CN116947208B - Anaerobic ammonia oxidation rapid starting and directional stimulation process capable of controlling potential step - Google Patents
Anaerobic ammonia oxidation rapid starting and directional stimulation process capable of controlling potential step Download PDFInfo
- Publication number
- CN116947208B CN116947208B CN202310933292.3A CN202310933292A CN116947208B CN 116947208 B CN116947208 B CN 116947208B CN 202310933292 A CN202310933292 A CN 202310933292A CN 116947208 B CN116947208 B CN 116947208B
- Authority
- CN
- China
- Prior art keywords
- time
- potential
- prediction
- anaerobic ammonia
- parameters
- 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.)
- Active
Links
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 title claims abstract description 171
- 238000007254 oxidation reaction Methods 0.000 title claims abstract description 92
- 230000003647 oxidation Effects 0.000 title claims abstract description 89
- 229910021529 ammonia Inorganic materials 0.000 title claims abstract description 84
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000008569 process Effects 0.000 title claims abstract description 21
- 230000000638 stimulation Effects 0.000 title claims abstract description 18
- 241000894006 Bacteria Species 0.000 claims abstract description 42
- 230000001276 controlling effect Effects 0.000 claims abstract description 13
- 230000001105 regulatory effect Effects 0.000 claims abstract description 7
- 238000005516 engineering process Methods 0.000 claims abstract description 6
- 230000009471 action Effects 0.000 claims description 47
- 238000004364 calculation method Methods 0.000 claims description 35
- 230000006870 function Effects 0.000 claims description 23
- 239000007788 liquid Substances 0.000 claims description 21
- 238000012544 monitoring process Methods 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 12
- 210000002569 neuron Anatomy 0.000 claims description 10
- QGZKDVFQNNGYKY-UHFFFAOYSA-O Ammonium Chemical compound [NH4+] QGZKDVFQNNGYKY-UHFFFAOYSA-O 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 6
- 238000013441 quality evaluation Methods 0.000 claims description 6
- 230000004044 response Effects 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 125000004122 cyclic group Chemical group 0.000 claims description 5
- 238000012546 transfer Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 4
- 230000000977 initiatory effect Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 12
- 238000006243 chemical reaction Methods 0.000 abstract description 9
- 241001453382 Nitrosomonadales Species 0.000 abstract description 7
- 230000004936 stimulating effect Effects 0.000 abstract description 2
- 238000005728 strengthening Methods 0.000 abstract description 2
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 20
- NHNBFGGVMKEFGY-UHFFFAOYSA-N Nitrate Chemical compound [O-][N+]([O-])=O NHNBFGGVMKEFGY-UHFFFAOYSA-N 0.000 description 17
- 229910002651 NO3 Inorganic materials 0.000 description 16
- 229910052757 nitrogen Inorganic materials 0.000 description 10
- 239000000758 substrate Substances 0.000 description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 7
- 230000009467 reduction Effects 0.000 description 6
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 5
- 229910052760 oxygen Inorganic materials 0.000 description 5
- 239000001301 oxygen Substances 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000001590 oxidative effect Effects 0.000 description 4
- 239000000126 substance Substances 0.000 description 4
- IOVCWXUNBOPUCH-UHFFFAOYSA-M Nitrite anion Chemical compound [O-]N=O IOVCWXUNBOPUCH-UHFFFAOYSA-M 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 description 2
- XKMRRTOUMJRJIA-UHFFFAOYSA-N ammonia nh3 Chemical compound N.N XKMRRTOUMJRJIA-UHFFFAOYSA-N 0.000 description 2
- 230000001580 bacterial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 239000003344 environmental pollutant Substances 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 229910052751 metal Inorganic materials 0.000 description 2
- 229910021645 metal ion Inorganic materials 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 239000007800 oxidant agent Substances 0.000 description 2
- 231100000719 pollutant Toxicity 0.000 description 2
- 230000003334 potential effect Effects 0.000 description 2
- 239000000523 sample Substances 0.000 description 2
- 239000010865 sewage Substances 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 description 1
- VBTPRWXFCXZKPQ-UHFFFAOYSA-N [O].NN Chemical compound [O].NN VBTPRWXFCXZKPQ-UHFFFAOYSA-N 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- WQZGKKKJIJFFOK-VFUOTHLCSA-N beta-D-glucose Chemical compound OC[C@H]1O[C@@H](O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-VFUOTHLCSA-N 0.000 description 1
- 239000006227 byproduct Substances 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 239000000356 contaminant Substances 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000005518 electrochemistry Effects 0.000 description 1
- -1 etc.) Substances 0.000 description 1
- 239000008103 glucose Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000002503 metabolic effect Effects 0.000 description 1
- 150000002739 metals Chemical class 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000033116 oxidation-reduction process Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000009257 reactivity Effects 0.000 description 1
- 230000027756 respiratory electron transport chain Effects 0.000 description 1
- 230000000284 resting effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 150000003384 small molecules Chemical class 0.000 description 1
- 238000003756 stirring Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
- 238000006276 transfer reaction Methods 0.000 description 1
Classifications
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F3/00—Biological treatment of water, waste water, or sewage
- C02F3/28—Anaerobic digestion processes
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F3/00—Biological treatment of water, waste water, or sewage
- C02F3/005—Combined electrochemical biological processes
-
- 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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F2101/00—Nature of the contaminant
- C02F2101/10—Inorganic compounds
- C02F2101/16—Nitrogen compounds, e.g. ammonia
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Microbiology (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Water Supply & Treatment (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Hydrology & Water Resources (AREA)
- Biodiversity & Conservation Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Organic Chemistry (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Environmental & Geological Engineering (AREA)
- General Physics & Mathematics (AREA)
- General Chemical & Material Sciences (AREA)
- Electrochemistry (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Purification Treatments By Anaerobic Or Anaerobic And Aerobic Bacteria Or Animals (AREA)
Abstract
The invention discloses an anaerobic ammonia oxidation rapid starting and directional stimulation process for controlling potential step, which comprises the following steps: constructing an anaerobic ammonia oxidation denitrification system based on a bioelectrochemical technology; in an anaerobic ammonia oxidation denitrification system, providing high potential and low potential for a cathode in a time interval to construct a potential step; determining a predicted parameter of the potential step based on the intelligent model; the potential step operation parameters are adjusted to be prediction parameters in real time through the anaerobic ammonia oxidation denitrification system, and the growth of anaerobic ammonia oxidation bacteria in the anaerobic ammonia oxidation denitrification system is regulated and controlled in real time through the adjusted potential step. The potential step is used for stimulating the growth of anaerobic ammonia oxidizing bacteria and strengthening denitrification reaction, the high potential and the low potential of time intervals are provided for the cathode, and the intelligent model is arranged, so that the operation parameters of the potential step can be adjusted in real time based on the intelligent model, the growth condition of the anaerobic ammonia oxidizing bacteria is ensured, and the denitrification effect is improved.
