A sewage treatment process optimization control method based on a multi-objective sparrow algorithm

Abstract

The application relates to a sewage treatment process optimization control method based on a multi-objective sparrow algorithm, which comprises the following steps: adopting a least square support vector machine to establish a soft measurement model of total energy consumption and effluent water quality, and taking the soft measurement model as an optimization objective function; improving the sparrow algorithm from five aspects of population initialization, finder position updating, alarm position updating, optimal position disturbance strategy and external archive updating mechanism to obtain a multi-objective sparrow algorithm; optimizing the optimization objective function by using the improved multi-objective sparrow algorithm, obtaining a Pareto solution set, and optimizing the set values of dissolved oxygen concentration and nitrate nitrogen concentration; the scheme is improved according to the sparrow algorithm, and is applied to the sewage treatment process to optimize the set values of dissolved oxygen concentration and nitrate nitrogen concentration, so that the effluent water quality is improved and the energy consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment optimization control, and in particular to a sewage treatment process optimization control method based on a multi-objective sparrow algorithm. Background Art

[0002] With the rapid development of industrialization in my country, the environmental pollution caused by the large-scale discharge of industrial wastewater has seriously affected the lives of residents and the development of enterprises. The activated sludge process mainly uses equipment such as blowers and reflux pumps to keep the dissolved oxygen concentration and nitrate nitrogen concentration within a reasonable range, so that the wastewater discharge meets the standards. However, this process will consume a lot of energy. How to ensure that the wastewater discharge meets the standards and effectively reduce energy consumption has become an urgent problem to be solved. The dissolved oxygen concentration S of the 5th unit of the sewage treatment biochemical reactor is O,5 and the nitrate nitrogen concentration S in unit 2 NO,2 They are key control variables in the sewage treatment process. They not only affect the effluent water quality, but are also the most important parameters for controlling total energy consumption. Total energy consumption and effluent water quality are a pair of conflicting performance indicators. If you want to improve the effluent water quality, then the total energy consumption will inevitably increase; conversely, if the total energy consumption is reduced, then the effluent water quality will inevitably deteriorate. Therefore, the optimization problem of dealing with total energy consumption and effluent water quality indicators is essentially a multi-objective optimization problem. The purpose of multi-objective optimization control of the sewage treatment process is to ensure that the effluent water quality meets the emission standards and achieve energy conservation and consumption reduction. To solve this multi-objective optimization problem, it is necessary to determine the relationship between dissolved oxygen concentration, nitrate nitrogen concentration, total energy consumption, and effluent water quality indicators. Therefore, optimizing the set values ​​of dissolved oxygen concentration and nitrate nitrogen concentration is an important means to implement optimized control of the sewage treatment process. Summary of the Invention

[0003] An embodiment of the present invention provides a sewage treatment process optimization control method based on a multi-objective sparrow algorithm, and applies it to the sewage treatment process to optimize the dissolved oxygen concentration and nitrate nitrogen concentration, so as to achieve the purpose of improving the effluent water quality and reducing energy consumption. In order to have a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary is not a general review, nor is it intended to identify key / important components or to describe the scope of protection of these embodiments. Its sole purpose is to present some concepts in a simple form as a preface to the detailed description that follows.

[0004] The present application provides a sewage treatment process optimization control method based on a multi-objective sparrow algorithm, comprising the following steps: using a least squares support vector machine to establish a soft measurement model of total energy consumption and effluent water quality, and using it as the optimization objective function; improving the sparrow algorithm from five aspects, including population initialization, discoverer position update, sentinel position update, optimal position perturbation strategy, and external archive update mechanism, to obtain a multi-objective sparrow algorithm; using the improved multi-objective sparrow algorithm to optimize the optimization objective function, obtain a Pareto solution set, and optimize the set values ​​of dissolved oxygen concentration and nitrate nitrogen concentration.

