# Differentiable modelling and optimization of multi-planar slicing for multi-axis additive manufacturing

## TL;DR

This paper introduces a new method for optimizing multi-planar slicing in additive manufacturing to reduce distortion in printed parts.

## Contribution

A novel differentiable formulation for multi-planar slicing in additive manufacturing is introduced, enabling gradient-based optimization.

## Key findings

- The multi-planar deposition approach reduces distortion by an order of magnitude compared to conventional planar strategies.
- The method was tested on complex geometries including holes, overhangs, and underhangs with promising results.

## Abstract

Multi-planar deposition, enabled by multi-axis additive manufacturing, provides an opportunity to address challenging issues in wire arc additive manufacturing, such as residual stresses and distortions. This strategy involves sequentially building sub-parts, by depositing material in each sub-part with a distinct printing direction. In this paper, we present a novel continuous and differentiable formulation to model the multi-planar slicing strategy. The strategy is parameterized using a pseudo-time field, which allows the part to be segmented into sub-parts. An orientation field is used to define the distinct printing direction for each sub-part. This differentiable formulation enables gradient-based optimization of the multi-planar slicing. We apply the method to reduce distortion in wire arc additive manufacturing. The method is tested on several numerical examples with complex geometries, including holes, overhangs, and underhangs. Numerical results show that the multi-planar deposition approach reduces distortion by an order of magnitude compared with the conventional planar strategy.