Description
Technical Field
The invention relates to the technical field of sewage treatment denitrification, in particular to an anaerobic ammonia oxidation rapid starting and directional stimulation process for controlling potential step.
Background
Anaerobic ammonia oxidation denitrification (Amx) is one of the most rapid denitrification modes at present, and as a dual-substrate denitrification reaction, ammonia Nitrogen (NH) must be simultaneously contained 4 + ) Nitrite (NO) 2 - ) Denitrification can be realized. But the Amx actually uses nitrate (NO 3 - ) The problem of low total nitrogen removal rate<60%) excess of NO 3 - The anaerobic ammonia oxidation reaction is affected, so that the overall denitrification efficiency of the system is reduced, and the problem that the actual application of the anaerobic ammonia oxidation is not negligible is solved. Excess of NO 3 - There are two sources, one is the byproduct of anammox and the other is the substrate competition of nitrite oxidizing bacteria against anammox bacteria, which will NO 2 - Oxidation to NO 3 - . At present, a denitrification approach is often adopted to solve NO in an anaerobic ammonia oxidation system 3 - Commonly used electron donors include small molecule carbon-containing organics (glucose, ethanol, etc.), metals (Zn, etc.), electrodes, etc. However, the anaerobic ammonia oxidation reaction has stricter limit on organic matters and dissolved oxygen, and the improper control of Chemical Oxygen Demand (COD) and dissolved oxygen is easy to cause the reduction of the anaerobic ammonia oxidation reaction, the discontinuous heterotrophic denitrification reaction and even the running instability. The reducing metal can generate metal ions, the anammox metabolic activity is partially reduced in a mode of reducing the hydrazine oxygen utilization activity, the anammox denitrification effect is reduced, and the sewage treatment cost is increased in the rear-end treatment process of the metal ions.
How to ensure the growth condition of anaerobic ammonia oxidizing bacteria and avoid the influence of other pollutants on the denitrification process is a problem to be solved in the present stage.
Disclosure of Invention
The present invention provides a rapid start and directional stimulation process for anaerobic ammoxidation with controlled potential steps to solve the above-described problems in the prior art.
The invention provides an anaerobic ammonia oxidation rapid start and directional stimulation process for controlling potential step, which comprises the following steps:
s100, constructing an anaerobic ammonia oxidation denitrification system based on a bioelectrochemical technology;
s200, in the anaerobic ammonia oxidation denitrification system, providing high potential and low potential for the cathode in a time-division mode, and constructing potential steps;
s300, determining a predicted parameter of the potential step based on the intelligent model;
s400, adjusting the potential step operation parameters into prediction parameters in real time through the anaerobic ammonia oxidation denitrification system, and regulating and controlling the growth of anaerobic ammonia oxidation bacteria in the anaerobic ammonia oxidation denitrification system in real time through the adjusted potential step.
Preferably, in S200, the step of providing the high potential and the low potential to the cathode for a time period, the step of constructing the potential includes:
s201, setting a first potential intensity of a high potential of the potential step and a first action time of the high potential, the first action time including: a first start time and a first duration; the first start time is a start time of supplying a high potential to the cathode, and the first duration is a hold time of continuously supplying the high potential to the cathode;
S202, setting a second potential intensity of a low potential and a second action time of a high potential of the potential step; the second action time includes: a second start time, which is a start time for supplying a low potential to the cathode, and a second duration, which is a hold time for continuously supplying a low potential to the cathode.
Preferably, in S300, determining the predicted parameter of the potential step based on the smart model includes:
s301, determining input parameters, wherein the input parameters comprise: the pH value of the liquid to be treated in the anaerobic ammonia oxidation denitrification system,Temperature, NH 4 + Concentration, NO 2 - Concentration, NO 3 - Concentration, ORP value, COD value, anaerobic ammonia oxidation bacteria growth kinetics;
s302, inputting the input parameters into the intelligent model, and calculating an output result by the intelligent model according to a prediction algorithm of the model, wherein the output result comprises: predicted potential intensity and predicted action time of the potential step;
s303, taking the predicted potential intensity and the predicted action time of the potential step as the predicted parameters of the potential step.
Preferably, the step S302 further includes: constructing an intelligent model;
the method for constructing the intelligent model comprises the following steps:
s3021, constructing a cyclic neural network (RNN) model, wherein the RNN model comprises: an input layer, a hidden layer and an output layer;
S3022, input variables of the input layer include: pH value, temperature and NH of liquid to be treated 4 + Concentration, NO 2 - Concentration, NO 3 - The concentration, ORP value, COD value and anammox bacteria growth dynamics are carried out, and input variables are normalized and embedded; setting a training set, a data quantity, training times and step sizes, carrying out predictive calculation on an input variable through a time sequence predictive calculation formula, and determining an output variable according to a predictive calculation result;
s3023, the output layer includes one or more neurons, each neuron representing an output variable, the output variable including: predicting potential intensity and predicting action time;
s3024, compressing an output variable into a response value range by adopting an activation function at the last layer of the RNN model;
s3025, performing parameter adjustment on the trained RNN model by using a model evaluation method to obtain an optimized intelligent model.
Preferably, in S3022, performing predictive computation on the input variable by using a computation formula of time sequence prediction, and determining the output variable according to a result of the predictive computation includes:
s3022-1, setting an input sequence with a length of T and a corresponding output sequence;
s3022-2, identifying a feature vector for each input variable, for predicting a scalar, the scalar being an output variable;
S3022-3, after the input variable of each time step is subjected to forward calculation of the hidden layer, predicting a predicted value of the next time step in an iterative calculation mode, wherein the predicted value is an output variable; the calculation formula of the forward calculation includes:
h t ＝f(x t ，h t-1 ；θ)，
wherein f represents the transfer function of the RNN, g represents the final prediction function or classifier, θ represents a learnable parameter, h t Indicating the implicit state of the current time step, h t-1 Indicating the implicit state, x, of the current time step t The input variable for the current time step,representing the final predicted output variable.
Preferably, the S400 includes:
s401, monitoring prediction parameters of an intelligent model in real time by an anaerobic ammonia oxidation denitrification system;
s402, if the prediction parameters of the intelligent model are different from the current operation parameters, the operation parameters are adjusted to the prediction parameters; if the predicted parameter is the same as the current operation parameter, a potential step is formed according to the current operation parameter.
Preferably, the S400 includes:
s403, setting a time node and an adjustment period for adjusting the operation parameters; the time node is related to the prediction time of the prediction parameters predicted by the intelligent model;
s404, monitoring the relation between the time node and the prediction time of the prediction parameter predicted by the intelligent model, and taking the time node as a final time node if the time node is before the prediction time and the time difference between the time node and the prediction time is within a set threshold range; if the time node is after the predicted time, readjusting the time node to ensure that the time node is before the predicted time;
S405, monitoring the relation between the adjustment period and the prediction period of the intelligent model, and if the adjustment period is matched with the prediction period of the intelligent model, setting a final adjustment period by taking a final time node as the period starting time;
s406, adjusting the potential step operation parameters to the prediction parameters according to the final time node and the final adjustment period in a period adjustment mode.
