Development of a forecasting system for supporting remediation design and process control based on NAPL‐biodegradation simulation and stepwise‐cluster analysis

Effective process control is crucial in implementing remediation actions for petroleum‐contaminated sites. However, in dealing with in situ bioremediation practices, difficulties exist in incorporating numerical simulation models that are needed for process forecasting within real‐time non‐linear optimization frameworks that are critical for supporting the process control. With such difficulties, it is desired that a statistical relationship between remediation system performance and operating condition be established. Nevertheless, in the remediation systems, many variables can be either continuous or discrete, and the relations among them can be either linear or non‐linear. These lead to complexities in the related multivaraite analyses. In this study, a forecasting system has been developed for supporting remediation design and process control based on techniques of NAPL‐biodegradation (non‐aqueous phase liquid biodegradation) simulation and stepwise‐cluster analysis (SCA). The results indicate that the developed system is effective in forecasting the effects of multiple cleanup actions under various conditions. The predicted benzene concentrations have acceptable error levels compared with the outputs of numerical simulation. An optimization model for obtaining optimum operating conditions is then proposed to illustrate how the SCA method can be used for supporting optimization of bioremediation operations. A unique contribution of this research is the development of a multivariate inference system associated with simulation and optimization efforts for tackling the complex in situ bioremediation practices.