AONN: An adjoint-oriented neural network method for all-at-once solutions of parametric optimal control problems

TL;DR

This paper introduces AONN, an adjoint-oriented neural network approach that efficiently solves parametric optimal control problems governed by PDEs, handling complex constraints without iterative PDE solutions.

Contribution

AONN is a novel neural network method that models control, adjoint, and state functions simultaneously, avoiding penalty terms and enabling rapid computation of solutions for various parameters.

Findings

AONN effectively handles complex constraints in parametric control problems.

The method significantly reduces computational cost compared to traditional approaches.

Numerical experiments demonstrate high accuracy and efficiency of AONN.

Abstract

Parametric optimal control problems governed by partial differential equations (PDEs) are widely found in scientific and engineering applications. Traditional grid-based numerical methods for such problems generally require repeated solutions of PDEs with different parameter settings, which is computationally prohibitive especially for problems with high-dimensional parameter spaces. Although recently proposed neural network methods make it possible to obtain the optimal solutions simultaneously for different parameters, challenges still remain when dealing with problems with complex constraints. In this paper, we propose AONN, an adjoint-oriented neural network method, to overcome the limitations of existing approaches in solving parametric optimal control problems. In AONN, the neural networks are served as parametric surrogate models for the control, adjoint and state functions to get the optimal solutions all at once. In order to reduce the training difficulty and handle complex constraints, we introduce an iterative training framework inspired by the classical direct-adjoint looping (DAL) method so that penalty terms arising from the Karush-Kuhn-Tucker (KKT) system can be avoided. Once the training is done, parameter-specific optimal solutions can be quickly computed through the forward propagation of the neural networks, which may be further used for analyzing the parametric properties of the optimal solutions. The validity and efficiency of AONN is demonstrated through a series of numerical experiments with problems involving various types of parameters.