An Adaptive Machine Learning Decision System for Flexible Predictive Maintenance

Process monitoring and Predictive Maintenance (PdM) are gaining increasing attention in most manufacturing environments as a means of reducing maintenance related costs and downtime. This is especially true in industries that are data intensive such as semiconductor manufacturing. In this paper an adaptive PdM based flexible maintenance scheduling decision support system, which pays particular attention to associated opportunity and risk costs, is presented. The proposed system, which employs Machine Learning and regularized regression methods, exploits new information as it becomes available from newly processed components to refine remaining useful life estimates and associated costs and risks. The system has been validated on a real industrial dataset related to an Ion Beam Etching process for semiconductor manufacturing.

In situ diagnostics and prognostics of solder fatigue in IGBT modules for electric vehicle drives

This paper proposes an in situ diagnostic and prognostic (D&P) technology to monitor the health condition of insulated gate bipolar transistors (IGBTs) used in EVs with a focus on the IGBTs' solder layer fatigue. IGBTs' thermal impedance and the junction temperature can be used as health indicators for through-life condition monitoring (CM) where the terminal characteristics are measured and the devices' internal temperature-sensitive parameters are employed as temperature sensors to estimate the junction temperature. An auxiliary power supply unit, which can be converted from the battery's 12-V dc supply, provides power to the in situ test circuits and CM data can be stored in the on-board data-logger for further offline analysis. The proposed method is experimentally validated on the developed test circuitry and also compared with finite-element thermoelectrical simulation. The test results from thermal cycling are also compared with acoustic microscope and thermal images. The developed circuitry is proved to be effective to detect solder fatigue while each IGBT in the converter can be examined sequentially during red-light stopping or services. The D&P circuitry can utilize existing on-board hardware and be embedded in the IGBT's gate drive unit.

Real-time temperature estimation for power MOSFETs considering thermal aging effects

This paper presents a novel real-time power-device temperature estimation method that monitors the power MOSFET's junction temperature shift arising from thermal aging effects and incorporates the updated electrothermal models of power modules into digital controllers. Currently, the real-time estimator is emerging as an important tool for active control of device junction temperature as well as online health monitoring for power electronic systems, but its thermal model fails to address the device's ongoing degradation. Because of a mismatch of coefficients of thermal expansion between layers of power devices, repetitive thermal cycling will cause cracks, voids, and even delamination within the device components, particularly in the solder and thermal grease layers. Consequently, the thermal resistance of power devices will increase, making it possible to use thermal resistance (and junction temperature) as key indicators for condition monitoring and control purposes. In this paper, the predicted device temperature via threshold voltage measurements is compared with the real-time estimated ones, and the difference is attributed to the aging of the device. The thermal models in digital controllers are frequently updated to correct the shift caused by thermal aging effects. Experimental results on three power MOSFETs confirm that the proposed methodologies are effective to incorporate the thermal aging effects in the power-device temperature estimator with good accuracy. The developed adaptive technologies can be applied to other power devices such as IGBTs and SiC MOSFETs, and have significant economic implications. 

An Integrated Toolkit for the Conservation of Stone-Built Heritage

Rapid in situ diagnosis of damage is a key issue in the preservation of stone-built cultural heritage. This is evident in the increasing number of congresses, workshops and publications dealing with this issue. With this increased activity has come, however, the realisation that for many culturally significant artefacts it is not possible either to remove samples for analysis or to affix surface markers for measurement. It is for this reason that there has been a growth of interest in non-destructive and minimally invasive techniques for characterising internal and external stone condition. With this interest has come the realisation that no single technique can adequately encompass the wide variety of parameters to be assessed or provide the range of information required to identify appropriate conservation. In this paper we describe a strategy to address these problems through the development of an integrated `tool kit' of measurement and analytical techniques aimed specifically at linking object-specific research to appropriate intervention. The strategy is based initially upon the acquisition of accurate three-dimensional models of stone-built heritage at different scales using a combination of millimetre accurate LiDAR and sub-millimetre accurate Object Scanning that can be exported into a GIS or directly into CAD. These are currently used to overlay information on stone characteristics obtained through a combination of Ground Penetrating Radar, Surface Permeametry, Colorimetry and X-ray Fluorescence, but the possibility exists for adding to this array of techniques as appropriate. In addition to the integrated three-dimensional data array provided by superimposition upon Digital Terrain Models, there is the capability of accurate re-measurement to show patterns of surface loss and changes in material condition over time. Thus it is possible to both record and base-line condition and to identify areas that require either preventive maintenance or more significant pre-emptive intervention. In pursuit of these goals the authors are developing, through a UK Government supported collaboration between University Researchers and Conservation Architects, commercially viable protocols for damage diagnosis, condition monitoring and eventually mechanisms for prioritizing repairs to stone-built heritage. The understanding is, however, that such strategies are not age-constrained and can ultimately be applied to structures of any age.

