Our mission at TREX-Ai is to transform the distribution grid for a sustainable future. The Mitacs Accelerate Entrepreneur Grant provided pivotal support over the last 15 months, and has allowed the team to make groundbreaking advancements in energy resource coordination. We’re excited to share our key milestones we’ve made with ALEX.
A Mission with Milestones: ALEX’s Synergistic Progress
Our objectives were ambitious but clear: demonstrate ALEX in digital twin, leveraging historical data to show its true potential in enhancing grid efficiency and resilience. Through a dedicated division of labor and the synergistic effort of our talented team, we achieved remarkable feats across four distinct milestones:
Simulation Excellence: Deploying ALEX with optimal actors was not just about creating a replica but about demonstrating a superior alternative to current DER integration approaches like Transactive Control.
Measuring Success: Developing ALEX’s performance metrics meant we could quantify its effectiveness and lay the groundwork for comparative analysis with existing systems.
Algorithmic Advancement: Identifying suitable Deep Reinforcement Learning (DRL) algorithms for ALEX was crucial. It allowed us to understand the system’s capabilities and potential as a robust energy management solution.
Agent Training: By training generalizing DRL agents for ALEX, we’re setting new standards in adaptable, intelligent energy management that can revolutionize how we approach grid distribution.
The Team's Triumphs:
Peter Atrazhev took on the challenge of Milestone 3, exploring collaborative AI in the Level Based Foraging Environment. His innovative approach to reward structures led to increased agent performance, with findings published in a notable 2022 international conference paper and then in a journal paper.
Daniel May and Steven Zhang tackled Milestones 1 and 2, diving deep into ALEX’s operational framework and market mechanisms. Their collaborative effort not only established a baseline for ALEX’s effectiveness but also unveiled crucial market properties needed for a balanced and truthful energy system. These findings lead to the publication of two journal articles (Reinforcement learning-driven local transactive energy market for distributed energy resources, Transactive Local Energy Markets Enable Community-Level Resource Coordination Using Individual Rewards).
Empirical Breakthroughs:
Our research has not just been theoretical. Empirical studies have also showcased the tangible benefits of ALEX:
Steven’s work shone light on how battery storage systems can increase local energy sufficiency, reducing grid reliance and energy bills significantly.
Daniel’s exploration of dynamic programming within ALEX has shown improvements in power quality metrics, highlighting ALEX’s ability to outperform non-coordinated DER systems.
Together, Daniel and Steven have proven that a local energy market facilitated by ALEX can lead to substantial economic benefits, with a community experiencing a remarkable 35.9% bill reduction compared to standard net billing methods.
Conclusion: Looking Ahead
Our journey through the Mitacs Accelerate Entrepreneur Grant has been one of discovery, innovation, and validation. The milestones reached are not just checkpoints but foundations for a new era of energy distribution. ALEX stands as a testament to what’s possible when innovation meets investment – a smarter, more sustainable grid that benefits all.
As we continue to refine and evolve ALEX, we extend our heartfelt thanks to Mitacs for their support. This journey has only just begun, and the future, powered by intelligent energy systems, is bright.
Join us as we continue to lead the charge in energy innovation. Stay tuned for more updates from TREX-Ai, where the grid of tomorrow is being shaped today.
