An integrated genetic algorithm-principal component analysis for improvement and estimation of gas consumption in Finland, Hungary, Ireland, Japan and Malaysia

This paper presents a genetic algorithm (GA)-principal component analysis (PCA) for long-term natural gas (NG) consumption prediction and improvement. Six models are proposed to forecast the annual gas demand. Around 27 GAs have been constructed and tested in order to find the best GA for gas consumption. The proposed models consist of input variables such as gross domestic product (GDP) and population (POP). All of trained GAs are then compared with each other respect to the mean absolute percentage error (MAPE). The GA model is capable of dealing both complexity and uncertainty in the data set. To show the applicability and superiority of the GA, actual gas consumptions in Finland, Hungary, Ireland, Japan and Malaysia from 1980 to 2007 are considered. With the aid of an autoregressive model, GDP and population are projected till 2015, and then with the projected GDP and population as inputs to the best GA model, gas consumption is predicted till 2015. Finally, we use the multivariate method of PCA in behaviour analysis of gas consumption in the selected countries. This method normalises the gas consumption by both population and GDP, and then the PCA procedure is run for efficiency assessment of the selected countries. PCA is used to examine the behaviour of gas consumption in the past and also to make insights for the forthcoming years.