Intelligent Bidding Strategies for Prosumers in Local Energy Markets Based on Reinforcement Learning

Abstract

Local energy markets (LEMs) are proposed in recent years as a way to enable local prosumers and community to trade their electricity and have control over their electrical related resources by ensuring that electricity is traded closer to where it is produced. However, literature is still scarce with the most optimal and effective trading strategies for LEM design. In this work, we propose two reinforcement learning based intelligent bidding strategies for prosumers and consumers trading within an LEM. Our proposed models were evaluated of their performance by testing them in a German real case scenario. The simulation results show that intelligent bidding strategies create additional self sufficiency and market savings to the local community compared to the baseline strategy where the agents make their trading decision randomly without an intelligent agent. Moreover, modelling the intelligent agents to perform towards a common goal creates more share of individual savings for the prosumers and consumers compared to the classical intelligent bidding strategies employed in this work.INDEX TERMS Bidding strategy, energy community, local energy markets, Markov decision process, peerto-peer, reinforcement learning.

Integrating Peer-to-Peer Energy Trading and Flexibility Market With Self-Sovereign Identity for Decentralized Energy Dispatch and Congestion Management

“…In FIGURE 9, the trading activity which is the average energy traded in a 15 min trading cycle of all prosumeragents, numbered 1 to 15, along with an external energy supplier represented as Seen over all test cases, prosumers experienced fluctuations in their financial gains. The wide spread of prices is also seen in the results of [13] where also a P2P-market is developed and analysed. The analysis also reveals that no prosumeragent experiences a negative FGP, thereby indicating that no prosumer-agent incurs financial losses.…”

“…However, this is to be expected when considering that the analysis focuses solely on low-voltage grids with PV-plants as the single energy source. Low-voltage grids inherently exhibit a high degree of dependency on external energy suppliers, a characteristic that is also shown by [13,46]. The discrepancy between test case 1 and the other test cases can be attributed to the timing of the scenarios.…”

“…To distinguish the quality of the grid forecast several key performance indicators must be introduced. With regard to the forecast of the node voltages, these are the average percentage voltage forecast error 𝑒𝑒 𝑈𝑈,𝑟𝑟𝑒𝑒𝑎𝑎,𝑘𝑘 at a node k as well as the average percentage voltage forecast error 𝑒𝑒 𝑈𝑈,𝑟𝑟𝑒𝑒𝑎𝑎 over all nodes 𝐾𝐾, which are shown in equation (13) and equation (14). The error per node k is calculated for a number of 𝑇𝑇 timesteps with the measured voltage value 𝑈𝑈 𝑘𝑘,𝑡𝑡 and the forecasted voltage value 𝑈𝑈 𝑘𝑘,𝑡𝑡 𝑝𝑝 for each timestep.…”

“…However, both grid control and financial transactions are highly sensitive, necessitating robust IT security measures [6]. In several research projects [12][13][14], market-based grid control serves as a supplementary mechanism for efficient energy dispatch, preventive congestion management, and promoting RES integration. However, if congestions cannot be addressed through market-based solutions, an automated system is required for curative congestion management.…”