Improved fault diagnosis in multivariate systems using regression-based reconstruction

Subspace monitoring has recently been proposed as a condition monitoring tool that requires considerably fewer variables to be analysed compared to dynamic principal component analysis (PCA). This paper analyses subspace monitoring in identifying and isolating fault conditions, which reveals that the existing work suffers from inherent limitations if complex fault senarios arise. Based on the assumption that the fault signature is deterministic while the monitored variables are stochastic, the paper introduces a regression-based reconstruction technique to overcome these limitations. The utility of the proposed fault identification and isolation method is shown using a simulation example and the analysis of experimental data from an industrial reactive distillation unit.

The use of process simulation methods in support of organisational learning in availability contracting

Traditional business models in the aerospace industry are based on a conventional supplier to customer relationship based on the design, manufacture and subsequent delivery of the physical product. Service provision, from the manufacturer's perspective, is typically limited to the supply of procedural documentation and the provision of spare parts to the end user as the product passes through the latter stages of its intended lifecycle. Challenging economic and political conditions have resulted in end users re-structuring their core business activities, particularly in the defence sector. This has resulted in the need for original equipment manufacturers (OEMs) to integrate and manage support service activities in partnership with the customer to deliver platform availability. This improves the probability of commercial sustainability for the OEM through shared operational risks while reducing the cost of platform ownership for the customer. The need for OEMs to evolve their design, manufacture and supply strategies by focusing on customer requirements has revealed a need for reconstruction of traditional internal behaviours and design methodologies. Application of organisational learning is now a well recognised principle for innovative companies to achieve long term growth and sustained technical development, and hence, greater market command. It focuses on the process by which the organisation's knowledge and value base changes, leading to improved problem solving ability and capacity for action. From the perspective of availability contracting, knowledge and the processes by which it is generated, used and retained, become primary assets within the learning organisation. This paper will introduce the application of digital methods to asset management by demonstrating how the process of learning can benefit from a digital approach, how product and process design can be integrated within a virtual framework and finally how the approach can be applied in a service context.

Extruder Melt Temperature Control with Fuzzy Logic

In polymer extrusion, the delivery of a melt which is homogenous in composition and temperature is paramount for achieving high quality extruded products. However, advancements in process control are required to reduce temperature variations across the melt flow which can result in poor product quality. The majority of thermal monitoring methods provide only low accuracy point/bulk melt temperature measurements and cause poor controller performance. Furthermore, the most common conventional proportional-integral-derivative controllers seem to be incapable of performing well over the nonlinear operating region. This paper presents a model-based fuzzy control approach to reduce the die melt temperature variations across the melt flow while achieving desired average die melt temperature. Simulation results confirm the efficacy of the proposed controller.

A research-based approach to exploring the adoption of novel UK-based concrete test methods in Chinese construction sector

As part of a UK-China science bridge project - a UK government funded initiative linking leading universities and businesses in selective partnering countries in 2009 a collaborative research programme was initiated between Queen's University and the Research Institute of High Performance Concrete (part of the Central Research Institute of Building and Construction) in Beijing.

For further details email b.magee@ulster.ac.uk

Going above and beyond: how sustainability culture and entrepreneurial orientation drive social sustainability supply chain practice adoption

Purpose
– This paper aims to examine what drives the adoption of different social sustainability supply chain practices. Research has shown that certain factors drive the adoption of environmental sustainability practices but few focus on social supply chain practices, delineate which practices are adopted or what drives their adoption.

Design/methodology/approach
– The authors examine the facilitative role of sustainability culture to explain the adoption of social sustainability supply chain practices: basic practices, consisting of monitoring and management systems and advanced practices, which are new product and process development and strategic redefinition. The authors then explore the role played by a firm’s entrepreneurial orientation in shaping and reinforcing the adoption of social sustainability supply chain practices. A survey of 156 supply chain managers in multiple industries in Ireland was conducted to test the relationship between the variables.

Findings
– The findings show that sustainability culture is positively related to all the practices, and entrepreneurial orientation impacts and moderates social sustainability culture in advanced social sustainability supply chain adoption.

Research limitations/implications
– As with any survey, this is a single point in time with a single respondent. Implications for managers include finding the right culture in the organisation to implement social sustainability supply chain management practices that go beyond monitoring to behavioural changes in the supply chain with implications beyond the dyad of buyer and supplier to lower tier suppliers and the community surrounding the supply chain.

Practical implications
– The implications for managers include developing and fostering cultural attributes in the organisation to implement social sustainability supply chain management practices that go beyond monitoring suppliers to behavioural changes in the supply chain with implications beyond the dyad of buyer and supplier to lower tier suppliers and the community surrounding the supply chain.

Originality/value
– This is the first time, to the authors’ knowledge, that cultural and entrepreneurial variables have been tested for social sustainability supply chain practices, giving them new insights into how and why social sustainability supply chain practices are adopted.