A sewage system regulation method based on water quality intelligent prediction

Abstract

The application discloses a sewage system regulation and control method based on water quality intelligent prediction, relates to the field of reflux ratio regulation and control in water treatment, and comprises the following steps: determining an auxiliary variable, constructing a prediction model of biochemical oxygen demand of effluent of an anoxic tank or a prediction model of nitrate nitrogen concentration of effluent of the anoxic tank; regulating and controlling the reflux ratio; and performing delay checking. When there is no signal of a water quality instrument, the application can temporarily predict actual effluent water quality data of the anoxic tank, participate in fine regulation and control of the reflux ratio, ensure that the system can continue to operate stably for more than 3 months when the instrument is unstable, and the deviation of a prediction mean value from an actual value is less than 10%. Through prediction of effluent water quality data of the anoxic zone, the application forms an associated logic of the reflux ratio and related water quality data, and then fine regulation and control of the reflux ratio is performed, so that the application avoids that the total nitrogen removal rate is limited due to excessively low reflux ratio or that chemical oxygen demand in the anoxic zone penetrates into the aerobic zone to affect nitrification, and also avoids that excessively high reflux ratio causes too high dissolved oxygen to be carried to affect the anoxic environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of reflux ratio control in water treatment, and in particular to a sewage system control method based on intelligent water quality prediction. BACKGROUND

[0002] Sewage treatment is based on AAO process, and the control of reflux ratio is related to the compliance of effluent TN and the cost of carbon source, which is an important parameter for operation and control. The effluent quality of the anoxic tank is an important basis for adjusting the reflux ratio, which requires additional instruments and probes, increasing the investment in instrument equipment of the sewage plant. With the development of artificial intelligence, soft measurement and other technologies, intelligent prediction is used to predict the effluent quality of the anoxic zone, and the reflux ratio is adjusted in real time, which is an effective means to solve the existing problems. In the prior art, methods have been proposed to solve this problem, but all have certain limitations.

[0003] In the aspect of water quality prediction: CN107664682A discloses a water quality soft measurement prediction method for ammonia nitrogen. The method forms a data sample set by obtaining data values of multiple water quality monitoring indicators including ammonia nitrogen in the water environment to be measured, and further divides it into a training set and a test set. A fuzzy neural network algorithm is used to train the training set to obtain an ammonia nitrogen soft measurement model. The model is tested using the test set, and model establishment and testing are repeatedly performed until the test results meet the preset conditions. The ammonia nitrogen soft measurement model that meets the preset conditions is used as the final soft measurement model.

[0004] CN118298962A discloses a method for predicting influent ammonia nitrogen concentration based on long short-term memory step sequence (LSTM). The method first constructs prediction input data and output data, and pre-processes the data. Then, the LSTM model is used to predict the ammonia nitrogen concentration, the model parameters are set to improve the prediction accuracy, and finally the influent ammonia nitrogen concentration of the sewage plant is simulated and the prediction performance of the model is evaluated.

[0005] CN117711521A discloses a method for predicting ammonia nitrogen content in compost using artificial intelligence model. The method obtains environmental parameters and ammonia nitrogen content information during the composting process of organic solid waste, and inputs these information into an elastic network to select the optimal environmental parameter combination, and then inputs the artificial neural network to predict the ammonia nitrogen content.

[0006] CN118886529A provides a water quality prediction method for sewage treatment plants based on an improved LSTM neural network model. The method first obtains water quality monitoring data of the sewage treatment plant, and performs outlier cleaning, smoothing processing and normalization on the data. Then, the genetic algorithm is used to optimize the LSTM model, and the water quality prediction model is obtained through training and testing.

