English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

While reinforcement learning (RL) is gaining popularity in energy systemscontrol, its real-world applications are limited due to the fact that theactions from learned policies may not satisfy functional requirements or befeasible for the underlying physical system. In this work, we propose PROjectedFeasibility (PROF), a method to enforce convex operational constraints withinneural policies. Specifically, we incorporate a differentiable projection layerwithin a neural network-based policy to enforce that all learned actions arefeasible. We then update the policy end-to-end by propagating gradients throughthis differentiable projection layer, making the policy cognizant of theoperational constraints. We demonstrate our method on two applications:energy-efficient building operation and inverter control. In the buildingoperation setting, we show that PROF maintains thermal comfort requirementswhile improving energy efficiency by 4% over state-of-the-art methods. In theinverter control setting, PROF perfectly satisfies voltage constraints on theIEEE 37-bus feeder system, as it learns to curtail as little renewable energyas possible within its safety set.

Enforcing Policy Feasibility Constraints through Differentiable Projection for Energy Optimization