CN112250246B - Intelligent azo dye wastewater detoxification and advanced treatment method - Google Patents

Intelligent azo dye wastewater detoxification and advanced treatment method Download PDF

Info

Publication number
CN112250246B
CN112250246B CN202010876463.XA CN202010876463A CN112250246B CN 112250246 B CN112250246 B CN 112250246B CN 202010876463 A CN202010876463 A CN 202010876463A CN 112250246 B CN112250246 B CN 112250246B
Authority
CN
China
Prior art keywords
azo dye
dye wastewater
neural network
detoxification
network model
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
Application number
CN202010876463.XA
Other languages
Chinese (zh)
Other versions
CN112250246A (en
Inventor
乔椋
远野
殷万欣
丁成
陈天明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yancheng Institute of Technology
Yancheng Institute of Technology Technology Transfer Center Co Ltd
Original Assignee
Yancheng Institute of Technology
Yancheng Institute of Technology Technology Transfer Center Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Yancheng Institute of Technology, Yancheng Institute of Technology Technology Transfer Center Co Ltd filed Critical Yancheng Institute of Technology
Priority to CN202010876463.XA priority Critical patent/CN112250246B/en
Publication of CN112250246A publication Critical patent/CN112250246A/en
Application granted granted Critical
Publication of CN112250246B publication Critical patent/CN112250246B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • C02F1/00—Treatment of water, waste water, or sewage
    • C02F1/46—Treatment of water, waste water, or sewage by electrochemical methods
    • C02F1/461—Treatment of water, waste water, or sewage by electrochemical methods by electrolysis
    • C02F1/46104—Devices therefor; Their operating or servicing
    • 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
    • C02F2101/30—Organic compounds
    • C02F2101/308—Dyes; Colorants; Fluorescent agents

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Microbiology (AREA)
  • Theoretical Computer Science (AREA)
  • Hydrology & Water Resources (AREA)
  • Environmental & Geological Engineering (AREA)
  • Water Supply & Treatment (AREA)
  • Organic Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Electrochemistry (AREA)
  • General Chemical & Material Sciences (AREA)
  • Treatment Of Water By Oxidation Or Reduction (AREA)
  • Purification Treatments By Anaerobic Or Anaerobic And Aerobic Bacteria Or Animals (AREA)

Abstract

The invention discloses an intelligent azo dye wastewater detoxification and advanced treatment method, wherein azo dye wastewater is treated by the action of point-by-point water inflow in an anaerobic sludge tower under the action of an electrode in an electrode area and the anaerobic hydrolysis acidification action in an anaerobic sludge area; the azo dye wastewater is uniformly distributed in the anaerobic sludge tower through point-by-point water feeding; and the electrode area and the sludge area are provided with detection sensors, the detection sensors acquire the azo dye reaction detoxification conditions of the corresponding areas in real time, the acquired data are used as input layer parameters of a neural network model, and the electrode potential and the water quantity of a water inlet pump are fed back and adjusted through the neural network model, so that the intelligent treatment of the azo dye wastewater detoxification is realized. The method ensures the high-efficiency hydrolysis of the azo dye wastewater through the multi-stage electrodes and the anaerobic hydrolysis acidification effect, and realizes the intelligent treatment of the azo dye wastewater by combining a neural network model.

