CN113674809A - Sewage treatment carbon source adding method based on predictive control - Google Patents
Sewage treatment carbon source adding method based on predictive control Download PDFInfo
- Publication number
- CN113674809A CN113674809A CN202111064097.9A CN202111064097A CN113674809A CN 113674809 A CN113674809 A CN 113674809A CN 202111064097 A CN202111064097 A CN 202111064097A CN 113674809 A CN113674809 A CN 113674809A
- Authority
- CN
- China
- Prior art keywords
- carbon source
- total nitrogen
- solution
- concentration
- water quality
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C10/00—Computational theoretical chemistry, i.e. ICT specially adapted for theoretical aspects of quantum chemistry, molecular mechanics, molecular dynamics or the like
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
Abstract
The invention provides a method for adding a carbon source for sewage treatment based on predictive control. The method comprises the following steps: establishing an energy consumption and water quality model, optimizing the energy consumption and water quality model by adopting an NSGA2 algorithm, and obtaining a solution set with the lowest energy consumption or the highest water quality as SO,5Concentration and SNO,2A set value of concentration; inputting the solution set with the lowest energy consumption or the highest water quality into a total nitrogen prediction model of the effluent of the sewage treatment to obtain a total nitrogen prediction value corresponding to the current input quantity; judging whether the total nitrogen predicted value exceeds the standard or not and adopting different carbon source fuzzy control rules to feed the carbon source, wherein the method specifically comprises the following steps: when the total nitrogen predicted value does not exceed the standard, adopting a carbon source fuzzy control rule, and passing through the error and the error change of the current concentration and the set valueThe oxygen conversion coefficient K of the fifth subarea is controlled by the quantityLa5And internal reflux quantity QaThereby realizing the tracking control of the total nitrogen concentration; and when the total nitrogen prediction value exceeds the standard, performing fuzzy control on the first and second partition external carbon sources by adopting a carbon source fuzzy control rule according to the total nitrogen prediction value.
Description
Technical Field
The invention relates to the technical field of sewage treatment, in particular to a method for adding a carbon source for sewage treatment based on predictive control.
Background
The Total nitrogen (Total nitrogen) concentration in the sewage treatment process is an important index for measuring the effluent quality, and the excessive discharge of the Total nitrogen can cause the serious consequences of the reduction of the dissolved oxygen concentration of a water body, the eutrophication of the water body, the death of biological poisoning and the like. Aiming at the problem, national emission standards for pollutants from municipal wastewater treatment plants (GB18918-2002) were introduced in 2002, and the daily highest allowable emission concentration of a basic control project for water pollutant emission is specified, wherein the first-class A emission standard of total nitrogen is 15mg/L, and the second-class A emission standard is 20 mg/L.
At present, the commonly used total nitrogen treatment process mainly comprises a traditional biological denitrification process, a novel biological denitrification process and a physical and chemical process. Wherein, the traditional biological denitrification process is A2/O, SBR and the like, has the advantages of classical process structure and high ammonia nitrogen removal rate, but has complex operation, long period, high treatment cost and easy influence of factors such as environmental temperature and the like; novel biological denitrification processes such as a short-range biological denitrification technology and a synchronous digestion denitrification technology have good removal effect when the concentration of ammonia nitrogen wastewater is low, but are easily influenced by conditions such as dissolved oxygen, pH value and the like; the physical and chemical methods such as precipitation, stripping and adsorption, and the chemical methods are expensive to purchase, and most sewage treatment plants cannot afford long-term treatment cost. In order to solve the problem that the total nitrogen concentration exceeds the standard, a technical measure of adding a carbon source to the anoxic tank to strengthen denitrification is usually adopted to ensure that the total nitrogen concentration of the effluent reaches the standard. The main feeding control modes are two types: one is a manual control mode, and workers manually control a carbon source according to the change trend of the effluent quality of sewage. The method is excessively dependent on manual experience and expert knowledge, and the input amount is constant within a certain time, so that the condition of excessive or insufficient carbon source addition is easily caused, and a series of risks such as excessive total nitrogen of effluent, increased cost, capacity of an aerobic tank occupied by the effluent are caused. The other type is a carbon source adding control mode based on feedforward-feedback, relevant indexes such as water quality of inlet water and water quality of outlet water are measured on line, a feedforward and feedback combination mode is adopted, and a PLC controller is used for automatically controlling the adding amount of the carbon source, but the time of the sewage treatment process is long, the affected factors are numerous, the amount of the carbon source is prone to being inaccurate, and the feedback link has certain hysteresis, so that the effluent of the urban sewage treatment reaches the standard, and the energy conservation and consumption reduction are not facilitated.
