-objective evolutionary algorithm, and its expression is:

[0010] min F(X)={f OCI (X), f EQI (X)}

[0011]

[0012] X=(x1, x2), l i ≤x i ≤u i , i = 1, 2

[0013] where, f OCI (X) and f EQI (X) respectively represent the effluent water quality and total energy consumption optimization functions, and s.t. gives the limiting conditions of the sewage treatment process for 5 water quality parameters; S Nh，e，avg represents the average concentration of effluent ammonia nitrogen, X is the decision vector, x1 and x2 are the dissolved oxygen concentration and nitrate nitrogen concentration respectively; u i and l i respectively represent the upper and lower bounds of each decision variable;

[0014] Among them, the expression of the effluent quality is as follows:

[0015]

[0016] In the formula, TSS, COD, S NKj , S NO , BOD5, and Q e respectively represent the concentration of total suspended solids, chemical oxygen demand, Kjeldahl nitrogen concentration, nitrate nitrogen concentration, 5-day biochemical oxygen demand, and the discharge of clear water;

[0017] The expression of the total energy consumption is as follows:

[0018] OCI = AE + PE + 3EC

[0019] In the formula: AE represents the aeration energy consumption, PE represents the pumping energy consumption, and EC represents the cost of carbon source, and its expression is as follows:

[0020]

[0021]

[0022]

[0023] Among them, T is the sampling period; t0 and t f respectively represent the start time and the end time; V i and K Lai respectively represent the volume and aeration volume of the i-th biochemical reaction tank; Q a , Q r and Q w respectively represent the internal return flow, external return flow, and excess sludge flow; q EC，j represents the flow rate of the additional carbon source added to the j-th reaction zone.

[0024] As a preferred implementation manner, the process of optimizing the set values of nitrate nitrogen concentration and dissolved oxygen concentration in the prediction model generates Pareto solutions, and the set values of dissolved oxygen concentration and nitrate nitrogen concentration are selected for each cycle's over-standard situation. The specific method is as follows:

[0025] Take the Pareto solutions obtained by optimization as the set values of dissolved oxygen concentration and nitrate nitrogen concentration in turn, substitute them into the prediction model, and retain and store the solutions that satisfy the average effluent quality constraint within this cycle in the solution set P A ;

[0026] If P A is not equal to the empty set, select the solution that can minimize the energy consumption from PA as the preferred solution;

[0027] If P AEqual to the empty set, select the solution that can make the water quality reach a relatively good level as the preferred solution.

[0028] As a preferred implementation, when the predicted output effluent ammonia nitrogen concentration exceeds the standard, increase the external reflux flow rate to dilute and adjust the ammonia nitrogen concentration, and dynamically optimize the set values of the dissolved oxygen concentration and nitrate nitrogen concentration. First, set S O5 to the maximum value within the adjustable range, and then perform a secondary optimization of the nitrate nitrogen concentration.

[0029] As a preferred implementation, the specific steps for the secondary optimization of the set values of the dissolved oxygen concentration and nitrate nitrogen concentration are as follows:

[0030] Judge the period p to which the current ammonia nitrogen concentration exceeding the standard belongs, where {p ∈ N * |1 ≤ p ≤ 112};

[0031] Load the energy consumption and water quality data corresponding to the dissolved oxygen concentration and nitrate nitrogen concentration collected in this period by the multi-objective optimization control for training the prediction model;

[0032] Fix one independent variable boundary of the multi-objective algorithm, let x1 = u1, where u1 represents the upper boundary of the dissolved oxygen concentration, so that the algorithm always maintains the maximum value of the dissolved oxygen concentration throughout the search process. By continuously optimizing the set value of the nitrate nitrogen concentration, find the Pareto solution set of energy consumption and water quality to achieve the purpose of secondary optimization of the nitrate nitrogen concentration;

[0033] Select the set value combination with lower energy consumption from the obtained Pareto solution set and use this set value combination as the updated set value.

[0034] As a preferred implementation, the adjustment formula for the external reflux flow rate is as follows:

[0035] Q r = (5 - S Nh5 ) × 50000

[0036] In the formula, S Nh5 is the ammonia nitrogen concentration in the fifth partition, and S Nh5 ≤ 5 mg / L.

