반복적 비선형 최소자승법을 이용한 IPCC 매립가스 발생량 모델의 변수 산정 (Determination of IPCC Landfill Gas Generation Model Parameters using Iterative Non-linear Least Square Method)

서지정보

· 발행기관 : 한국폐기물자원순환학회
· 수록지 정보 : 한국폐기물자원순환학회지 / 36권 / 3호 / 308 ~ 320페이지
· 저자명 : 박진규, 김란희, 이남훈

초록

The UNFCCC on annual inventories require that Annex I Parties use the 2006 IPCC Guidelines for National GreenhouseGas Inventories. The 2006 IPCC Guidelines provide a landfill gas generation model to estimate methane emissions froma landfill. There are several parameters in the IPCC landfill gas generation model. However, a major challenge in IPCClandfill gas generation modeling is the estimation of these parameter values. Thus, this study describes a method todetermine these parameter values from the fitting of an iterative non-linear least square method (INLSM) to actual landfillgas data, using Generalized Reduced Gradient (GRG) and bootstrap methods. In this study, actual landfill gas data wascollected at Sudokwon Landfill Site 2, and curve fitting was carried out to find the parameter values (DOC, DOCf, andk) for each waste component as well as OX values. We used the SOLVER of Microsoft Excel, which employs the INLSMfitting routine, to produce the optimal goodness-of-fit between the actual landfill gas data and the IPCC modeling results.
This process was repeated several times to ensure that changes in the parameter values resulted in minimized sums ofsquares residuals (SSR) and a maximized coefficient of determination (R2). It should be noted that INLSM can neverprovide absolute certainty of having found the global SSR minimum unless a systematic sampling of all possiblecombinations of parameter values is carried out. Thus, mean values and ranges (95% confidence level) for the parameterswere calculated using a bootstrapping method. The results of the study show that a combination of INLSM andbootstrapping can be used appropriately to estimate IPCC landfill gas generation model parameters.

영어초록

The UNFCCC on annual inventories require that Annex I Parties use the 2006 IPCC Guidelines for National GreenhouseGas Inventories. The 2006 IPCC Guidelines provide a landfill gas generation model to estimate methane emissions froma landfill. There are several parameters in the IPCC landfill gas generation model. However, a major challenge in IPCClandfill gas generation modeling is the estimation of these parameter values. Thus, this study describes a method todetermine these parameter values from the fitting of an iterative non-linear least square method (INLSM) to actual landfillgas data, using Generalized Reduced Gradient (GRG) and bootstrap methods. In this study, actual landfill gas data wascollected at Sudokwon Landfill Site 2, and curve fitting was carried out to find the parameter values (DOC, DOCf, andk) for each waste component as well as OX values. We used the SOLVER of Microsoft Excel, which employs the INLSMfitting routine, to produce the optimal goodness-of-fit between the actual landfill gas data and the IPCC modeling results.
This process was repeated several times to ensure that changes in the parameter values resulted in minimized sums ofsquares residuals (SSR) and a maximized coefficient of determination (R2). It should be noted that INLSM can neverprovide absolute certainty of having found the global SSR minimum unless a systematic sampling of all possiblecombinations of parameter values is carried out. Thus, mean values and ranges (95% confidence level) for the parameterswere calculated using a bootstrapping method. The results of the study show that a combination of INLSM andbootstrapping can be used appropriately to estimate IPCC landfill gas generation model parameters.

참고자료

· 없음