Description

Intelligent azo dye wastewater detoxification and advanced treatment method
Technical Field
The invention belongs to the technical field of sewage treatment, and particularly relates to a detoxification treatment method for azo dye wastewater.
Background
The azo dye wastewater has high chroma, strong toxicity and difficult degradation, and is regarded as one of the wastewater to be treated urgently. At present, biochemical treatment is still adopted in the main processes of wastewater treatment systems of most enterprises in China. Toxic organic matters such as azo, nitro and the like contained in the dye wastewater are often main factors causing difficult degradation, poor biodegradability and the like of the wastewater, and the process reaches the standard and reduces emission to face serious challenges. Therefore, solving the toxicity problem of the dye wastewater is a prerequisite for reaching the discharge standard.
The traditional pretreatment method of the azo dye, such as flocculation precipitation, membrane treatment, adsorption and the like, can not completely eliminate the toxicity of the azo dye, and is easy to generate dangerous waste and cause secondary pollution. At present, the pretreatment detoxification method can be mostly classified into a chemical oxidation method and a biochemical method. The chemical oxidation method is characterized in that chemical agents with strong oxidizing property, such as Fenton reagent, ozone, iron-carbon filler and the like, are added in a pretreatment stage, and the method has the advantages of high detoxification reaction speed, high energy consumption, high cost and the like. With the regulation of the state to the chemical industry, middle and small chemical enterprises with backward processes and equipment are eliminated, the price of chemical oxidants is gradually increased, and the chemical oxidation method is only suitable for small-discharge industrial wastewater.
The biological method is to utilize the hydrolytic acidification way of microbes to remove some toxic substances or inhibitory substances and change the structure of some refractory organic matters into easily degradable substances. Because the microorganism has the characteristic of adaptability, the reaction rate is relatively slow, but the cost is far lower than that of a chemical oxidation method, and the method has no energy consumption, has great advantages for treating the wastewater with large discharge amount, and is a pretreatment process widely adopted in the field of wastewater treatment. If the tolerance and the degradation capability of the hydrolytic acidification process to toxic pollutants can be improved, the dependence on chemical agents is greatly reduced, and the operation cost of industrial wastewater is reduced. The research idea of strengthening pollutant degradation and transformation by using the extracellular electron transfer effect of microorganisms is provided internationally, and the azo dye is decolored and subjected to ring opening by using a microbial electrolysis cell, so that the electron transfer path of the microorganisms can be directionally regulated and controlled, the metabolic pathway between the microorganisms and the pollutants is activated, and the complexity or the oxidability of the pollutants difficult to degrade is reduced, so that the functions of decoloring, detoxifying, azo bond breaking and the like of the pollutants are strengthened, the biological inhibition of the toxic pollutants is blocked, and the biochemical performance of the wastewater is greatly improved.
The environmental protection industry is used as a terminal node for industrial pollution treatment, the pollution is removed at high quality, the intelligent process is accelerated, the control automation, the detection real-time and the unmanned operation of the reaction process are realized, and the environmental protection industry becomes an important development direction for environmental management. The traditional azo dye wastewater treatment process usually needs long-time debugging and long-term detection on treatment effect, generally needs to consume a large amount of manpower and material resources, can not detect the change of internal azo molecules in time due to equipment type selection in some processes, can only meet the starting of the process by detecting the running experience of water data combined process, has less detection on the ring-opening detoxification process of azo dyes, and does not have the function of intelligently regulating and controlling the running of equipment.
Disclosure of Invention
In order to solve the technical problems mentioned in the background technology, the invention provides an intelligent azo dye wastewater detoxification and advanced treatment method.
In order to achieve the technical purpose, the technical scheme of the invention is as follows:
an intelligent azo dye wastewater detoxification and advanced treatment method is characterized in that: azo dye wastewater is treated by the action of point-by-point water feeding in an anaerobic sludge tower under the action of an electrode in an electrode area and the anaerobic hydrolysis acidification action in an anaerobic sludge area respectively; the azo dye wastewater is uniformly distributed in the anaerobic sludge tower through point-by-point water feeding; the electrode area comprises an anode area and a cathode area, and the anaerobic sludge area comprises anaerobic sludge and a separation net layer attached by the sludge; and the electrode area and the sludge area are provided with detection sensors, the detection sensors acquire the azo dye reaction detoxification conditions of the corresponding areas in real time, the acquired data are used as input layer parameters of a neural network model, and the electrode potential and the water quantity of a water inlet pump are fed back and adjusted through the neural network model, so that the intelligent treatment of the azo dye wastewater detoxification is realized.