Therefore, how to optimally add a carbon source to reduce the total nitrogen concentration and realize optimal control, energy conservation, consumption reduction and peak value inhibition is still a difficult point to be solved urgently on the premise of reaching the effluent standard in the sewage treatment process.
Disclosure of Invention
In order to solve the problem that the total nitrogen of the effluent of the sewage treatment exceeds the standard, the invention provides a method for adding a carbon source for the sewage treatment based on prediction control.
The invention provides a sewage treatment carbon source adding method based on predictive control, which comprises the following steps:
step 1: establishing an energy consumption and water quality model, optimizing the energy consumption and water quality model by adopting an NSGA2 algorithm, and obtaining a solution set with the lowest energy consumption or the highest water quality as a fifth subarea dissolved oxygen SO,5Concentration and second partition nitrate nitrogen SNO,2A set value of concentration;
step 2: inputting the solution set with the lowest energy consumption or the highest water quality into a constructed sewage treatment effluent total nitrogen prediction model to obtain a total nitrogen prediction value corresponding to the current input quantity;
and step 3: judging whether the total nitrogen predicted value exceeds the standard or not and adopting different carbon source fuzzy control rules to feed the carbon source, wherein the method specifically comprises the following steps: and when the total nitrogen prediction value exceeds the standard, performing fuzzy control on the first and second partition external carbon sources by adopting a carbon source fuzzy control rule according to the total nitrogen prediction value.
Further, step 1 specifically includes:
step 1.1: selecting F1And F2Respectively an energy consumption objective function and a water quality objective function, f1And f2Describing the optimization problem of energy consumption and water quality by adopting formulas (1) to (3) respectively as an energy consumption model and a water quality model:
X=(x1,x2)
li≤xi≤ui,i=1,2 (2)
△=100CNtot
wherein, Delta is an overproof penalty term CNtotPredicting the effluent total nitrogen superscalar value; f. ofNtotRepresenting a total nitrogen prediction model of effluent of sewage treatment; scIs a set total nitrogen concentration upper limit value; x represents the solution of the energy consumption and water quality optimization problem, XiDenotes the i-th decision variable, l, in the solutioni、uiAre decision variables x, respectivelyiA lower and an upper bound of (1), wherein x1And x2Each specifically represents the fifth region dissolved oxygen SO,5Concentration and second partition nitrate nitrogen SNO,2Concentration;
step 1.2: let p solve XpIs (F)1(Xp),F2(Xp) Calculate the p-th solution X after normalization according to equation (4)pThen choosing the non-dominant solution X ═ and (c) with the smallest current sparsityx1,x2) As a sparse solution:
wherein, SP (X)p) Representing the p-th solution X after normalizationpSparsity of WpIs the sum target vector (F) in the target function space1(Xp),F2(Xp) The Euclidean distance between the target vectors is less than r, r is a positive real number less than 1, and N is the set initial population number;
step 1.3: performing cross variation after selecting the sparse solution to avoid falling into local optimum, specifically performing local search by using a limit optimization variation method according to formulas (5) to (8) to generate a local solution; meanwhile, carrying out mutation operation by using a random immigration strategy according to a formula (9) and a formula (10) to generate [0.2N ] local solutions;