Preferably, in S403, the setting of the time node and the adjustment period of the operation parameter adjustment includes:
s4031, obtaining a prediction rule of the intelligent model, wherein the prediction rule comprises: the prediction starting time and the prediction ending time of each prediction are used for determining a prediction period according to the prediction starting time and the prediction ending time;
s4032, setting an adjustment period to be an integer multiple of a prediction period; adjusting the multiple value according to the growth condition of the adjusted anaerobic ammonium oxidation bacteria, and determining an adjustment period according to the adjusted multiple value and the predicted period data;
s4033, setting a predetermined amount according to the determined adjustment period, and setting the time node to a time point having a predetermined amount with respect to the start time of the adjustment period.
Preferably, after the step S400, the step S500 includes setting an orientation parameter, and comprehensively adjusting an operation parameter of the potential step according to the orientation parameter and the prediction parameter;
The S500 includes:
s501, in a high potential phase of the potential step, detecting NO in the liquid to be treated 2 - Concentration; according to NO 2 - The concentration is set to a first adjustment parameter which is used for adjusting the potential intensity and the action time of the high potential so as to lead NO in the liquid to be treated 2 - The concentration is maintained within a set first range;
s502, in the low potential phase of the potential step,detection of NO in a liquid to be treated 3 - Concentration and N 2 Concentration according to NO 3 - Concentration and N 2 Setting a second adjustment parameter for adjusting the potential intensity and the action time of the low potential to enable NO in the liquid to be treated 3 - The concentration is kept within a second range, and N 2 The concentration is kept within a set third range;
s503, carrying out weighted average calculation on the first adjustment parameter and the second adjustment parameter to obtain an orientation parameter;
s504, comprehensively adjusting the operation parameters of the potential step according to the orientation parameters and the prediction parameters.
Preferably, in S504, the step of adjusting the operation parameters of the potential step according to the orientation parameter and the prediction parameter includes:
s5041, setting a comprehensive adjustment period, wherein the comprehensive adjustment period comprises a first stage and a second stage, the first stage is a stage for adjusting the operation parameters only through the prediction parameters, and the second stage is a stage for comprehensively adjusting the operation parameters of the potential step through the orientation parameters and the prediction parameters;
S5042, adjusting the potential step to form the potential step in a cycle mode of the comprehensive adjustment cycle;
s5043, monitoring first comparison data of the growth condition of the corresponding anaerobic ammonia oxidation bacteria when the first stage is the current stage; monitoring second comparison data of the growth condition of the corresponding anaerobic ammonia oxidation bacteria when the second stage is the current stage;
s5044, the first comparison data and the second comparison data are used as the basis of the quality evaluation of the growth condition of the anaerobic ammonium oxidation bacteria, and the corresponding stages of the comparison data with high quality evaluation are marked; the operating parameters of the potential steps are adjusted using the adjustment scheme of the corresponding stage.
Compared with the prior art, the invention has the following advantages:
the invention provides an anaerobic ammonia oxidation rapid start and directional stimulation process for controlling potential step, which comprises the following steps: constructing an anaerobic ammonia oxidation denitrification system based on a bioelectrochemical technology; in an anaerobic ammonia oxidation denitrification system, providing high potential and low potential for a cathode in a time interval to construct a potential step; determining a predicted parameter of the potential step based on the intelligent model; the potential step operation parameters are adjusted to be prediction parameters in real time through the anaerobic ammonia oxidation denitrification system, and the growth of anaerobic ammonia oxidation bacteria in the anaerobic ammonia oxidation denitrification system is regulated and controlled in real time through the adjusted potential step. The potential step is used for stimulating the growth of anaerobic ammonia oxidizing bacteria and strengthening denitrification reaction, the high potential and the low potential of time intervals are provided for the cathode, and the intelligent model is arranged, so that the operation parameters of the potential step can be adjusted in real time based on the intelligent model, the growth condition of the anaerobic ammonia oxidizing bacteria is ensured, and the denitrification effect is improved.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a flow chart of an anaerobic ammonia oxidation rapid initiation and directional stimulation process for controlling potential steps in an embodiment of the present invention;
FIG. 2 is a flow chart of a method of constructing a smart model in an embodiment of the present invention;
FIG. 3 is a diagram showing an example of the structure of an RNN model based on time-series input according to an embodiment of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
The embodiment of the invention provides an anaerobic ammonia oxidation rapid start and directional stimulation process for controlling potential step, referring to fig. 1, the process method comprises the following steps:
s100, constructing an anaerobic ammonia oxidation denitrification system based on a bioelectrochemical technology;
s200, in the anaerobic ammonia oxidation denitrification system, providing high potential and low potential for the cathode in a time-division mode, and constructing potential steps;
s300, determining a predicted parameter of the potential step based on the intelligent model;
s400, adjusting the potential step operation parameters into prediction parameters in real time through the anaerobic ammonia oxidation denitrification system, and regulating and controlling the growth of anaerobic ammonia oxidation bacteria in the anaerobic ammonia oxidation denitrification system in real time through the adjusted potential step.
The working principle of the technical scheme is as follows: the scheme adopted by the embodiment is that an anaerobic ammonia oxidation denitrification system is firstly constructed based on a bioelectrochemistry technology; then in the anaerobic ammonia oxidation denitrification system, providing high potential and low potential for the cathode in time intervals to construct potential steps; secondly, determining a prediction parameter of the potential step based on the intelligent model; finally, the potential step operation parameters are adjusted to be prediction parameters in real time through the anaerobic ammonia oxidation denitrification system, and the growth of anaerobic ammonia oxidation bacteria in the anaerobic ammonia oxidation denitrification system is regulated and controlled in real time through the adjusted potential step.
Combining anaerobic ammonia oxidation denitrification (Amx) with bioelectrochemical system (Bio-electrochemical system, BES) and providing potential effective for NO3 abatement - The total nitrogen removal rate is accumulated and improved, and the anaerobic ammonia oxidation bacteria attached to the surface of the cathode can expand an electron transfer chain and accelerate growth under the electrifying action. However, the optimal growth potential of the anaerobic ammonium oxidation bacteria is in the range of V=0.145-0.225 VSHE, and NO3 is continuously provided - The reduction potential (v=0.433 VSHE) affects amb growth, and long-term potential accumulation and maintenance of the reduction potential reduces amb reactivity, resulting in NO2 - Accumulating and destroying the structure of the cathode biomembrane, reducing the nitrogen concentration gradient between the water environment and the biological phase, and generating NO2 - With NO3 - Is a co-accumulation of (a). Although stirring by adding waterEquation can provide substrate and ease accumulation for Amx, but this approach can additionally increase operating power costs.
