Fuzzy-based novel risk and reward definition applied for optimal generation-mix estimation

Abstract Every generation source is an asset with unique reward and risk characteristics. Hence, the estimation of optimal generation mix can be considered to be an asset allocation problem. Reward is the profit from the investment, while risk is any uncertainty with the potential to cause financial loss. The risk associated with renewable energy generation is high, owing to the variable nature of solar radiation and wind. In this paper, the optimal generation mix problem is solved by considering the economic and power production parameters, simultaneously, using a novel fuzzy-generated index. Optimizing reward and risk, which are defined based on the index created, requires a multi-objective formulation and the deciding factor for the optimal point of operation is the risk aversion capability of the investor. The said index also facilitates the application of theories of finance for the optimal generation mix estimation of solar, wind and hydro powered system. Efficient frontiers defining risk as standard deviation (Markowitz mean-variance Theory), Mean Absolute Deviation (MAD) and Conditional Value at Risk (CVaR) are plotted to solve for the optimal generation mix. Monte-Carlo simulation is used to improve the robustness of the model. The results obtained show potential for significant contribution to the adopted investor-strategy in generation planning.