The study aims to characterize the engine knock phenemenon in a heavy fueled reciprocating engine. The effect of knock on the pressure and mechanical vibrations is observed, several known knock indices compared, and relevant frequencies identified.
High speed data aquisition is used to observe a knock event in the time-frequency domain of pressure and vibration signals. The non-stationary pressure signature is shown to be best suited to time-frequency analysis while the vibrational component was best observed using Welch power spectral density distributions.
Known pressure based time series knock intensity indices are compared to a smoothed reallocated Wigner-Ville energy distribution based method. The mean signal energy over a crank angle window yielded the highest correlation.
Accelerometer power bands of a series of intermittently knocking cycles are studied to indentify knock related vibration frequencies. Parasitic mechanical noise prevented adequate cycle-to-cycle prediction of pressure based knock intensity indices, but identified target frequency bands.
Over an operating range, an array of vibrations power bands are related to pressure based knock detection using regression analysis. Several frequency bands are identified, for each axis, for engine block vibrations based knock detection. The influence of engine order frequencies was quantified and found to have 1 and 6% impact on the correlation coefficient.
| Date | 11 Nov 2015 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Patrice Seers (Supervisor) |
|---|
Fleurent-Wilson, É. B. (Author),
Seers (Supervisor),
11 Nov 2015Student thesis: Master's thesis › Master in Engineering: Mechanical Engineering