Search

In this study, the adaptive network-based fuzzy inference system (ANFIS) was used to generate the fuzzy rules for discussing the effects of gas and electricity price on residential sector’s energy consumption rate (RSECR) in Taiwan. Twenty seven fuzzy rules of Sugeno type were generated from ANFIS. It was found that the fuzzy rules did not reveal the significance when the situation was extreme, i.e. one input variable was low but another one was high. When all three input variables fell in the low degree of membership, the output RSECR values tended to be “high”. When all three input variables fell in the high degree of membership, the output RSECR values tended to be “low” even “negative”. It suggested that higher energy price would inhibit RSEC.

In this study, the back-propagation neural network (BPNN) was used to predict the number of low-income households (NLIH) in Taiwan, taking the seasonally adjusted annualized rates (SAAR) for real gross domestic product (GDP) as input variables. The results indicated that the lowest mean absolute percentage error (MAPE), mean squared error (MSE), root mean squared error (RMSE), and highest correlation coefficient (R) for training and testing were 4.759 % versus 19.343 %, 24429972.268 versus 781839890.859, 4942.669 versus 27961.400, and 0.945 versus 0.838, respectively.

This study adopted analytical hierarchical process (AHP) to evaluate the significance of the criterion of policy regarding promoting biomass energy, taking environmental social groups’ points of view into consideration. The analytical results of the study are as follows. The experts from environmental social groups suggested that the most important major criteria are in the order as follows: policy criterion layer, technical criterion layer, economy criterion layer, and energy education criterion layer. As for the global weights of the criteria, the criteria with top five weights are in the order as follows: “active development of green energy industry by government”, “supply and consumption method of low-carbon and low-pollution energy”, “increase of percentage of power generation from renewable energy in total power supply”, “encourage manufacturers to develop new biomass power generation technology”, and “continually develop the technology projects which has potential and prospects in the future”.

The analytical hierarchical process (AHP) was adopted in this study to determine the priority of the policy criterion related to promoting wind energy, considering the environmental practitioners’ points of view. The results were described as follows. The practitioners from environmental organizations suggested that the most important major criteria are in the following order: policy criterion layer, technical criterion layer, economy criterion layer, and energy education criterion layer. For the global weights of the criteria, the criteria with top five weights were in the order as follows: “active development of green energy industry by government”, “supply and consumption method of low-carbon and low-pollution energy”, “increase of percentage of power generation from renewable energy in total power supply”, “encourage manufacturers to develop new wind power generation technology”, and “continually develop the technology projects which has potential and prospects in the future”.

This study adopted analytical hierarchical process (AHP) to evaluate the significance of the policy criterion for promoting solar energy, regarding social publics’ points of view. The results of the study are described as follows. The social publics suggested that the most important major criteria are in the following order: policy, technique, educational promotion, and economy incentives. For the global weights of the criteria, the criteria with top five weights are in the following order: “patents for solar energy products“ (0.183), “encourage R&D of new techniques and reduce power cost“ (0.168), “improve power saving technique“ (0.095), “encourage the firms to develop new techniques of solar energy power“ (0.082), and “increase percentage of solar energy power in total power capacity“ (0.075).

In this study, seven types of first-order and one-variable grey differential equation model (abbreviated as GM (1, 1) model) were used to predict the hardness of four groundwater monitoring stations in Kaohsiung City of Taiwan. The mean absolute percentage error (MAPE) was used to evaluate the predicting performance. The results indicated the minimum MAPE of 4.71 %, 3.15 %, 2.66 %, and 16.63 % could be achieved when predicting hardness of Fonsi, Datung, Shaukang, and Chihsien stations, respectively. According to the results, it revealed that GM (1, 1) was an efficiently early warning tool for providing groundwater quality information to the competent authority.