[0005] As a preferred implementation, the optimization objective function is:

[0006] minf OCI (x),f EQI (x)

[0007]

[0008] Among them, f OCI (X) and f EQI (X) represents the optimization objective function of total energy consumption and effluent water quality, respectively. The constraints are the limit values ​​of the five effluent water quality parameters. S Nh,e,avg Indicates the average concentration of ammonia nitrogen in the effluent, x=[S O ,S NO ] is the optimization vector;

[0009] The expression of effluent water quality is as follows:

[0010]

[0011] In the formula, TSS, COD, S NKj 、S NO , BOD5, and Q e They represent the concentration of suspended solids, chemical oxygen demand, Kjeldahl nitrogen concentration, nitrate nitrogen concentration, 5-day biochemical oxygen demand and clean water discharge respectively;

[0012] The expression of total energy consumption is as follows:

[0013] OCI=AE+PE

[0014] Where: AE represents aeration energy consumption, PE represents pumping energy consumption, and the expression is as follows:

[0015]

[0016]

[0017] Where T is the sampling period; t0 and t f Represents the start time and end time respectively; Vi and K Lai Respectively represent the volume and aeration volume of the i-th biochemical reaction tank; Q a , Q r and Q w Represent the internal return flow, external return flow and excess sludge flow respectively; q EC,j Represents the flow rate of the external carbon source added to the jth reaction zone.

[0018] As a preferred implementation, the improvement of the sparrow algorithm from the perspective of population initialization includes initializing the population using a good point set, assuming G S is the unit cube of S-dimensional Euclidean space, that is, x∈G S ,X=(x1,x2,x3,...,x S ), where 0≤x i ≤1, i=1,2,...,S,G S The point r in the S ), let r∈G S , in the form of P n (k)={(r1(k),...,r S (k)),k=1,2,...,n} satisfy Where C(r,ε) is a constant related only to r and ε (ε>0), then P n (k) is a good point set, r is a good point, and the specific steps of initializing the population using the good point set include:

[0019] Generate a good point set X={x1,x2,...x i ,...,x k}(i=1,2,...,k), where k is the population size;

[0020] x i ={x i1 ,x i2 ,...x ij ,...,x id Any dimensional component in} is d is the spatial dimension;

[0021] Map the good point set to the search space: x ij =Lb j +mod(x ij ,1)×(Ub j -Lb j ), where x ij For the location of the sparrow, Ub j , Lb j Denotes the upper and lower bounds of the j-th dimension.

[0022] As a preferred embodiment, the finder location update adopts the following formula:

[0023]

[0024] w=r1-r2·(t / MaxIt) 2

[0025] Among them, X i is the position of the discoverer, R2(R2∈[0,1]) and ST(ST∈[0.5,1]) are the early warning and safety values ​​respectively, Q is a random number that obeys the normal distribution, L is a 1×d matrix with all internal elements being 1, r1 and r2 are two weight coefficients, and w is the decreasing inertia weight.

[0026] As a preferred embodiment, the position update of the sentinel adopts the following formula:

[0027]

[0028] w1=r3+r4·(t / MaxIt) 2

[0029] w2=r5-r6·(t / MaxIt) 2

[0030] Decreasing inertia weight w2, increasing inertia weight w1, r3, r4, r5, r6 are all weight coefficients, f i is the fitness value of the current sparrow individual, f g is the current global optimal fitness value, X z For the position of the sentinel, X best is the current global optimal position, X worst This is the current global worst position.

[0031] As a preferred embodiment, the optimal position disturbance strategy includes the following steps: m , disturb the optimal position, the transition probability P m The calculation formula is:

[0032]

[0033] Where t is the number of iterations; MaxIt is the maximum number of iterations.

[0034] As a preferred implementation, the calculation formula for disturbing the optimal position is:

[0035]

[0036] X'=rc ct+X best ,r c ∈(0,1)

[0037] Among them, ct is the disturbance term, r c is a random number between [0,1].

[0038] As a preferred embodiment, the external archive update mechanism includes comparing the relationship between individuals in the previous generation population and individuals in the new population to update the population. The specific steps are as follows: if an individual in the previous generation population dominates an individual in the new population, the position and fitness value of the corresponding individual in the previous generation population is replaced with the position and fitness value of the new population; if an individual in the new population dominates an individual in the previous generation population, it is replaced with a small probability to maintain the diversity of the population;

[0039] If neither of them dominates, they will be replaced with a 50% probability.