Further, the azo dye wastewater is treated in an anaerobic sludge tower by the following steps:
step 1: the azo dye wastewater is uniformly mixed under the action of a water homogenizer, then enters each stage of electrode area under the action of a water pump, and the micromolecule COD in the inlet water is oxidized into CO at the anode 2 And H + While generating electrons e - The azo dye macromolecules firstly receive electrons generated by the anode at the cathode for accelerated ring opening, and part of the micromolecules generated after the ring opening are used as COD (chemical oxygen demand) to be oxidized at the anode again to realize the electron e - The sewage flows into an anaerobic sludge area along with the sewage;
step 2: in an anaerobic sludge zone, azo dye macromolecules which are subjected to ring opening are further decomposed into micromolecules through anaerobic hydrolysis acidification, wherein a part of micromolecules enter an anode of a next-stage electrode zone as COD to be oxidized, and electrons e are realized - Self-feeding of (1);
and step 3: when the azo dye wastewater enters the top reaction zone for treatment, a part of the azo dye wastewater flows back to the water homogenizer for secondary treatment, and the rest of the azo dye wastewater enters the subsequent treatment process.
Furthermore, the separation net layers are arranged above the electrodes to ensure that water inlet smoothly reacts with the inside of sludge on the surfaces of the electrodes, and the increase and decrease of the total amount of the sludge are realized by adjusting the number of the separation net layers.
Furthermore, the cathode and the anode of the electrode area are connected with an external power supply, the provided electric potential strength is the electric potential strength which enables azo dye molecules to be decomposed or opened by the electrochemical action, and the increase and decrease of the electric potential in the system are realized by adjusting the number of the cathode and the anode of the electrode area.
Further, the neural network model is an RBF neural network model.
Further, the construction method of the RBF neural network model is as follows:
(1) The method comprises the following steps of (1) arranging monitoring data of each sensor in the azo treatment process, taking the concentration of azo dyes, the concentration of azo dye detoxification products, the COD concentration, the pH value of each position of a reactor, the temperature, the local hydraulic retention time, the total hydraulic retention time and the reflux as input parameters of a RBF neural network model, and carrying out normalization operation on the input parameters:
Figure BDA0002652714780000031
in the above formula, x' is a normalized value, x is a value of an input parameter, a is a set of each input variable, and min and max respectively represent a minimum value and a maximum value in the set;
and selecting the concentration of azo dye detoxification products, the COD concentration, the pH value, the temperature, the local hydraulic retention time, the total hydraulic retention time and the reflux quantity of each position of the reactor in the next time period as the output of the RBF neural network model;
(2) Constructing an RBF neural network model:
(201) Determining the input parameter as X, X = [ X ] according to the data obtained in the step 1 1 ,x 2 ,...,x n ] T N is the number of input layer units;
(202) Determining an output vector Y = [ Y ] 1 ,y 2 ,...,y q ]Q is the number of output layer units;
(203) Initializing connection weight W of hidden layer value output layer k ＝[w k1 ,w k2 ,...,w kp ] T P is the number of hidden layer units, k =1,2, \ 8230;, q, w kj The initialization method of (1) is as follows:
Figure BDA0002652714780000041
in the above formula, mink is the minimum value of all expected outputs in the kth output neuron in the training set, maxk is the maximum value of expected outputs in the kth output neuron in the training set, j =1,2, \ 8230;, p;
(204) Initializing the Central parameter C of hidden layer neurons j ＝[c j1 ,c j2 ,...,c jn ] T Wherein c is ji The initialization method of (1) is as follows:
Figure BDA0002652714780000042
in the above formula, min i is the minimum value of the central parameters of the hidden layer neuron, max i is the maximum value of the central parameters of the hidden layer neuron, i =,1,2, \8230;, n,
(205) Initializing Width vector D for hidden layer neurons j ＝[d j1 ,d j2 ,...,d jn ] T Wherein d is ji The initialization method of (1) is as follows:
Figure BDA0002652714780000051
in the above formula, d f For the width adjustment factor, i =,1,2, \8230;, N, N 1 In order to input the number of samples,
Figure BDA0002652714780000052
the kth output neuron for the ith set of samples;
(3) Training an RBF neural network model:
(301) Computing the output z of the jth neuron of the hidden layer j ：
Figure BDA0002652714780000053
In the above formula, | | · | |, represents the euclidean norm;