[0037] As a preferred implementation, when the predicted output water nitrogen concentration exceeds the standard, control the total amount of externally added carbon source according to the prediction result, and its calculation formula is:

[0038] q EC = (S Ntot，e - 18) * 2

[0039] Among them, S Ntot，e is the total nitrogen concentration, q ECis the total amount of externally added carbon source, and the adjustable range for controlling the total amount of externally added carbon source is 4 - 7m 3 / d.

[0040] As a preferred embodiment, the percentage P of the water quality exceeding the standard time is selected as the evaluation index of the inhibition strategy, that is, the total duration T of the water quality exceeding the standard c and the total operation time T z The ratio is expressed as P = T c / T z .

[0041] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0042] Prediction models are established for the effluent ammonia nitrogen and the effluent total nitrogen respectively, and then the multi-objective evolutionary algorithm is used to optimize the set values of S O5 and S NO2 . Substitute the set values into the model to predict S Nh，e and S Ntot，e . If the prediction does not exceed the standard, multi-objective optimization control is adopted. If it exceeds the standard, the corresponding over-standard inhibition strategy is added to inhibit S Nh，e and S Ntot，e . At the same time, the set values of S O5 and S NO2 are optimized again to ensure effective reduction of energy consumption while improving water quality.

[0043] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present invention and used together with the specification to explain the principles of the present invention.

[0045] Figure 1 shows the overall optimization control process of the method according to an exemplary embodiment;

[0046] Figure 2 shows the variation of five effluent water quality parameters under multi-objective optimization control according to an exemplary embodiment;

[0047] Figure 3 shows the tracking effect of multi-objective optimization control of SO5 according to an exemplary embodiment;

[0048] Figure 4 shows the tracking effect of multi-objective optimization control of SNO2 according to an exemplary embodiment;

[0049] Figure 5It is a curve showing the concentration changes of BOD5, COD, and TSS during the process of adding an over-standard suppression strategy control according to an exemplary embodiment;

[0050] Figure 6 It is a comparison chart before and after S Nh，e under the condition of adding a strategy to suppress the over-standard of effluent water quality according to an exemplary embodiment;

[0051] Figure 7 It is a comparison chart before and after S Ntot，e under the condition of adding a strategy to suppress the over-standard of effluent water quality according to an exemplary embodiment;

[0052] Figure 8 It is a real-time tracking control curve of the PID controller for the concentration of S O5 under the condition of adding a strategy to suppress the over-standard of effluent water quality according to an exemplary embodiment;

[0053] Figure 9 It is a real-time tracking control curve of the PID controller for the concentration of S NO2 under the condition of adding a strategy to suppress the over-standard of effluent water quality according to an exemplary embodiment. Detailed implementation manners

[0054] The following description and drawings fully illustrate the specific embodiments herein, enabling those skilled in the art to practice them. Parts and features of some embodiments may be included in or substituted for parts and features of other embodiments. The scope of the embodiments herein includes the entire scope of the claims and all available equivalents of the claims. Herein, terms such as "first", "second", etc. are only used to distinguish one element from another, without requiring or implying any actual relationship or order between these elements. In fact, the first element can also be called the second element, and vice versa. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a structure, device or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such structure, device or equipment. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the structure, device or equipment including the said element. The various embodiments herein are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0055] As used herein, the terms "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. They are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation on the present invention. In the description of the present invention, unless otherwise specified and defined, the terms "mounted", "connected", and "coupled" should be understood in a broad sense. For example, they can be mechanical connections or electrical connections, or can be the communication inside two elements, can be directly connected, or can be indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.

[0056] As used herein, unless otherwise specified, the term "a plurality of" means two or more.

[0057] As used herein, the character " / " indicates that the objects before and after are in an "or" relationship. For example, A / B means: A or B.

[0058] As used herein, the term "and / or" is an associative relationship describing an object, indicating that three relationships can exist. For example, A and / or B means: A or B, or, the three relationships of A and B.

[0059] Without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other.

