In this paper, we propose a recurrent neural network in order to predict the time series of the gas energy consumption in a conditioning equipment. We, in particular, characterize the time series data by using various method and find that the data contain a non-linear correlation. Based on the finding that the time series of the gas consumption contained a considerable amount of the non-linear correlation, we adopt the Elman neural network as a non-linear model for the prediction of the gas consumption. By tuning the parameters in the model, we demonstrate that the proposed model predicted the gas consumption with the relative errors and the average errors less than 1% and 0.5N㎥/h - 3.2N㎥/h respectively. The results of this study can be used to the gas energy management system in terms of the effective control of the conditioning energy in the dry process.