Potential step (Potential Step Method) is a potential scanning method used in electrochemistry to study the chemical reactions at the electrode surface. As the electrode surface potential suddenly changes, a potential step phenomenon is generated on the cathode surface, and the characteristic of electrode current/potential response change is observed. The potential step may provide different potentials for Amx by controlling the stepwise potential change to provide NO3 under high potential conditions - Reduction to NO2 - Providing enough substrate for Amx to realize denitrification, promoting Amx denitrification to realize nitrogen removal under low potential condition and generating NO3 - Providing substrate for the subsequent increase of potential, realizing partial electronic circulation. By measuring parameters such as reaction kinetic parameters of the electrode surface, rate constants of charge transfer reactions, charge transfer coefficients and the like and combining Amx reaction kinetics, the effect of potential step conditions and potential alternation time on nitrogen removal can be judged. Therefore, the potential step can further improve the Amx denitrification effect, and has excellent performance and strong operability.
The potential step action time and the potential intensity can be directly controlled by depending on the electrochemical workstation, but the scanning speed and the step potential and the holding time of the electrochemical workstation in each scanning potential range need to be manually controlled, the initial point position, the scanning section sampling interval, the step section scanning interval, the resting time, the period and the sensitivity are regulated and controlled, and the manual control parameters further influence the Amx bacteria growth and the nitrogen removal effect. Although pH, temperature, ORP, COD, NH4 can be obtained by a nitrogen detection probe + 、NO2 - With NO3 - However, the manual control mode has certain hysteresis, and the potential cannot be adjusted in time according to Amx growth dynamics and nitrogen removal effect.
Therefore, the intelligent model is arranged, and the operation parameters of the potential step can be adjusted in real time based on the intelligent model, so that the denitrification effect is improved according to the growth condition of anaerobic ammonia oxidation bacteria.
The beneficial effects of the technical scheme are as follows: scheme general provided by the embodimentThe overpotential step stimulates the growth of anammox bacteria and strengthens denitrification reaction, and provides high and low electric potentials for the cathode in different time periods, and the high electric potential leads NO3 to be generated - Reduction to NO2 - Providing a substrate for anaerobic ammoxidation; low potential ensures anaerobic ammoxidation growth, and leads nitrogen to be removed to generate N 2 And a small amount of NO3 - . And through setting up intelligent model, can carry out real-time adjustment to the operating parameter of potential step based on intelligent model to guarantee according to anaerobic ammonia oxidation bacterial's growth condition, improve the denitrification effect.
In another embodiment, in S200, providing the high potential and the low potential to the cathode for a time period, constructing a potential step, includes:
s201, setting a first potential intensity of a high potential of the potential step and a first action time of the high potential, the first action time including: a first start time and a first duration; the first start time is a start time of supplying a high potential to the cathode, and the first duration is a hold time of continuously supplying the high potential to the cathode;
S202, setting a second potential intensity of a low potential and a second action time of a high potential of the potential step; the second action time includes: a second start time, which is a start time for supplying a low potential to the cathode, and a second duration, which is a hold time for continuously supplying a low potential to the cathode.
The working principle of the technical scheme is as follows: the scheme adopted by the embodiment is that a high potential and a low potential are provided for the cathode in a time interval, and a potential step is constructed, comprising: setting a first potential intensity of a high potential of the potential step and a first action time of the high potential, the first action time including: a first start time and a first duration; the first start time is a start time of supplying a high potential to the cathode, and the first duration is a hold time of continuously supplying the high potential to the cathode; setting a second potential intensity of the low potential and a second action time of the high potential of the potential step; the second action time includes: a second start time, which is a start time for supplying a low potential to the cathode, and a second duration, which is a hold time for continuously supplying a low potential to the cathode. The high potential intensity and the action time of a specific potential step and the low potential intensity and the action time of a specific potential step are set, so that the growth of anammox bacteria is ensured.
In another embodiment, in S300, determining the predicted parameter of the potential step based on the smart model includes:
s301, determining input parameters, wherein the input parameters comprise: pH value, temperature and NH4 of liquid to be treated in anaerobic ammonia oxidation denitrification system + Concentration, NO2 - Concentration, NO3 - Concentration, ORP value, COD value, anaerobic ammonia oxidation bacteria growth kinetics;
s302, inputting the input parameters into the intelligent model, and calculating an output result by the intelligent model according to a prediction algorithm of the model, wherein the output result comprises: predicted potential intensity and predicted action time of the potential step;
s303, taking the predicted potential intensity and the predicted action time of the potential step as the predicted parameters of the potential step.
The working principle of the technical scheme is as follows: the scheme adopted by the embodiment is to determine the prediction parameters of the potential step based on an intelligent model, and the scheme comprises the following steps: determining input parameters, the input parameters comprising: pH value, temperature and NH of liquid to be treated in anaerobic ammonia oxidation denitrification system 4 + Concentration, NO 2 - Concentration, NO 3 - Concentration, ORP value, COD value, anaerobic ammonia oxidation bacteria growth kinetics; inputting the input parameters into the intelligent model, and calculating an output result by the intelligent model according to a prediction algorithm of the model, wherein the output result comprises: predicted potential intensity and predicted action time of the potential step; taking the predicted potential intensity and the predicted action time of the potential step as the predicted parameters of the potential step.
In order to improve analysis of multi-level data and realize real-time regulation and control, related data can be identified and analyzed through an artificial intelligent network model, such as a multi-layer perceptron Model (MLP), a cyclic neural network model (RNN) or a Convolutional Neural Network (CNN), an intelligent model is constructed, potential strength and action time of potential steps are conveniently predicted, potential strength and action time of high and low potential of a cathode are further adjusted in real time according to a predicted result, manual control is not needed, time lag is avoided, and meanwhile, the problems of low denitrification efficiency and the like caused by errors of manual adjustment are avoided in an automatic adjustment mode.
The beneficial effects of the technical scheme are as follows: the proposal provided by the embodiment is adopted to stimulate the growth of anammox bacteria and strengthen denitrification reaction through potential step, the cathode is provided with high and low potentials in different time periods, and the high potential leads NO3 - Reduction to NO2 - Providing a substrate for anaerobic ammoxidation; low potential ensures anaerobic ammoxidation growth, and leads nitrogen to be removed to generate N 2 And a small amount of NO3 - . And through setting up intelligent model, can carry out real-time adjustment to the operating parameter of potential step based on intelligent model to guarantee according to anaerobic ammonia oxidation bacterial's growth condition, improve the denitrification effect.