[0060] Please refer to Figure 1 , this embodiment provides a multi-objective optimization control method for the sewage treatment process of an over-standard suppression strategy, including the following steps:

[0061] Utilize the time series modeling ability of the long short-term memory network to establish prediction models for the effluent ammonia nitrogen concentration and total nitrogen concentration, and optimize the set values of the nitrate nitrogen concentration and dissolved oxygen concentration in the prediction models;

[0062] Judge the over-standard situation of water quality parameters according to the output of the prediction model. If the water quality parameters exceed the standard, control the external reflux and the added carbon source, and perform secondary optimization on the set values of the dissolved oxygen concentration and nitrate nitrogen concentration.

[0063] This method first establishes prediction models for the effluent ammonia nitrogen and effluent total nitrogen respectively, and then uses a multi-objective evolutionary algorithm to optimize the set values of S O5 and S NO2 . Substitute the set values into the model to predict S Nh，e and S Ntot，e . If the prediction is not over-standard, multi-objective optimization control is adopted. If it exceeds the standard, corresponding over-standard suppression strategies are added to S Nh，e and SNtot，e Suppress it while performing secondary optimization on the set values of S O5 and S NO2 to ensure effective reduction of energy consumption while improving water quality. The overall optimization control process is as Figure 1 shown.

[0064] Table 1 presents the accuracy comparison between the prediction model adopted in this scheme and other prediction models. Among them, ANN is the artificial neural network; FNN is the feedforward neural network; LSTM is the long short-term memory neural network; the root mean square error (RMSE) is used as the index to evaluate the performance of the model. It can be seen from Table 1 that the accuracy of the adopted prediction model has been greatly improved compared with the other two models.

[0065] Table 1 shows the accuracy comparison of the prediction models as follows:

[0066]

[0067] The process of optimizing the set values of nitrate nitrogen concentration and dissolved oxygen concentration in the prediction model is carried out using a multi-objective evolutionary algorithm, and its expression is:

[0068] min F(X) = {f OCI (X), f EQI (X)}

[0069]

[0070] X = (x1, x2), l i ≤ x i ≤ u i , i = 1, 2

[0071] where f OCI (X) and f EQI (X) represent the effluent water quality and total energy consumption optimization functions respectively, and s.t. gives the limiting conditions of the sewage treatment process for 5 water quality parameters; S Nh，e，avg represents the average concentration of ammonia nitrogen in the effluent, X is the decision vector, x1 and x2 are the dissolved oxygen concentration and nitrate nitrogen concentration respectively; u i and l i represent the upper and lower bounds of each decision variable respectively.

[0072] Multi-objective optimization control is essentially to optimize two key and conflicting indicators: effluent water quality (EQI) and total energy consumption (OCI) to achieve a relative balance.

[0073] Among them, the expression of the effluent water quality is as follows:

[0074]

[0075] In the formula, TSS, COD, S NKj , S NO , BOD5, and Q e respectively represent the concentration of total suspended solids, chemical oxygen demand, Kjeldahl nitrogen concentration, nitrate nitrogen concentration, 5-day biochemical oxygen demand, and the discharge of clear water;

[0076] The expression of the total energy consumption is as follows:

[0077] OCI = AE + PE + 3EC

[0078] In the formula: AE represents the aeration energy consumption, PE represents the pumping energy consumption, and EC represents the cost of carbon source, and its expression is as follows:

[0079]

[0080]

[0081]

[0082] Among them, T is the sampling period; t0 and t f respectively represent the start time and the end time; V i and K Lai respectively represent the volume and the aeration volume of the i-th biochemical reaction tank; Q a , Q r and Q w respectively represent the internal reflux flow rate, the external reflux flow rate, and the excess sludge flow rate; q EC，j represents the flow rate of the additional carbon source added to the j-th reaction zone.

[0083] S O5 and S NO2 's set values directly affect S Nh，e and S Ntot，e , and K Lai and Q a affect the setting of their concentrations. Therefore, selecting the correct set values can effectively reduce the energy consumption and improve the effluent quality. In this paper, the improved multi-objective evolutionary algorithm based on decomposition and dynamic population multi-neighborhood (I-MOEA / D) is used to optimize the set values of S O5 and S NO2 , and the PID controller achieves real-time tracking control of the set values by adjusting K Lai and Q a to achieve the goal of improving water quality and reducing energy consumption.