[0007] The problem with this type of patent is that:

[0008] ①Data volume required is large, investment cost is high. The ammonia nitrogen prediction model completely relies on neural network training, requires massive data support, increases the requirement for operation data; CN107664682A mentions that in the implementation, 24 water quality indexes are selected for modeling through three modeling; CN117711521A mentions that the environmental information to be predicted in the organic solid waste composting process includes days, temperature, total carbon, total nitrogen, compost volume, pH value, moisture content, carbon-nitrogen ratio, seed germination rate, conductivity and many other parameters; the influence factors of ammonia nitrogen concentration in CN118298962A include PH, suspended solids (SS), water temperature (WT), air temperature, air pressure, relative humidity, rainfall, wind speed, average total cloud cover and visibility and many other indexes. Too many indexes required need the support of instrument probes, which further increases the investment cost;

[0009] ②Low data utilization rate. CN107664682A optimizes the model through repeated training and testing, which increases the development cost and may also cause the model training period to be too long, affecting the speed and efficiency of actual application. The excessive data requirement also makes data collection and processing complex, reducing the response speed and flexibility of the system.

[0010] ③Poor prediction accuracy. The input data is completely based on instrument measured data, which cannot exclude the relevance confusion caused by poor instrument measurement accuracy and thus affect the model accuracy. And a single model is difficult to handle complex and variable water quality, especially when facing different types of sewage, the prediction accuracy may decrease.

[0011] ④Lack of feedback mechanism. Most of the existing technologies lack effective feedback mechanism, and cannot self-correct and optimize according to the difference between actual operation data and prediction results. This defect causes the model to gradually lose accuracy in long-term operation, especially when facing sudden water quality changes, it cannot adjust the prediction strategy in time. CN107664682A and CN118298962A have undergone multiple model training and testing, but have not introduced a feedback loop of actual operation data, making it difficult to achieve continuous improvement.

[0012] In terms of reflux ratio regulation: CN115448531B discloses a method for correcting the internal and external reflux ratios of an A2 / O process. First, the external reflux ratio is calculated based on known data of the sewage treatment line, and the initial sampling period is calculated by combining the initially given internal reflux ratio. Then, the sludge in the anoxic tank is sampled at the same interval within the sampling period to obtain the variation of the phosphate concentration of the sludge in the anoxic tank within the sampling period. Meanwhile, the phosphate concentrations of the effluent of the anaerobic tank, the internal reflux sewage, and the effluent of the anoxic tank are detected respectively to calculate the actual internal reflux ratio. The actual internal reflux ratio is used to correct the sampling period again, and the sampling and detection are performed again to further correct the internal reflux ratio, and the iteration continues until the variation coefficient of the internal reflux ratio measured by the two samplings is less than 0.1, and the correction of the internal reflux ratio is completed. The patent has the following defects: first, the method relies on actual sampling and detection data, and the correction process is time-consuming, which is difficult to meet the real-time regulation requirements; second, the sampling and detection process may be affected by various factors, which limits the accuracy of the correction result; third, the method lacks effective response strategies when the instrument fails or the data is abnormal, and the system stability and robustness are insufficient.

[0013] In summary, although the existing water quality concentration prediction methods have achieved monitoring and prediction functions to some extent, there are still problems such as large data demand, low data utilization rate, poor prediction accuracy, and lack of feedback mechanism in actual application. In terms of reflux ratio regulation, the method relies on actual sampling and detection data, and the correction process is time-consuming, and may be affected by various factors, which limits the accuracy of the correction result. Therefore, the next improvement direction should focus on developing an intelligent regulation system that can accurately predict the effluent water quality of the anoxic tank, adjust the reflux ratio in real time, and has high stability and robustness. Such a system should be able to make full use of existing data, improve data utilization rate, reduce investment cost, and quickly adjust prediction and regulation strategies when facing sudden water quality changes to ensure stable and standard effluent water quality and effective control of operating costs. SUMMARY

[0014] In view of the problems of the prior art, the present application provides a sewage system regulation method based on water quality intelligent prediction.

[0015] The present application adopts the following technical solutions:

[0016] A sewage system regulation method based on water quality intelligent prediction, comprising the following steps:

[0017] Step 1: Determine the auxiliary variable; if the biochemical oxygen demand instrument of the anoxic tank effluent has no signal, then determine the first auxiliary variable as the nitrate nitrogen concentration of the anoxic tank effluent, the reflux ratio of the anoxic tank, the oxidation-reduction potential, the pH value, and the water temperature, and obtain the historical data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable;