Xp=(x′1,x′2),p=1,2,...,N (5)
x′i=xi+α·βmax(xi),i=1,2 (6)
βmax(xi)=max[xi-li,ui-xi],i=1,2 (8)
XXk=(xx′1,xx′2),k=1,2,...,[0.2N] (9)
xx′i=γxi,i=1,2,0<γ<1.2 (10)
wherein, XpDenotes the solution after Limit-optimized mutation update, x'iRepresents the ith decision variable in the solution after the updating of the extreme optimization variation, h and gamma are random numbers, q represents the shape parameter, betamax(xi) For the current decision variable xiMaximum value of variability, XXkRepresenting random immigration strategy variationNew solution, xx'iThe ith decision variable in the solution after the random immigration strategy variation updating;
step 1.4: performing non-dominated sorting and congestion distance calculation on all the solutions obtained in the step 1.3, selecting N solutions from the solutions to form a next generation population, and iteratively executing the step 1.2 to the step 1.4 until a preset iteration step number is reached; the selection rule for forming the next generation population is to preferentially select a non-dominant solution, and when the number of solutions is less than N, a suboptimal solution is selected;
step 1.5: the N non-dominated solutions finally obtained in the step 1.4 are brought into a constructed sewage treatment effluent total nitrogen prediction model, and if the total nitrogen prediction value does not exceed the standard, a solution set with the lowest energy consumption is selected as a fifth subarea dissolved oxygen SO,5Concentration and second partition nitrate nitrogen SNO,2A set value of concentration; if the predicted value of the total nitrogen exceeds the standard, selecting the solution set with the highest water quality as the dissolved oxygen S of the fifth subareaO,5Concentration and second partition nitrate nitrogen SNO,2The set value of the concentration.
Further, the input of the sewage treatment effluent total nitrogen prediction model further comprises total nitrogen of inflow, inflow flow and current effluent total nitrogen.
Further, the carbon source fuzzy control rule comprises:
if SNtot>19.7mg/L, setting the first partition carbon source qEC1=5m3D, second partition carbon source qEC2=2m3/d;
If 19.0. ltoreq.SNtotLess than or equal to 19.7mg/L, and a first subarea external carbon source q is arrangedEC1=4m3D, second partition carbon source qEC2=2m3/d;
If 18.0. ltoreq.SNtotLess than or equal to 18.9mg/L, and a first subarea external carbon source q is arrangedEC1=4m3D, second partition carbon source qEC2=1m3/d;
If 17.0. ltoreq.SNtotLess than or equal to 17.9mg/L, and a first subarea is provided with an external carbon source qEC1=3m3D, second partition carbon source qEC2=0m3/d;
If SNtot<17.0mg/L, setting the first partition external carbon source qEC1=0m3D, second partition carbon source qEC2=0m3/d。
The invention has the beneficial effects that:
the method utilizes the neural network to establish a total nitrogen prediction model, predicts whether the concentration of the ammonia nitrogen in the effluent exceeds the standard in advance on the premise of obtaining the lowest energy consumption or the highest water quality, and finely controls the carbon source feeding in the sewage treatment anoxic tank by adopting a fuzzy rule if the concentration of the ammonia nitrogen exceeds the standard, thereby avoiding the problems of excessive or insufficient carbon source generated by manual feeding and feedback feeding. Compared with the prior art, the invention overcomes the problems of insufficient measurement precision, delayed control of effluent concentration and the like caused by complex sewage treatment process and long reaction time, and has good application prospect.