[0084] From Appendix Figure 3 and Appendix Figure 4It can be seen that although the concentrations of nitrate nitrogen and dissolved oxygen are constantly changing, the PID controller can still ensure tracking control with high accuracy. Table 2 shows the comparison of energy consumption and water quality obtained by different optimization control methods under sunny weather conditions, as well as the comparison of the average values of five effluent indicators. Among them, Openloop is open-loop control; APSO is an optimization control method based on the combination of an improved particle swarm algorithm and a neural network; ESN is an optimization control method based on a state echo network; NSGA2-DLS is an optimization control method based on the combination of a density-based local search NSGA2 algorithm and a neural network; combined appendix Figure 2 And from Table 2, it can be seen that although S Nh，e and S Ntot，e The average values are lower than the constraint conditions, but there are relatively serious peak over-standard situations for these two parameters during the entire sewage treatment process; in contrast, the concentrations of BOD5, COD, and TSS are far lower than the constraint conditions and there are no over-standard phenomena during the process.

[0085] Table 2 shows the comparison of energy consumption and effluent water quality of different optimization control methods under sunny weather as follows:

[0086]

[0087] After the multi-objective evolutionary algorithm performs one optimization on the set values of S O5 and S NO2 A series of Pareto solutions will be generated during the optimization process. Although the effluent water quality meets the discharge standards during the entire optimization control process, there are phenomena where the average effluent water quality exceeds the standard in some optimization cycles. Therefore, for the over-standard situations in each cycle, the most suitable S O5 and S NO2 Set values are selected to effectively reduce energy consumption on the basis of improving the effluent water quality. The specific process is as follows:

[0088] The process of optimizing the set values of nitrate nitrogen concentration and dissolved oxygen concentration in the prediction model generates Pareto solutions. For the over-standard situations in each cycle, the set values of dissolved oxygen concentration and nitrate nitrogen concentration are selected. The specific method is as follows:

[0089] Take the Pareto solutions obtained from the optimization as the set values of dissolved oxygen concentration and nitrate nitrogen concentration in turn, substitute them into the prediction model, and retain and store the solutions that meet the average effluent water quality constraints in this cycle into the solution set P A ;

[0090] If P A is not an empty set, select the solution that can minimize the energy consumption from PA as the preferred solution;

[0091] If P A is an empty set, select the solution that can make the water quality relatively better as the preferred solution.

[0092] The method for selecting the set value can pre-judge whether the water quality meets the standard, and select the preferred solution as S according to the water quality compliance situation O5 and S NO2 set value, which can reduce energy consumption as much as possible on the premise of ensuring that the water quality meets the standard. Although this method can improve the effluent water quality, it cannot completely avoid S Nh，e and S Ntot，e peak over-standard phenomenon. Therefore, it is necessary to design an inhibition strategy to deal with the over-standard problem. To verify the advantages and disadvantages of the inhibition strategy, based on the evaluation indexes of energy consumption and water quality, the percentage P of water quality over-standard time is selected as the evaluation index, that is, the total over-standard time T c of the total running time T z ratio, and its expression is:

[0093] P = T c / T z .

[0094] Since S Nh，e and S Ntot，e are affected by the S O5 and S NO2 set values, this paper selects S O5 , S NO2 , influent flow rate, influent ammonia nitrogen concentration and current effluent ammonia nitrogen concentration as the input variables of the S Nh，e prediction model; S O5 , S NO2 , influent flow rate, influent total nitrogen concentration and current effluent total nitrogen concentration as the input variables of the S Ntot，e prediction model. After multiple experiments, the set value range of S O5 is set at 1-2.8 mg / l, and the set value range of S NO2 is set at 0.5-2 mg / l. The grid search method is used to search for the S O5 and S NO2 set values, and the search step size is 0.1. A total of 300 groups of set value combinations are obtained, and they are run on the BSM1 platform for 14 days with a sampling interval of 15 minutes. A total of 403200 groups of data are obtained, of which 360000 groups are used as the training set and 43200 groups are used as the test set.

[0095] Use the long short-term memory network modeling method for S Nh，e and S Ntot，eA prediction model is established, with 2 layers of LSTM hidden layers, the number of neurons in each layer set to 64, and the initial learning rate set to 0.1. When selecting input variables, the dissolved oxygen concentration and nitrate nitrogen concentration are added to the established model, the temperature variable is removed, and due to the obvious lag in the sewage treatment process, the current ammonia nitrogen concentration and total nitrogen concentration are also added to the input variables of the prediction model. Among the 403,200 groups of data obtained by sampling, 360,000 groups of data are selected as the training set, and the remaining 43,200 groups of data are used as the test set.

