Autonomous Electrochemistry Platform with Real-Time Normality Testing of Voltammetry Measurements Using ML

TL;DR

The paper addresses the fragmentation of instrument-software ecosystems by proposing an autonomous electrochemistry computing platform (ICE) that links synthesis, delivery, testing, and analytics across multiple facilities. It combines a SwingXL-based synthesis workstation, a Kuka mobile robot, and remote compute with ML-driven real-time normality checking of voltammetry data, using a Gaussian process regression-based feature extractor and a fused-classifier ensemble to detect abnormal CV profiles. The key contributions include the cross-facility ICE architecture, remote orchestration of tasks, a fixed-size 10-d feature representation for variable-length I-V data, and a generalization-guaranteed ML framework for CV normality detection. Experimental results demonstrate effective integration, real-time data transfer, and accurate abnormality detection across multiple failure modes, underpinned by analytical generalization equations. This work enables long-running, autonomous electrochemistry campaigns across distributed labs, potentially accelerating catalyst discovery and optimization.

Abstract

Electrochemistry workflows utilize various instruments and computing systems to execute workflows consisting of electrocatalyst synthesis, testing and evaluation tasks. The heterogeneity of the software and hardware of these ecosystems makes it challenging to orchestrate a complete workflow from production to characterization by automating its tasks. We propose an autonomous electrochemistry computing platform for a multi-site ecosystem that provides the services for remote experiment steering, real-time measurement transfer, and AI/ML-driven analytics. We describe the integration of a mobile robot and synthesis workstation into the ecosystem by developing custom hub-networks and software modules to support remote operations over the ecosystem's wireless and wired networks. We describe a workflow task for generating I-V voltammetry measurements using a potentiostat, and a machine learning framework to ensure their normality by detecting abnormal conditions such as disconnected electrodes. We study a number of machine learning methods for the underlying detection problem, including smooth, non-smooth, structural and statistical methods, and their fusers. We present experimental results to illustrate the effectiveness of this platform, and also validate the proposed ML method by deriving its rigorous generalization equations.

Figures (14)

• Figure 1: ORNL Electrochemistry ecosystem consists of instruments and computing facilities connected over networks.
• Figure 2: ACL facility map generated by Kuka robot using its LiDAR range sensor system, which is used for navigation.
• Figure 3: Electrochemistry ecosystem integration design
• Figure 4: Electrochemistry workflow components
• Figure 5: I-V measurements under normal, and reference, working, counter disconnected experimental conditions.