Deep Learning Explicit Differentiable Predictive Control Laws for Buildings

Drgoňa, Ján; Tuor, Aaron; Skomski, Elliott; Vasisht, Soumya; Vrabie, Draguna

doi:10.1016/j.ifacol.2021.08.518

Cited by 23 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…b) Differentiable optimization: Recently, there has been an emergence on the intersection of constrained optimization and deep learning we could label as learning to optimize methods. Supported problem types range from quadratic programs [18], stochastic optimization [19], submodular optimization [20], optimal control problems [21], or even combinatorial optimization problems [22], to name just a few. Reference [23] shows enforcement of hard constraints in learning the solution of constrained nonlinear optimization problems via deep neural networks.…”

Section: Related Workmentioning

confidence: 99%

Neuro-physical dynamic load modeling using differentiable parametric optimization

Abhyankar¹,

Drgoňa²,

August³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In this work, we investigate a data-driven approach for obtaining a reduced equivalent load model of distribution systems for electromechanical transient stability analysis. The proposed reduced equivalent is a neurophysical model comprising of a traditional ZIP load model augmented with a neural network. This neuro-physical model is trained through differentiable programming. We discuss the formulation, modeling details, and training of the proposed model set up as a differential parametric program. The performance and accuracy of this neurophysical ZIP load model is presented on a medium-scale 350-bus transmission-distribution network.

show abstract

Section: Related Workmentioning

confidence: 99%

Neuro-physical dynamic load modeling using differentiable parametric optimization

Abhyankar¹,

Drgoňa²,

August³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…DPC brings forth the idea of offline computation of explicit predictive control policies by leveraging automatic differentiation (AD) of the constrained optimization problem for direct computation of the policy gradients. In our previous work, we have demonstrated the scalability of the DPC framework on systems, including uncertainties and nonlinear constraints [32], [33]. Contributions.…”

Section: Introductionmentioning

confidence: 99%

Neural Lyapunov Differentiable Predictive Control

Mukherjee¹,

Drgoňa²,

Tuor³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

We present a learning-based predictive control methodology using the differentiable programming framework with probabilistic Lyapunov-based stability guarantees. The neural Lyapunov differentiable predictive control (NLDPC) learns the policy by constructing a computational graph encompassing the system dynamics, state and input constraints, and the necessary Lyapunov certification constraints, and thereafter using the automatic differentiation to update the neural policy parameters. In conjunction, our approach jointly learns a Lyapunov function that certifies the regions of state-space with stable dynamics. We also provide a sampling-based statistical guarantee for the training of NLDPC from the distribution of initial conditions. Our offline training approach provides a computationally efficient and scalable alternative to classical explicit model predictive control solutions. We substantiate the advantages of the proposed approach with simulations to stabilize the double integrator model and on an example of controlling an aircraft model.

show abstract

“…Differentiable predictive control (DPC) is an unsupervised learning-based method for learning explicit neural control laws for model predictive control (MPC) problems [3], [4]. DPC alleviates the computational burden of online MPC by learning a receding horizon controller offline.…”

Section: Introductionmentioning

confidence: 99%

Differentiable Predictive Control with Safety Guarantees: A Control Barrier Function Approach

Cortez¹,

Drgoňa²,

Tuor³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

We develop a novel form of differentiable predictive control (DPC) with safety and robustness guarantees based on control barrier functions. DPC is an unsupervised learning-based method for obtaining approximate solutions to explicit model predictive control (MPC) problems. In DPC, the predictive control policy parametrized by a neural network is optimized offline via direct policy gradients obtained by automatic differentiation of the MPC problem. The proposed approach exploits a new form of sampled-data barrier function to enforce offline and online safety requirements in DPC settings while only interrupting the neural network-based controller near the boundary of the safe set. The effectiveness of the proposed approach is demonstrated in simulation.

show abstract

Deep Learning Explicit Differentiable Predictive Control Laws for Buildings

Cited by 23 publications

References 19 publications

Neuro-physical dynamic load modeling using differentiable parametric optimization

Neuro-physical dynamic load modeling using differentiable parametric optimization

Neural Lyapunov Differentiable Predictive Control

Differentiable Predictive Control with Safety Guarantees: A Control Barrier Function Approach

Contact Info

Product

Resources

About