Among the many safety hazards facing chainsaw operators, the phenomenon known as kickback is the most dangerous. Kickback occurs when the chain at the tip of the chainsaw is caused to stop abruptly, and transfers the energy of the cutting chain to motion of the saw. The saw will rotate backward toward the operator rapidly. The limited amount of published research on the topic of chainsaw kickback was conducted to develop standardized testing for consumer chainsaws. Modern chainsaws are equipped with safety measures such as low-kickback cutting chains and hand-guard braking mechanisms. These mechanisms have greatly improved the safety of chainsaws, but their inherent mechanical simplicity leaves room for improvement.
The current work presents the research that analyzed the possible methods for detecting kickback electronically. Phase 1 of this work utilized a set of two accelerometers and a single gyroscope to determine if it is possible to distinguish a kickback event from normal cutting operations. A method for applying weighting coefficients to the three sensor readings, then summing the three signal values was optimized to obtain the greatest margin between kickback and normal cutting. The result of this study was that kickback is most easily identified by using only a gyroscope and setting a threshold. Phase 2 focused on detecting kickback as early as possible. Three methods were attempted: Signal Differentiation, a Simplified Bag of Words method, and applying a Support Vector Machine with selective undersampling and a stack of classifier vectors. Signal differentiation, while detecting the kickback events earlier, also suffered from many false positives. The Bag of Words method was unsuccessful in creating results different than the threshold method from Phase 1. The Support Vector Machine classification was able to detect kickback an average of 19.4 ms before the simple threshold method with no occurrence of either false positives or false negatives. This method is the most reliable and provides the greatest likelihood of detecting kickback early. / Graduation Date: 2013
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/36071 |
| Date | 27 November 2012 |
| Creators | Arnold, Drew D. |
| Contributors | Parmigiani, John |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.002 seconds