Deterministic and Multiobjective Optimisation Methods in Thermo-Wind Dispatch Problems

• Grant number: 25/08615-1
• Support Opportunities: Regular Research Grants
• Start date: August 01, 2026
• End date: July 31, 2028
• Field of knowledge: Engineering - Production Engineering - Operational Research
• Principal Investigator: Antonio Roberto Balbo
• Grantee: Antonio Roberto Balbo
• Host Institution: Faculdade de Ciências (FC). Universidade Estadual Paulista (UNESP). Campus de Bauru. Bauru , SP, Brazil
• City of the host institution: Bauru
• Associated researchers: Dylan Francis Jones ; Leonardo Nepomuceno

Abstract

The production of wind energy stands out in Brazil because it is of great environmental importance, helping to reduce polluting gases emitted into the atmosphere by replacing thermoelectric units as a complementary source to hydroelectric plants. This project proposes a new solution approach for Thermo-Wind Dispatch Problems, which are non-linear and multimodal optimisation problems that combine the production of energy from thermal and wind sources, subject to operational constraints relating to the generation system. The development of this approach exploits classical optimisation methods in conjunction with a method based on auxiliary dynamic systems, described in the literature for restricted optimisation problems. In the proposed solution approach, the original problem is replaced by an auxiliary dynamic system, obtained in such a way that its equilibrium points correspond to the local optimal solutions of the original optimisation problem. To determine these solutions, the primal-dual interior point method will be used, which makes it possible to deal with restricted problems and determine these solutions at each cycle generated by the aforementioned system. A probabilistic strategy based on the Weibull probability distribution will also be considered to deal with the wind generation problem, which is stochastic and dependent on wind speed. The proposed solution approach will be applied to determine multiple local optimum points for the Thermo-Wind Dispatch Problem, which is a multimodal optimisation problem containing multiple local optimum points. In this context, the method identifies several minimum points and, in the end, selects the best of them, with the aim of improving its performance in solving the problem in question. An extension of this method considers multi-objective optimisation methods, such as progressive bounded constraints and goal programming, to determine efficient or Pareto-optimal solutions to the Multi-objective Environmental Thermo-Wind Dispatch Problem, which considers environmental aspects in the formulation of the problem, seeking to minimise production costs in conflict with the reduction of polluting gas emissions, which have an impact on energy production. After the computational implementation of the proposed methodology, computational tests will be carried out on these problems to evaluate its efficiency and performance in comparison with other meta-heuristic and deterministic methods previously proposed in the literature. (AU)