In another embodiment, the step S302 further includes: constructing an intelligent model;
referring to fig. 2, the method for constructing the intelligent model includes:
s3021, constructing a cyclic neural network (RNN) model, wherein the RNN model comprises: an input layer, a hidden layer and an output layer;
s3022, input variables of the input layer include: pH value, temperature and NH of liquid to be treated 4 + Concentration, NO2 - Concentration, NO3 - The concentration, ORP value, COD value and anammox bacteria growth dynamics are carried out, and input variables are normalized and embedded; setting a training set, a data quantity, training times and step sizes, carrying out predictive calculation on an input variable through a time sequence predictive calculation formula, and determining an output variable according to a predictive calculation result;
s3023, the output layer includes one or more neurons, each neuron representing an output variable, the output variable including: predicting potential intensity and predicting action time;
s3024, compressing an output variable into a response value range by adopting an activation function at the last layer of the RNN model;
s3025, performing parameter adjustment on the trained RNN model by using a model evaluation method to obtain an optimized intelligent model.
The working principle of the technical scheme is as follows: the scheme adopted by the embodiment further comprises: constructing an intelligent model; the method for constructing the intelligent model comprises the following steps: constructing a cyclic neural network (RNN) model, wherein the RNN model comprises: an input layer, a hidden layer and an output layer; input variables for the input layer include: pH value, temperature and NH of liquid to be treated 4 + Concentration, NO2 - Concentration, NO3 - The concentration, ORP value, COD value and anammox bacteria growth dynamics are carried out, and input variables are normalized and embedded; setting a training set, a data quantity, training times and step sizes, carrying out predictive calculation on an input variable through a time sequence predictive calculation formula, and determining an output variable according to a predictive calculation result; the output layer includes one or more neurons, each neuron representing an output variable comprising: predicting potential intensity and predicting action time; at the last layer of the RNN model, an activation function is adopted to compress the output variable into the value range of the response; and carrying out parameter adjustment on the trained RNN model by adopting a model evaluation method to obtain an optimized intelligent model.
It should be noted that RNNs, as a type of sequence-based neural network, may be used to process time-series data or time-dependent data. Predicting potential and action time of potential step by RNN, and using pH and NH4 + 、NO2 - 、NO3 - The concentration and the growth dynamics of the anaerobic ammonia oxidizing bacteria are taken as input parameters, and the potential and the action time of the potential step are predicted.
ORP value (oxidation-reduction potential) is an important index in water quality, and although it cannot independently reflect the quality of water, it can integrate other water quality indexes to reflect the ecological environment in the aquarium system. It characterizes the relative extent of oxidizing or reducing nature of the mediator. COD (chemical oxygen demand): under certain conditions, when a water sample is treated by a certain strong oxidant, the consumption of the oxidant is needed. It reflects the pollution degree of substances in water, and the greater the chemical oxygen demand is, the more serious the pollution of organic matters in water is.
The RNN model is composed of an input layer, a hidden layer, and an output layer. In the problem of time series prediction, input data is required to be input in a sequence of time steps, each time step representing a sequence of pieces of data.
Input parameters of the input layer: input parameters include pH, temperature, ORP, COD, NH4+, NO2 - 、NO3 - Concentration and growth kinetics of anammox bacteria require treatment and embedding of these variables. The processing mode can be realized through a data normalization approach, and the normalization formula is as follows:
wherein: r represents data obtained after normalization of a certain parameter; rt represents the data of the parameter at a certain moment; rmin and Rmax represent the minimum and maximum values, respectively, of the parameter over a certain period of time.
Output layer: the output layer may be one or more neurons, where each neuron represents an output variable (e.g., potential and time of action of a potential step). Thus, at the last layer of the RNN, a sigmoid or softmax activation function may be used to compress the output variable into the range of values of the response.
Model training and optimization: and judging a model training result through Mean Square Error (MSE) or Mean Absolute Error (MAE), and adjusting the model to obtain higher prediction precision and result.
Therefore, parameters such as a training set, data volume, training times, step length and the like are defined, then the input layer is trained through a model, and after the parameters of the output layer are obtained, the parameters of the input layer are analyzed through MSE and MAE.
The beneficial effects of the technical scheme are as follows: the scheme provided by the embodiment is adopted to predict the pH value, the temperature, ORP, COD, NH & lt 4 & gt and NO2 contained in the anaerobic ammonia oxidation reactor which is operated for one stage by a potential step method through RNN time sequence - With NO3 - The concentration of pollutant is used as input parameter to predict the potential strength and the potential action time of the potential step of the next time sequence, and the potential strength and the potential action time directly act on the electrochemical workstation to control electricityThe potential step intensity and the action time further strengthen the anaerobic ammonia oxidation and improve the denitrification effect.
In another embodiment, in S3022, performing a predictive calculation on an input variable according to a calculation formula of time sequence prediction, and determining an output variable according to a result of the predictive calculation includes:
s3022-1, setting an input sequence with a length of T and a corresponding output sequence;
s3022-2, identifying a feature vector for each input variable, for predicting a scalar, the scalar being an output variable;
s3022-3, after the input variable of each time step is subjected to forward calculation of the hidden layer, predicting a predicted value of the next time step in an iterative calculation mode, wherein the predicted value is an output variable; the calculation formula of the forward calculation includes:
h t ＝f(x t ，h t-1 ；θ)，
Wherein f represents the transfer function of the RNN, g represents the final prediction function or classifier, θ represents a learnable parameter, h t Indicating the implicit state of the current time step, h t-1 Indicating the implicit state, x, of the current time step t The input variable for the current time step,representing the final predicted output variable.
The working principle of the technical scheme is as follows: the scheme adopted in this embodiment is a specific structure of an RNN model, please refer to fig. 2, fig. 3 is an exemplary diagram of an RNN model structure based on time series input, fig. 3, x in the diagram t For input time step data, h t-1 In the hidden state of the last time step, h t Is the hidden state of the current time step, y t Is the output predicted value.
Assuming an input sequence of length T and a corresponding output sequence, x 1:T ＝{x 1 ，...，x T Sum y 1:T ＝{y 1 ，...，y T }. Wherein each input data x t May be a feature vector for predicting a scalar, each feature vector may contain temperature, COD, pH, related contaminant concentrations, etc. Each output data yt is typically a scalar, the output data being the potential of the potential step and the duration of action.
Let the implicit state (h idden state) at each time t be h t The forward calculation formula commonly used by RNNs is:
h t ＝f(x t ，h t-1 ；θ)，
where f represents the transfer function of the RNN, g represents the final prediction function or classifier, and θ represents a learnable parameter.
In RNN, a commonly used activation function is the tanh or ReLU function. Wherein, the tanh function is a variant of the s igmoid function, the value range of which is [ -1,1], and the tanh function is easier to train a large-scale deep learning model because the range of the tanh function is wider than that of the s igmoid function.
Common loss functions used by RNNs include MSE (mean square error), cross entropy loss functions. In the task of timing prediction, a commonly used loss function is the Mean Square Error (MSE), which represents the distance between the predicted value and the true value, namely:
the beneficial effects of the technical scheme are as follows: the proposal provided by the embodiment predicts the potential and the action time of the potential step through RNN, and uses pH and NH 4 + 、NO 2 - 、NO 3 - The concentration and the growth dynamics of the anaerobic ammonia oxidizing bacteria are taken as input parameters, and the potential and the action time of the potential step are predicted.
