In power network management, the heat capacity of transmission lines originally arises from the line temperature constraint. The line temperatures are affected not only by the Joule heat but also by the thermal environment. This motivates us to exploit weather forecast data to improve the power management performance. The goal of this paper is to propose a stochastic model predictive control scheme for this purpose. In particular, we compensate for the probabilistic uncertainty by means of chance constraint optimization. The effectiveness of the proposed control scheme is examined through numerical simulation of power grids under the area dependent uncertainty and transmission line failure. (C) 2016, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.