(302) Calculating output vector Y = [ Y ] 1 ,y 2 ,...,y q ]In which
Figure BDA0002652714780000054
(303) And (3) iteratively calculating each parameter:
Figure BDA0002652714780000055
Figure BDA0002652714780000056
Figure BDA0002652714780000057
in the above formula, w kj (t)、c ji (t)、d ji (t) w for the t-th iteration respectively kj 、c ji 、d ji A value of (d); eta is a learning factor; e is RBF nerveThe network evaluation function is a function of the network evaluation,
Figure BDA0002652714780000058
wherein o is lk For the expected output value of the kth output neuron at the l input sample, y lk For the net output value of the kth output neuron at the l input sample, N 2 As the number of input samples, α is the learning rate;
(304) Calculating the root mean square error RMS of the RBF neural network:
Figure BDA0002652714780000059
if RMS is less than or equal to epsilon, epsilon is a preset threshold value, the training is finished, otherwise, the step (303) is returned.
Adopt the beneficial effect that above-mentioned technical scheme brought:
(1) The azo dye molecule receives electrons generated by the anodic oxidation of COD, and the accelerated ring opening is realized under the action of a microbial cathode, so that the accelerated ring opening and the detoxification of the azo dye molecule at the cathode are realized;
(2) Part of micromolecular substances generated after the azo dye molecules are subjected to accelerated ring opening and detoxification can be used as COD to be continuously oxidized at the anode to generate electrons; through the combination of the point-divided water feeding effect and the hydrolytic acidification, the detoxified azo dye molecules are further decomposed and converted by the microbial hydrolytic acidification in the process, and the generated micromolecule substances are continuously oxidized at the anode as COD (chemical oxygen demand), so that the electron self-supply in the process is realized;
(3) The method utilizes the neural network model to analyze the input parameters and the function weights of the hidden layer and the output layer, carries out prediction on the parameters of the output layer, adjusts the parameters of process operation equipment in real time, and realizes the intellectualization and the unmanned realization of the accelerated detoxification of the azo dye wastewater;
(4) The invention predicts the decomposition path and decomposition product of azo dye molecules in a system through the electrode-anaerobic biological action by utilizing the neural network model learning and real-time analysis of process operation data, predicts the content and molecular structure of intermediate products which are difficult to detect, and feeds back the intermediate products to the neural network model to optimize the process.
Drawings
FIG. 1 is a process flow diagram of the present invention;
FIG. 2 is a schematic diagram of the mechanism by which the electrode of the present invention effects detoxification of azo dye molecules;
FIG. 3 is a schematic diagram of a single stage reaction zone configuration according to the present invention;
FIG. 4 is a block diagram of a neural network model in the present invention;
FIG. 5 is a flow chart of a neural network model in the present invention.
Detailed Description
The technical scheme of the invention is explained in detail in the following with the accompanying drawings.
The invention designs an intelligent azo dye wastewater detoxification and advanced treatment method, as shown in figure 1, azo dye wastewater is treated by the action of point-by-point water feeding in an anaerobic sludge tower under the electrode action of an electrode region and the anaerobic hydrolysis acidification action of an anaerobic sludge region respectively; azo dye wastewater is uniformly distributed in an anaerobic sludge tower through point-by-point water inlet; the electrode area comprises an anode area and a cathode area which are connected with each other, and the anaerobic sludge area comprises anaerobic sludge and a separation net layer to which the sludge is attached; arranging detection sensors in the electrode area and the sludge area, uploading the azo dye reaction detoxification condition of the corresponding area to a data collection center in real time by the detection sensors, wherein the collected data comprises but is not limited to the concentration of the azo dye in the reactor, the concentration of azo dye molecule detoxification products, the COD concentration, the pH value and the temperature of the reactor; the data analysis center takes the collected data as the input layer parameters of the neural network model, and the electrode potential and the water quantity of the water inlet pump are fed back and adjusted through the neural network model, so that the intelligent treatment of the azo dye wastewater detoxification is realized.