[0018] If the effluent nitrate nitrogen instrument of the anoxic tank has no signal, the second auxiliary variable is determined as the biochemical oxygen demand of the effluent of the anoxic tank, the reflux ratio of the anoxic tank, the oxidation-reduction potential, the pH value and the water temperature, and the historical data of the effluent nitrate nitrogen concentration of the anoxic tank and the second auxiliary variable are obtained;

[0019] Step 2: Model construction; a function relationship between the effluent biochemical oxygen demand of the anoxic tank and the first auxiliary variable is constructed according to the historical data of the effluent biochemical oxygen demand of the anoxic tank and the first auxiliary variable, forming an effluent biochemical oxygen demand prediction model of the anoxic tank; the first auxiliary variable at the time to be predicted is input into the effluent biochemical oxygen demand prediction model of the anoxic tank, and the effluent biochemical oxygen demand of the anoxic tank at the time to be predicted is obtained as the prediction data;

[0020] A function relationship between the effluent nitrate nitrogen concentration of the anoxic tank and the second auxiliary variable is constructed according to the historical data of the effluent nitrate nitrogen concentration of the anoxic tank and the second auxiliary variable, forming an effluent nitrate nitrogen concentration prediction model of the anoxic tank; the second auxiliary variable at the time to be predicted is input into the effluent nitrate nitrogen concentration prediction model of the anoxic tank, and the effluent nitrate nitrogen concentration value of the anoxic tank at the time to be predicted is obtained as the prediction data;

[0021] Step 3: Reflux ratio control; if the effluent nitrate nitrogen concentration value of the anoxic tank is <1 mg / L and the effluent biochemical oxygen demand of the anoxic tank is <50 mg / L, the reflux ratio is maintained unchanged;

[0022] If the effluent nitrate nitrogen concentration value of the anoxic tank is <1 mg / L and the effluent biochemical oxygen demand of the anoxic tank is ≥50 mg / L, the reflux ratio is increased by 10%;

[0023] If the effluent nitrate nitrogen concentration value of the anoxic tank is ≥1 mg / L, the reflux ratio is decreased by 10%;

[0024] Step 4: Delay check; after the prediction data is run, 25%-50% of the actual hydraulic retention time of the anoxic tank is taken as the check interval, the value of the prediction data corresponding to 25% of the actual hydraulic retention time of the anoxic tank is taken as C1, and the value of the prediction data corresponding to 50% of the actual hydraulic retention time of the anoxic tank is taken as C2;

[0025] When the prediction data is the effluent nitrate nitrogen concentration value of the anoxic tank, if |C1-C2|≤10% C1, the effluent nitrate nitrogen concentration value of the anoxic tank is unchanged; when C2-C1>10% C1, the compensation coefficient k is set to 1.2, and the final prediction value is 1.2*the effluent nitrate nitrogen concentration value of the anoxic tank; when C1-C2>10% C1, the compensation coefficient k is set to 0.8, and the final prediction value is 0.8*the effluent nitrate nitrogen concentration value of the anoxic tank;

[0026] When the predicted data is the biochemical oxygen demand of the anoxic tank effluent, if |C1-C2|≤10%C1, the biochemical oxygen demand of the anoxic tank effluent is unchanged; when C2-C1>10%C1, the compensation coefficient k is set to 0.8, and the final predicted value is 0.8*the biochemical oxygen demand of the anoxic tank effluent; when C1-C2>10%C1, the compensation coefficient k is set to 1.2, and the final predicted value is 1.2*the biochemical oxygen demand of the anoxic tank effluent.

[0027] Preferably, the functional relationship between the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable is constructed by a multiple form regression algorithm, and the functional relationship between the nitrate nitrogen concentration of the anoxic tank effluent and the second auxiliary variable is constructed by a multiple form regression algorithm.

[0028] The present application has the beneficial effects of:

[0029] 1) System stability. When there is no signal from the water quality instrument, the actual anoxic tank effluent water quality data can be temporarily predicted to participate in fine control of the reflux ratio, ensuring that the system can continue to operate stably for more than 3 months when the instrument is unstable, and the deviation of the predicted mean value from the actual value is less than 10%.