Drawings
FIG. 1 is a schematic flow chart of a method for adding a carbon source for sewage treatment based on predictive control according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a model for predicting total nitrogen in effluent from sewage treatment according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a control process of carbon source control rules according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of the total nitrogen concentration variation curve of effluent without adding carbon source provided by the embodiment of the present invention;
FIG. 5 is a schematic diagram of a curve of the total nitrogen concentration of the effluent after the carbon source is added according to the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1 and 3, an embodiment of the present invention provides a method for adding a carbon source for sewage treatment based on predictive control, including the following steps:
s101: establishing an energy consumption and water quality model, optimizing the energy consumption and water quality model by adopting an NSGA2 algorithm, and obtaining a solution set with the lowest energy consumption or the highest water quality as a fifth subarea dissolved oxygen SO,5Concentration and second partition nitrate nitrogen SNO,2A set value of concentration;
specifically, an energy consumption water quality model is established by adopting a feedforward neural network. The input variables of the energy consumption model and the water quality model are the concentration and S of the suspended matter in the effluentNtot、SO,5And SNO,2. Data acquisition regardless of time delay, data is acquired every 15 minutes, SNO,2Concentration set value and SO,5The concentration set value takes different values within a value range and is tested on a BSM1 platform.
As an implementation manner, this step specifically includes the following sub-steps:
s1011: selecting F1And F2Respectively an energy consumption objective function and a water quality objective function, f1And f2Describing the optimization problem of energy consumption and water quality by adopting formulas (1) to (3) respectively as an energy consumption model and a water quality model:
X=(x1,x2)
li≤xi≤ui,i=1,2 (2)
△=100CNtot
wherein, Delta is an overproof penalty term CNtotPredicting the effluent total nitrogen superscalar value; f. ofNtotRepresenting a total nitrogen prediction model of effluent of sewage treatment; scThe value in the embodiment is 17.0mg/L for the set upper limit value of the total nitrogen concentration; x represents energy consumptionAnd solution of water quality optimization problem, xiDenotes the i-th decision variable, l, in the solutioni、uiAre decision variables x, respectivelyiA lower and an upper bound of (1), wherein x1And x2Each specifically represents the fifth region dissolved oxygen SO,5Concentration and second partition nitrate nitrogen SNO,2Concentration;
s1012: let p solve XpIs (F)1(Xp),F2(Xp) Calculate the p-th solution X after normalization according to equation (4)pThen selecting the non-dominant solution X ═ X (X) with the minimum current sparsity1,x2) As a sparse solution:
wherein, SP (X)p) Representing the p-th solution X after normalizationpSparsity of WpIs the sum target vector (F) in the target function space1(Xp),F2(Xp) The Euclidean distance between the target vectors is less than r, and r is a positive real number less than 1; n is the set initial population number, which is taken as 100 in this example.
S1013: performing cross variation after selecting the sparse solution to avoid falling into local optimum, specifically performing local search by using a limit optimization variation method according to formulas (5) to (8) to generate a local solution; meanwhile, carrying out mutation operation by using a random immigration strategy according to a formula (9) and a formula (10) to generate [0.2N ] local solutions;
Xp=(x′1,x′2),p=1,2,...,N (5)
x′i=xi+α·βmax(xi),i=1,2 (6)
βmax(xi)=max[xi-li,ui-xi],i=1,2 (8)
XXk=(xx′1,xx′2),k=1,2,...,[0.2N] (9)
xx′i=γxi,i=1,2,0<γ<1.2 (10)
wherein, XpDenotes the solution after Limit-optimized mutation update, x'iRepresents the ith decision variable in the solution after the updating of the extreme optimization variation, h and gamma are random numbers, q represents the shape parameter, betamax(xi) For the current decision variable xiMaximum value of variability, XXkRepresents a solution after the variation update of the random immigration policy, xx'iThe ith decision variable in the solution after the random immigration strategy variation updating;
s1014: performing non-dominated sorting and congestion distance calculation on all the solutions obtained in the step S1013, selecting N solutions from the solutions to form a next generation population, and iteratively executing the steps S1012 to S1014 until a preset iteration step number is reached; the selection rule for forming the next generation population is to preferentially select a non-dominant solution, and when the number of solutions is less than N, a suboptimal solution is selected;
step 1.5: the N non-dominated solutions finally obtained in the step S1014 are brought into the constructed sewage treatment effluent total nitrogen prediction model, and if the total nitrogen prediction value does not exceed the standard, the solution set with the lowest energy consumption is selected as the fifth subarea dissolved oxygen SO,5Concentration and second partition nitrate nitrogen SNO,2A set value of concentration; if the predicted value of the total nitrogen exceeds the standard, selecting the solution set with the highest water quality as the dissolved oxygen S of the fifth subareaO,5Concentration and second partition nitrate nitrogen SNO,2The set value of the concentration.
