Optimisation of tool life through novel data acquisition and decision making techniques

Hill, Jacob L. 2020. Optimisation of tool life through novel data acquisition and decision making techniques. PhD Thesis, Cardiff University. Item availability restricted.

Preview	PDF - Accepted Post-Print Version Download (42MB) | Preview
	PDF (Cardiff University Electronic Publication Form) - Supplemental Material Restricted to Repository staff only Download (1MB)

Abstract

Variations in the operation and management of machine tool cutting processes will cause deviations in the quality of the manufactured parts. Current process management approaches combat these variations using combinations of pre- and/or post- process operator centred actions. The experience of the Author, and indications from involved industry partners, indicates that the associated ”conservative” approaches to tool life management is costing between five and ten percent of the money spent on cutting tools, here amounting to two million pounds per annum. The additional cost of quality arising from process related variations cannot be accurately assessed. This research enables the real time assessment of CNC milling cutting processes and the management of process variations. Innovative systems, programs and algorithms are developed through the course of this research project for the on-line monitoring of cutting tool health. These innovations include: the development of a cross-section area model to indicate variable metal removal in milling processes, the conversion of limited load data into process energy consumption, the engineering of an embedded tool wear data acquisition program, the application of an offline cubic change-point detection algorithm to quantitatively identify changes in cutting tool wear behaviour, the implementation of the density evaluation and separation algorithm to enable the separation of cutting and non-cutting process control signals, the development of novel Dispersion Plots, and the development of novel 3D process plots for illustrating instantaneous cutting tool condition. In support of these innovations specially defined methods of signal analysis are deployed to acquire information for the assessment of enabled and complex health features. The approach is autonomous and based upon learning from the acquisition and analysis of information directly from the machine controller. This approach limits the impact on the operation and availability of the machine tool and mitigates any further impact on the capacity of the machine tools in question. Decision making is enabled within the deployed diagnostic techniques. This provides the opportunity for plant-wide tool condition status monitoring and data visualisation. The deployed approach enables researchers to engineer machine systems that can provide more accurate, reliable and repeatable machine operations, with less waste and better managed processes. It is shown that there is significant value in the process control data that was acquired throughout this study. The data is used to show the deployed cutting tool condition based on current and imminent machining requirements. It is also deployed to estimate the expected end of useful life for specific cutting tools and to generate innovative models of the cutting process. These models will enable Engineers to improve the cutting processes and to optimise the assessment of cutting tool condition and life.

Item Type:	Thesis (PhD)
Date Type:	Completion
Status:	Unpublished
Schools:	Engineering
Uncontrolled Keywords:	CNC; Cutting tools; Process management; Digital signal processing; Control data; Tool condition monitoring.
Date of First Compliant Deposit:	28 May 2021
Last Modified:	04 Aug 2022 01:51
URI:	https://orca.cardiff.ac.uk/id/eprint/141623

Actions (repository staff only)

Edit Item