Method for predicting discharge level of effluent from decentralized sewage treatment facilities

Abstract

A method for predicting a discharge level of an effluent from decentralized sewage treatment facilities, the method including: measuring the conductivity of an influent, the conductivity and suspended solids concentration of an effluent of a plurality of decentralized sewage treatment facilities; repeatedly measuring a pH, a concentration of COD, a concentration of ammonia nitrogen, the concentration of total phosphorus of the effluent of each of the plurality of decentralized sewage treatment facilities; calculating average values of the pH, the concentration of COD, the concentration of ammonia nitrogen, the concentration of total phosphorus; comparing the average values with a local sewage discharge standard, and determining a discharge level of the effluent; constructing a predictive model; and sampling an influent and an effluent of a sewage treatment facility, measuring the conductivity of an influent, the conductivity and suspended solids concentration of the effluent, inputting the obtained data to the predictive model.

Claims

1. A method for treating sewage and discharging treated sewage into the environment, comprising: 1) proving a first plurality of decentralized sewage treatment facilities, treating sewage using the first plurality of decentralized sewage treatment facilities; 2) randomly selecting a second plurality of decentralized sewage treatment facilities from the first plurality of decentralized sewage treatment facilities as a training dataset; measuring a conductivity of an influent, a conductivity and suspended solids concentration of an effluent of each of the second plurality of decentralized sewage treatment facilities; repeatedly measuring a pH, a concentration of COD, a concentration of ammonia nitrogen (NH.sub.3—N), and a concentration of total phosphorus of the effluent from a discharge outlet of each of the second plurality of decentralized sewage treatment facilities; calculating average values of the pH, the concentration of COD, the concentration of ammonia nitrogen, and the concentration of total phosphorus; comparing the average values of the pH, the concentration of COD, the concentration of ammonia nitrogen, and the concentration of total phosphorus with a local sewage discharge standard, thereby determining a discharge level of the effluent of each of the second plurality of decentralized sewage treatment facilities, wherein the discharge level is a number, and the number is a first preset number when all of the average values of the pH, the concentration of COD, the concentration of ammonia nitrogen, and the concentration of total phosphorus meet the local sewage discharge standard, and the number is a second preset number when one or more of the average values of the pH, the concentration of COD, the concentration of ammonia nitrogen, and the concentration of total phosphorus do not meet the local sewage discharge standard; 3) inputting the conductivity of the influent, the conductivity and suspended solids concentration of the effluent of each of the second plurality of decentralized sewage treatment facilities to a support vector machine, employing the discharge level of the effluent of each of the second plurality of decentralized sewage treatment facilities as an output value, to train the training dataset, thereby constructing a predictive model to predict a discharge level of an effluent from a sewage treatment facility sample; and 4) sampling an influent and an effluent of a sewage treatment facility from the first plurality of decentralized sewage treatment facilities, measuring a conductivity of the influent of the sewage treatment facility, a conductivity and suspended solids concentration of the effluent, inputting the conductivity of the influent, the conductivity and the suspended solids concentration of the effluent into the predictive model obtained in 2), thereby obtaining a predictive result of a discharge level of the effluent of the sewage treatment facility, and discharging the effluent of the sewage treatment facility into the environment when the predictive result of the discharge level of the effluent of the sewage treatment facility is the first preset number.

2. The method of claim 1, wherein the first plurality of decentralized sewage treatment facilities is an anaerobic/anoxic/oxic (A.sup.2O) treatment system, a constructed wetland system, a sequencing batch reactor (SBR) treatment system, a biological aerated filter (BAF) system, or a combination thereof.

3. The method of claim 1, wherein in 1), each of the first plurality of decentralized sewage treatment facilities comprises a regulating pool provided with a lifting pump; and measuring the conductivity of the influent, the conductivity and suspended solids concentration of the effluent of each of the second plurality of decentralized sewage treatment facilities comprising: starting the lifting pump; 15 mins later, synchronously measuring the conductivity of the influent in the regulating pool and the conductivity and suspended solids concentration of the effluent in the discharge outlet of each of the second plurality of decentralized sewage treatment facilities.