[0030] 2) High effluent water quality. Through prediction of the anoxic zone effluent water quality data, the correlation logic between the reflux ratio and related water quality data is formed, and the reflux ratio is fine-tuned to avoid the situation that the total nitrogen removal rate is limited due to too low reflux ratio or the chemical oxygen demand in the anoxic zone penetrates into the aerobic zone to affect nitrification, and also avoid the situation that too high reflux ratio leads to too high dissolved oxygen carried to affect the anoxic environment. The total nitrogen in the effluent can be stably below 5 mg / L, meeting the current optimal TN discharge standard.

[0031] 3) Saving of drug consumption. Through prediction of the anoxic zone effluent water quality data, combined with reflux ratio adjustment, the utilization of raw water carbon source is strengthened, and the amount of external carbon source is reduced by more than 50%, achieving reduction of drug consumption. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 The figure is the total nitrogen in and out of the water of Example 1. DETAILED DESCRIPTION

[0033] The specific embodiments of the present application will be further described below in combination with the drawings and specific examples:

[0034] Example 1

[0035] In combination with Figure 1 , a certain sewage treatment plant, the total nitrogen standard of the effluent is 5 mg / L, and the biochemical oxygen demand instrument of the anoxic tank effluent at the predicted time is out of signal.

[0036] Step 1: Using the nitrate and nitrogen concentration in the effluent from the anoxic tank, the reflux ratio in the anoxic tank, the oxidation-reduction potential, the pH value, and the water temperature as the first auxiliary variables, obtain 60 days of historical data on the biochemical oxygen demand in the effluent from the anoxic tank and the first auxiliary variables.

[0037] Step 2: Based on 60 days of historical data on the biochemical oxygen demand (BOD) of the anoxic tank effluent and the first auxiliary variable, a multi-form regression algorithm is used to construct a functional relationship between the BOD of the anoxic tank effluent and the first auxiliary variable, forming a prediction model for the BOD of the anoxic tank effluent. The first auxiliary variable at the time to be predicted is input into the prediction model for the BOD of the anoxic tank effluent at the time to be predicted, and the BOD of the anoxic tank effluent at the time to be predicted is obtained as the prediction data.

[0038] Step 3: If the nitrate concentration in the effluent from the anoxic tank is <1 mg / L and the biochemical oxygen demand in the effluent from the anoxic tank at the predicted time is <50 mg / L, then maintain the reflux ratio unchanged.

[0039] If the nitrate concentration in the effluent from the anoxic tank is <1 mg / L, and the biochemical oxygen demand in the effluent from the anoxic tank at the predicted time is ≥50 mg / L, then the reflux ratio should be increased by 10%.

[0040] If the nitrate concentration in the effluent from the anoxic tank is ≥1 mg / L, then the reflux ratio should be reduced by 10%.

[0041] Step 4: After running the predicted data, use 25%-50% of the actual hydraulic retention time in the anoxic tank as the verification range. The value of biochemical oxygen demand (BOD) in the effluent of the anoxic tank corresponding to 25% of the actual hydraulic retention time is C1, and the value of biochemical oxygen demand (BOD) in the effluent of the anoxic tank corresponding to 50% of the actual hydraulic retention time is C2.

[0042] If |C1-C2|≤10%C1, the biochemical oxygen demand (BOD) of the effluent from the anoxic tank remains unchanged; if C2-C1>10%C1, the compensation coefficient k is set to 0.8, and the final predicted value is 0.8 * BOD of the effluent from the anoxic tank; if C1-C2>10%C1, the compensation coefficient k is set to 1.2, and the final predicted value is 1.2 * BOD of the effluent from the anoxic tank.

[0043] like Figure 1 As shown, the water plant's operating data after more than 400 days of operation showed that the total nitrogen in the effluent using the method of this invention was stable at 3.80±0.81 mg / L, achieving a compliance rate of 100%, and the calculated C / N ratio for nitrogen removal was 4.46±0.78.

[0044] Example 2

[0045] At a wastewater treatment plant, the total nitrogen standard for effluent is 15 mg / L. At the time of prediction, the nitrate nitrogen meter in the anoxic tank of the biological treatment section of the plant showed no signal.

[0046] Step 1: The biochemical oxygen demand of the effluent of the anoxic tank, the reflux ratio of the anoxic tank, the oxidation-reduction potential, the pH value and the water temperature are taken as the second auxiliary variables, and the 60-day historical data of the nitrate nitrogen concentration of the effluent of the anoxic tank and the second auxiliary variables are obtained.

