Reinforcement Learning Applied to Multi Agent Modelling, the Case of the Iranian Power Market
With increasing competition in the wholesale Electricity markets and advances in behavioral economics in recent years, the multi-agent modeling approach has been applied widely to simulate the outcome of the markets. The electricity market consists of power generating agents that compete over produc...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | fas |
| Published: |
Allameh Tabataba'i University Press
2017-12-01
|
| Series: | Pizhūhishnāmah-i Iqtiṣād-i Inirzhī-i Īrān |
| Subjects: | |
| Online Access: | https://jiee.atu.ac.ir/article_9993_12cc07b387a584e06b813671b051b95f.pdf |
| _version_ | 1797369227308433408 |
|---|---|
| author | Mohammadreza Asghari Oskoei Farhad Fallahi Meysam Doostizadeh saeed Moshiri |
| author_facet | Mohammadreza Asghari Oskoei Farhad Fallahi Meysam Doostizadeh saeed Moshiri |
| author_sort | Mohammadreza Asghari Oskoei |
| collection | DOAJ |
| description | With increasing competition in the wholesale Electricity markets and advances in behavioral economics in recent years, the multi-agent modeling approach has been applied widely to simulate the outcome of the markets. The electricity market consists of power generating agents that compete over production in daily auction conducted by an independent system operator (ISO). The market clearing mechanism can be seen as a static game that repeats every hour. In this game, an agent proposes her price for the next day and the ISO chooses the best proposals that minimizes the total costs given the demand and the technical constraints. Agents are also assumed to learn from the outcomes and adjust their biding strategy accordingly. In this paper, we develop an agent-based model for the day-ahead and pay-as-bid electricity market in Iran. The objective is to compare the outcome of the market measured by the agents profit and the time to converge using three different strategies: greedy, random and reinforcement learning. The simulation results indicate that the reinforcement learning leads to higher profits with a faster convergence rate than the other two strategies. |
| first_indexed | 2024-03-08T17:43:35Z |
| format | Article |
| id | doaj.art-273524baa56b48beafdc9451bdd2403c |
| institution | Directory Open Access Journal |
| issn | 2423-5954 2476-6437 |
| language | fas |
| last_indexed | 2024-03-08T17:43:35Z |
| publishDate | 2017-12-01 |
| publisher | Allameh Tabataba'i University Press |
| record_format | Article |
| series | Pizhūhishnāmah-i Iqtiṣād-i Inirzhī-i Īrān |
| spelling | doaj.art-273524baa56b48beafdc9451bdd2403c2024-01-02T10:48:22ZfasAllameh Tabataba'i University PressPizhūhishnāmah-i Iqtiṣād-i Inirzhī-i Īrān2423-59542476-64372017-12-0172514010.22054/jiee.2018.99939993Reinforcement Learning Applied to Multi Agent Modelling, the Case of the Iranian Power MarketMohammadreza Asghari Oskoei0Farhad Fallahi1Meysam Doostizadeh2saeed Moshiri3Computer Science Dept, Allameh Tabataba'i UniversityResearch Fellow in Niro Research Inst.,Assistant Prof., Lorestan University,Associate Prof., Allameh Tabataba’i UniversityWith increasing competition in the wholesale Electricity markets and advances in behavioral economics in recent years, the multi-agent modeling approach has been applied widely to simulate the outcome of the markets. The electricity market consists of power generating agents that compete over production in daily auction conducted by an independent system operator (ISO). The market clearing mechanism can be seen as a static game that repeats every hour. In this game, an agent proposes her price for the next day and the ISO chooses the best proposals that minimizes the total costs given the demand and the technical constraints. Agents are also assumed to learn from the outcomes and adjust their biding strategy accordingly. In this paper, we develop an agent-based model for the day-ahead and pay-as-bid electricity market in Iran. The objective is to compare the outcome of the market measured by the agents profit and the time to converge using three different strategies: greedy, random and reinforcement learning. The simulation results indicate that the reinforcement learning leads to higher profits with a faster convergence rate than the other two strategies.https://jiee.atu.ac.ir/article_9993_12cc07b387a584e06b813671b051b95f.pdfagent based modelelectricity marketreinforcement learninggame theoryiran |
| spellingShingle | Mohammadreza Asghari Oskoei Farhad Fallahi Meysam Doostizadeh saeed Moshiri Reinforcement Learning Applied to Multi Agent Modelling, the Case of the Iranian Power Market Pizhūhishnāmah-i Iqtiṣād-i Inirzhī-i Īrān agent based model electricity market reinforcement learning game theory iran |
| title | Reinforcement Learning Applied to Multi Agent Modelling, the Case of the Iranian Power Market |
| title_full | Reinforcement Learning Applied to Multi Agent Modelling, the Case of the Iranian Power Market |
| title_fullStr | Reinforcement Learning Applied to Multi Agent Modelling, the Case of the Iranian Power Market |
| title_full_unstemmed | Reinforcement Learning Applied to Multi Agent Modelling, the Case of the Iranian Power Market |
| title_short | Reinforcement Learning Applied to Multi Agent Modelling, the Case of the Iranian Power Market |
| title_sort | reinforcement learning applied to multi agent modelling the case of the iranian power market |
| topic | agent based model electricity market reinforcement learning game theory iran |
| url | https://jiee.atu.ac.ir/article_9993_12cc07b387a584e06b813671b051b95f.pdf |
| work_keys_str_mv | AT mohammadrezaasgharioskoei reinforcementlearningappliedtomultiagentmodellingthecaseoftheiranianpowermarket AT farhadfallahi reinforcementlearningappliedtomultiagentmodellingthecaseoftheiranianpowermarket AT meysamdoostizadeh reinforcementlearningappliedtomultiagentmodellingthecaseoftheiranianpowermarket AT saeedmoshiri reinforcementlearningappliedtomultiagentmodellingthecaseoftheiranianpowermarket |