As an implementation mode, a neural network is adopted to construct a sewage treatment water total nitrogen prediction model, as shown in fig. 2, the sewage treatment water total nitrogen prediction model comprises an input layer, a hidden layer and an output layer; wherein, the input layer contains p neurons, and the output of the input layer neurons is:the hidden layer is output asc is the center vector and σ is the width. Output layerw is the output weight. The input of the sewage treatment effluent total nitrogen prediction model comprises total nitrogen entering water, water entering flow, current effluent total nitrogen, dissolved oxygen concentration and nitrate nitrogen concentration, and the output of the sewage treatment effluent total nitrogen prediction model is an effluent total nitrogen concentration prediction value.
As an implementation mode, when the total nitrogen prediction model of the effluent of sewage treatment is trained, data sampled by a BSM1 experiment platform are selected as model training test data. Dissolved oxygen SO,5The concentration set value is set to be 1.4-2.4mg/l, and nitrate nitrogen S is addedNO,2The concentration set value is set to be 0.5-1.5mg/l, after the set value is set, the BSM1 model runs for 14 days, sampling is carried out once every 15 minutes, and the 181390 group data are obtained in a combination mode of different set values. Wherein 163251 sets of data were used as training samples and 18139 sets of data were used as test samples. A neural network structure of 5-50-1 is selected through an empirical method, a sample sampling normalization method is adopted, a gradient algorithm is adopted in a learning algorithm, the learning rate is 0.1, the maximum learning step number is 4000 steps, and RMSE is a performance evaluation index to verify the network accuracy.
As an implementation manner, the parameter S in the above optimization algorithmO,5The concentration set point is set at 1.4-2.4mg/l, SNO,2The concentration set point was set at 0.5-1.5 mg/l. The number N of the initial population is 100, the maximum function call frequency is 30, the crossover probability is 0.9, the mutation probability is 0.01, and the shape parameter q is 11.
S102: inputting the solution set with the lowest energy consumption or the highest water quality into a constructed sewage treatment effluent total nitrogen prediction model to obtain a total nitrogen prediction value corresponding to the current input quantity;
s103: judging whether the total nitrogen predicted value exceeds the standard or not and adopting different carbon source fuzzy control rules to feed the carbon source, wherein the method specifically comprises the following steps: when the total nitrogen predicted value does not exceed the standard, adopting a conventional tracking control method, and setting the total nitrogen by predicting the current concentration of the total nitrogen and the total nitrogenThe error and the error variation of the value respectively control the oxygen conversion coefficient K of the fifth subareaLa5And internal reflux quantity QaThereby realizing the tracking control of the total nitrogen concentration; when the total nitrogen prediction value exceeds the standard, a carbon source fuzzy control rule is adopted, fuzzy control is carried out on the first partition external carbon source and the second partition external carbon source according to the total nitrogen prediction value, the carbon source is increased, the denitrification effect can be promoted, and therefore nitrogen elements are removed.