[0047] Step 2: According to the 60-day historical data of the nitrate nitrogen concentration of the effluent of the anoxic tank and the second auxiliary variables, the functional relationship between the nitrate nitrogen concentration of the effluent of the anoxic tank and the second auxiliary variables is formed by a multi-form regression algorithm to form a nitrate nitrogen concentration prediction model of the effluent of the anoxic tank. The second auxiliary variable at the time to be predicted is input into the nitrate nitrogen concentration prediction model of the effluent of the anoxic tank, and the nitrate nitrogen concentration value of the effluent of the anoxic tank at the time to be predicted is obtained as the prediction data.

[0048] Step 3: If the nitrate nitrogen concentration value of the effluent of the anoxic tank at the time to be predicted is less than 1 mg / L, and the biochemical oxygen demand of the effluent of the anoxic tank is less than 50 mg / L, the reflux ratio is maintained unchanged.

[0049] If the nitrate nitrogen concentration value of the effluent of the anoxic tank at the time to be predicted is less than 1 mg / L, and the biochemical oxygen demand of the effluent of the anoxic tank is greater than or equal to 50 mg / L, the reflux ratio is increased by 10%.

[0050] If the nitrate nitrogen concentration value of the effluent of the anoxic tank at the time to be predicted is greater than or equal to 1 mg / L, the reflux ratio is decreased by 10%.

[0051] Step 4: After the prediction data is run, 25%-50% of the actual hydraulic retention time of the anoxic tank is taken as the checking interval, the value of the nitrate nitrogen concentration of the effluent of the anoxic tank corresponding to 25% of the actual hydraulic retention time of the anoxic tank is taken as C1, and the value of the nitrate nitrogen concentration of the effluent of the anoxic tank corresponding to 50% of the actual hydraulic retention time of the anoxic tank is taken as C2.

[0052] If |C1-C2|≤10% C1, the nitrate nitrogen concentration value of the effluent of the anoxic tank is unchanged; when C2-C1>10% C1, the compensation coefficient k is set to 1.2, and the final prediction value is 1.2*the nitrate nitrogen concentration value of the effluent of the anoxic tank; when C1-C2>10% C1, the compensation coefficient k is set to 0.8, and the final prediction value is 0.8*the nitrate nitrogen concentration value of the effluent of the anoxic tank.

[0053] The 200-day operation data of the water plant show that the total nitrogen of the effluent is 12.69±0.85 mg / L, the compliance rate is 100%, and the required C / N for denitrification is 4.89±1.26.

[0054] Comparative Example 1

[0055] A sewage pilot system, the total nitrogen standard of the effluent is 15 mg / L, and the biochemical oxygen demand instrument of the anoxic tank of the pilot biochemical section has no signal.

[0056] Step 1: The nitrate concentration of the anoxic tank effluent, the anoxic tank reflux ratio, the oxidation-reduction potential, the pH value and the water temperature are used as the first auxiliary variable, and the 60-day history data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable are obtained.

[0057] Step 2: According to the 60-day history data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable, a function relationship between the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable is constructed by a multi-form regression algorithm to form a biochemical oxygen demand prediction model of the anoxic tank effluent. The first auxiliary variable at the time to be predicted is input into the biochemical oxygen demand prediction model of the anoxic tank effluent to obtain the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted as the prediction data.

[0058] Step 3: If the nitrate concentration of the anoxic tank effluent is < 2 mg / L and the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted is < 50 mg / L, the reflux ratio is maintained unchanged.

[0059] If the nitrate concentration of the anoxic tank effluent is < 2 mg / L and the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted is ≥ 50 mg / L, the reflux ratio is increased by 10%.

[0060] If the nitrate concentration of the anoxic tank effluent is ≥ 2 mg / L, the reflux ratio is decreased by 10%.

[0061] Step 4: After the prediction data is run, 25%-50% of the actual hydraulic retention time of the anoxic tank is used as the checking interval, the value of the biochemical oxygen demand of the anoxic tank effluent corresponding to 25% of the actual hydraulic retention time of the anoxic tank is C1, and the value of the biochemical oxygen demand of the anoxic tank effluent corresponding to 50% of the actual hydraulic retention time of the anoxic tank is C2.