[0096] When it is predicted that S Nh，e exceeds the standard and S Nh，e ≤ 6 mg / L, increase the external reflux flow rate to make Q r = 120,000 m 3 / d. By increasing Q r , dilute the ammonia nitrogen concentration in the first partition. This dilution process continues until the peak ammonia nitrogen concentration reaches the fifth reaction unit, and then reduce Q r to increase the hydraulic residence time between units and give full play to the nitrification reaction. Adjust Q Nh5 according to the ammonia nitrogen concentration (S r ) in the fifth partition. While adjusting Q O5 and S NO2 , perform dynamic optimization on S O5 . First, set S O5 * to the maximum value within the adjustable range. Secondly, perform secondary optimization on the nitrate nitrogen concentration. If the original set value of nitrate nitrogen is directly used without treatment and is directly combined with the updated dissolved oxygen concentration for optimization control, it will inevitably affect the energy consumption and effluent quality. Therefore, perform secondary optimization on the nitrate nitrogen concentration to find the optimal set values of S O5 and S NO2 (S O5 * and S NO2 * ), which plays a certain role in improving the effluent quality and reducing the energy consumption.

[0097] When the predicted output ammonia nitrogen concentration of the prediction model exceeds the standard, increase the external reflux flow rate to dilute and adjust the ammonia nitrogen concentration, and perform dynamic optimization on the set values of the dissolved oxygen concentration and nitrate nitrogen concentration. First, set S O5 to the maximum value within the adjustable range, and secondly, perform secondary optimization on the nitrate nitrogen concentration.

[0098] The specific steps for performing secondary optimization on the set values of the dissolved oxygen concentration and nitrate nitrogen concentration are as follows:

[0099] Judge the period p to which the current ammonia nitrogen concentration exceeding the standard belongs, where {p ∈ N * |1 ≤ p ≤ 112};

[0100] Load the energy consumption and water quality data corresponding to the dissolved oxygen concentration and nitrate nitrogen concentration collected in this cycle for multi-objective optimization control, and use them to train the prediction model;

[0101] Fix the boundary of an independent variable of the multi-objective algorithm, let x1 = u1, where u1 represents the upper boundary of the dissolved oxygen concentration, so that the algorithm always maintains the maximum value of the dissolved oxygen concentration throughout the search process. By continuously optimizing the set value of the nitrate nitrogen concentration, find the Pareto solution set of energy consumption and water quality, and achieve the purpose of secondary optimization of the nitrate nitrogen concentration;

[0102] Select the set value combination with lower energy consumption from the Pareto solution set obtained by optimization, and use this set value combination as the set value after updating the PID controller.

[0103] In addition, Q r The adjustable upper bound of should not be set too high, because too large Q r may cause more ammonia nitrogen and total nitrogen to flow into the biochemical reaction tank. Therefore, the adjustable range of Q r is set to 0 - 100000m 3 / d, and S Nh5 will not exceed 5mg / L. After multiple experiments, the adjustment formula of Q r is designed as follows:

[0104] The adjustment formula of the external reflux flow is as follows:

[0105] Q r =(5 - S Nh5 )×50000

[0106] In the formula, S Nh5 is the ammonia nitrogen concentration in the fifth partition, and S Nh5 ≤5mg / L.

[0107] When it is predicted that S Nh，e exceeds the standard and S Nh，e > 6mg / L, it is necessary to adjust the external carbon source flow rates q EC1 and q EC2 of the first and second units, that is, q EC1 = q EC2 = 5, to meet the denitrification requirements of the denitrification reaction, promote the absorption of nitrate and ammonia nitrogen at the same time, improve the denitrification effect and reduce the total nitrogen concentration. Other control methods are the same as when S Nh，e ≤6mg / L.

