A function approximation approach to Reinforcement Learning for solving unit commitment problem with Photo voltaic sources

Jasmin, E A; Ahamed, T. P. Imthias; Remani, T.

doi:10.1109/pedes.2016.7914428

Cited by 15 publications

(16 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Guided tree search was shown to scale better than unguided tree search with increasing number of generators, while at the same time producing solutions with similar operating costs. Existing research either considered small power systems when applying RL [8,[12][13][14], or simplified the problem to achieve tractability [15]. Our approach used RL to train a policy to reduce the action space, allowing for application to larger problem sizes, and showed that operating costs did not increase as a result.…”

Section: Discussionmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

“…Q-learning, a popular class of RL methods, has been applied to the UC problem in [12][13][14]. These papers have applied tabular Qlearning [12] and function approximation [13,14] with applications to problems of up to 10 generators. In [13], the results are validated against by comparison with Lagrangian Relaxation for a small deterministic problems while elsewhere Q-learning is not compared with the state of the art.…”

Section: Literature Reviewmentioning

confidence: 99%

“…These papers have applied tabular Qlearning [12] and function approximation [13,14] with applications to problems of up to 10 generators. In [13], the results are validated against by comparison with Lagrangian Relaxation for a small deterministic problems while elsewhere Q-learning is not compared with the state of the art. The Q-learning methods applied so far are only applied to relatively small power systems and suffer from curses of dimensionality in the state space when tabular methods are used [12] and action space due to its combinatorial structure.…”

Section: Literature Reviewmentioning

confidence: 99%

See 3 more Smart Citations

Applying reinforcement learning and tree search to the unit commitment problem

Mars

O’Sullivan

2021

Applied Energy

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

See 2 more Smart Citations

Applying reinforcement learning and tree search to the unit commitment problem

Mars

O’Sullivan

2021

Applied Energy

View full text Add to dashboard Cite

“…More work should be done right now. Motivated by these research challenges, intend of the suggested research is to develop a hybrid metaheuristics research algorithm for the solution of PBUCP of electrical power sector considering power demand of renewable energy source and plug-in charging vehicles [15] [16].…”

Section: Introductionmentioning

confidence: 99%

A New Solution to Profit Based Unit Commitment Problem Considering PEVs/BEVs and Renewable Energy Sources

Nandi

Kamboj

2020

E3S Web Conf.

View full text Add to dashboard Cite

Daily load demand for industrial, residential and commercial sectors are changing day by day. Also, inclusion of e-mobility has totally effected the operations of realistic power sector. Hence, to meet this time varying load demand with minimum production cost is very challenging. The proposed research work focuses on the mathematical formulation of profit based unit commitment problem of realistic power system considering the impact of battery electric vehicles, hybrid electric vehicles and plug in electric vehicles and its solution using Intensify Harris Hawks Optimizer (IHHO). The coordination of plants with each other is named as Unit commitment of plants in which the most economical patterns of the generating station is taken so as to gain low production cost with higher reliability. But with the increase in industrialization has affected the environment badly so to maintain the balance between the generation and environment a new thinking of generating low cost power with high reliability by causing less harm to environment i.e. less emission of flue gases is adopted by considering renewable energy sources.

show abstract

ACanis lupusinspired upgraded Harris hawks optimizer for nonlinear, constrained, continuous, and discrete engineering design problem

Nandi

Kamboj

2020

Numerical Meth Engineering

View full text Add to dashboard Cite

Recently established Harris hawks optimization (HHO) has natural behavior for finding an optimum solution in global search space without getting trapped in previous convergence. However, the exploitation phase of the current Harris hawks optimizer algorithm is poor. In the present research, an improved version of the HHO algorithm, which combines Harris hawks optimizer with Canis lupus inspire grey wolf optimizer (GWO) and named as hHHO-GWO algorithm, has been proposed to find the solution of various optimization problems such as nonlinear, nonconvex, and highly constrained engineering design problem. In the proposed research, the phase of exploration and exploitation of the existing HHO algorithm has been further improved using GWO algorithm and its performance has been tested for various benchmarks problems including CEC2005 (unimodal, multimodal, and fixed dimensions functions), multimodal functions with variable dimensions, and CEC-BC-2017 test functions. Further, the developed hybrid optimizer has been tested for 11 different engineering design and optimization problems and experimental results of hHHO-GWO have been compared with other optimizer.

show abstract

A function approximation approach to Reinforcement Learning for solving unit commitment problem with Photo voltaic sources

Cited by 15 publications

References 28 publications

Applying reinforcement learning and tree search to the unit commitment problem

Applying reinforcement learning and tree search to the unit commitment problem

A New Solution to Profit Based Unit Commitment Problem Considering PEVs/BEVs and Renewable Energy Sources

ACanis lupusinspired upgraded Harris hawks optimizer for nonlinear, constrained, continuous, and discrete engineering design problem

Contact Info

Product

Resources

About