DeepAI AI Chat
Log In Sign Up

Contracting Neural-Newton Solver

by   Samuel Chevalier, et al.

Recent advances in deep learning have set the focus on neural networks (NNs) that can successfully replace traditional numerical solvers in many applications, achieving impressive computing gains. One such application is time domain simulation, which is indispensable for the design, analysis and operation of many engineering systems. Simulating dynamical systems with implicit Newton-based solvers is a computationally heavy task, as it requires the solution of a parameterized system of differential and algebraic equations at each time step. A variety of NN-based methodologies have been shown to successfully approximate the dynamical trajectories computed by numerical time domain solvers at a fraction of the time. However, so far no previous NN-based model has explicitly captured the fact that any predicted point on the time domain trajectory also represents the fixed point of the numerical solver itself. As we show, explicitly capturing this property can lead to significantly increased NN accuracy and much smaller NN sizes. In this paper, we model the Newton solver at the heart of an implicit Runge-Kutta integrator as a contracting map iteratively seeking this fixed point. Our primary contribution is to develop a recurrent NN simulation tool, termed the Contracting Neural-Newton Solver (CoNNS), which explicitly captures the contracting nature of these Newton iterations. To build CoNNS, we train a feedforward NN and mimic this contraction behavior by embedding a series of training constraints which guarantee the mapping provided by the NN satisfies the Banach fixed-point theorem; thus, we are able to prove that successive passes through the NN are guaranteed to converge to a unique, fixed point.


page 1

page 2

page 3

page 4


Optimal Power Flow Based on Physical-Model-Integrated Neural Network with Worth-Learning Data Generation

Fast and reliable solvers for optimal power flow (OPF) problems are attr...

Local error quantification for efficient neural network dynamical system solvers

Neural Networks have been identified as potentially powerful tools for t...

A Quasi-Newton method for physically-admissible simulation of Poiseuille flow under fracture propagation

Coupled hydro-mechanical processes are of great importance to numerous e...

Proximal Implicit ODE Solvers for Accelerating Learning Neural ODEs

Learning neural ODEs often requires solving very stiff ODE systems, prim...

Towards Hardware Implementation of Neural Network-based Communication Algorithms

There is a recent interest in neural network (NN)-based communication al...

Analysis of a new implicit solver for a semiconductor model

We present and analyze a new iterative solver for implicit discretizatio...