4. The method of claim 1, wherein in 2), prior to inputting data to the support vector machine, the method comprises inputting the conductivity of the influent, the conductivity and suspended solids concentration of the effluent of each of the second plurality of decentralized sewage treatment facilities to a mapminmax function: y=(x−x.sub.min)/(x.sub.max−x.sub.min), where y refers to a normalized measured data of the conductivity of the influent, the conductivity of the effluent, or the suspended solids concentration of the effluent; x refers to a real-time measured data of the conductivity of the influent, the conductivity of the effluent, or the suspended solids concentration of the effluent; x.sub.min is a minimum value of x, and x.sub.max is a maximum value of x; when the discharge level of the effluent satisfies the local sewage discharge standard, the discharge level is recorded as 1; when the discharge level of the effluent fails to satisfy the local sewage discharge standard, the discharge level is recorded as −1.

5. The method of claim 1, wherein in 3), the method further comprises preliminarily predicting the discharge level of the effluent of the sewage treatment facility after measuring the conductivity of an influent, the conductivity and suspended solids concentration of the effluent; the preliminary predicting is implemented as follows: i) when the suspended solids concentration of the effluent is greater than a standard value, showing the discharge level of the effluent fails to satisfy the local sewage discharge standard; and ii) when the suspended solids concentration of the effluent is less than or equal to a standard value, inputting the conductivity of the influent, the conductivity and the suspended solids concentration of the effluent into the predictive model obtained in 2).

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a flow chart of a method for predicting a discharge level of an effluent from a sewage treatment facility according to one embodiment of the disclosure;

(2) FIG. 2 shows a rough selection result of the optimal penalty parameter c and the optimal kernel parameter g according to one embodiment of the disclosure;

(3) FIG. 3 shows a fine selection result of the optimal penalty parameter c and the optimal kernel parameter g according to one embodiment of the disclosure;

(4) FIG. 4 is a comparison chart of predicted results and actual results according to one embodiment of the disclosure; and

(5) FIG. 5 is a comparison chart of predicted results and actual results according to another embodiment of the disclosure.

DETAILED DESCRIPTION OF THE EMBODIMENTS

(6) To further illustrate the invention, embodiments detailing a method for predicting a discharge level of an effluent from a sewage treatment facility are described below. It should be noted that the following embodiments are intended to describe and not to limit the disclosure.

Example 1

(7) A method for predicting a discharge level of an effluent from a sewage treatment facility is described as follows.

(8) (1) 164 decentralized sewage treatment facilities were selected in rural areas of the Yangtze River Delta, including an anaerobic/anoxic/oxic (A.sup.2O) treatment system, a constructed wetland system, a sequencing batch reactor (SBR) treatment system and a biological aerated filter (BAF) system, with a treatment scale of 5-160 t/d. The systems each comprised a regulating pool provided with a lifting pump, and a sewage treatment device provided with a discharge outlet at the water outlet thereof. The rural sewage was from a septic tank, kitchen sinks and laundry facilities. The main water pollutant parameter in the rural sewage were COD, total nitrogen, ammonia nitrogen, total phosphorus and suspended solids.

(9) First, the conductivity of the influent, and the conductivity and suspended solids concentration of the effluent of each of the sewage treatment facilities were measured. Specifically, after 15 min on starting up the lifting pump, the influent from the regulating pool and the effluent from the discharge outlet were sampled and the conductivity of the influent, the conductivity and suspended solids concentration of the effluent were measured. 15 min and 30 min later, the sampling and measurement process were repeated twice. The average values of the conductivity of the influent, and the conductivity and suspended solids concentration of the effluent were calculated.

