An intelligent machine monitoring system for energy prediction using a Gaussian Process regression

Abstract

Abstract-Recent advances in machine automation and sensing technology offer new opportunities for continuous condition monitoring of an operating machine. This paper describes an intelligent machine monitoring framework that integrates and utilizes data collection, management, and analytics to derive an adaptive predictive model for the energy usage of a milling machine. This model is designed using a Gaussian Process (GP) regression algorithm, which is a flexible regression method that also provides an uncertainty estimate. To improve computational efficiency, we propose a Collective Gaussian Process (CGP) in which the overall energy prediction is made by constructing local GP models weighted by probability distribution functions obtained using the Gaussian Mixture Model (GMM) technique. Finally, we demonstrate the ability of the proposed monitoring framework to construct an energy prediction model to predict the energy used to machine a part.

Prediction of granular time-series energy consumption for manufacturing jobs from analysis and learning of historical data

“…The library is then able to return the most relevant historical profile via exact nearest neighbor. Similar work on relating process data with energy consumption information has shown it can be very useful for energy data processing [8], [11], however the machine conditions have not been considered until now.…”

“…Furthermore, a number of modeling approaches seen require machine specific values which, in an active production line, can be difficult to decipher. Work by Bhinge et al avoids this by solely using process parameters [8]. While this improves applicability, all the approaches reviewed do not consider the machine as a dynamic system.…”

A Pedagogical Module Framework to Improve Scaffolded Active Learning in Manufacturing Engineering Education

“…The instantaneous positions can be extracted from the block of code being processed in real-time. This information can be condensed and a block-by-block simulation can be demonstrated in real-time as the processing is unfolding in the machine tool (Bhinge et al, 2014).…”

“…Prior research has shown that real-time data can be used for machine tool monitoring to support sustainable manufacturing education. The MTConnect standard has emerged to efficiently extract real-time data and develop machine learning models for equipment and process characterization, and energy prediction and monitoring (Vijayaraghavan and Dornfeld, 2010;Bhinge et al, 2014). The MTConnect standard is an interoperability standard that facilitates archiving, accessing, and retrieving operational data from manufacturing equipment (MTConnect Institute, 2015).…”

Cloud-enabled prognosis for manufacturing

“…26 shows a typical MTConnect architecture deployed on the shop floor for remote monitoring. Using this architecture, data from the machine tool controller (e.g., actual position, actual and commanded feed, actual and commanded speed) and a variety of sensors (e.g., power meters, thermocouples, accelerometers) can be combined for many different remote monitoring, diagnosis, and prognosis applications, such as preventive maintenance [249], process planning verification [224], accurate cycle time estimation, tool position verification [225], and energy monitoring [223] and prediction [23]. Teti et al [205] provided an extensive list of industrial efforts for remote monitoring and diagnosis.…”