Pré-Seleção de turbinas hidráulicas utilizando AHP e Fuzzy-VIKOR

Abstract

Small hydroelectric plants are of great importance in the scenario of renewable energy generation in Brazil. In order to better reproduce the solution to be used in the construction, the pre-selection of the hydraulic turbine becomes essential in the elaboration of the basic project of these enterprises. The traditional method of selecting turbines has an extremely technical approach, without considering the aspects of investor preference, this sometimes ends up requiring rework with revisions to the basic design. The main objective of this dissertation is to develop a pre-selection model for hydraulic turbines using a multicriteria approach, using the AHP (Analytic Hierarchy Process) and Fuzzy-VIKOR (VlseKriterijumska Optimizacija I Kompromisno Resenje) methods for situations in which technical data drop height and flow, available for installation of the project, represent conflict in the choice of existing configurations. In the methodology, a structured approach, based on the Proknow-C method (Knowledge Development Process -Constructivist), was used to select the study references. A multicriteria model has been proposed for use in cases of drop values of 20 to 35 m and flow rates of 10 and 30 m³/s. For the construction of the model, the AHP methods were used to define the importance of the criteria (weights), based on expert assessments in hydroelectric projects. The Fuzzy-VIKOR method was used, to classify alternatives based on expert judgments regarding performance of hydraulic turbines in the Simple Horizontal Francis (FHS), Double Horizontal Francis (FHD), Vertical Francis (FV), Kaplan Upstream S-Type (KSM) and Kaplan Downstream S-Type (KSJ) variants against the criteria. As main results of the dissertation, sixteen new hydraulic turbine selection criteria were presented in the pre-selection stage, which were divided into four groups of criteria: Performance (C1), Turbine-Generator Cost (C2), Peripheral Equipment Cost (C3) and the group of criteria Cost of Construction (C4). The result of applying the developed model was admitted as a general solution, so a case study was carried out to verify the applicability of the model. In the case study, the investor's preferences regarding the relationship between criteria, using the AHP method, and the result of the performance analysis of the alternatives provided by the specialists were used in the application of the Fuzzy-VIKOR method. The application of the model for a general solution, presented the following solution: A1 > A4 > A2 > A5 > A3. Based on the evaluation of the indicators of acceptable advantage and acceptable stability of the Fuzzy-VIKOR method, onlythe alternative A1 -Simple Horizontal Francis (FHS) is accepted as a single pre-selection compromise solution. For the case study, three different scenarios were evaluated about the decision strategy: neutrality regarding the preference for the benefitsassociated with the decision or minimizing the losses associated with the choice (v = 0.5); the scenario for prioritizing the benefits associated with the decision (v = 0.7); and the scenario of minimizing the losses associated with the choice (v = 0.3). The results for the three scenarios evaluated were the same: A1 > A2 > A3 > A4 > A5. The verification of compliance with the indicators of acceptable advantage and acceptable stability of the Fuzzy-VIKOR method presents the alternatives A1 -Simple Horizontal Francis (FHS) and A2 -Double Horizontal Francis (FHD), as a pre-selection compromise solution.