End-to-End Differentiable Predictive Control with Guaranteed Constraint Satisfaction and feasibility for Building Demand Response

TL;DR

This paper introduces an end-to-end differentiable predictive control framework using transformer models and online constraint tightening, achieving near-perfect constraint satisfaction in building demand response with theoretical guarantees.

Contribution

It proposes a novel E2E-DPC method that models complex building dynamics and guarantees constraint satisfaction without offline terminal set computation.

Findings

Over 99% reduction in constraint violations.

Maintains near-optimal electricity costs.

Validated in high-fidelity building simulations.

Abstract

The high energy consumption of buildings presents a critical need for advanced control strategies like Demand Response (DR). Differentiable Predictive Control (DPC) has emerged as a promising method for learning explicit control policies, yet conventional DPC frameworks are hindered by three key limitations: the use of simplistic dynamics models with limited expressiveness, a decoupled training paradigm that fails to optimize for closed-loop performance, and a lack of practical safety guarantees under realistic assumptions. To address these shortcomings, this paper proposes a novel End-to-End Differentiable Predictive Control (E2E-DPC) framework. Our approach utilizes an Encoder-Only Transformer to model the complex system dynamics and employs a unified, performance-oriented loss to jointly train the model and the control policy. Crucially, we introduce an online tube-based constraint tightening method that provides theoretical guarantees for recursive feasibility and constraint satisfaction without requiring complex offline computation of terminal sets. The framework is validated in a high-fidelity EnergyPlus simulation, controlling a multi-zone building for a DR task. The results demonstrate that the proposed method with guarantees achieves near-perfect constraint satisfaction - a reduction of over 99% in violations compared to the baseline - at the cost of only a minor increase in electricity expenditure. This work provides a deployable, performance-driven control solution for building energy management and establishes a new pathway for developing verifiable learning-based control systems under milder assumptions.