Feature extraction on an intelligent polycrystalline diamond insert clock testing method and prediction of product performance

Abstract

Increasing applications of polycrystalline diamond inserts in rock drilling is visibly seen as the discovery of new oil wells and tunnelling projects for metro lines are on a continuous rise around the world. As a result, the market consumption of polycrystalline diamond inserts is increasing severely. However, the sudden increased requirement of polycrystalline diamond inserts has also triggered a global competition as new players are increasing each day. The prevailing situation offers a dynamic challenge to the manufacturers to successfully stay in the business, and hence, enhancing product quality has become an essential requirement. In other words, to stay competitive and to remain ahead of the pack, it is critical to build up innovative testing capabilities of the polycrystalline diamond insert so as to pre-empty the undesirable functional characteristics of the polycrystalline diamond insert as well as to proactively engage the production floor for ensuring high product quality. This manuscript unveils a developed intelligent polycrystalline diamond insert testing platform that would link the failure characteristics of the polycrystalline diamond insert to the fracture mechanics through the study of process digitisation of the tool–work interface. An experimental set-up was developed, which incorporates a dynamometer, acoustic emission and accelerometer, for the digitisation of data signals in a feature extraction engine. The feature extraction engine in turn is used to monitor the failure of polycrystalline diamond cutting inserts during machining. The raw data fed through the feature extraction engine were used to identify the progression of failure in terms of flank wear or tool life for the polycrystalline diamond cutting inserts. The system comprises three key elements, which are (a) sensing and conditioning, (b) information extraction and (c) performance and failure analysis. The results of this experiment build the feature extraction engine that tracks the progression of flank wear in the polycrystalline diamond cutting inserts with reasonable accuracy. Furthermore, the break-in testing of a randomly selected insert from the production floor was applied to the feature extraction engine platform to predict the produce performance. This method also alerts the in-process manufacturing stages and enabled to considerably reduce the production scrap of polycrystalline diamond inserts.