(10) Referring to Class II in the standard of “Discharge Standard of Water Pollutants for Sewage treatment Facilities” (DB 33/973-2015) implemented in Zhejiang Province of China, the monitored water quality indicators included pH, COD, ammonia nitrogen, total phosphorus (viewed as phosphorus (P)) and suspended solids (SS), and the corresponding numeric criteria were 6-9, 100 mg/L, 25 mg/L, 3 mg/L, and 30 mg/L, respectively. The concentrations of the water pollutants in the discharge outlet were measured thrice at intervals. The average value of the concentration of each water pollutants were calculated and compared with the local sewage discharge standard, thereby showing the discharge level of the effluent from the sewage treatment facilities. When the discharge level of the effluent satisfied the local sewage discharge level, the discharge level was recorded as 1; when the discharge level of the effluent failed to satisfy the local sewage discharge standard, the discharge level was recorded as −1.

(11) (2). The conductivity of the influent, and the conductivity and suspended solids concentration of the effluent of each of the plurality of decentralized sewage treatment facilities were input to a support vector machine, and the discharge level of the effluent of each of the plurality of decentralized sewage treatment facilities was employed as an output value. 154 sets of data were randomly selected from 164 sets of data as a training dataset, and the training dataset was trained using the support vector machine, thereby constructing a predictive model to predict the discharge level of an effluent from each of the plurality of decentralized sewage treatment facilities.

(12) Specifically, the conductivity of the influent, and the conductivity and suspended solids concentration of the effluent of each of the plurality of decentralized sewage treatment facilities were input to a mapminmax function for normalization, and then input to the support vector machine.

(13) The formula for the mapminmax function was: y=(x−x.sub.min)/(x.sub.max−x.sub.min) (1), where v referred to a normalized measured data of the conductivity of the influent, and the conductivity and suspended solids concentration of the effluent; x referred to a real-time measured data of the conductivity of the influent, and the conductivity and suspended solids concentration of the effluent; x.sub.min was a minimum value of x, and x.sub.max was a maximum value of x.

(14) The training was to construct a predictive model to predict the discharge level of the effluent of the sewage treatment facilities using LIBSVM toolbox, and the training comprised optimization of penalty parameter c and Radial Basis Function (RBF) kernel parameter g.

(15) The optimization comprised modifying the penalty parameter c and kernel parameter g twice with SVMcgForClass function, thereby acquiring an optimal solution to the penalty parameter c and the kernel function parameter g.

(16) The first optimization was a rough selection that determined the penalty parameter c within a variation range of [2.sup.−10,2.sup.10] and the kernel parameter g within a variation range of [2.sup.−10,2.sup.10].

(17) (3) The remaining 10 sets of data were employed as a prediction set. The conductivity of the influent, and the conductivity and suspended solids concentration of the effluent of each of the remaining 10 sewage treatment facilities were measured and implemented a preliminary judgment.

(18) A method for the preliminary judgment was to determine whether the effluent satisfied the discharge standard by comparing the suspended solids concentration of the effluent with a standard value stipulated by the local sewage discharge standard.

(19) When the suspended solids concentration of the effluent was greater than the standard value, showing that the discharge level of the effluent failed to satisfy the local sewage discharge standard; when the suspended solids concentration of the effluent was less than or equal to a standard value, the conductivity of the influent, and the conductivity and suspended solids concentration of the effluent were input to the predictive model obtained in 2), thereby obtaining a predictive result of a discharge level of the effluent of the sewage treatment facility.

(20) Predictive results: the actual discharge level determined by the measured values of the effluents of 8 sewage treatment facilities accorded with the predictive result, showing that the prediction was correct. But the predictive result of each of the 2 sewage treatment facilities was different from the measured values, showing that the prediction was wrong. The prediction accuracy was 80%.

Example 2

(21) The effluent samples and the prediction method in this example were the same as that in Example 1, except that the local sewage discharge standard adopted the class I B standard of “Discharge standard of pollutants for municipal wastewater treatment plant”.

(22) Predictive results: the actual discharge level determined by the measured values of the effluents of 8 sewage treatment facilities accorded with the predictive result, showing that the prediction was correct. But the predictive result of 2 sewage treatment facilities was different from the measured values, showing that the prediction was wrong. The prediction accuracy was 80%.

(23) It will be obvious to those skilled in the art that changes and modifications may be made, and therefore, the aim in the appended claims is to cover all such changes and modifications.