Tool Condition Monitoring (TCM) methods have shown significant potential to automatically detect worn tools without intervention in the machining process, thus decreasing machine downtime and improving reliability and part quality. Previous research on TCM systems have used a wide variety of time-domain and frequency-domain features extracted from cutting force related parameters as well as mechanical and acoustical vibrations to infer the wear state of tools. This project concerns the process of drilling thousands of tight-tolerance holes on tubesheets and baffles of heat exchangers using large diameter indexable insert drills on a horizontal boring machine. To address the issues involved in the process, the aim of this research is to develop a non-intrusive, indirect, online TCM system on the horizontal boring machine to monitor the drill wear and hole quality while drilling. The specific objectives are to establish an indirect TCM system for the drilling process, to develop models to predict tool wear and the machining accuracy of the drilled holes, and to develop an optimum tool replacement strategy.
The TCM system developed used two cutting-force related signals on the horizontal boring machine, namely the spindle motor current and the axial feed motor current. Features extracted from these data streams, as well as the machining parameters, the cutting speed and the feed rate, and the number of holes drilled with the current inserts, are the inputs to a series of models to predict the tool wear state and the hole diameter. The first model is an autoregressive model that allows the prediction of the extracted features for the next hole before it is drilled. As each hole is drilled, this model is updated with the most recent data to improve the accuracy of the prediction. The predicted values for the features are then used as inputs to the second and third models which are surface response models, one to estimate the tool wear state and one to estimate the hole diameter.
A tool replacement strategy based on applying limits to the predicted hole diameter was also developed. Adjusting these limits allows the strategy to be tuned for either hole accuracy or tool life depending on the requirements of a specific application. Tuning the replacement strategy for tool life resulted in a significant 44% increase in tool life and a non-trivial reduction in machine down time due to fewer tool changes while holding a hole diameter tolerance of ±0.1mm. The TCM system ensured that not a single over tolerance hole would have been drilled which is critically important since over tolerance holes can result in a scrapped workpiece.
The proposed 3-model TCM system shows promise in being able to significantly reduce the risk of drilling out of tolerance holes while at the same time increasing tool life and correspondingly decreasing tool change time. The models are able to accurately predict the insert flank wear and as well as the actual hole diameter within acceptable error. The TCM system could be implemented in an industrial settingwith minimal revision and since it is an indirect system there would be no intrusion into the manufacturing operation.
One limitation of the TCM system as proposed is that it is only capable of detecting gradual tool wear and not catastrophic tool failure, a limitation that was known from the outset but was not investigated as it was beyond the scope of this project. The proposed TCM system would allow the integration of additional functionality to instantaneously detect catastrophic tool failure.
Finally, for use in a production environment, the developed models need to be implemented on a standalone device that requires essentially no operator input to monitor continuous drilling operations for tubesheet and baffle applications. This implementation could include automatic detection of the machining parameters using frequency analysis of the motor signals.
| Identifer | oai:union.ndltd.org:USASK/oai:ecommons.usask.ca:10388/ETD-2013-09-1240 |
| Date | 2013 September 1900 |
| Contributors | Chen, Daniel |
| Source Sets | University of Saskatchewan Library |
| Language | English |
| Detected Language | English |
| Type | text, thesis |
Page generated in 0.0076 seconds