[0108] When it is predicted that S Ntot，e exceeds the standard but S Nh，e < 6mg / L, according to the prediction results, adjust the external carbon source (q EC1 ) of the first unit and the external carbon source (q EC2) is controlled, and the adjustable upper bounds of both are 5. The total amount of externally added carbon source (q EC ) has an adjustable range of 4 - 7 m 3 / d, and its calculation formula is as shown in Equation Q r =(5 - S Nh5 )×50000. When q EC ≤5, then let q EC1 = q EC ; when q EC >5, then q EC1 = 5, q EC2 = q EC - q EC1 .

[0109] When the predicted output effluent nitrogen concentration exceeds the standard according to the prediction model, the total amount of externally added carbon source is controlled according to the prediction result, and its calculation formula is:

[0110] q EC =(S Ntot，e - 18)*2

[0111] where S Ntot，e is the total nitrogen concentration, q EC is the total amount of externally added carbon source, and the adjustable range for controlling the total amount of externally added carbon source is 4 - 7 m 3 / d.

[0112] When S Ntot，e is predicted to meet the standard and the total nitrogen concentration in the fifth unit is less than 13.5 mg / l, terminate the over - standard inhibition control strategy for the total nitrogen concentration and continue to use multi - objective optimization control.

[0113] Select the percentage P of water quality over - standard time as the evaluation index of the inhibition strategy, that is, the ratio of the total over - standard time T c to the total operation time T z , and its expression is

[0114] P = T c / T z .

[0115] It can be seen from Figure 5 that the impact on the three effluent indexes is relatively small when the strategy of inhibiting the over - standard of effluent water quality is introduced, and the concentrations of BOD5, COD, and TSS can still be maintained within the constraint conditions.

[0116] It can be seen from Appendix Figure 6 and Appendix Figure 7 that after introducing the optimized strategy of inhibiting the peak over - standard of effluent water quality, the concentrations of S Nh，e and S Ntot，e can be effectively reduced, ensuring S Nh，e and S Ntot，eAlways within the constraint range, there is no over-standard phenomenon throughout the process: P(S Nh，e ) and P(S Ntot，e ) are both 0%, and this solution has achieved good results in terms of energy consumption and water quality. Among them, OCI: 4725.6, EQI: 5161.9. Therefore, the optimized control strategy proposed by this solution for the problem of over-standard effluent quality in the sewage treatment process can suppress S Nh，e and S Ntot，e peak over-standard and achieve the goal of reducing energy consumption.

[0117] From Figure 8 it can be seen that the tracking control effect of the PID controller on the S O5 concentration remains good, while from Figure 9 it can also be seen that although the PID controller will have an overshoot phenomenon when tracking and controlling the S NO2 concentration, the overshoot time is short and it can quickly track the S NO2 concentration again.

[0118] In one embodiment, a computer device is provided. The computer device can be a server. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store static information and dynamic information data. The network interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it realizes the steps in the above method embodiment.

[0119] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0120] It should be noted that the above description is only some embodiments of the present application and an explanation of the technical principles applied. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features with similar functions disclosed in the present application.

[0121] In addition, although the operations are depicted in a particular order, this should not be construed as requiring that the operations be performed in the particular order shown or in sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although a number of specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present application. Certain features described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, the various features described in the context of a single embodiment can also be implemented separately or in any suitable sub-combination in multiple embodiments.

[0122] Although the subject matter has been described in language specific to structural features and / or methodological acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims.

Claims

1. A multi-objective optimal control method for sewage treatment process with over-standard suppression strategy, characterized in that, It includes the following steps: Utilize the time series modeling ability of the long short-term memory network to establish a prediction model for the effluent ammonia nitrogen concentration and total nitrogen concentration, and optimize the set values of the nitrate nitrogen concentration and dissolved oxygen concentration in the prediction model; Judge the over-standard situation of water quality parameters according to the output of the prediction model. If the water quality parameters exceed the standard, control the external reflux and the added carbon source, and perform secondary optimization on the set values of the dissolved oxygen concentration and nitrate nitrogen concentration; The specific steps for performing secondary optimization on the set values of the dissolved oxygen concentration and nitrate nitrogen concentration are as follows: Determine the period to which the current ammonia nitrogen concentration exceeding the standard belongs p , where ; Load the energy consumption and water quality data corresponding to the dissolved oxygen concentration and nitrate nitrogen concentration collected in this cycle for multi-objective optimization control to train the prediction model; Fix the boundary of an independent variable of the multi-objective algorithm, and let , where represents the upper boundary of the dissolved oxygen concentration, enabling the algorithm to always maintain the maximum dissolved oxygen concentration throughout the search process. By continuously optimizing the set value of the nitrate nitrogen concentration, the Pareto solution set of energy consumption and water quality is found, achieving the purpose of secondary optimization of the nitrate nitrogen concentration; Select the set value combination with lower energy consumption from the obtained Pareto solution set by optimization, and use this set value combination as the updated set value.