[0040] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0041] This solution improves the sparrow algorithm. This solution improves the multi-objective sparrow algorithm. First, the sparrow population is initialized using the good point set to make the population distribution more uniform and expand the search space. The optimal guidance strategy and dynamic inertia weight are introduced into the discoverer position update formula to improve the discoverer's origin convergence problem and balance the algorithm's global exploration and local development capabilities. Secondly, to improve the algorithm's local escape capability, the sentinel position update strategy is improved and the optimal position sparrow individual is disturbed. Finally, the population update method is introduced during the external archive update process to accelerate the algorithm convergence speed and increase population diversity.

[0042] It is also applied to the sewage treatment process to optimize the dissolved oxygen concentration and nitrate nitrogen concentration, so as to improve the effluent quality and reduce energy consumption.

[0043] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0045] Figure 1 is a diagram showing a multi-objective optimization control structure for sewage treatment according to an exemplary embodiment;

[0046] Figure 2 is a random number initialization population shown according to an exemplary embodiment;

[0047] Figure 3 is a tent mapping initialization population shown according to an exemplary embodiment;

[0048] Figure 4 is a population initialized by a good point set according to an exemplary embodiment;

[0049] Figure 5 is a schematic diagram of external archive update according to an exemplary embodiment;

[0050] Figure 6 is a flow chart of an MSIMOSSA algorithm according to an exemplary embodiment;

[0051] Figure 7 is a diagram showing an SO,5 tracking control effect according to an exemplary embodiment;

[0052] Figure 8 is a diagram showing an SNO,2 tracking control effect according to an exemplary embodiment;

[0053] Figure 9 is a graph of SO,5 tracking error according to an exemplary embodiment;

[0054] Figure 10 is a graph showing an SNO,2 tracking error according to an exemplary embodiment;

[0055] Figure 11 The figure shows the change process of five effluent water quality parameters according to an exemplary embodiment. DETAILED DESCRIPTION

[0056] The following description and accompanying drawings sufficiently illustrate the specific embodiments herein to enable those skilled in the art to practice them. Portions and features of some embodiments may be included in or substituted for portions and features of other embodiments. The scope of the embodiments herein includes the entire scope of the claims, including all available equivalents thereof. Herein, the terms "first," "second," and the like are used solely to distinguish one element from another and do not require or imply any actual relationship or order between these elements. In practice, the first element can also be referred to as the second element, and vice versa. Furthermore, the terms "comprise," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a structure, device, or apparatus comprising a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such structure, device, or apparatus. Without further limitation, an element defined by the phrase "comprising a..." does not preclude the presence of other identical elements in the structure, device, or apparatus comprising the element. The various embodiments herein are described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. Similar or identical parts between the various embodiments can be referenced to each other.

[0057] The terms "longitudinal," "transverse," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," and "outside" used herein to indicate orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely for ease of description and simplification. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and are therefore not to be construed as limitations on the present invention. In the description herein, unless otherwise specified or limited, the terms "mounted," "connected," and "connected" are to be understood broadly. For example, they may refer to mechanical or electrical connections, internal communication between two components, direct connections, or indirect connections through an intermediary. The specific meanings of the above terms will be understood by those skilled in the art based on the specific circumstances.

[0058] As used herein, unless otherwise specified, the term "plurality" means two or more.

[0059] In this document, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B means: A or B.

[0060] In this article, the term "and / or" is used to describe the association relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or, A and B.

[0061] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0062] Please refer to Figures 1-11 This embodiment provides a sewage treatment process optimization control method based on a multi-objective sparrow algorithm, comprising the following steps: using a least squares support vector machine to establish a soft measurement model of total energy consumption and effluent water quality, and using it as the optimization objective function; improving the sparrow algorithm from five aspects, including population initialization, discoverer position update, sentinel position update, optimal position perturbation strategy, and external archive update mechanism, to obtain a multi-objective sparrow algorithm; using the improved multi-objective sparrow algorithm to optimize the optimization objective function, obtain a Pareto solution set, and optimize the set values ​​of dissolved oxygen concentration and nitrate nitrogen concentration.