[0062] If |C1-C2|≤10% C1, the biochemical oxygen demand of the anoxic tank effluent is unchanged; when C2-C1>10% C1, the compensation coefficient k is set to 0.8, and the final prediction value is 0.8*the biochemical oxygen demand of the anoxic tank effluent; when C1-C2>10% C1, the compensation coefficient k is set to 1.2, and the final prediction value is 1.2*the biochemical oxygen demand of the anoxic tank effluent.

[0063] The 200-day running data of the pilot test show that the application of the method of Comparative Example 1 has an effluent TN of 12.59±0.59 mg / L, a compliance rate of 94.41%, and a required C / N for denitrification of 5.23±1.27.

[0064] Comparative Example 2

[0065] The wastewater pilot test system in Comparative Example 1 has an effluent total nitrogen standard of 15 mg / L, and the biochemical oxygen demand instrument of the anoxic tank in the biochemical section of the pilot test has no signal.

[0066] Step 1: The nitrate concentration of the anoxic tank effluent, the anoxic tank reflux ratio, the oxidation-reduction potential, the pH value and the water temperature are used as the first auxiliary variable, and the 60-day history data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable are obtained.

[0067] Step 2: According to the 60-day history data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable, a function relationship between the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable is constructed by a multi-form regression algorithm to form a biochemical oxygen demand prediction model of the anoxic tank effluent. The first auxiliary variable at the time to be predicted is input into the biochemical oxygen demand prediction model of the anoxic tank effluent to obtain the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted as the prediction data.

[0068] Step 3: If the nitrate concentration of the anoxic tank effluent is <1 mg / L and the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted is <60 mg / L, the reflux ratio is maintained unchanged.

[0069] If the nitrate concentration of the anoxic tank effluent is <1 mg / L and the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted is ≥60 mg / L, the reflux ratio is increased by 10%.

[0070] If the nitrate concentration of the anoxic tank effluent is ≥1 mg / L, the reflux ratio is decreased by 10%.

[0071] Step 4: After the prediction data is run, 25%-50% of the actual hydraulic retention time of the anoxic tank is used as the checking interval, the value of the biochemical oxygen demand of the anoxic tank effluent corresponding to 25% of the actual hydraulic retention time of the anoxic tank is C1, and the value of the biochemical oxygen demand of the anoxic tank effluent corresponding to 50% of the actual hydraulic retention time of the anoxic tank is C2.

[0072] If |C1-C2|≤10% C1, the biochemical oxygen demand of the anoxic tank effluent is unchanged; when C2-C1>10% C1, the compensation coefficient k is set to 0.8, and the final prediction value is 0.8*the biochemical oxygen demand of the anoxic tank effluent; when C1-C2>10% C1, the compensation coefficient k is set to 1.2, and the final prediction value is 1.2*the biochemical oxygen demand of the anoxic tank effluent.

[0073] The 200-day running data of the pilot test show that the application of the method has an effluent TN of 12.09±0.46 mg / L, a compliance rate of 100%, but the required C / N for denitrification is 5.51±0.72.

[0074] Comparative Example 3

[0075] The total nitrogen standard of the effluent of the sewage pilot test system in Comparative Example 1 is 15 mg / L, and the biochemical oxygen demand instrument of the anoxic tank of the biochemical section of the pilot test has no signal.

[0076] Step 1: The nitrate concentration of the anoxic tank effluent, the anoxic tank reflux ratio, the oxidation-reduction potential, the pH value and the water temperature are used as the first auxiliary variable, and the 60-day history data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable are obtained.

[0077] Step 2: According to the 60-day history data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable, a function relationship between the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable is constructed by a multi-form regression algorithm to form a biochemical oxygen demand prediction model of the anoxic tank effluent. The first auxiliary variable at the time to be predicted is input into the biochemical oxygen demand prediction model of the anoxic tank effluent to obtain the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted as the prediction data.

[0078] Step 3: If the nitrate concentration of the anoxic tank effluent is <1 mg / L and the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted is <50 mg / L, the reflux ratio is maintained unchanged.

