Predicting ecological environment for a residential community using intelligent hybrid system

In this paper, intelligent hybrid system (IHS), in which fuzzy reasoning, Genetic Algorithm (GA) and neural network based on fuzzy rule (NNFR) are combined, is employed to incorporate diverse residential environment aspects and predict ecological environment levels. Based on connotation of composite ecosystem, the affected ecological environment aspects (and here the input patterns for the IHS) in such predictions are extended to consider eight random factors (and here linguistic variables): afforested environment, atmospheric environment, water environment, sound environment, traffic environment, energy supply, waste disposal, and civilization environment. The predicted result (and here the output pattern for the IHS) is represented by one linguistic variable, i.e. ecological environment level for residential community. Consequently, the IHS enables us to predict successfully the ecological environment levels of a residential community in Guiyang city of China, and the effectiveness of the IHS is verified and compared with in situ measurement.