[0063] The least squares support vector machine (LSSVM) is used to train S O5 、S NO2 Modeling is performed with OCI and EQI; then, the sparrow algorithm is improved. First, the sparrow population is initialized with the good point set to make the population distribution more uniform and improve the search space. The optimal guidance strategy and dynamic inertia weight are introduced into the discoverer position update formula to improve the origin convergence problem of the discoverer and balance the global exploration and local development capabilities of the algorithm. In order to improve the local escape ability of the algorithm, the sentinel position update strategy is improved, and the sparrow individuals in the optimal position are disturbed. At the same time, the population update method is introduced in the external archive update process to accelerate the convergence speed of the algorithm and increase the population diversity; secondly, the improved algorithm is used to optimize the set values ​​of SO5 and SNO2; finally, in order to ensure that the total energy consumption of the system is effectively reduced under the premise of meeting the effluent water quality standards, a set of Pareto solution sets are obtained through the proposed algorithm. The total energy consumption of the system corresponding to each solution in the solution set is compared, and the solution with the lowest total energy consumption is selected as the current optimization set value, which is used as the controller tracking set value.

[0064] The multi-objective optimization control structure of sewage treatment consists of three parts: sewage treatment process module, multi-objective optimization module, and bottom tracking control module. Figure 1 As shown. The control structure mainly includes the following contents: first, total energy consumption, outlet water quality and S O,5 、S NO,2 There is no clear mathematical relationship between them, and offline collection of S is required. O,5 、S NO,2 , total energy consumption, and effluent water quality process data, LSSVM is used to establish a soft measurement model of total energy consumption and effluent water quality, and it is used as the optimization objective function; then the improved multi-objective sparrow algorithm is used to optimize the optimization objective function. The algorithm has good convergence and diversity, so it can ensure that the Pareto solution set obtained is relatively accurate and diverse; finally, the algorithm obtains S O,5 and S NO,2The optimized setting value is obtained by using PID tracking control, specifically by adjusting the dissolved oxygen conversion coefficient (K La5 ) and internal reflux (Q a ) respectively control the dissolved oxygen concentration and nitrate nitrogen concentration

[0065] To rationally compare control strategies and optimization methods, the International Water Quality Association and the European Commission for Scientific and Technological Cooperation jointly developed the Benchmark Simulation Model 1 (BSM1). The BSM1 model primarily consists of a biochemical reactor and a secondary sedimentation tank. The biochemical reactor has five reaction units: the first two are anaerobic, and the last three are aerobic, performing denitrification and nitrification, respectively. Denitrification reduces nitrate to gaseous nitrogen, while nitrification oxidizes ammonia nitrogen to nitrate.

[0066] In order to objectively evaluate the performance of control strategies and optimization methods, the overall cost index (OCI) and effluent quality index (EQI) were used as evaluation criteria. The OCI includes aeration energy (AE) and pumping energy (PE).

[0067] The optimization objective function is:

[0068] minf OCI (x),f EQI (x)

[0069]

[0070] Among them, f OCI (X) and f EQI (X) represents the optimization objective function of total energy consumption and effluent water quality, respectively. The constraints are the limit values ​​of the five effluent water quality parameters. S Nh,e,avg Indicates the average concentration of ammonia nitrogen in the effluent, x=[S O ,S NO ] is the optimization vector;

[0071] The expression of effluent water quality is as follows:

[0072]

[0073] In the formula, TSS, COD, S NKj 、S NO , BOD5, and Q e They represent the concentration of suspended solids, chemical oxygen demand, Kjeldahl nitrogen concentration, nitrate nitrogen concentration, 5-day biochemical oxygen demand and clean water discharge respectively;

[0074] The expression of total energy consumption is as follows:

[0075] OCI=AE+PE

[0076] Where: AE represents aeration energy consumption, PE represents pumping energy consumption, and the expression is as follows:

[0077]

[0078]

[0079] Where T is the sampling period; t0 and t f Represents the start time and end time respectively; 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 Represent the internal return flow, external return flow and excess sludge flow respectively; q EC,j Represents the flow rate of the external carbon source added to the jth reaction zone.

