Journal
ELECTRIC POWER SYSTEMS RESEARCH
Volume 187, Issue -, Pages -Publisher
ELSEVIER SCIENCE SA
DOI: 10.1016/j.epsr.2020.106474
Keywords
Distributed generation allocation; Electricity market; Meta-heuristic optimization; Reinforcement learning
Categories
Funding
- Gonbad Kavous University [6-226]
Ask authors/readers for more resources
Independent investors can install distributed generations (DGs) in power networks as long as they follow the systems' rules and frameworks. Logically, they try to maximize their profits, and network owners/operators usually cannot directly control their decisions regarding the DG location and size. However, there are inherent interactions between their decisions and the systems' technical and economic characteristics. In this paper, a new approach is presented for optimal DG placement and sizing from the standpoint of independent investors, with specific consideration given to interactions with pool-based electricity markets. The proposed approach is founded on the innovative combination of the meta-heuristic optimization model and the reinforcement learning-based market simulation strategy. This approach is implemented using the particle swarm optimization algorithm and the Q-learning technique, and it is comprehensively examined in the IEEE 30-bus system with consideration for different pricing rules and DG types. The results confirm that there are significant inter-dependencies between the optimal decision of DG investors and the strategic bidding behavior of market players. Also, they indicate that the proposed approach is useful for independent investors in identifying the most valuable DG location and size as well as for the systems' policy-makers in investigating the long-term effects of specific rules.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available