Causal Mechanism Transfer Network for Time Series Domain Adaptation in Mechanical Systems

TL;DR

This paper introduces Causal Mechanism Transfer Network (CMTN), a novel approach for time series domain adaptation in mechanical systems that leverages invariant causal mechanisms to improve model transferability across different equipment with limited data.

Contribution

The paper proposes CMTN, a new method that captures and transfers causal mechanisms in time series data, addressing limitations of existing domain adaptation techniques for dynamic and causal data.

Findings

CMTN outperforms state-of-the-art methods in case studies.

Effective transfer of causal mechanisms improves model reliability.

Reduces data requirements for new systems.

Abstract

Data-driven models are becoming essential parts in modern mechanical systems, commonly used to capture the behavior of various equipment and varying environmental characteristics. Despite the advantages of these data-driven models on excellent adaptivity to high dynamics and aging equipment, they are usually hungry to massive labels over historical data, mostly contributed by human engineers at an extremely high cost. The label demand is now the major limiting factor to modeling accuracy, hindering the fulfillment of visions for applications. Fortunately, domain adaptation enhances the model generalization by utilizing the labelled source data as well as the unlabelled target data and then we can reuse the model on different domains. However, the mainstream domain adaptation methods cannot achieve ideal performance on time series data, because most of them focus on static samples and even the existing time series domain adaptation methods ignore the properties of time series data, such as temporal causal mechanism. In this paper, we assume that causal mechanism is invariant and present our Causal Mechanism Transfer Network(CMTN) for time series domain adaptation. By capturing and transferring the dynamic and temporal causal mechanism of multivariate time series data and alleviating the time lags and different value ranges among different machines, CMTN allows the data-driven models to exploit existing data and labels from similar systems, such that the resulting model on a new system is highly reliable even with very limited data. We report our empirical results and lessons learned from two real-world case studies, on chiller plant energy optimization and boiler fault detection, which outperforms the existing state-of-the-art method.