[0080] The Sparrow Algorithm is a novel intelligent optimization algorithm that simulates the foraging process of sparrows. During their foraging process, sparrows are divided into three groups: discoverers, joiners, and sentinels. Discoverers are individuals with the highest fitness scores, while the others are joiners. The identities of discoverers and joiners are dynamic and can be switched. A subset of discoverers and joiners are randomly selected as sentinels. Discoverers are responsible for providing foraging areas and directions for the population, and their number ranges from 10% to 20% of the population. The formula for updating the discoverer position is as follows:

[0081]

[0082] Among them, X i is the position of the discoverer, R2(R2∈[0,1]) and ST(ST∈[0.5,1]) are the early warning and safety values ​​respectively, Q is a random number that obeys the normal distribution, and L is a 1×d matrix with all internal elements being 1.

[0083] Joiners always monitor the discoverers. Joiners with poor food sources will compete for food with discoverers with better food sources, while joiners with better food sources will conduct extensive searches. The position update formula is as follows:

[0084]

[0085] Among them, X j is the joiner position, X p is the optimal position currently occupied by the discoverer, X worstis the current global worst position, A is a 1×d matrix, each element in the matrix is ​​randomly assigned to 1 or -1, A + =A T (AA T ) -1 .

[0086] The sentinels are responsible for alerting danger. When they detect danger, they will quickly flee their current position and flee to the vicinity of the sparrow individual at the current global optimal position. Their number generally accounts for 10% to 20% of the population. The position update formula is as follows:

[0087]

[0088] Among them, X z For the position of the sentinel, X best is the current global optimal position, β is the step size control parameter, and is a random number that obeys a normal distribution with a mean of 0 and a variance of 1. K∈[-1,1] is a random number, and f i is the fitness value of the current sparrow individual, f g and f w are the current global best and worst fitness values ​​respectively, and ε is a very small constant to prevent the denominator from being zero.

[0089] Based on the sparrow algorithm, this scheme improves it from five aspects, including population initialization, discoverer and sentinel position update, optimal position perturbation strategy and external archive update mechanism, and proposes a multi-objective sparrow algorithm that integrates multiple strategies.

[0090] The improvement of the sparrow algorithm from the perspective of population initialization includes using the best point set to initialize the population, setting G S is the unit cube of S-dimensional Euclidean space, that is, x∈G S ,X=(x1,x2,x3,...,x S ), where 0≤x i ≤1, i=1,2,...,S,G S The point r in the S ), let r∈G S , in the form of P n (k)={(r1(k),...,r S (k)),k=1,2,...,n} satisfy Where C(r,ε) is a constant related only to r and ε (ε>0), then P n(k) is the set of good points, and r is the good point. The algorithm's optimization performance is related to the diversity of the initial population. The better the diversity, the faster the algorithm converges and the higher the accuracy. In the basic sparrow algorithm, the population is initialized using random numbers. This method is prone to uneven population distribution and poor diversity, causing the algorithm to easily fall into local optimality. The population initialized based on the good point set is more evenly distributed and has better initial population diversity, which can effectively improve the algorithm's performance. The specific steps for initializing the population using the good point set include:

[0091] Generate a good point set X={x1,x2,...x i ,...,x k}(i=1,2,...,k), where k is the population size;

[0092] x i ={x i1 ,x i2 ,...x ij ,...,x id Any dimensional component in} is d is the spatial dimension;

[0093] Map the good point set to the search space: x ij =Lb j +mod(x ij ,1)×(Ub j -Lb j ), where x ij For the location of the sparrow, Ub j , Lb j Denotes the upper and lower bounds of the j-th dimension.

[0094] In order to verify the superiority of the population initialized by the good point set, it is compared with the population initialized by random numbers and the population initialized by tent mapping. The random number initialization method generates the initial population by randomly generating individuals in the solution space. The tent mapping initialization method first uses tent mapping to generate a chaotic sequence, and then maps the sequence to the solution space to generate the initial population. Figure 2 -Attached Figure 4 By comparison, the population generated based on the initialization of the good point set is more evenly distributed and can better ensure population diversity.