In another embodiment, the S400 includes:
s401, monitoring prediction parameters of an intelligent model in real time by an anaerobic ammonia oxidation denitrification system;
s402, if the prediction parameters of the intelligent model are different from the current operation parameters, the operation parameters are adjusted to the prediction parameters; if the predicted parameter is the same as the current operation parameter, a potential step is formed according to the current operation parameter.
The working principle of the technical scheme is as follows: the scheme adopted by the embodiment is that the anaerobic ammonia oxidation denitrification system monitors the prediction parameters of the intelligent model in real time; if the prediction parameters of the intelligent model are different from the current operation parameters, the operation parameters are adjusted to the prediction parameters; if the predicted parameter is the same as the current operation parameter, a potential step is formed according to the current operation parameter.
And determining whether the predicted parameters are the same as the current operating parameters or not in a real-time monitoring mode so as to ensure that the anaerobic ammonia oxidation denitrification system adjusts the operating parameters in real time according to the predicted parameters, and further ensure the denitrification efficiency in a mode of automatically adjusting according to the predicted results.
In another embodiment, the S400 includes:
s403, setting a time node and an adjustment period for adjusting the operation parameters; the time node is related to the prediction time of the prediction parameters predicted by the intelligent model;
s404, monitoring the relation between the time node and the prediction time of the prediction parameter predicted by the intelligent model, and taking the time node as a final time node if the time node is before the prediction time and the time difference between the time node and the prediction time is within a set threshold range; if the time node is after the predicted time, readjusting the time node to ensure that the time node is before the predicted time;
s405, monitoring the relation between the adjustment period and the prediction period of the intelligent model, and if the adjustment period is matched with the prediction period of the intelligent model, setting a final adjustment period by taking a final time node as the period starting time;
S406, adjusting the potential step operation parameters to the prediction parameters according to the final time node and the final adjustment period in a period adjustment mode.
The working principle of the technical scheme is as follows: according to the scheme adopted by the embodiment, the time node and the adjustment period are further guaranteed, the final time node and the adjustment period are formed through advanced setting and subsequent optimization, the real-time performance of the automatic adjustment operation parameters is further improved, manual monitoring is not needed, the operation parameters can be adjusted according to the prediction parameters after prediction is completed, the mode of periodic circulation is adopted, the automatic adjustment is comprehensively guaranteed, a time difference is set between the time node and the prediction time in the setting of the time node, namely, the operation parameters can be automatically adjusted in a certain time period, frequent parameter prediction calculation can be avoided, and the calculated amount is increased. Therefore, the scheme provided by the embodiment saves the calculated amount and can ensure the real-time adjustment of the operation parameters.
In another embodiment, in S403, setting a time node and an adjustment period of the operation parameter adjustment includes:
S4031, obtaining a prediction rule of the intelligent model, wherein the prediction rule comprises: the prediction starting time and the prediction ending time of each prediction are used for determining a prediction period according to the prediction starting time and the prediction ending time;
s4032, setting an adjustment period to be an integer multiple of a prediction period; adjusting the multiple value according to the growth condition of the adjusted anaerobic ammonium oxidation bacteria, and determining an adjustment period according to the adjusted multiple value and the predicted period data;
s4033, setting a predetermined amount according to the determined adjustment period, and setting the time node to a time point having a predetermined amount with respect to the start time of the adjustment period.
The working principle of the technical scheme is as follows: the scheme adopted by the embodiment is that the operation parameter adjustment is performed after the completion of one prediction period, but the adjustment period is set, and after a few prediction periods, the operation parameter is adjusted according to the adjustment period, so that the problem of frequent operation control caused by frequent adjustment of the operation parameter under the condition that the prediction parameter is less or is not changed after a plurality of continuous prediction periods is avoided, the preset time of the set time node before the initial time of the adjustment period, namely the preset amount is set, and the adjusted time node is ensured to be before the adjustment period starts.
In another embodiment, after S400, the method includes S500, setting an orientation parameter, and comprehensively adjusting an operation parameter of the potential step according to the orientation parameter and the prediction parameter;
the S500 includes:
s501, in a high potential phase of the potential step, detecting NO in the liquid to be treated 2 - Concentration; according to NO 2 - The concentration is set to a first adjustment parameter which is used for adjusting the potential intensity and the action time of the high potential so as to lead NO in the liquid to be treated 2 - The concentration is maintained within a set first range;
s502, in the low potential stage of the potential step, detecting NO in the liquid to be treated 3 - Concentration and N 2 Concentration according to NO 3 - Concentration and N 2 Setting a second adjustment parameter for adjusting the potential intensity and the action time of the low potential to enable NO in the liquid to be treated 3 - The concentration is kept within a second range, and N 2 The concentration is kept within a set third range;
s503, carrying out weighted average calculation on the first adjustment parameter and the second adjustment parameter to obtain an orientation parameter;
s504, comprehensively adjusting the operation parameters of the potential step according to the orientation parameters and the prediction parameters.
The working principle of the technical scheme is as follows: the embodiment adopts the scheme that the operation parameter is adjusted according to the predicted parameter, and the embodiment adds the orientation parameter which is set by detecting the concentration of the orientation parameter in the liquid to be treated, and the orientation parameter is NO 2 - Concentration, NO 3 - Concentration and N 2 Concentration, since the directional parameter will directly affect the denitrification efficiency, the directional parameter is used as the optimized adjustment parameter to be further combined with the pre-treatmentAnd the measured parameters are comprehensively used for adjusting the operation parameters, so that the denitrification efficiency is further improved.
In another embodiment, in S504, adjusting the operation parameters of the potential step according to the orientation parameter and the prediction parameter includes:
s5041, setting a comprehensive adjustment period, wherein the comprehensive adjustment period comprises a first stage and a second stage, the first stage is a stage for adjusting the operation parameters only through the prediction parameters, and the second stage is a stage for comprehensively adjusting the operation parameters of the potential step through the orientation parameters and the prediction parameters;
s5042, adjusting the potential step to form the potential step in a cycle mode of the comprehensive adjustment cycle;
s5043, monitoring first comparison data of the growth condition of the corresponding anaerobic ammonia oxidation bacteria when the first stage is the current stage; monitoring second comparison data of the growth condition of the corresponding anaerobic ammonia oxidation bacteria when the second stage is the current stage;
s5044, the first comparison data and the second comparison data are used as the basis of the quality evaluation of the growth condition of the anaerobic ammonium oxidation bacteria, and the corresponding stages of the comparison data with high quality evaluation are marked; the operating parameters of the potential steps are adjusted using the adjustment scheme of the corresponding stage.
