Optimization for energy purchase strategy at demand side through cost-probability-comparison

This paper presents a demand side energy purchase strategies' selection methodology. In order to take into account the reflection of the local load forecast fluctuation which is impossible to avoid, the method of total time on test transform is applied. So the optimal solution can be obtained over a wide rang of cost probabilities' comparison. The costs based on one strategy stand for the possible costs calculated based on the possible local loads which are modified according to the experience from the local load forecast. By using this method, not only how much electricity the retailer should purchase from the long term market, but also the suitable purchase quantities from each generator can be calculated. In addition, compared with the other methods, this graphic methodology can generate a group of preferential solutions for the retailer instead of one optimal solution. Because it is not for sure that all of the generation companies that the retailer wish to sign the contract with will agree this business, the final strategy should be the first agreed solution between the retailer and all generation companies. In short, this cost probability comparison method can provide a reasonable, valid and flexible energy purchase strategy.