The azo dye wastewater is uniformly mixed under the action of a water homogenizer, enters a reactor at a specific flow rate and a water inlet site under the action of a water pump after initial calculation, and as shown in figure 2, the azo dye is firstly decomposed in an electrode area under the action of an electrode and then enters an anaerobic sludge area to be decomposed again under the action of hydrolysis and acidification; azo dye wastewaterIn the electrode area, small molecule COD in the inlet water is oxidized into CO at the anode 2 And H + While generating electrons e - The contained azo dye macromolecules firstly receive electrons generated by the anode at the cathode for accelerated ring opening, and a part of the micromolecules generated after the ring opening can be used as COD for oxidizing at the anode again to realize e - The sewage flows into an anaerobic sludge area along with the sewage; in the anaerobic sludge area, the ring-opened azo dye macromolecules are further decomposed into small molecules through anaerobic hydrolytic acidification. At the moment, along with the upward flowing of the azo wastewater, a part of small molecules can be used as COD and enter an anode region of a next-stage electrode to be oxidized, so that e is realized - Self-feeding of (2). The continuity of the azo wastewater treatment in the anaerobic sludge tower is ensured by the two times of e-self-supply modes generated by the electrodes and the hydrolysis acidification of the anaerobic sludge.
From the angle of a multistage electrode-anaerobic sludge tower, 1 treatment unit is formed by 3 layers of electrodes and 2 layers of anaerobic sludge, as shown in figure 1, the inlet water in the area a passes through 5 treatment units respectively, the inlet water in the area b passes through 4 treatment units respectively, the inlet water in the area b passes through 3 treatment units respectively, the inlet water in the area c passes through 3 treatment units respectively, the inlet water in the area d passes through 2 treatment units respectively, and the inlet water is treated by at least 2 treatment units; and finally, after azo wastewater fed by the treatment unit e is treated, part of the azo wastewater flows back to the water homogenizer to be treated again, and the rest of the azo wastewater is discharged into a subsequent treatment process to be subjected to advanced treatment. The treatment unit can adjust the electrode and the sludge separation net layer to realize increase and decrease according to the actual sewage concentration and content.
As shown in figure 3, the sludge of the anaerobic sludge tower is attached to the surface of the sludge separation mesh layer and is arranged above the electrode to ensure that the inlet water smoothly reacts with the inside of the sludge on the surface of the electrode. The effective content of the sludge and the distribution of the sludge in the anaerobic sludge tower are ensured by the action of the separation net layer. The sludge separation net layer is formed by overlapping a plurality of separation layers, the separation layers are fixed in a certain number in a staggered overlapping mode to form a sludge separation layer, and the sludge separation layer is fixed with biomass through film hanging in advance, so that the sludge quantity is maintained, namely the anaerobic sludge cannot fall off or settle along with gravity under the action of water flow. The increase and decrease of the total amount of the sludge are realized by adjusting the number of the separation net layers.
The cathode and the anode of the electrode area are connected with an external power supply, and the provided electric potential strength is the electric potential strength which enables azo dye molecules to be decomposed or opened by electrochemical action. The increase and decrease of the electric potential in the system can be realized by adjusting the number of the cathodes and the anodes in the electrode area.
In this embodiment, an RBF neural network model is taken as an example to explain an approach of the neural network model to realize intelligence. The RBF neural network model is one way to achieve intelligence, but should not be construed as limiting the invention. The intelligent process is realized through a RBF neural network model, the internal relation among pollutants is automatically analyzed, the reaction condition of the next time sequence is predicted, the current operation condition is adjusted, and the potential, the flow of a water inlet pump and the related reaction operation parameter indexes are fed back and adjusted, so that the efficient detoxification and advanced treatment of the azo dye are realized. In the present invention, the structure and training process of the RBF neural network model are shown in fig. 4 and 5.
Through the action of point-by-point water inlet, a multistage electrode and anaerobic microorganisms fixed on a sludge separation net layer, the mixture stays in an anaerobic sludge tower for enough time to realize the accelerated ring opening of azo dyes, the breakage of azo bonds and the detoxification and decoloration of the azo dyes, thereby blocking the inhibition of toxic pollutants on organisms and improving the biodegradability of wastewater; the method is combined with the learning analysis of the neural network model on the real-time data, optimizes the model structure, reduces the prediction error, realizes the high-efficiency prediction of the output value, forms a system for the cooperative application of the environment-friendly process and the neural network model, accelerates the high-efficiency detoxification of the azo dye wastewater under the electrochemical-biological action and provides an auxiliary effect for the subsequent deep treatment of the azo dye
The embodiments are only for illustrating the technical idea of the present invention, and the technical idea of the present invention is not limited thereto, and any modifications made on the basis of the technical scheme according to the technical idea of the present invention fall within the scope of the present invention.

Claims (5)

1. An intelligent azo dye wastewater detoxification and advanced treatment method is characterized in that: azo dye wastewater is treated by the action of point-by-point water feeding in an anaerobic sludge tower under the action of an electrode in an electrode area and the anaerobic hydrolysis acidification action in an anaerobic sludge area respectively; the azo dye wastewater is uniformly distributed in the anaerobic sludge tower through point-by-point water feeding; the electrode area comprises an anode area and a cathode area, and the anaerobic sludge area comprises anaerobic sludge and a separation net layer attached by the sludge; the electrode area and the sludge area are provided with detection sensors, the detection sensors acquire the azo dye reaction detoxification conditions of the corresponding areas in real time, the acquired data are used as input layer parameters of a neural network model, and the electrode potential and the water quantity of a water inlet pump are fed back and adjusted through the neural network model, so that the intelligent treatment of the azo dye wastewater detoxification is realized;
the treatment steps of the azo dye wastewater in the anaerobic sludge tower are as follows:
step 1: the azo dye wastewater is uniformly mixed under the action of a water homogenizer, then enters each stage of electrode area under the action of a water pump, and the micromolecule COD in the inlet water is oxidized into CO at the anode 2 And H + While generating electrons e - The azo dye macromolecules firstly receive electrons generated by an anode at a cathode to accelerate ring opening, and part of micromolecules generated after ring opening are used as COD to be oxidized at the anode again to realize the electron e - The sewage flows into an anaerobic sludge area along with the sewage;
step 2: in the anaerobic sludge zone, azo dye macromolecules which are subjected to ring opening are further decomposed into small molecules through anaerobic hydrolytic acidification, wherein a part of the small molecules are used as COD to enter the anode of the next-stage electrode zone for oxidation, and electrons e are realized - Self-feeding of (1);
and step 3: when the azo dye wastewater enters the top reaction zone for treatment, a part of the azo dye wastewater flows back to the water homogenizer for secondary treatment, and the rest of the azo dye wastewater enters the subsequent treatment process.
2. The intelligent azo dye wastewater detoxification and advanced treatment method according to claim 1, wherein the method comprises the following steps: the separation net layers are arranged above the electrodes to ensure that water inlet smoothly reacts with the inside of sludge on the surfaces of the electrodes, and the increase and decrease of the total amount of the sludge are realized by adjusting the number of the separation net layers.
3. The intelligent azo dye wastewater detoxification and advanced treatment method according to claim 1, wherein the method comprises the following steps: the cathode and the anode of the electrode area are connected with an external power supply, the provided electric potential strength is the electric potential strength which enables azo dye molecules to be decomposed or opened by the electrochemical action, and the increase and decrease of the electric potential in the system are realized by adjusting the number of the cathode and the anode of the electrode area.
4. The intelligent azo dye wastewater detoxification and advanced treatment method according to claim 1, wherein the method comprises the following steps: the neural network model is an RBF neural network model.
5. The intelligent azo dye wastewater detoxification and advanced treatment method according to claim 4, wherein the method comprises the following steps: the construction method of the RBF neural network model comprises the following steps:
(1) The method comprises the following steps of (1) arranging monitoring data of each sensor in the azo treatment process, taking the concentration of azo dyes, the concentration of azo dye detoxification products, the COD concentration, the pH value of each position of a reactor, the temperature, the local hydraulic retention time, the total hydraulic retention time and the reflux as input parameters of a RBF neural network model, and carrying out normalization operation on the input parameters:
Figure FDA0003791577310000021
in the above formula, x' is a normalized value, x is a value of an input parameter, a is a set of each input variable, and min and max respectively represent a minimum value and a maximum value in the set;
and selecting the concentration of azo dye detoxification products, the COD concentration, the pH value, the temperature, the local hydraulic retention time, the total hydraulic retention time and the reflux quantity of each position of the reactor in the next time period as the output of the RBF neural network model;
(2) Constructing an RBF neural network model:
(201) Determining the input parameters asX，X＝[x 1 ,x 2 ,...,x n ] T N is the number of input layer units;
(202) Determining an output vector Y = [ Y ] 1 ,y 2 ,...,y q ]Q is the number of output layer units;
(203) Initializing connection weight W of hidden layer value output layer k ＝[w k1 ,w k2 ,...,w kp ] T P is the number of hidden layer units, k =1,2, \ 8230;, q, w kj The initialization method of (1) is as follows:
Figure FDA0003791577310000031
in the above formula, mink is the minimum value of all expected outputs in the kth output neuron in the training set, maxk is the maximum value of expected outputs in the kth output neuron in the training set, j =1,2, \ 8230;, p;
(204) Initializing the Central parameter C of hidden layer neurons j ＝[c j1 ,c j2 ,...,c jn ] T Wherein c is ji The initialization method of (1) is as follows:
Figure FDA0003791577310000032
in the above formula, min i is the minimum value of the central parameters of the hidden layer neuron, max i is the maximum value of the central parameters of the hidden layer neuron, i =,1,2, \8230;, n,
(205) Initializing Width vector D for hidden layer neurons j ＝[d j1 ,d j2 ,...,d jn ] T Wherein d is ji The initialization method of (1) is as follows:
Figure FDA0003791577310000033
in the above formula, d f For the width adjustment factor, i =,1,2, \8230;, N, N 1 As the number of input samples，
Figure FDA0003791577310000034
The kth output neuron for the ith set of samples;
(3) Training an RBF neural network model:
(301) Computing the output z of the jth neuron of the hidden layer j ：
Figure FDA0003791577310000035
In the above formula, | | · | |, represents the euclidean norm;
(302) Calculating output vector Y = [ Y ] 1 ,y 2 ,...,y q ]Wherein
Figure FDA0003791577310000036
k＝1,2,…,q；
(303) And (3) iteratively calculating each parameter:
Figure FDA0003791577310000037
Figure FDA0003791577310000041
Figure FDA0003791577310000042
in the above formula, w kj (t)、c ji (t)、d ji (t) w for the t-th iteration respectively kj 、c ji 、d ji A value of (d); eta is a learning factor; e is an RBF neural network evaluation function,
Figure FDA0003791577310000043
wherein o is lk For the kth output neuron at the l input sampleDesired output value at present, y lk For the net output value of the kth output neuron at the l input sample, N 2 As the number of input samples, α is the learning rate;
(304) Calculating the root mean square error RMS of the RBF neural network:
Figure FDA0003791577310000044
if the RMS is less than or equal to the epsilon and the epsilon is a preset threshold value, the training is finished, otherwise, the step (303) is returned.
CN202010876463.XA 2020-08-27 2020-08-27 Intelligent azo dye wastewater detoxification and advanced treatment method Active CN112250246B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010876463.XA CN112250246B (en) 2020-08-27 2020-08-27 Intelligent azo dye wastewater detoxification and advanced treatment method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010876463.XA CN112250246B (en) 2020-08-27 2020-08-27 Intelligent azo dye wastewater detoxification and advanced treatment method

Publications (2)

Publication Number Publication Date
CN112250246A CN112250246A (en) 2021-01-22
CN112250246B true CN112250246B (en) 2023-03-21

Family

ID=74224607

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010876463.XA Active CN112250246B (en) 2020-08-27 2020-08-27 Intelligent azo dye wastewater detoxification and advanced treatment method

Country Status (1)

Country Link
CN (1) CN112250246B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113087030B (en) * 2021-03-02 2022-08-02 天津大学 Water resource double-circulation feedback control system and method
CN115477447B (en) * 2022-10-10 2023-10-27 江苏科易达环保科技股份有限公司 Hydrolysis acidification coupling strengthening azo dye wastewater detoxification synergistic method and device

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103951056A (en) * 2014-05-20 2014-07-30 哈尔滨工业大学 Treatment device capable of intensifying degrading of wastewater containing azo dyes by microbial electric auxiliary system and water treatment method
CN104843862A (en) * 2014-12-31 2015-08-19 广东省微生物研究所 Process for intensifying anaerobic degradation and transformation of azo dyes based on breath of microbial electrode
CN105776738A (en) * 2016-03-22 2016-07-20 中国环境科学研究院 Method and device for pre-treating organic wastewater
CN105776775A (en) * 2016-05-12 2016-07-20 武汉理工大学 Anaerobic-autotrophic nitrogen removal-ozone oxidation coupling landfill leachate whole flow process zero emission processing process

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103951056A (en) * 2014-05-20 2014-07-30 哈尔滨工业大学 Treatment device capable of intensifying degrading of wastewater containing azo dyes by microbial electric auxiliary system and water treatment method
CN104843862A (en) * 2014-12-31 2015-08-19 广东省微生物研究所 Process for intensifying anaerobic degradation and transformation of azo dyes based on breath of microbial electrode
CN105776738A (en) * 2016-03-22 2016-07-20 中国环境科学研究院 Method and device for pre-treating organic wastewater
CN105776775A (en) * 2016-05-12 2016-07-20 武汉理工大学 Anaerobic-autotrophic nitrogen removal-ozone oxidation coupling landfill leachate whole flow process zero emission processing process

Also Published As

Publication number Publication date
CN112250246A (en) 2021-01-22

Similar Documents

Publication Publication Date Title
Gajda et al. Recent advancements in real-world microbial fuel cell applications
CN103744293B (en) Wastewater treatment monitoring method and system based on fuzzy neural network
US9747544B2 (en) Method and system for wastewater treatment based on dissolved oxygen control by fuzzy neural network
CN112250246B (en) Intelligent azo dye wastewater detoxification and advanced treatment method
Lee et al. Neural network modeling for on-line estimation of nutrient dynamics in a sequentially-operated batch reactor
Ardakani et al. Microbial fuel cells (MFCs) in integration with anaerobic treatment processes (AnTPs) and membrane bioreactors (MBRs) for simultaneous efficient wastewater/sludge treatment and energy recovery-A state-of-the-art review
CN109052865B (en) Sewage treatment device and method
CN112250166B (en) Printing and dyeing wastewater efficient detoxification gradient treatment method based on intelligent SBR electrode process
Gadkari et al. Understanding the interdependence of operating parameters in microbial electrosynthesis: a numerical investigation
Tsekouras et al. Microbial fuel cell for wastewater treatment as power plant in smart grids: Utopia or reality?
KR102348977B1 (en) Integrated managing platform system for biogas producing plant
CN105174676A (en) Electro-Fenton source sludge reduction method and system based on electrochemical sludge pretreatment
Sulaiman et al. Modeling anaerobic process for wastewater treatment: new trends and methodologies
CN112250161B (en) Self-adaptive potential controller of microbial electrolysis cell for azo wastewater treatment and control method thereof
Arumugam et al. Two-phase anaerobic digestion model of a tannery solid waste: experimental investigation and modeling with ANFIS
CN105836879A (en) An anaerobic electrically-conductive ceramic membrane biological reacting device effectively controlling membrane pollution and a method
Ma et al. Electrochemical membrane bioreactors for sustainable wastewater treatment: principles and challenges
Iyappan et al. Electrochemical treatment of tannery effluent using a battery integrated DC-DC converter and solar PV power supply—An approach towards environment and energy management
CN111377535A (en) Anaerobic forward osmosis membrane bioreactor process for oxidizing ammonia nitrogen by coupling electrochemistry
Pérez et al. Operationalization of a microbial electrolysis cell: The interaction of the primary factors for energy storage efficiency
Huang et al. Modeling and optimization of the activated sludge process
CN108147524A (en) A kind of hydrolysis acidification coupling microorganism electrolysis unit and its processing method
Lin et al. Prediction of wastewater treatment system based on deep learning
Manokaran et al. Feed-forward back-propagation neural network for the electro-oxidation of distillery effluent
Aremu et al. Development of optimum operating parameters for bioelectricity generation from sugar wastewater using response surface methodology

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
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20211221

Address after: 224051 No. 1 hope road middle road, Ting Hu District, Yancheng City, Jiangsu

Applicant after: YANCHENG INSTITUTE OF TECHNOLOGY

Applicant after: Yancheng Institute of Technology Technology Transfer Center Co.,Ltd.

Address before: 224051 No. 1 hope road middle road, Ting Hu District, Yancheng City, Jiangsu

Applicant before: YANCHENG INSTITUTE OF TECHNOLOGY

GR01 Patent grant
GR01 Patent grant