[0095] In the original discoverer position update strategy, when R2 < ST, the discoverer is prone to converge to the origin in the later stage of iteration. Moreover, since the discoverer guides the population to search, it is easy for the algorithm to fall into local optimum. When the sparrow individuals do not detect danger, the original position update formula does not utilize the global optimum position, and there is a lack of sufficient and effective communication among the discoverers. The improved formula incorporates an optimal guidance strategy to enable the rapid transmission of the optimal position information among the discoverers. At the same time, a decreasing inertia weight w is introduced, enabling the algorithm to have a strong global search ability in the early stage of iteration and a strong local development ability in the later stage of iteration. The updated position of the discoverer after improvement adopts the following formula:

[0096]

[0097] w = r1 - r2·(t / MaxIt) 2

[0098] where, X i is the position of the discoverer, R2 (R2 ∈ [0, 1]) and ST (ST ∈ [0.5, 1]) are the early warning value and the safety value respectively, Q is a random number subject to normal distribution, L is a 1×d matrix with all internal elements being 1, r1 and r2 are two weight coefficients, and w is a decreasing inertia weight.

[0099] For the vigilant individuals at the edge of the population, that is, individuals with f i ≠ f g , a decreasing inertia weight w2 is adopted to make them gradually approach the optimal position and accelerate the convergence of the algorithm; for the vigilant individuals at the center of the population, an increasing inertia weight w1 is adopted to enhance the ability of the algorithm to jump out of local optimum in the later stage of iteration. At the same time, in the multi-objective optimization problem, the fitness values f i and f w are multi-dimensional vectors, and the position of the vigilant individual cannot be calculated through formula . Therefore, its update formula is improved, and the updated position of the vigilant individual adopts the following formula:

[0100]

[0101] w1 = r3 + r4·(t / MaxIt) 2

[0102] w2 = r5 - r6·(t / MaxIt) 2

[0103] The decreasing inertia weight w2, the increasing inertia weight w1, r3, r4, r5, and r6 are all weight coefficients, f i is the fitness value of the current sparrow individual, f g is the fitness value of the current global best, X zFor the position of the sentinel, X best is the current global optimal position, X worst This is the current global worst position.

[0104] The improved position update formulas of the discoverer and the vigilant both introduce the global optimal position. Since the algorithm gradually converges to the optimal individual in the later iteration, if the optimal individual is the local optimal individual, the algorithm will fall into the local optimal state. To solve this problem, this paper adopts a perturbation strategy to perturb the optimal position through the conversion probability to improve the ability of the algorithm to jump out of the local optimal state. The optimal position perturbation strategy includes the conversion probability P m , disturb the optimal position, the transition probability P m The calculation formula is:

[0105]

[0106] Where t is the number of iterations; MaxIt is the maximum number of iterations.

[0107] From the above formula, we can see that the conversion probability decreases as the number of iterations increases. m When it is greater than the random number r generated between [0,1], the optimal position is not disturbed. m If the value is large, the probability of perturbation is small, so that the algorithm can quickly find the global optimal solution in the early stage; otherwise, the optimal position is disturbed, and the algorithm will find the optimal solution in the later stage of P iteration. m The smaller the value, the greater the probability of disturbance. The calculation formula for disturbing the optimal position is:

[0108]

[0109] X'=r c ct+X best ,r c ∈(0,1)

[0110] Among them, ct is the disturbance term, r c is a random number between [0,1].

[0111] The external archive update mechanism includes comparing the relationship between individuals in the previous generation population and individuals in the new population to update the population. The specific steps are: if the individuals in the previous generation population dominate the individuals in the new population, the positions and fitness values ​​of the corresponding individuals in the previous generation population are replaced with the positions and fitness values ​​of the new population; if the individuals in the new population dominate the individuals in the previous generation population, then they are replaced with a small probability to maintain the diversity of the population; if they do not dominate each other, they are replaced with a 50% probability.

[0112] Through this population update method, individuals with high fitness can proceed to the next iteration, which improves the offspring utilization rate and convergence speed, and improves the diversity of the population to a certain extent.