2. The multi-objective optimization control method for the sewage treatment process of the over-standard suppression strategy according to claim 1, wherein, The process of optimizing the set values of the nitrate nitrogen concentration and dissolved oxygen concentration in the prediction model is carried out using a multi-objective evolutionary algorithm, and its expression is: Among them, and respectively represent the effluent water quality and the total energy consumption optimization function, s.t. gives the limit conditions of the sewage treatment process for 5 water quality parameters; represents the average concentration of ammonia nitrogen in the effluent, X is the decision vector, x 1 and x 2 are the dissolved oxygen concentration and the nitrate nitrogen concentration respectively; and respectively represent the upper and lower bounds of each decision variable; Among them, the expression of the effluent water quality is as follows: Wherein, TSS, COD, 、 、 and represent the concentration of suspended solids, chemical oxygen demand, Kjeldahl nitrogen concentration, nitrate nitrogen concentration, 5-day biochemical oxygen demand, and the discharge of clean water, respectively; The expression of the total energy consumption is as follows: In the formula: AE represents the aeration energy consumption, PE represents the pumping energy consumption, and EC represents the carbon source cost, and its expression is as follows: Among them, T is the sampling period; and respectively represent the start time and the end time; and respectively represent the volume and the aeration rate of the i th biochemical reaction tank; and respectively represent the internal return flow rate, the external return flow rate and the excess sludge flow rate; represents the flow rate of the additional carbon source added to the j th reaction zone.

3. The multi-objective optimization control method for sewage treatment process with over-standard suppression strategy according to claim 2, characterized in that, The process of optimizing the set values of the nitrate nitrogen concentration and dissolved oxygen concentration in the prediction model generates Pareto solutions, and select the set values of the dissolved oxygen concentration and nitrate nitrogen concentration for each cycle's over-standard situation. The specific method is as follows: Take the Pareto solutions obtained by optimization as the setpoint values of dissolved oxygen concentration and nitrate nitrogen concentration in turn, substitute them into the prediction model, and retain and store the solutions that meet the average effluent water quality constraints within this period in the solution set. ; If is not equal to the empty set, select the solution from that can minimize the energy consumption as the preferred solution; If is equal to the empty set, select the solution that can make the water quality reach a relatively good level as the preferred solution.

4. The multi-objective optimization control method for sewage treatment process of over-standard suppression strategy according to claim 3, characterized in that When the predicted output ammonia nitrogen concentration of the prediction model exceeds the standard, increase the external reflux flow rate to dilute and adjust the ammonia nitrogen concentration, and dynamically optimize the set values of the dissolved oxygen concentration and nitrate nitrogen concentration. First, set to the maximum value within the adjustable range, and then perform a secondary optimization of the nitrate nitrogen concentration.

5. The multi-objective optimization control method for sewage treatment process with over-standard suppression strategy according to claim 1, characterized in that The adjustment formula for the external reflux flow rate is as follows: In the formula, is the ammonia nitrogen concentration in the fifth partition, and ≤ 5 mg / L.

6. The multi-objective optimization control method for sewage treatment process with over-standard suppression strategy according to claim 5, wherein, When the predicted output effluent nitrogen concentration of the prediction model exceeds the standard, control the total amount of the added carbon source according to the prediction result, and its calculation formula is: Among them, is the total nitrogen concentration, is the total amount of externally added carbon source, and the adjustable range for controlling the total amount of externally added carbon source is 4 to 7 .

7. The multi-objective optimization control method for sewage treatment process with over-standard suppression strategy according to claim 6, characterized in that Select the percentage of water quality exceeding the standard time P as the evaluation index of the inhibition strategy, that is, the total duration of water quality exceeding the standard and the total operation time The ratio is expressed as: 。

Images