Meta-heurísticas aplicadas ao modelo AC no planejamento da expansão de sistemas de transmissão
Carregando...
Data
2019-05-31
Autores
Melchor Gutiérrez, José Nicolás
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Os sistemas elétricos do futuro próximo terão um comportamento dinâmico porque grandes fontes renováveis de geração, principalmente geração eólica e solar, serão construídas nos sistemas de potência de vários países do mundo incluindo o Brasil. A incerteza introduzida pelas fontes renováveis de energia é um problema que deve ser solucionado. Para resolver o problema da incerteza é necessário criar uma rede mais flexível. A alocação de armazenadores de energia no sistema de potência será necessária para lidar com as novas incertezas e para manter o funcionamento adequado do sistema. Adicionalmente, o sistema de transmissão deve-se tornar muito mais ativo através do uso de novas tecnologias flexíveis que permitirão ter um controle ótimo do sistema. As novas tecnologias incluem Sistemas Flexíveis de Transmissão em Corrente Alternada (FACTS), sistemas de transmissão de corrente continua (HVDC) e outras tecnologias que permitirão controlar as direções dos fluxos de potência sem alterar a geração ou demanda do sistema. Portanto, nesta tese são estudadas diferentes representações matemáticas para os problemas de fluxo de potência ótimo (FPO), planejamento da transmissão de longo prazo (PERT) e alocação de fontes de reativos (AFR) usando as equações de fluxo de potência de corrente alternada (CA). Adicionalmente, são propostas novas formulações para representar os problemas de fluxo de potência ótimo multiperíodo para sistemas de transmissão ativos que incluem FACTS, sistemas HVDC e baterias. Também foram desenvolvidas técnicas heurísticas e meta-heurísticas para resolver o problema de planejamento da expansão de redes de transmissão estático e multiestágio de forma eficiente. Os sistemas testes usados no problema de FPO de CA são os sistemas IEEE-30, IEEE-118 e IEEE-300. Os sistemas testes usados no problema PERT de CA foram os sistemas Garver, IEEE-24. IEEE-118, Sul Brasileiro e Norte-nordeste Brasileiro. As modelagens exatas foram implementadas em AMPL e as técnicas heurísticas e meta-heurísticas foram implementadas na linguagem C++. As diferentes formulações matemáticas têm sido resolvidas com os pacotes de otimização CPLEX, Knitro e Baron. Os resultados mostram a boa qualidade das diferentes representações não lineares para o problema de FPO de CA. Adicionalmente, a efetividade das técnicas de solução propostas para o problema de PERT de CA é demostrada.
The future power systems will have a dynamic behaviour due to a significant amount of renewable generation, especially wind and solar generation, will be installed in the power systems of many countries in the world including Brazil.The uncertainty introduced by the renewable energy resources is a problem that has to be solved. The development of a flexible network is necessary to solve the aforementioned problem. The installation of energy storage in the power system will be necessary to manage the new uncertainties and to maintain the adequate system’s operation. Additionally, the transmission system will become more active though new flexible technologies, which allows having an optimal control of the power system. The new technologies include Flexible AC Transmission System (FACTS), High Voltage Direct current (HVDC) transmission systems and other technologies, which allow controlling the power flow direction without changing the power injections of generation or demand. Therefore, in this thesis are studied different mathematical formulations using the AC power flow equations for the optimal power flow (OPF) problem, the transmission expansion planning (TEP) problem and the reactive power planning (RPP) problem. Furthermore, two additional contributions have been made. The first contribution is the development of new mathematical formulations for the multiperiod optimal power flow for active transmission systems that includes FACTS, HVDC and batteries. The second contribution is the development of effective and efficient heuristics and metaheuristics to solve the static and multistage AC TEP. The test systems that has been used to assess the AC OPF formulations are the IEEE 30, IEEE 118 and IEEE 300. Moreover, the Garver, IEEE 24, IEEE 118, South Brazilian and North-Northeast Brazilian power systems have been used to test the formulations and solution techniques for the AC TEP problem. The mathematical formulations have been implemented in the mathematical programming language AMPL and the heuristics and metaheuristics have been implemented in C++. The different mathematical formulations have been solved with the solver CPLEX, Knitro and Baron. The results show the good quality of the nonlinear formulations to represent the AC OPF problem. Additionally, the effectiveness of the solution techniques for the AC TEP problem is proven.
The future power systems will have a dynamic behaviour due to a significant amount of renewable generation, especially wind and solar generation, will be installed in the power systems of many countries in the world including Brazil.The uncertainty introduced by the renewable energy resources is a problem that has to be solved. The development of a flexible network is necessary to solve the aforementioned problem. The installation of energy storage in the power system will be necessary to manage the new uncertainties and to maintain the adequate system’s operation. Additionally, the transmission system will become more active though new flexible technologies, which allows having an optimal control of the power system. The new technologies include Flexible AC Transmission System (FACTS), High Voltage Direct current (HVDC) transmission systems and other technologies, which allow controlling the power flow direction without changing the power injections of generation or demand. Therefore, in this thesis are studied different mathematical formulations using the AC power flow equations for the optimal power flow (OPF) problem, the transmission expansion planning (TEP) problem and the reactive power planning (RPP) problem. Furthermore, two additional contributions have been made. The first contribution is the development of new mathematical formulations for the multiperiod optimal power flow for active transmission systems that includes FACTS, HVDC and batteries. The second contribution is the development of effective and efficient heuristics and metaheuristics to solve the static and multistage AC TEP. The test systems that has been used to assess the AC OPF formulations are the IEEE 30, IEEE 118 and IEEE 300. Moreover, the Garver, IEEE 24, IEEE 118, South Brazilian and North-Northeast Brazilian power systems have been used to test the formulations and solution techniques for the AC TEP problem. The mathematical formulations have been implemented in the mathematical programming language AMPL and the heuristics and metaheuristics have been implemented in C++. The different mathematical formulations have been solved with the solver CPLEX, Knitro and Baron. The results show the good quality of the nonlinear formulations to represent the AC OPF problem. Additionally, the effectiveness of the solution techniques for the AC TEP problem is proven.
Descrição
Palavras-chave
Algoritmo genético especializado, Planejamento da expansão de redes de transmissão, Alocação de fontes de reativos, Fluxo de potência ótimo, Optimal power flow, Reactive power planning, Specialized genetic algorithm, Transmission expansion planning