Production estimates of coalbed methane by type-2 fuzzy logic systems

Reasonable prediction estimates of coalbed methane (CBM) has important significance for economic development. Production estimates of CBM had recognized to be complex, nonlinear, and adherent with uncertainty, so it is necessary to analyze accurately coalbed gas potential production capacity by adopting other mathematics methods. In this paper, the using of type-2 fuzzy logic system (T2FLS) as a novel approach for forecasting production of CBM has been investigated and implemented. Considering there are a lot of parameters that affect the model's performance, the authors adopted RBF neural network combined with the algorithm of Mean Impact Value to select the effect parameters. T2FLS production forecast method was applied to CBM wells of Hancheng mine. Comparative studies have been carried out to compare the performance of the proposed method with those earlier used methods. Empirical results show the proposed method has notable advantage in generalization, stability and consistency.