신경회로망을 이용한 순환식 돈분폐수 처리시스템의 모니터링

A recycling reactor system operated under sequential anoxic and oxic conditions for the swine wastewater has been developed, in which piggery slurry is fermentatively and aerobically treated and then part of the effluent recycled to the pigsty. This system significantly removes offensive smells (at both pigsty and treatment plant), BOD and other loads, and appears to be costeffective for the small-scale farms. The most dominant heterotrophs were Alcaligenes faecalis, Brevundimonas diminuta and Streptococcus sp. in order while lactic acid bacteria were dominantly observed in the anoxic tank. We propose a novel monitoring system for a recycling piggery slurry treatment system through neural networks. Here we tried to model treatment process for each tank(influent, fermentation, aeration, first sedimentation and fourth sedimentation tanks) in the system based on population densities of heterotrophic and lactic acid bacteria. Principle component analysis(PCA) was first applied to identify a relation between input(microbial densities and parameters for the treatment such as population densities of heterotrophic and lactic acid bacteria, suspended solids (SS), COD, $NH_3-N$, ortho-P, and total-P) and output, and then multilayer neural networks were employed to model the treatment process for each tank. PCA filtration of input data as microbial densities was found to facilitate the modeling procedure for the system monitoring even with a relatively lower number of input. Neural networks independently trained for each treatment tank and their subsequent combinatorial data analysis allowed a successful prediction of the treatment system for at least two days.

본 연구에서는 순환식 돈분 폐수 처리 시스템에서의 미생물 분포에 따른 폐수 처리 효과를 모델링하기 위해 신경회로망과 PCA를 이용하는 새로운 방법을 제안하였다. PCA 분석 결과를 바탕으로 신경회로망의 최적 입력 조건을 찾고, 실측 데이터를 이용하여, 폐수 처리 시스템의 각 탱크를 별도로 학습함으로써 비교적 적은 수의 데이터에도 불구하고 정확한 모델링 결과를 얻었다. 제안한 시스템은 폐수 처리 시스템의 효과적인모니터링 시스템으로 사용할 수 있으며, 향후 실제 돈분 처리 시스템에서 원하는 기준의 방류수를 얻기 위한 최적의 입력조건 (미생물밀도 등)을 결정하는데 있어서 에뮬레이터로 사용될 수 있을 것으로 기대된다.