[0079] If the nitrate concentration of the anoxic tank effluent is <1 mg / L and the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted is ≥50 mg / L, the reflux ratio is increased by 10%.

[0080] If the nitrate concentration of the anoxic tank effluent is ≥1 mg / L, the reflux ratio is decreased by 10%.

[0081] Step 4: After the prediction data is run, 10%-25% of the actual hydraulic retention time of the anoxic tank is used as the checking interval, the value of the biochemical oxygen demand of the anoxic tank effluent corresponding to 10% of the actual hydraulic retention time of the anoxic tank is C1, and the value of the biochemical oxygen demand of the anoxic tank effluent corresponding to 25% of the actual hydraulic retention time of the anoxic tank is C2.

[0082] If |C1-C2|≤10% C1, the biochemical oxygen demand of the anoxic tank effluent is unchanged; when C2-C1>10% C1, the compensation coefficient k is set to 0.8, and the final prediction value is 0.8*the biochemical oxygen demand of the anoxic tank effluent; when C1-C2>10% C1, the compensation coefficient k is set to 1.2, and the final prediction value is 1.2*the biochemical oxygen demand of the anoxic tank effluent.

[0083] The 200-day running data of the pilot test show that the application of the method has an effluent TN of 11.55±0.54 mg / L, a compliance rate of 100%, but a required C / N for denitrification of 5.34±0.74.

[0084] Comparative Example 4

[0085] The total nitrogen standard of the effluent of the sewage pilot test system in Comparative Example 1 is 15 mg / L, and the biochemical oxygen demand instrument of the anoxic tank of the biochemical section of the pilot test has no signal.

[0086] Step 1: The nitrate concentration of the anoxic tank effluent, the anoxic tank reflux ratio, the oxidation-reduction potential, the pH value and the water temperature are taken as the first auxiliary variable, and the 60-day history data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable are obtained.

[0087] Step 2: According to the 60-day history data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable, a function relationship between the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable is constructed by a multi-form regression algorithm to form a biochemical oxygen demand prediction model of the anoxic tank effluent. The first auxiliary variable at the time to be predicted is input into the biochemical oxygen demand prediction model of the anoxic tank effluent, and the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted is obtained as the prediction data.

[0088] Step 3: If the nitrate concentration of the anoxic tank effluent is less than 1 mg / L, and the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted is less than 50 mg / L, the reflux ratio is maintained unchanged.

[0089] If the nitrate concentration of the anoxic tank effluent is less than 1 mg / L, and the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted is greater than or equal to 50 mg / L, the reflux ratio is increased by 10%.

[0090] If the nitrate concentration of the anoxic tank effluent is greater than or equal to 1 mg / L, the reflux ratio is decreased by 10%.

[0091] Step 4: After the prediction data is run, 50%-75% of the actual hydraulic retention time of the anoxic tank is taken as the checking interval, the value of the biochemical oxygen demand of the anoxic tank effluent corresponding to 50% of the actual hydraulic retention time of the anoxic tank is taken as C1, and the value of the biochemical oxygen demand of the anoxic tank effluent corresponding to 75% of the actual hydraulic retention time of the anoxic tank is taken as C2.

[0092] If |C1-C2|≤10% C1, the biochemical oxygen demand of the anoxic tank effluent is unchanged; when C2-C1>10% C1, the compensation coefficient k is set to 0.8, and the final prediction value is 0.8*the biochemical oxygen demand of the anoxic tank effluent; when C1-C2>10% C1, the compensation coefficient k is set to 1.2, and the final prediction value is 1.2*the biochemical oxygen demand of the anoxic tank effluent.

[0093] The 200-day running data of the pilot test show that the application of the method has an effluent TN of 12.07±1.18 mg / L, a compliance rate of 91.26%, and a required C / N for denitrification of 4.66±1.67.

[0094] Of course, the above description is not a limitation on the present application, and the present application is not limited to the above examples. Changes, modifications, additions or substitutions made by those skilled in the art within the essential scope of the present application should also be within the protection scope of the present application.

