Energy planning involves multiple actors (authorities, community, investors, and operators) and multiple objectives most of the time in perpetual conflict. The Analysis Hierarchy Process (AHP) which consists of a Multi-Criteria Decision Analysis method (MCDA) theory of measurement through pairwise comparisons and relies on the judgments of experts to derive priority scales was used to evaluate the energy resources of Idjwi island in the Democratic Republic of Congo (DRC). The weights obtained from AHP analysis were used to rank the seven alternative energy resources for the island using the Technique of Order Preferences to Similarity to Ideal Solution (TOPSIS). The combination of the two MCDA methods reduces the uncertainty and reinforce the reliability of the decision by combining the advantages of both methods. Two surveys were conducted in this research, one for the rural community in Idjwi, in order to determine their preferences, priority, economic situation for an energy project. The second survey was conducted for the energy experts in the region was to analyze the importance of the criteria. This thesis aims at ranking the different energy resources for rural electrification in the Island. Energy planning in many developing countries is still not well-handled and depend mostly on the priorities of actual authorities. To minimize the risk of investing in wrong sources of energy, energy planning is needed which includes many actors and factors. For example, the exclusion of the local communities in the process of electrification projects can lead to failures of many mini-grid projects in villages. Based on the preferences and priorities of the community, the small hydropower and solar photovoltaic sources were the first alternatives to power Idjwi Island. An assessment of the renewable energy resources of the DRC is also included.

This paper proposes a revised Lambda-Gamma algorithm applied to the solution of a short-term hydrothermal planning problem. It is a Lambda-Gamma algorithm, which, in addition to the Newton-Raphson method for solving the coordination equations, also specifies the direction of adjustment of the Lambda and Gamma parameters according to the convergence conditions. This gives it the particularity of being more explicit, detailed and easy to implement. Next, the performance of this revised algorithm is compared with that of the Newton-Raphson Approximate algorithm for hydrothermal planning. From this comparison, it is found that the proposed algorithm outperforms and is more robust than the approximate Newton-Raphson algorithm given its convergence to a global minimum even when the practical constraints of generator power limits are taken into account. In addition, the conformity of the results obtained with those of the literature, allowed to validate the algorithm.