In this thesis we analyse stochastic control problems for the valuation of a natural gas storage facility and natural gas-fi red power station while taking into account their operating characteristics. In the gas storage problem the underlying spot price dynamics is a mean-reverting process driven by a time-inhomogeneous Levy process. For the power valuation both electricity and gas spot price processes exhibit mean-reverting spikes and Markov regime-switches. The Levy regime-switching model incorporates the effects of demand-supply uctuations in energy markets and abrupt economic disruptions or business cycles. We make use of skewed Levy copulas to model the dependence risk of electricity and gas jumps. The corresponding HJB equations are solved by an explicit finite difference method. The numerical approach gives us both the values of the facilities and their optimal operating strategies depending on the gas and electricity prices, storage level, current temperature of the boiler and time. The surfaces of control strategies and contract values are obtained by implementing the numerical method for particular examples.