Abstract
The population is sharply growing in the last decade, resulting in non-potential power requests in dense urban areas, especially with the traditional power grid where the system is not compatible with the infrequent changes. Smart grids have shown strong potential to effectively mitigate and smooth power consumption curves to avoid shortages by adjusting and forecasting the cost function in real-time in response to consumption fluctuations to achieve the desired objectives. The main challenge for the smart grid designers is to reduce the cost and Peak to Average Ratio (PAR) while maintaining the desired satisfaction level. This article presents the development and evaluation of a Multi-Agent Reinforcement Learning Algorithm for efficient demand response in Smart Grid (MARLA-SG). Also, it shows a simple and flexible way of choosing state elements to reduce the possible number of states, regardless of the device type, range of operation, and maximum allowable delay. It also produces a simple way to represent the reward function regardless of the used cost function. SARSA (State-Action-Reward-State-Action) and Q-learning schemes are used and attained PAR reduction of 9.6%, 12.16%, and an average cost reduction of 10.2%, 7.8%, respectively.
Highlights
Traditional power grids are no longer able to deal with the enormous increase in the number of users and the massive load of modern devices, which results in either a power shortage or extensive raise in power prices to force the reduction of users’ consumption [1]
Changing the frequent consumption patterns of users, called demand response [2], is done either by rearranging devices operating hours, which will affect the satisfaction level of the customers [3], [4], or by giving incentives for using chargeable devices during offpeak hours to compensate the heavy load on the grid during peak hours [5], or using different price-based programs to redistribute the consumption during peak and off-peak hours and reduce the gap between them, which results in a more smooth consumption curve [6]–[8]
We propose Multi-Agent Reinforcement Learning Algorithm for efficient demand response in Smart Grid (MARLA-SG) which represents the reinforcement learning elements in a manner that decreases the storage capacity required for learning
Summary
Traditional power grids are no longer able to deal with the enormous increase in the number of users and the massive load of modern devices, which results in either a power shortage or extensive raise in power prices to force the reduction of users’ consumption [1]. At the start of each time slot, the agent observes the state parameters (e.g., devices’ requested range, remaining slots to the end of the job, current delay and cost level, etc.) of the environment, and takes an action (e.g., switch on/off the device) depending on simple calculations or complex or even stochastic calculations, It receives a feedback reward as a measure of the state-action pair (e.g., a satisfaction level or cost or a function of both) for the actions taken, and adjusts its policy until it converges to the optimal mapping from states to optimal actions that maximize the cumulative reward on the long run (e.g., maximizes satisfaction or minimizes cost on the long run depending on the user’s needs).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