Claims

1. A sewage system regulation method based on water quality intelligent prediction, characterized in that, The method comprises the following steps: Step 1: determining auxiliary variables; If there is no signal of the biochemical oxygen demand instrument of the anoxic tank effluent, the first auxiliary variable is determined as the nitrate nitrogen concentration of the anoxic tank effluent, the reflux ratio of the anoxic tank, the oxidation-reduction potential, the pH value and the water temperature, the historical data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable are obtained; If there is no signal of the nitrate nitrogen instrument of the anoxic tank effluent, the second auxiliary variable is determined as the biochemical oxygen demand of the anoxic tank effluent, the reflux ratio of the anoxic tank, the oxidation-reduction potential, the pH value and the water temperature, the historical data of the nitrate nitrogen concentration of the anoxic tank effluent and the second auxiliary variable are obtained; Step 2: constructing a model; According to the historical data of the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable, a functional relationship between the biochemical oxygen demand of the anoxic tank effluent and the first auxiliary variable is constructed, and a biochemical oxygen demand prediction model of the anoxic tank effluent is formed; the first auxiliary variable at the time to be predicted is input into the biochemical oxygen demand prediction model of the anoxic tank effluent, and the biochemical oxygen demand of the anoxic tank effluent at the time to be predicted is obtained as the prediction data; According to the historical data of the nitrate nitrogen concentration of the anoxic tank effluent and the second auxiliary variable, a functional relationship between the nitrate nitrogen concentration of the anoxic tank effluent and the second auxiliary variable is constructed, and a nitrate nitrogen concentration prediction model of the anoxic tank effluent is formed; the second auxiliary variable at the time to be predicted is input into the nitrate nitrogen concentration prediction model of the anoxic tank effluent, and the nitrate nitrogen concentration value of the anoxic tank effluent at the time to be predicted is obtained as the prediction data; Step 3: reflux ratio control; if the nitrate nitrogen concentration value of the anoxic tank effluent is less than 1 mg / L and the biochemical oxygen demand of the anoxic tank effluent is less than 50 mg / L, the reflux ratio is maintained unchanged; If the nitrate nitrogen concentration value of the anoxic tank effluent is less than 1 mg / L and the biochemical oxygen demand of the anoxic tank effluent is greater than or equal to 50 mg / L, the reflux ratio is increased by 10%; If the nitrate nitrogen concentration value of the anoxic tank effluent is greater than or equal to 1 mg / L, the reflux ratio is decreased by 10%; Step 4: delay check; after the prediction data is run, 25%-50% of the actual hydraulic retention time of the anoxic tank is taken as the check interval, the value of the prediction data corresponding to 25% of the actual hydraulic retention time of the anoxic tank is taken as C1, and the value of the prediction data corresponding to 50% of the actual hydraulic retention time of the anoxic tank is taken as C2; When the prediction data is the nitrate nitrogen concentration value of the anoxic tank effluent, if |C1-C2|≤10% C1, the nitrate nitrogen concentration value of the anoxic tank effluent is unchanged; when C2-C1>10% C1, the compensation coefficient k is set to 1.2, and the final prediction value is 1.2*nitrate nitrogen concentration value of the anoxic tank effluent; when C1-C2>10% C1, the compensation coefficient k is set to 0.8, and the final prediction value is 0.8*nitrate nitrogen concentration value of the anoxic tank effluent; When the prediction data is the biochemical oxygen demand of the anoxic tank effluent, if |C1-C2|≤10% C1, the biochemical oxygen demand of the anoxic tank effluent is unchanged; when C2-C1>10% C1, the compensation coefficient k is set to 0.8, and the final prediction value is 0.8*biochemical oxygen demand of the anoxic tank effluent; when C1-C2>10% C1, the compensation coefficient k is set to 1.2, and the final prediction value is 1.2*biochemical oxygen demand of the anoxic tank effluent. 2.The sewage system regulation method based on water quality intelligent prediction of claim 1, wherein, The function relationship between the biochemical oxygen demand of the effluent of the anoxic tank and the first auxiliary variable is constructed by a multiple form regression algorithm, and the function relationship between the nitrate nitrogen concentration of the effluent of the anoxic tank and the second auxiliary variable is constructed by a multiple form regression algorithm.

Images