The working principle of the technical scheme is as follows: according to the scheme adopted by the embodiment, one period is formed by arranging two stages, the growth condition of the anaerobic ammonia oxidation bacteria is further detected according to the period circulation of the two stages, and a more suitable stage is further selected according to the growth condition of the anaerobic ammonia oxidation bacteria, so that the adjustment scheme of the more suitable stage can be correspondingly adopted. The growth condition of the anaerobic ammonia oxidation bacteria is used as feedback, the better growth of the anaerobic ammonia oxidation bacteria is ensured, the operation parameters can be selected to be adjusted only through the prediction parameters, and the operation parameters of the potential step can be adjusted by comprehensively selecting the orientation parameters and the prediction parameters. The method is a more optimized scheme, ensures the growth condition of anaerobic ammonia oxidation bacteria, realizes rapid directional stimulation of the growth of anaerobic ammonia oxidation bacteria, and improves the denitrification efficiency.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (8)
1. An anaerobic ammonia oxidation rapid initiation and directional stimulation process for controlling a potential step, comprising:
S100, constructing an anaerobic ammonia oxidation denitrification system based on a bioelectrochemical technology;
s200, in the anaerobic ammonia oxidation denitrification system, providing high potential and low potential for the cathode in a time-division mode, and constructing potential steps;
s300, determining a predicted parameter of the potential step based on the intelligent model;
s400, adjusting potential step operation parameters into prediction parameters in real time through an anaerobic ammonia oxidation denitrification system, and regulating and controlling the growth of anaerobic ammonia oxidation bacteria in the anaerobic ammonia oxidation denitrification system in real time through the adjusted potential step;
in S300, determining a predicted parameter of the potential step based on the smart model includes:
s301, determining input parameters, wherein the input parameters comprise: pH value, temperature and NH of liquid to be treated in anaerobic ammonia oxidation denitrification system 4 + Concentration, NO 2 - Concentration, NO 3 - Concentration, ORP value, COD value, anaerobic ammonia oxidation bacteria growth kinetics;
s302, inputting the input parameters into the intelligent model, and calculating an output result by the intelligent model according to a prediction algorithm of the model, wherein the output result comprises: predicted potential intensity and predicted action time of the potential step;
s303, taking the predicted potential intensity and the predicted action time of the potential step as the predicted parameters of the potential step;
The step S302 further includes: constructing an intelligent model;
the method for constructing the intelligent model comprises the following steps:
s3021, constructing a cyclic neural network (RNN) model, wherein the RNN model comprises: an input layer, a hidden layer and an output layer;
s3022, input variables of the input layer include: pH value, temperature and NH of liquid to be treated 4 + Concentration, NO 2 - Concentration, NO 3 - The concentration, ORP value, COD value and anammox bacteria growth dynamics are carried out, and input variables are normalized and embedded; setting a training set, a data quantity, training times and step sizes, carrying out predictive calculation on an input variable through a time sequence predictive calculation formula, and determining an output variable according to a predictive calculation result;
s3023, the output layer includes one or more neurons, each neuron representing an output variable, the output variable including: predicting potential intensity and predicting action time;
s3024, compressing an output variable into a response value range by adopting an activation function at the last layer of the RNN model;
s3025, performing parameter adjustment on the trained RNN model by using a model evaluation method to obtain an optimized intelligent model.
2. The rapid start-up and directional stimulation process of anaerobic ammonia oxidation with controlled potential step according to claim 1, wherein in S200, providing a high potential and a low potential to the cathode for a time period, creating a potential step, comprises:
S201, setting a first potential intensity of a high potential of the potential step and a first action time of the high potential, the first action time including: a first start time and a first duration; the first start time is a start time of supplying a high potential to the cathode, and the first duration is a hold time of continuously supplying the high potential to the cathode;
s202, setting a second potential intensity of a low potential of the potential step and a second action time of the low potential; the second action time includes: a second start time, which is a start time for supplying a low potential to the cathode, and a second duration, which is a hold time for continuously supplying a low potential to the cathode.
3. The rapid start-up and directional stimulation process for controlling potential step anaerobic ammonia oxidation according to claim 1, wherein in S3022, the predictive calculation is performed on the input variable by using a calculation formula of time sequence prediction, and the determining the output variable according to the result of the predictive calculation comprises:
s3022-1, setting an input sequence with a length of T and a corresponding output sequence;
s3022-2, identifying a feature vector for each input variable, for predicting a scalar, the scalar being an output variable;
S3022-3, after the input variable of each time step is subjected to forward calculation of the hidden layer, predicting a predicted value of the next time step in an iterative calculation mode, wherein the predicted value is an output variable; the calculation formula of the forward calculation includes:
h t ＝f(x t ，h t-1 ；θ)，
wherein f represents the transfer function of the RNN, g represents the final prediction function or classifier, θ represents a learnable parameter, h t Indicating the implicit state of the current time step, h t-1 Indicating the implicit state, x, of the current time step t The input variable for the current time step,representing the final predicted output variable.
4. The rapid initiation and directional stimulation process of anaerobic ammonia oxidation with controlled potential step according to claim 1, wherein S400 comprises:
s401, monitoring prediction parameters of an intelligent model in real time by an anaerobic ammonia oxidation denitrification system;
s402, if the prediction parameters of the intelligent model are different from the current operation parameters, the operation parameters are adjusted to the prediction parameters; if the predicted parameter is the same as the current operation parameter, a potential step is formed according to the current operation parameter.
5. The rapid initiation and directional stimulation process of anaerobic ammonia oxidation with controlled potential step according to claim 1, wherein S400 comprises:
S403, setting a time node and an adjustment period for adjusting the operation parameters; the time node is related to the prediction time of the prediction parameters predicted by the intelligent model;
s404, monitoring the relation between the time node and the prediction time of the prediction parameter predicted by the intelligent model, and taking the time node as a final time node if the time node is before the prediction time and the time difference between the time node and the prediction time is within a set threshold range; if the time node is after the predicted time, readjusting the time node to ensure that the time node is before the predicted time;
s405, monitoring the relation between the adjustment period and the prediction period of the intelligent model, and if the adjustment period is matched with the prediction period of the intelligent model, setting a final adjustment period by taking a final time node as the period starting time;
s406, adjusting the potential step operation parameters to the prediction parameters according to the final time node and the final adjustment period in a period adjustment mode.
6. The rapid start-up and directional stimulation process for anaerobic ammonia oxidation with controlled potential step according to claim 5, wherein in S403, the time node and adjustment period for the operation parameter adjustment is set, comprising:
S4031, obtaining a prediction rule of the intelligent model, wherein the prediction rule comprises: the prediction starting time and the prediction ending time of each prediction are used for determining a prediction period according to the prediction starting time and the prediction ending time;
s4032, setting an adjustment period to be an integer multiple of a prediction period; adjusting the multiple value according to the growth condition of the adjusted anaerobic ammonium oxidation bacteria, and determining an adjustment period according to the adjusted multiple value and the predicted period data;
s4033, setting a predetermined amount according to the determined adjustment period, and setting the time node to a time point having a predetermined amount with respect to the start time of the adjustment period.
