This paper proposed a grouting power time series prediction model, based on fuzzy information granulation(FIG)and grey wolf optimized support vector machine(GWO-SVM). The model establishment consisted of three major steps. First, the granular computing method was introduced to decompose the original numerical time series into a series of information granules for reducing the scale of the data. Second, based on the theory of fuzzy sets, the fuzzy set operator was used to carry out the fuzzy calculation for each information granular, so that the obtained fuzzy information granular can reasonably represent the original numerical point set. Finally, the support vector machine was used as a predictive tool, optimized by the grey wolf optimizer algorithm, and the information granules can be predicted quickly, accurately and robustly. Taking an actual project as an example, this model was used to predict the fluctuation range and trend of grouting power time series during the grouting process. After the performance evaluation and comparative analysis, the feasibility and effectiveness of the prediction model were verified.