According to the national sewage treatment discharge standard, the discharge standard of B class is executed on the effluent of a certain sewage treatment plant, in order to ensure that the effluent reaches the standard, the total nitrogen warning concentration is 17mg/L, and if the total nitrogen warning concentration exceeds the standard, a carbon source needs to be added in advance according to a prediction control method. As an embodiment, the carbon source fuzzy control rule comprises:
if SNtot>19.7mg/L, setting the first partition carbon source qEC1=5m3D, second partition carbon source qEC2=2m3/d;
If 19.0. ltoreq.SNtotLess than or equal to 19.7mg/L, and a first subarea external carbon source q is arrangedEC1=4m3D, second partition carbon source qEC2=2m3/d;
If 18.0. ltoreq.SNtotLess than or equal to 18.9mg/L, and a first subarea external carbon source q is arrangedEC1=4m3D, second partition carbon source qEC2=1m3/d;
If 17.0. ltoreq.SNtotLess than or equal to 17.9mg/L, and a first subarea is provided with an external carbon source qEC1=3m3D, second partition carbon source qEC2=0m3/d;
If SNtot<17.0mg/L, setting the first partition external carbon source qEC1=0m3D, second partition carbon source qEC2=0m3/d。
And switching back to the conventional tracking control when the total nitrogen predicted value output by the prediction model is lower than 17mg/L and the total nitrogen of the fifth subarea is lower than 13.5 mg/L.
In order to verify the effectiveness of the carbon source adding method based on prediction control, the invention also carries out analysis and simulation through BSM and practical experiments. As can be seen from the graphs in FIGS. 4 and 5, after the carbon source is fed in advance by using the predictive control, the total nitrogen concentration of the effluent of the sewage treatment is obviously reduced at the peak value, the overproof discharge of the total nitrogen is inhibited, the total nitrogen control effect of the sewage treatment is better, the energy consumption is lower, and the national discharge standard of the sewage treatment is reached.
The method utilizes the neural network to establish a total nitrogen prediction model, predicts whether the total nitrogen concentration of the effluent exceeds the standard in advance on the premise of obtaining the lowest energy consumption or the highest water quality, and finely controls the carbon source feeding in the sewage treatment anoxic tank by adopting a fuzzy rule if the total nitrogen concentration exceeds the standard, thereby avoiding the problems of excessive or insufficient carbon source generated by manual feeding and feedback feeding. Compared with the prior art, the invention overcomes the problems of insufficient measurement precision, delayed control of effluent concentration and the like caused by complex sewage treatment process and long reaction time, and has good application prospect.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (4)
1. A sewage treatment carbon source adding method based on predictive control is characterized by comprising the following steps:
step 1: establishing an energy consumption and water quality model, optimizing the energy consumption and water quality model by adopting an NSGA2 algorithm, and obtaining a solution set with the lowest energy consumption or the highest water quality as a fifth subarea dissolved oxygen SO,5Concentration and second partition nitrate nitrogen SNO,2A set value of concentration;
step 2: inputting the solution set with the lowest energy consumption or the highest water quality into a constructed sewage treatment effluent total nitrogen prediction model to obtain a total nitrogen prediction value corresponding to the current input quantity;
and step 3: judging whether the total nitrogen predicted value exceeds the standard or not and adopting different carbon source fuzzy control rules to feed the carbon source, wherein the method specifically comprises the following steps: and when the total nitrogen prediction value exceeds the standard, performing fuzzy control on the first and second partition external carbon sources by adopting a carbon source fuzzy control rule according to the total nitrogen prediction value.
2. The method for adding the carbon source for sewage treatment based on the predictive control as claimed in claim 1, wherein the step 1 specifically comprises:
step 1.1: selecting F1And F2Respectively an energy consumption objective function and a water quality objective function, f1And f2Describing the optimization problem of energy consumption and water quality by adopting formulas (1) to (3) respectively as an energy consumption model and a water quality model:
wherein, Delta is an overproof penalty term CNtotPredicting the effluent total nitrogen superscalar value; f. ofNtotRepresenting a total nitrogen prediction model of effluent of sewage treatment; scIs a set total nitrogen concentration upper limit value; x represents the solution of the energy consumption and water quality optimization problem, XiDenotes the i-th decision variable, l, in the solutioni、uiAre decision variables x, respectivelyiA lower and an upper bound of (1), wherein x1And x2Each specifically represents the fifth region dissolved oxygen SO,5Concentration and second partition nitrate nitrogen SNO,2Concentration;
step 1.2: let p solve XpIs (F)1(Xp),F2(Xp) Calculate the p-th solution X after normalization according to equation (4)pThen selecting the non-dominant solution X ═ X (X) with the minimum current sparsity1,x2) As a sparse solution:
wherein, SP (X)p) Representing the p-th solution X after normalizationpSparsity of WpIs the sum target vector (F) in the target function space1(Xp),F2(Xp) The Euclidean distance between the target vectors is less than r, r is a positive real number less than 1, and N is the set initial population number;
step 1.3: performing cross variation after selecting the sparse solution to avoid falling into local optimum, specifically performing local search by using a limit optimization variation method according to formulas (5) to (8) to generate a local solution; meanwhile, carrying out mutation operation by using a random immigration strategy according to a formula (9) and a formula (10) to generate [0.2N ] local solutions;
Xp=(x′1,x′2),p=1,2,...,N (5)
x′i=xi+α·βmax(xi),i=1,2 (6)
βmax(xi)=max[xi-li,ui-xi],i=1,2 (8)
XXk=(xx′1,xx′2),k=1,2,...,[0.2N] (9)
xx′i=γxi,i=1,2,0<γ<1.2 (10)
wherein, XpDenotes the solution after Limit-optimized mutation update, x'iRepresents the ith decision variable in the solution after the updating of the extreme optimization variation, h and gamma are random numbers, qDenotes the shape parameter, betamax(xi) For the current decision variable xiMaximum value of variability, XXkRepresents a solution after the variation update of the random immigration policy, xx'iThe ith decision variable in the solution after the random immigration strategy variation updating;
step 1.4: performing non-dominated sorting and congestion distance calculation on all the solutions obtained in the step 1.3, selecting N solutions from the solutions to form a next generation population, and iteratively executing the step 1.2 to the step 1.4 until a preset iteration step number is reached; the selection rule for forming the next generation population is to preferentially select a non-dominant solution, and when the number of solutions is less than N, a suboptimal solution is selected;
step 1.5: the N non-dominated solutions finally obtained in the step 1.4 are brought into a constructed sewage treatment effluent total nitrogen prediction model, and if the total nitrogen prediction value does not exceed the standard, a solution set with the lowest energy consumption is selected as a fifth subarea dissolved oxygen SO,5Concentration and second partition nitrate nitrogen SNO,2A set value of concentration; if the predicted value of the total nitrogen exceeds the standard, selecting the solution set with the highest water quality as the dissolved oxygen S of the fifth subareaO,5Concentration and second partition nitrate nitrogen SNO,2The set value of the concentration.
3. The method for adding the carbon source to the wastewater treatment based on the predictive control as claimed in claim 1, wherein the inputs of the model for predicting the total nitrogen in the wastewater treatment effluent further include total nitrogen in the influent, the influent flow rate and the current total nitrogen in the effluent.