7. The rapid start-up and directional stimulation process for anaerobic ammonia oxidation with controlled potential step according to claim 1, wherein after S400, comprising S500, setting a directional parameter, and comprehensively adjusting the operation parameters of the potential step according to the directional parameter and the prediction parameter;
the S500 includes:
s501, in a high potential phase of the potential step, detecting NO in the liquid to be treated 2 - Concentration; according to NO 2 - The concentration is set to a first adjustment parameter which is used for adjusting the potential intensity and the action time of the high potential so as to lead NO in the liquid to be treated 2 - The concentration is maintained within a set first range;
S502, in the low potential stage of the potential step, detecting NO in the liquid to be treated 3 - Concentration and N 2 Concentration according to NO 3 - Concentration and N 2 Setting a second adjustment parameter for adjusting the potential intensity and the action time of the low potential to enable NO in the liquid to be treated 3 - The concentration is kept within a second range, and N 2 The concentration is kept within a set third range;
s503, carrying out weighted average calculation on the first adjustment parameter and the second adjustment parameter to obtain an orientation parameter;
s504, comprehensively adjusting the operation parameters of the potential step according to the orientation parameters and the prediction parameters.
8. The rapid start-up and directional stimulation process for anaerobic ammonia oxidation with controlled potential step according to claim 7, wherein in S504, the operation parameters of the potential step are adjusted according to the directional parameters and the predicted parameters, comprising:
s5041, setting a comprehensive adjustment period, wherein the comprehensive adjustment period comprises a first stage and a second stage, the first stage is a stage for adjusting the operation parameters only through the prediction parameters, and the second stage is a stage for comprehensively adjusting the operation parameters of the potential step through the orientation parameters and the prediction parameters;
s5042, adjusting the potential step to form the potential step in a cycle mode of the comprehensive adjustment cycle;
S5043, monitoring first comparison data of the growth condition of the corresponding anaerobic ammonia oxidation bacteria when the first stage is the current stage;
monitoring second comparison data of the growth condition of the corresponding anaerobic ammonia oxidation bacteria when the second stage is the current stage;
s5044, the first comparison data and the second comparison data are used as the basis of the quality evaluation of the growth condition of the anaerobic ammonium oxidation bacteria, and the corresponding stages of the comparison data with high quality evaluation are marked; the operating parameters of the potential steps are adjusted using the adjustment scheme of the corresponding stage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310933292.3A CN116947208B (en) | 2023-07-26 | 2023-07-26 | Anaerobic ammonia oxidation rapid starting and directional stimulation process capable of controlling potential step |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310933292.3A CN116947208B (en) | 2023-07-26 | 2023-07-26 | Anaerobic ammonia oxidation rapid starting and directional stimulation process capable of controlling potential step |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116947208A CN116947208A (en) | 2023-10-27 |
| CN116947208B true CN116947208B (en) | 2024-01-02 |
Family
ID=88445974
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310933292.3A Active CN116947208B (en) | 2023-07-26 | 2023-07-26 | Anaerobic ammonia oxidation rapid starting and directional stimulation process capable of controlling potential step |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116947208B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110330096A (en) * | 2019-06-24 | 2019-10-15 | 浙江大学 | Electrochemical couple anaerobic organism geochemistry microcosm experimental rig and application method |
| CN111573834A (en) * | 2020-05-22 | 2020-08-25 | 盐城工学院 | Reactor based on short-cut denitrification electrode is in coordination with anaerobic ammonia oxidation denitrogenation |
| CN113683188A (en) * | 2021-09-13 | 2021-11-23 | 江苏大学 | Method and device for electrochemically domesticating anaerobic ammonium oxidation bacteria |
-
2023
- 2023-07-26 CN CN202310933292.3A patent/CN116947208B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110330096A (en) * | 2019-06-24 | 2019-10-15 | 浙江大学 | Electrochemical couple anaerobic organism geochemistry microcosm experimental rig and application method |
| CN111573834A (en) * | 2020-05-22 | 2020-08-25 | 盐城工学院 | Reactor based on short-cut denitrification electrode is in coordination with anaerobic ammonia oxidation denitrogenation |
| CN113683188A (en) * | 2021-09-13 | 2021-11-23 | 江苏大学 | Method and device for electrochemically domesticating anaerobic ammonium oxidation bacteria |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116947208A (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9747544B2 (en) | Method and system for wastewater treatment based on dissolved oxygen control by fuzzy neural network | |
| CN109669352B (en) | Oily sewage treatment process optimization control method based on self-adaptive multi-target particle swarm | |
| Feng et al. | Using microbial fuel cell output metrics and nonlinear modeling techniques for smart biosensing | |
| CN106651032B (en) | Method for intelligently controlling neural network in ammoxidation process by SBR (sequencing batch reactor) method | |
| Ostace et al. | Extension of Activated Sludge Model No 1 with two-step nitrification and denitrification processes for operation improvement. | |
| KR100683477B1 (en) | Method for controlling activated sludge process based on prediction using module-typed model | |
| CN113189881A (en) | Multi-objective optimization control method and system for sewage treatment | |
| CN113448245A (en) | Deep learning-based dissolved oxygen control method and system in sewage treatment process | |
| CN116947208B (en) | Anaerobic ammonia oxidation rapid starting and directional stimulation process capable of controlling potential step | |
| CN113111576B (en) | Mixed coding particle swarm-long-short-term memory neural network-based effluent ammonia nitrogen soft measurement method | |
| CN112250166B (en) | Printing and dyeing wastewater efficient detoxification gradient treatment method based on intelligent SBR electrode process | |
| CN112250246B (en) | Intelligent azo dye wastewater detoxification and advanced treatment method | |
| CN111290275B (en) | Sewage treatment optimization control method based on reinforcement learning particle swarm algorithm | |
| CN113568311A (en) | Intelligent sewage treatment optimization control method based on knowledge information | |
| KR100912032B1 (en) | Apparatus and method for controlling activated sludge process | |
| Yang et al. | Multivariate optimization of the electrochemical degradation for COD and TN removal from wastewater: An inverse computation machine learning approach | |
| CN115481720A (en) | Coagulant dosing amount prediction method based on BR-NARX neural network | |
| CN111798134B (en) | Method for improving methane yield of sewage treatment plant based on data-driven model | |
| Keskitalo et al. | Artificial neural network ensembles in hybrid modelling of activated sludge plant | |
| Caraman et al. | Predictive control of a wastewater treatment process | |
| Mihály et al. | Optimization of the Wastewater Treatment Plant Recycle Flowrates Using Artificial Neural Networks | |
| CN117420759A (en) | Urban sewage treatment process robust control method based on self-adaptive two-type model neural network | |
| CN114967780B (en) | Desulfurization system pH value control method and system based on predictive control | |
| CN115713977A (en) | Online self-learning random configuration network effluent ammonia nitrogen concentration real-time prediction method | |
| Xu et al. | Intelligent Dosing Applying Long Short-Term Memory Algorithm |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |