A method and system for determining aeration time of a sewage treatment pool

Abstract

The present invention provides a method and system for determining the aeration time of a sewage treatment pool, belonging to the technical field of sewage treatment, comprising: obtaining water quality index data and water quantity data of an SBR reaction pool; inputting the water quality index data and water quantity data into a trained RNN neural network, and using the RNN neural network to output the required aeration time; inputting the water quality index data and water quantity data into an empirical formula, and using the empirical formula to calculate the required aeration time; if the difference between the required aeration time output by the RNN neural network and the required aeration time calculated by the empirical formula is less than a threshold value, according to t 曝气 =(t RNN +t fuzzy ) / 2 as the aeration time for aeration; if the difference between the aeration time required by the RNN neural network output and the aeration time required by the empirical formula is greater than the threshold, aeration is performed according to the aeration time required by the empirical formula. This method can accurately and conveniently control the aeration time.

Description

Technical Field

[0001] The invention belongs to the technical field of sewage treatment, and in particular relates to a method and system for determining the aeration time of a sewage treatment pool. Background Art

[0002] SBR reactors are widely used in sewage treatment plants. In the sewage treatment process, due to the lack of carbon source in domestic sewage and the low efficiency of biochemical reaction treatment, carbon source supplementation is often required. At present, the carbon source is mostly replenished manually after the problem of insufficient carbon source is found, or PLC is used to replenish carbon source regularly and quantitatively. Both cannot accurately control the aeration volume and aeration time, and often cause waste due to excessive aeration, increasing operating costs. On the other hand, the phosphorus content in domestic sewage is often relatively high. Relying solely on polyphosphate bacteria for phosphorus removal often fails to meet the discharge standards. It is necessary to add flocculants to assist in phosphorus removal. The flocculant replenishment process is also an important process to ensure that the sewage can meet the standard discharge after treatment. Usually, flocculants (generally prepared with a mass concentration of 0.2%) are added to the clarification sedimentation tank to achieve chemical phosphorus removal. The general flocculant is polyaluminum chloride (PAC) or polyacrylamide (PAM) to form phosphate precipitation, and then remove phosphorus in the water through mud and water separation. Therefore, how to accurately control the aeration volume, achieve high phosphorus removal efficiency, and reduce operating costs is a problem that needs to be solved urgently.

[0003] SBR reaction tanks are widely used in sewage treatment plants. In the process of sewage treatment, due to insufficient aeration, the growth rate of microorganisms is slow, and the efficiency of sewage nitrogen and phosphorus removal is low. Excessive aeration can easily cause microorganisms to oxidize and die, resulting in sludge bulking. Similarly, the biochemical reaction treatment efficiency is low and the carbon source supply amount increases.

[0004] Due to the nonlinearity, hysteresis, and uncertainty of aeration volume in the aeration process during the sewage treatment process, at present, the aeration method in the sewage treatment process usually adopts PLC timed quantitative aeration, or adopts dissolved oxygen sensor for feedback aeration, but the hysteresis of dissolved oxygen sensor affected by microbubbles in the water is very obvious, and the data fluctuates greatly, and the aeration volume cannot be effectively controlled. Therefore, how to accurately control the aeration time and aeration volume to achieve efficient nitrogen and phosphorus removal, low carbon energy saving, and reduce carbon source replenishment is a sewage treatment process problem that needs to be solved urgently.

[0005] In short, the existing technology cannot accurately and conveniently control the aeration volume. Summary of the invention

[0006] In order to overcome the above-mentioned deficiencies in the prior art, the present invention provides a method and system for determining the aeration time of a sewage treatment tank.

[0007] In order to achieve the above object, the present invention provides the following technical solutions:

[0008] A method and system for determining aeration time of a sewage treatment pool, comprising:

[0009] Obtain water quality index data, water volume data, and water temperature data of the sequencing batch activated sludge process SBR reaction tank;

[0010] Input water quality index data, water volume data, and water temperature data into the trained RNN neural network, and use the RNN neural network to output the time required for aeration;

[0011] Input water quality index data and water quantity data into the empirical formula, and use the empirical formula to calculate the time required for aeration;

[0012] If the difference between the aeration time required by the RNN neural network output and the aeration time required by the empirical formula is less than the threshold, the aeration time is calculated according to t 曝气 =(t RNN +t fuzzy ) / 2 as the aeration time for aeration; t 曝气 is the aeration time, t RNN is the time required for aeration of the RNN neural network output, t fuzzy The time required for aeration is calculated using the empirical formula;

[0013] If the difference between the aeration time required by the RNN neural network output and the aeration time required by the empirical formula is greater than a threshold, aeration is performed according to the aeration time required by the empirical formula.

[0014] Further, the calculation of the required aeration time using an empirical formula includes:

[0015] Calculate the ratio of COD to NH:

[0016] COD / NH=n

[0017] When n≥3, the aeration time T 曝气 for:

[0018] T 曝气 =(COD-50) / 0.34*MLSS / 1000

[0019] Among them, MLSS is the sludge concentration.

[0020] Further, training the RNN neural network includes:

[0021] The water quality index data, water volume data, water temperature data and the corresponding aeration time are input into the RNN neural network as a training set, and the model parameters of the RNN neural network are adjusted so that the RNN neural network outputs a value equal to the aeration time in the training set, and the RNN neural network training is completed.

[0022] Furthermore, the water quality index data includes:

[0023] Sewage inlet turbidity, COD, ammonia concentration effluent index, effluent COD content, ammonia nitrogen NH content, turbidity.

[0024] Furthermore, it also includes: after each aeration, the aeration time, water quality index data, water volume and water temperature data after the water quality meets the standard are used as database data. When the water quality index data, water volume and water temperature data are detected to be the same as the data in the database next time, the aeration time in the database is directly retrieved for aeration.

[0025] Furthermore, it also includes: using the aeration time, water quality index data, water volume and water temperature data after the water quality reaches the standard in the database as a training set to train and optimize the RNN neural network.

[0026] Further, the method further includes: after aeration is completed, calculating the turbidity ss of the water, and if the turbidity ss>9 mg / L, adding a flocculant to the SBR reaction tank of the sequential batch activated sludge process to purify the sewage, including:

[0027] Calculate the amount of flocculant added:

[0028] C=(V*S*C0) / (Q)

[0029] Where, C is the amount of flocculant added, V is the volume of water, S is the concentration of suspended matter or colloidal matter, C0 is the amount of flocculant required per unit mass of suspended matter; Q is the amount of water;

[0030] Add flocculant according to the added amount of flocculant, turn on the stirring pump at the same time, stir and coagulate, wait until the flocculant addition is completed, and continue stirring for 20 minutes before ending the stirring.

[0031] A system for determining aeration time of a sewage treatment tank, characterized by comprising:

[0032] The sensor module is used to obtain water quality index data, water volume data, and water temperature data of the sequencing batch activated sludge process SBR reaction tank;

[0033] The calculation unit module is used to input water quality index data, water volume data, and water temperature data into the trained RNN neural network, and use the RNN neural network to output the time required for aeration; input water quality index data and water volume data into the empirical formula, and use the empirical formula to calculate the time required for aeration;

[0034] The aeration module is used to calculate the aeration time required according to t if the difference between the aeration time required by the RNN neural network output and the aeration time required by the empirical formula is less than the threshold. 曝气 =(t RNN +t fuzzy) / 2 as the aeration time for aeration; if the difference between the aeration time output by the RNN neural network and the aeration time calculated by the empirical formula is greater than the threshold, aeration is performed according to the aeration time calculated by the empirical formula.

[0035] Furthermore, the sensor module comprises:

[0036] Ammonia nitrogen sensor, used to detect the ammonia nitrogen content in sewage;

[0037] COD sensor, used to detect sewage COD data;

[0038] Activated sludge sensor, used to detect the sludge concentration in the SBR reaction tank;

[0039] Turbidity sensor, used to detect turbidity value;

[0040] Temperature sensor, used to detect the water temperature of the SBR reaction tank;

[0041] Liquid metering pump, used to measure the water volume in SBR reaction tank.

[0042] The method and system for determining the aeration time of a sewage treatment pool provided by the present invention have the following beneficial effects:

[0043] The present invention inputs the data obtained from the SBR reaction tank into the RNN neural network and the empirical formula, and obtains the aeration time by comprehensively comparing the data output by the RNN neural network and the data calculated by the empirical formula. The two synergistically improve the robustness of the system. Compared with the prior art, the aeration amount can be accurately and conveniently controlled. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiment of the present invention and its design scheme, the following briefly introduces the drawings required for this embodiment. The drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 A schematic diagram of an aeration time determination system for a sewage treatment pool according to the present invention;

[0046] Figure 2 It is a structural diagram of the RNN neural network and fuzzy computing intelligent aeration system of the present invention;

[0047] Figure 3 This is the RNN recurrent neural network structure diagram of the present invention;

[0048] Figure 4 This is a schematic diagram of the fuzzy control program for carbon source replenishment and flocculant addition according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to enable those skilled in the art to better understand the technical solution of the present invention and implement it, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the scope of protection of the present invention.

[0050] Example:

[0051] The present invention provides a method and system for determining the aeration time of a sewage treatment pool, specifically Figure 1 As shown, it includes: obtaining water quality index data and water quantity data of the SBR reaction tank; inputting the water quality index data and water quantity data into the trained RNN neural network, and using the RNN neural network to output the aeration time required; inputting the water quality index data and water quantity data into the empirical formula, and using the empirical formula to calculate the aeration time required; if the difference between the aeration time required output by the RNN neural network and the aeration time required calculated by the empirical formula is less than a threshold, according to t 曝气 =(t RNN +t fuzzy ) / 2 as the aeration time for aeration; if the difference between the aeration time output by the RNN neural network and the aeration time calculated by the empirical formula is greater than the threshold, aeration is performed according to the aeration time calculated by the empirical formula.

[0052] The following are implementation details of the present invention:

[0053] 1. Aeration system:

[0054] The aeration system consists of four units: sensor unit, calculation and control unit (including RNN neural network module, fuzzy fuzzy calculation module and signal interaction and control module), aeration unit and bioreactor unit. Figure 1 shown.

[0055] (1) Sensor unit: including ammonia nitrogen sensor, COD sensor, sludge concentration sensor, activated sludge sensor, temperature sensor, liquid level sensor, etc. Figure 1 The main sensor.

[0056] (2) Aeration unit: includes aeration pump, aeration pipe (disc) and gas flow meter.

[0057] (3) Computing and control unit: including information processing module, RNN neural network module, fuzzy fuzzy computing module and signal interaction and control module.

[0058] a) Information processing module: It processes the input signals from sensors such as ammonia nitrogen, COD, sludge concentration MLSS, reaction temperature T, and liquid level through filtering and normalization. The ammonia nitrogen, COD, and sludge concentration are NH, COD, and MLSS respectively, as shown in Program 1.

[0059] RNN neural network calculation module: It inputs the processed signals into the artificial neural network model for calculation to obtain the required aeration time, calculates the start and end times of aeration, and controls the aeration volume through the start and end times. Let the aeration time calculated by the neural network model be t-RNN.

[0060] b) Carbon source fuzzy calculation module: The amount of supplementary carbon source is calculated using the empirical formula:

[0061] t-fuzzy = (COD - 50) / 0.34 * MLSS / 1000 min.

[0062] c) Let m = 丨t-RNN - t-fuzzy丨 / t-RNN; when n ≤ 0.2, aeration is carried out according to t aeration = (t-RNN + t-fuzzy) / 2. If n > 0.2, then aeration is carried out according to t aeration = t-fuzzy.

[0063] e) Signal output and control module: It outputs the start and end times (δt1) of the operation of the aeration pump and the start and end times (δt2) of the agitator, as shown in Program 2.

[0064] (4) The biological reactor unit consists of five parts: equipment such as a stirring device, activated sludge, a feed pump, and a discharge pump.

[0065] II. Program logic diagram:

[0066] (1) Schematic diagram of the operation logic of the RNN and fuzzy models

[0067] Through the sensor unit, water quality index data, water volume and liquid level data, etc. are obtained in real time. The data is transmitted to the calculation and control unit. Through information processing, it enters the RNN neural network model and the Fuzzy fuzzy calculation module for calculation respectively to obtain the calculated aeration volume. Through the RNN network and fuzzy calculation signal interaction interface, identification and judgment are carried out to obtain the aeration volume respectively, and then through the linkage control of the aeration unit, intelligent on-demand aeration is realized.

[0068] When the RNN network model works, first it receives data information, implements task scheduling, trains the RNN trainer, establishes the RNN network model with the trained data, and predicts the aeration volume.

[0069] The fuzzy calculation module works. When the module obtains data, it first performs data screening and averaging, and then checks whether there is already stored aeration process data in the database. If there is, the data is directly called. If not, the data is included in the empirical formula for calculation. After the calculation, reasoning, size comparison, and logical judgment are performed.

[0070] When the RNN network model and the fuzzy calculation are completed, the calculation data are interacted and the best solution is selected. In the first case, if the two values ​​are close and the difference is within 20%, aeration is performed according to t aeration = (t-RNN + t-fuzzy) / 2. In the second case, if the difference between the two values ​​is large and the difference is greater than 20%, the aeration amount t is calculated according to the fuzzy fuzzy calculation. 曝气 =t fuzzy Aeration is performed.

[0071] (2) Schematic diagram of RNN neural network model calculation

[0072] RNN neural network calculation process: by reading the water volume Q (SBR reaction tank level sensor calculates Q), reaction tank temperature T, sewage effluent concentration COD and ammonia nitrogen, reading sludge concentration MLSS, and calculating at cycle interval t through deep recurrent neural network (reading data and calculating once every t minutes), Q3 is continuously performed according to actual needs, and instructions are sent only after the effluent COD, ammonia nitrogen and turbidity meet the standards, and the effluent pump is turned on to drain water.

[0073] (3) Fuzzy control program logic of aeration process:

[0074] Control Flow:

[0075] The first step is water intake: start the water intake pump to let water into the SBR reaction tank. When the SBR liquid level reaches the high water level h1, stop water intake.

[0076] The second step is to read and calculate: the turbidity sensor reads the sludge concentration, the COD sensor reads the COD value, and the formula is the time required for aerobic aeration Taeration = (COD-50) / 0.34*MLSS / 1000min. At the same time, the ammonia nitrogen sensor reads the NH value and calculates the C / N ratio, COD / NH = n.

[0077] The third step is to supply carbon source and aerate: when n≥3, start the aeration pump to aerate the reaction pool; if n<3, calculate the carbon source supply, the calculation formula is: carbon source supply ρ*L=(3-n)*(NH) / 0.8COD, ρ is the COD equivalent per unit volume. After the supply is completed, measure the COD and NH values ​​again. When it is determined that n≥3, start the aeration pump to aerate the reaction pool. If n is still<3, calculate the carbon source supply and supply again until n≥3 is met. Start the aeration pump to aerate the reaction pool. When the aeration time reaches Taeration, stop aeration.

[0078] Step 4: Stirring anaerobic reaction: Anaerobic reaction t3 = 40min. Step 5: Let it stand to determine whether CODNH meets the standard: After the anaerobic reaction is over, let it stand for a while (about 10 minutes), read the sensor value, and determine whether COD≤45 and NH≤14. If both are true, it meets the standard, stop stirring, and start precipitation.

[0079] Step 6, replenishing carbon source and circulating treatment (this step is not necessarily required): If one of the above is no, then cycle to determine and calculate the C / N ratio, and replenish carbon source and aeration reaction and anaerobic reaction again until the standard is met.

[0080] Step 7: Let the mixture stand and settle. Stop stirring and start settling. The settling time reaches t4.

[0081] Step 8. Determine whether ss meets the standard: read the SS sensor data. If the turbidity ss ≤ 9 mg / L, start the drainage pump to drain water to the MBR pool until the liquid level in the SBR reaction pool reaches the minimum level h2, then stop draining.

[0082] Step 9, add flocculant to remove phosphorus: If the turbidity ss>9mg / L, calculate the flocculant dosage, the calculation formula is: empirical formula C = (V*S*C0) / (Q) to calculate the flocculant dosage. After calculation, start the flocculant metering pump, add flocculant, and turn on the stirring pump at the same time to stir and coagulate. After the flocculant addition is completed, continue stirring for 20 minutes, end stirring, and let the sedimentation time reach t4. Read the SS sensor turbidity value again until ss≤9mg / L.

[0083] Step 10: After the turbidity reaches the standard, drain: start the drainage pump until the liquid level in the SBR reaction tank reaches the minimum level h2, then stop draining. After the drainage is completed, read whether the liquid level in the regulating tank has reached the minimum level h3 (at least enough to fill the SBR reaction tank). If so, start the water inlet pump and enter the initial step. If not, idle the equipment until the liquid level in the regulating tank reaches the minimum level h3, and then proceed to the next cycle.

[0084] Procedure 1:

[0085] 1.1 Wavelet denoising procedure:

[0086] importpywt

[0087] import numpy as np

[0088] #Define wavelet denoising function

[0089] defwavelet_denoise(data):

[0090] #Select wavelet function

[0091] wavelet = 'db4'

[0092] #Set threshold selection rules

[0093] mode='soft'

[0094] # Number of decomposition layers

[0095] level=2

[0096] # Decompose the data by wavelet

[0097] coeffs=pywt.wavedec(data,wavelet,level=level)

[0098] #Denoise coefficients according to threshold rules

[0099] threshold=np.sqrt(2*np.log(len(data)))

[0100] denoised_coeffs = []

[0101] for coefficients:

[0102] denoised_coeffs.append(pywt.threshold(coeff,threshold,mode=mode))#Reconstruct the denoised coefficients to obtain the denoised data

[0103] denoised_data=pywt.waverec(denoised_coeffs,wavelet)

[0104] return denoised_data

[0105] #Original data

[0106] data=[[124,228,16.6],

[0107] [96,177,24.3],

[0108] [107,231,20.4]]

[0109] # Perform wavelet denoising on each column of data

[0110] denoised_data=np.zeros((3,3))

[0111] for i in range(3):

[0112] denoised_data[:,i]=wavelet_denoise(data[:,i])

[0113] print("Data after denoising:")

[0114] print(denoised_data)

[0115] 1.2 Data normalization procedure:

[0116] import numpy as np

[0117] #Define the maximum and minimum normalization function

[0118] defmin_max_normalize(data):

[0119] # Calculate the maximum and minimum values ​​of each column

[0120] min_values=np.min(data,axis=0)

[0121] max_values=np.max(data,axis=0)

[0122] #Normalization

[0123] normalized_data=(data-min_values) / (max_values-min_values)returnnormalized_data

[0124] #Original data

[0125] data=[[124,228,16.6],

[0126] [96,177,24.3],

[0127] [107,231,20.4]]

[0128] #Normalization

[0129] normalized_data=min_max_normalize(data)

[0130] print("Normalized data:")

[0131] print(normalized_data)

[0132] Procedure 2:

[0133] ```Python

[0134] import numpy as np

[0135] # Input data

[0136] data=np.array([[3408,124,228,16.6,2.7,10.20,5.1,12.4,3.6,0.07,512 7,2.14],[3877,96,177,24.3,1.9,10.20,2.8,13.9,3.5,0.06,4924,2.09],

[0137] [3884,208,304,18.3,1.8,10.20,3.6,19.6,3.3,0.03,5663,2.22],

[0138] [3810,168,300,22.0,1.4,11.30,3.6,19.2,4.1,0.05,5864,2.21],

[0139] [3192,122,204,20.8,1.8,10.30,3.6,17.8,4.9,0.02,5564,2.13],

[0140] [3977,286,268,17.3,2.6,10.30,6.1,14.9,2.1,0.08,5754,2.07],

[0141] [3417,385,220,17.4,2.5,10.30,1.5,18.8,3.7,0.09,5524,1.92],

[0142] [3126,759,316,24.7,2.9,10.30,1.8,21.9,3.6,0.10,5143,1.86],

[0143] [3116,1180,420,27.2,3.6,10.30,2.8,12.1,7.9,0.01,6592,2.04],

[0144] [3410,138,212,27.3,2.0,11.20,2.3,24.7,4.5,0.03,5894,2.14],

[0145] [3550,246,436,19.3,1.9,11.30,1.5,15.3,3.3,0.04,5427,2.07],

[0146] [3644,241,228,19.7,2.7,12.30,5.9,12.4,4.7,0.10,5576,2.23],

[0147] [3492,51,440,21.1,2.3,10.20,2.2,25.9,3.9,0.10,5806,2.21],

[0148] [3658,93,480,23.4,2.5,10.20,3.5,31.3,3.5,0.12,5864,2.34],

[0149] [3708,86,430,22.7,2.4,10.20,1.9,29.5,3.2,0.13,5478,2.12],

[0150] [3419,79,752,21.3,2.2,10.20,5.5,21.3,3.1,0.10,5547,2.27],

[0151] [3463,83,443,11.2,2.2,11.30,3.5,25.0,2.5,0.10,5964,2.09],

[0152] [2943,71,385,16.4,2.1,10.30,1.5,26.5,3.1,0.09,5857,2.14],

[0153] [3167,69,395,17.4,2.5,10.30,2.5,27.0,3.2,0.20,5741,2.01],

[0154] [3416,85,435,20.9,2.3,10.30,3.1,23.5,3.5,0.25,6014,1.93],

[0155] [3366,153,216,3.5,2.4,10.30,4.8,25.6,0.8,0.20,6328,1.89]])

[0156] #Normalization

[0157] normalized_data=data / np.max(data,axis=0)

[0158] #Define RNN model

[0159] #Define the dimensions of the input layer, hidden layer, and output layer

[0160] input_dim=12

[0161] hidden_dim=16

[0162] output_dim=1

[0163] # Initialize weights

[0164] synapse_0 = 2 * np.random.random ((input_dim, hidden_dim)) - 1 synapse_1 = 2 * np.random.random ((hidden_dim, output_dim)) - 1 # Training

[0165] num_epochs = 50000

[0166] for epoch in range(num_epochs):

[0167] #Previous Item Propagation

[0168] layer_0=normalized_datalayer_1=1 / (1+np.exp(-(np.dot(layer_0,synapse_0))))

[0169] layer_2=np.dot(layer_1,synapse_1)

[0170] # Calculate error

[0171] layer_2_error=layer_2-layer_0

[0172] ifepoch%10000==0:

[0173] print("Error:",np.mean(np.abs(layer_2_error)))

[0174] #Backward Propagation

[0175] layer_2_delta=layer_2_errorlayer_1_error=np.dot(layer_2_delta,synapse_1.T)

[0176] layer_1_delta=layer_1_error*(layer_1*(1-layer_1))

[0177] # Update weights

[0178] synapse_1-=np.dot(layer_1.T,layer_2_delta)

[0179] synapse_0-=np.dot(layer_0.T,layer_1_delta)

[0180] #Get the prediction results

[0181] predicted_output=np.dot(1 / (1+np.exp(-(np.dot(normalized_data,synapse_0)))),synapse_1)

[0182] #Denormalization

[0183] predicted_output*=np.max(data,axis=0)

[0184] # Output prediction results

[0185] print("Predicted output:\n",predicted_output)

[0186] ```

[0187] The embodiments described above are only preferred specific implementation modes of the present invention, and the protection scope of the present invention is not limited thereto. Any simple changes or equivalent replacements of the technical solutions that can be obviously obtained by any technician familiar with the field within the technical scope disclosed in the present invention belong to the protection scope of the present invention.

Claims

1. A method for determining the aeration time of a sewage treatment tank, characterized in that: include: Obtain water quality index data, water volume data, and water temperature data of the sequencing batch activated sludge process SBR reaction tank; Input water quality index data, water volume data, and water temperature data into the trained RNN neural network, and use the RNN neural network to output the time required for aeration; Input water quality index data and water quantity data into the empirical formula, and use the empirical formula to calculate the time required for aeration; If the difference between the aeration time required by the RNN neural network output and the aeration time required by the empirical formula is less than the threshold, the aeration time is calculated according to t 曝气 =(t RNN +t fuzzy ) / 2 as the aeration time for aeration; t 曝气 is the aeration time, t RNN is the time required for aeration of the RNN neural network output, t fuzzy The time required for aeration is calculated using the empirical formula; If the difference between the aeration time required by the RNN neural network output and the aeration time required by the empirical formula is greater than the threshold, aeration is performed according to the aeration time required by the empirical formula; After aeration, calculate the turbidity ss of the water. If the turbidity ss> 9 mg / L, add flocculants to the SBR reactor to purify the sewage, including: Calculate the amount of flocculant added: C=(V*S*C0) / (Q) Where, C is the amount of flocculant added, V is the volume of water, S is the concentration of suspended matter or colloidal matter, C0 is the amount of flocculant required per unit mass of suspended matter; Q is the amount of water; Wherein, the method of calculating the required aeration time using an empirical formula includes: Calculate the ratio of COD to NH: COD / NH=n When n≥3, the aeration time T 曝气 for: T 曝气 =(COD-50) / 0.34*MLSS / 1000 Among them, MLSS is the sludge concentration.

2. The method for determining the aeration time of a sewage treatment pool according to claim 1, characterized in that: Training the RNN neural network includes: The water quality index data, water volume data, water temperature data and the corresponding aeration time are input into the RNN neural network as a training set, and the model parameters of the RNN neural network are adjusted so that the RNN neural network outputs a value equal to the aeration time in the training set, and the RNN neural network training is completed.

3. The method for determining the aeration time of a sewage treatment pool according to claim 2, characterized in that: The water quality index data include: Sewage inlet turbidity, COD, ammonia concentration effluent index, effluent COD content, ammonia nitrogen NH content, turbidity.

4. A method for determining the aeration time of a sewage treatment pool according to claim 3, characterized in that: Also includes: After each aeration, the aeration time, water quality index data, water volume and water temperature data after the water quality meets the standard are used as database data. When the water quality index data, water volume and water temperature data are detected to be the same as the data in the database next time, the aeration time in the database is directly retrieved for aeration.

5. A method for determining the aeration time of a sewage treatment pool according to claim 4, characterized in that: Also includes: The aeration time, water quality index data, water volume and water temperature data after the water quality reaches the standard in the database are used as the training set to train and optimize the RNN neural network.

6. A system for determining aeration time of a sewage treatment tank, characterized in that: include: The sensor module is used to obtain water quality index data, water volume data, and water temperature data of the sequencing batch activated sludge process SBR reaction tank; The calculation unit module is used to input water quality index data, water volume data, and water temperature data into the trained RNN neural network, and use the RNN neural network to output the time required for aeration; Input water quality index data and water quantity data into the empirical formula, and use the empirical formula to calculate the time required for aeration; The aeration module is used to calculate the aeration time required according to t if the difference between the aeration time required by the RNN neural network output and the aeration time required by the empirical formula is less than the threshold. 曝气 =(t RNN +t fuzzy ) / 2 as the aeration time for aeration, t RNN is the time required for aeration of the RNN neural network output, t fuzzy The aeration time required is calculated by the empirical formula. If the difference between the aeration time required by the RNN neural network output and the aeration time required by the empirical formula is greater than the threshold, aeration is performed according to the aeration time required by the empirical formula. After aeration, calculate the turbidity ss of the water. If the turbidity ss> 9 mg / L, add flocculants to the SBR reactor to purify the sewage, including: Calculate the amount of flocculant added: C=(V*S*C0) / (Q) Where, C is the amount of flocculant added, V is the volume of water, S is the concentration of suspended matter or colloidal matter, C0 is the amount of flocculant required per unit mass of suspended matter; Q is the amount of water; Wherein, the method of calculating the required aeration time using an empirical formula includes: Calculate the ratio of COD to NH: COD / NH=n When n≥3, the aeration time T 曝气 for: T 曝气 =(COD-50) / 0.34*MLSS / 1000 Among them, MLSS is the sludge concentration.

7. The aeration time determination system for a sewage treatment pool according to claim 6, characterized in that: The sensor module comprises: Ammonia nitrogen sensor, used to detect the ammonia nitrogen content in sewage; COD sensor, used to detect sewage COD data; Activated sludge sensor, used to detect the sludge concentration in the SBR reaction tank; Turbidity sensor, used to detect turbidity value; Temperature sensor, used to detect the water temperature of the SBR reaction tank; Liquid metering pump, used to measure the water volume in SBR reaction tank.

Images