4. The method for adding the carbon source in the sewage treatment based on the predictive control as claimed in claim 1, wherein the fuzzy control rule of the carbon source comprises:
if SNtot>19.7mg/L, setting the first partition carbon source qEC1=5m3D, second partition carbon source qEC2=2m3/d;
If 19.0. ltoreq.SNtotLess than or equal to 19.7mg/L, and a first subarea external carbon source q is arrangedEC1=4m3D, second partition carbon source qEC2=2m3/d;
If 18.0. ltoreq.SNtotLess than or equal to 18.9mg/L, and a first subarea external carbon source q is arrangedEC1=4m3D, second partition carbon source qEC2=1m3/d;
If 17.0. ltoreq.SNtotLess than or equal to 17.9mg/L, and a first subarea is provided with an external carbon source qEC1=3m3D, second partition carbon source qEC2=0m3/d;
If SNtot<17.0mg/L, setting the first partition external carbon source qEC1=0m3D, second partition carbon source qEC2=0m3/d。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111064097.9A CN113674809A (en) | 2021-09-10 | 2021-09-10 | Sewage treatment carbon source adding method based on predictive control |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111064097.9A CN113674809A (en) | 2021-09-10 | 2021-09-10 | Sewage treatment carbon source adding method based on predictive control |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113674809A true CN113674809A (en) | 2021-11-19 |
Family
ID=78549100
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111064097.9A Pending CN113674809A (en) | 2021-09-10 | 2021-09-10 | Sewage treatment carbon source adding method based on predictive control |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113674809A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115353200A (en) * | 2022-07-12 | 2022-11-18 | 苏州水星环保工业系统有限公司 | Sewage plant effluent TN intelligent control system and method |
-
2021
- 2021-09-10 CN CN202111064097.9A patent/CN113674809A/en active Pending
Non-Patent Citations (1)
Title |
---|
栗三一 等: "污水处理决策优化控制", 自动化学报, vol. 44, no. 12, pages 2198 - 2209 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115353200A (en) * | 2022-07-12 | 2022-11-18 | 苏州水星环保工业系统有限公司 | Sewage plant effluent TN intelligent control system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Qiao et al. | Adaptive fuzzy neural network control of wastewater treatment process with multiobjective operation | |
CN106873379B (en) | Sewage treatment optimal control method based on iterative ADP algorithm | |
CN108549234B (en) | Multi-objective optimization control method based on dynamic variable values | |
CN109669352B (en) | Oily sewage treatment process optimization control method based on self-adaptive multi-target particle swarm | |
AU2021101438A4 (en) | Adaptive control method and system for aeration process | |
Bo et al. | Online adaptive dynamic programming based on echo state networks for dissolved oxygen control | |
Wang et al. | Intelligent optimal tracking with asymmetric constraints of a nonlinear wastewater treatment system | |
CN106802563B (en) | A kind of sewage procedure optimization control method based on drosophila optimization and LSSVM | |
Xie et al. | Simulation and optimization of a full-scale Carrousel oxidation ditch plant for municipal wastewater treatment | |
CN108427268B (en) | Sewage treatment optimization control method based on knowledge and data information decision | |
CN108088974B (en) | Soft measurement method for effluent nitrate nitrogen in anaerobic simultaneous denitrification methanogenesis process | |
Qiao et al. | Neural network on‐line modeling and controlling method for multi‐variable control of wastewater treatment processes | |
CN110032755B (en) | Multi-objective optimization method for urban sewage treatment process under multiple working conditions | |
Do et al. | A design of higher-level control based genetic algorithms for wastewater treatment plants | |
US20210087074A1 (en) | Type-2 fuzzy neural network-based cooperative control method for wastewater treatment process | |
Han et al. | Fuzzy neural network-based model predictive control for dissolved oxygen concentration of WWTPs | |
CN113674809A (en) | Sewage treatment carbon source adding method based on predictive control | |
CN113189881A (en) | Multi-objective optimization control method and system for sewage treatment | |
Chen et al. | Smart energy savings for aeration control in wastewater treatment | |
WO2023165001A1 (en) | Composite sewage denitrification process device and operation parameter optimization method | |
CN103792844B (en) | A kind of sewage treatment control method based on orderly cluster | |
Santín et al. | New approach for regulation of the internal recirculation flow rate by fuzzy logic in biological wastewater treatments | |
Du et al. | Multi-objective pity beetle algorithm based optimal control of wastewater treatment process | |
Kalogeropoulos et al. | Economic oriented dynamic matrix control of wastewater treatment plants | |
Fu et al. | Dissolved oxygen control system based on the TS fuzzy neural network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |