About

The several components in a hydraulic system that is of interest in the ESTV project is shown. The components are Pump, Shaft, Bearing, Cylinder, Valve, Filter, and Electric Motor. The progress related to each component in terms of literature, simulation model, test rig and validation, Single fault classification, and multi-fault classification is shown using three colors. Green is for completed, yellow for in progress and red for did not start respectively. The current progress is in pumps, valve, cylinder, and filter.

The ESTV project focuses on intelligent techniques in condition monitoring of Electro-hydraulic systems. The mind map shows the components of interest and the progress.

Background

Electromechanical energy conversion systems, or powertrains are an essential part of many industries. E.g., wind turbines, electric vehicles, and hydraulic units contain powertrains, which consist of electrical machines, gearboxes, shafts, and other components. Failures occurring at any place of a powertrain incur significant economic loss and endanger the safety of humans and the environment.

Goals

Goal.1: Develop a multi-signal fault diagnosis framework based on high-dimensional non-linear prediction models, such as deep nets or networked exponential families, and transfer learning.

Goal.2: Build a diagnosis system, which can locate the failure, segregate between different simultaneous failures at different loads, and estimate the remaining useful life time (RUL) of different components.

Goal.3: Develop a physics-based method for generating training data, based on data augmentation from a reduced amount of experiments.

The utmost goal is an intelligent and accurate condition monitoring system for electromechanical energy conversion powertrains.

Impact

The proposal aims at developing condition monitoring methods and systems for powertrains. Within the proposal several simulation environments and methods will be developed. The condition monitoring systems can be easily deployed in different sectors of industry and thus provide it with a mean for economic growth. Furthermore, it impacts the well-being through secure use of the energy conversion systems and electric vehicles as well as reduced need for scheduled maintenance, e.g., in offshore wind farms, which is very risky for the personnel. The developed simulations methods will make it possible to design robust, efficient, and affordable powertrains. The research on Intelligent Systems will increase the understanding of how these systems behave when used in energy conversion systems and thus release the anxiety of the population regarding AI in general.

Funding

This work was enabled by the financial support of Academy of Finland (project ESTV) and internal funding from the Department of Automation Technology and Mechanical Engineering, IHA group at Tampere University, Finland.

Partners and co-operators

Aalto University

VTT Technical Research Centre of Finland