[0113] The updated population is used as the next generation population, and non-dominated sorting is performed with the optimal solution set archive to generate a new elite external archive, which replaces the previous generation optimal solution set archive when the next generation is updated. This cycle is repeated until the termination condition is met, and the final elite archive is output. The specific update process is as follows Figure 5 As shown. Figure 6 This is a flowchart of a multi-objective algorithm integrating multiple strategies. The confidence intervals for the algorithm's weight coefficients are determined based on prior knowledge, and then a grid search algorithm is used to verify and optimize the final weight coefficients. The overall process of the proposed algorithm is as follows: the population is initialized using a set of good points, and the globally optimal and worst sparrow individuals are selected simultaneously; the globally optimal sparrow individual is perturbed and iteratively updated using an improved position update formula; an elite external archive is generated using a population update method, and the archive size is maintained using an adaptive grid method. The algorithm's time complexity is on the same order of magnitude as NSGAII.

[0114] Tables 1 and 2 show the numerical simulation performance of the proposed algorithm. Inverted Generational Distance (IGD) and Hypervolume (HV) are used as performance indicators to evaluate the convergence and diversity of the algorithm. Tables 1 and 2 show that the proposed algorithm has improved convergence and distribution compared to the comparison algorithms. Therefore, it can be concluded that the proposed algorithm performs better.

[0115] Table 1 Average values ​​of IGD

[0116]

[0117] Table 2 HV average values

[0118]

[0119] Table 3 Parameter settings

[0120]

[0121] The algorithm and PID controller parameter settings are shown in Table 3. Figure 7 -Attached Figure 10 For multi-objective optimization control S O5set and S NO2set From the tracking effect and tracking error diagram, it can be seen that the set values ​​of the dissolved oxygen concentration in the fifth zone and the nitrate nitrogen concentration in the second zone will be continuously adjusted along with the reaction process, and the PID controller can track and control them more accurately.

[0122] Attachment Figure 11 The figure shows the change process of five effluent water qualities in 14 days under multi-objective optimization control of sewage treatment. It can be seen that within 14 days, BOD5, COD and TSS are all lower than the effluent water quality parameter indicators, which are 10mg / L, 100mg / L and 30mg / L respectively. However, it can also be seen that S Nh and S Ntot There were peak values ​​exceeding the standard within 14 days. Since the average values ​​for 14 days were 3.88 mg / L and 15.60 mg / L, respectively, which were lower than the respective index values ​​of 4 mg / L and 18 mg / L, the discharge conditions were met.

[0123] Table 4 shows the energy consumption comparison and five effluent quality comparisons obtained by different optimization control methods under clear weather conditions. As can be seen from Table 4, the MSIMOSSA-based optimization control method can effectively reduce energy consumption while ensuring effluent quality, proving the effectiveness of this method.

[0124] Table 4 Comparison of energy consumption and effluent water quality of different optimization control methods under clear weather

[0125]

[0126] In one embodiment, a computer device is provided, which may be a server. The computer device includes a processor, a memory, and a network interface connected via a system bus. 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 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 an external terminal via a network connection. When the computer program is executed by the processor, the steps of the above-mentioned method embodiment are implemented.

[0127] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the 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-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. 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. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0128] It should be noted that the above description is only an illustration of some embodiments of the present application and the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in this application is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also includes other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, the above-mentioned features can be replaced with (but not limited to) technical features with similar functions disclosed in this application.

[0129] In addition, although adopting specific order to describe each operation, this should not be interpreted as requiring these operations to be executed in the specific order shown or in sequential order.Under certain environment, multitasking and parallel processing may be advantageous.Similarly, although comprising some specific implementation details in the above discussion, these should not be interpreted as limiting the scope of the application.Some features described in the context of separate embodiment can also be implemented in a single embodiment in combination.On the contrary, the various features described in the context of a single embodiment also can be implemented in multiple embodiments individually or in the mode of any suitable subcombination.

[0130] Although the subject matter has been described in language specific to structural features and / or methodological logical 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 sewage treatment process optimization control method based on a multi-objective sparrow algorithm, characterized in that: The following steps are involved: The soft sensing model of total energy consumption and effluent water quality was established using least squares support vector machine and used as the optimization objective function. The multi-objective sparrow algorithm is obtained by improving the sparrow algorithm from five aspects: population initialization, discoverer position update, sentinel position update, optimal position perturbation strategy, and external archive update mechanism. The improved multi-objective sparrow algorithm is used to optimize the optimization objective function, obtain the Pareto solution set, and optimize the set values ​​of dissolved oxygen concentration and nitrate nitrogen concentration; the optimization objective function is: Among them, f OCI (X) and f EQI (X) represents the optimization objective function of total energy consumption and effluent water quality, respectively. The constraints are the limit values ​​of the five effluent water quality parameters. S Nh,e,avg Indicates the average concentration of ammonia nitrogen in the effluent, x=[SO,S NO ] is the optimization vector; The expression of effluent water quality is as follows: In the formula, TSS, COD, S NKj 、S NO , BOD5, and Q e They represent the concentration of suspended solids, chemical oxygen demand, Kjeldahl nitrogen concentration, nitrate nitrogen concentration, 5-day biochemical oxygen demand and clean water discharge respectively; The expression of total energy consumption is as follows: OCI=AE+PE Where: AE represents aeration energy consumption, PE represents pumping energy consumption, and the expression is as follows: Where T is the sampling period; t0 and t f Represents the start time and end time respectively; 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 They represent internal recirculation flow, external recirculation flow and excess sludge flow respectively; The improvement of the sparrow algorithm from the perspective of population initialization includes initializing the population with a good point set, assuming that GS is the unit cube of the S-dimensional Euclidean space, that is, x∈G S ,X=(x1,x2,x3,...,x S ), where 0≤x i ≤1, i=1,2,...,S,G S The point r in the S ), let r∈G S , in the form of P n (k)={(r1(k),...,r S (k)),k=1,2,...,n} satisfy Where C(r,ε) is a constant that is only related to r and ε (ε>0), then Pn(k) is called the good point set, r is a good point, and the specific steps of initializing the population using the good point set include: Generate a good point set X={x1,x2,...x i ,...,x k }(i=1,2,...,k), where k is the population size; x i ={x i1 ,x i2 ,...x ij ,...,x id Any dimensional component in} is , d is the spatial dimension; Map the good point set to the search space: x ij =Lb j +mod(x ij ,1)×(Ub j -Lb j ), where x ij For the location of the sparrow, Ub j , Lb j represents the upper and lower bounds of the j-th dimension; The finder location is updated using the following formula: Among them, X i is the position of the discoverer, R2(R2∈[0,1]) and ST(ST∈[0.5,1]) are the early warning person and the safety value respectively, Q is a random number that obeys the normal distribution, L is a 1×d matrix with all internal elements set to 1, r1 and r2 are two weight coefficients, and w is the decreasing inertia weight; The sentinel position is updated using the following formula: Decreasing inertia weight w2, increasing inertia weight w1, r3, r4, r5, r6 are all weight coefficients, f i is the fitness value of the current sparrow individual, f g is the current global optimal fitness value, X z For the position of the sentinel, X best is the current global optimal position, X worst is the current global worst position; The optimal position perturbation strategy includes perturbing the optimal position through the conversion probability Pm, and the calculation formula of the conversion probability Pm is: Where t is the number of iterations; MaxIt is the maximum number of iterations; The calculation formula for disturbing the optimal position is: Among them, ct is the disturbance term, r c is a random number between [0,1].

2. The sewage treatment process optimization control method based on the multi-objective sparrow algorithm according to claim 1 is characterized in that: The external archive update mechanism involves comparing the relationships between individuals in the previous generation population and individuals in the new population to update the population. The specific steps are: If an individual from the previous generation of population dominates an individual from the new generation of population, the position and fitness value of the corresponding individual from the previous generation of population will be replaced by the position and fitness value of the new population; if an individual from the new population dominates an individual from the previous generation of population, then the replacement will be made with a small probability to maintain the diversity of the population; if neither of them dominates each other, the replacement will be made with a 50% probability.

Images