This paper provides a methodology to assess the optimal multi-agent architecture for collaborative prognostics in modern fleets of assets. The use of multi-agent systems has been shown to improve the ability to predict equipment failures by enabling machines with communication and collaborative learning capabilities. Different architectures have been postulated for industrial multi-agent systems in general. A rigorous analysis of the implications of their implementation for collaborative prognostics is essential to guide industrial deployment. In this paper, we investigate the cost and reliability implications of using different multi-agent systems architectures for collaborative failure prediction and maintenance optimization in large fleets of industrial assets. Results show that purely distributed architectures are optimal for high-value assets, while hierarchical architectures optimize communication costs for low-value assets. This enables asset managers to design and implement multi-agent systems for predictive maintenance that significantly decrease the whole-life cost of their assets.
In recent years, the development of neutral helium microscopes has gained increasing interest. The low energy, charge neutrality, and inertness of the helium atoms makes helium microscopy an attractive candidate for the imaging of a range of samples. The simplest neutral helium microscope is the so-called pinhole microscope. It consists of a supersonic expansion helium beam collimated by two consecutive apertures (skimmer and pinhole), which together determine the beam spot size and hence the resolution at a given working distance to the sample. Due to the high ionization potential of neutral helium atoms, it is difficult to build efficient helium detectors. Therefore, it is crucial to optimize the microscope design to maximize the intensity for a given resolution and working distance. Here we present an optimization model for the helium pinhole microscope system. We show that for a given resolution and working distance, there is a single intensity maximum. Further we show that with present-day state-of-the-art detector technology (ionization efficiency 1×10-3), a resolution of the order of 600 nm at a working distance of 3 mm is possible. In order to make this quantification, we have assumed a Lambertian reflecting surface and calculated the beam spot size that gives a signal 100 cts/s within a solid angle of 0.02π sr, following an existing design. Reducing the working distance to the micron range leads to an improved resolution of around 40 nm
We present the first steps towards real-time distributed collaborative prognostics enabled by an implementation of the Weibull Time To Event-Recurrent Neural Network (WTTE-RNN) algorithm. In our system, assets determine their time to failure (TTF) in real-time according to an asset-specific model that is obtained in collaboration with other similar assets in the asset fleet. The presented approach builds on the emergent field of similarity analysis in asset management, and extends it to distributed collaborative prognostics. We show how through collaboration between assets and distributed prognostics, competitive time to failure estimates can be obtained. 1
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.