Data Analysis and Compression Techniques for Megabyte - Data Pde Experiments

Pulsed detonation engines have the potential to provide thrust over a wide operating range for a fraction of the cost of conventional turbine engines. These promised returns have given rise to several vigorous research programs in academia, industry, and government labs. To capture the ultra-fast events comprising a detonation wave, researchers are forced to use megahertz range data acquisition systems for relatively long testing intervals, producing gigabytes of purely numeric test data. Since human inspection of such enormous data sets is virtually impossible, a computational framework of highly automated analysis tools is necessary to facilitate the interpretation and analysis of pulsed detonation engine experiments. A variety of software algorithms can be applied to produce meaningful output for rapid human assimilation. Dynamic pressure transducer data can be filtered to remove nonlinear drifting of the baseline, scanned to locate each captured detonation, and condensed to produce small, highly representative point sets for subsequent analysis. Wave speeds, firing frequencies, thrust estimates, and output shaft work can all be automatically calculated through the use of intelligent, adaptive analysis routines. Data from different sensors, including high speed imaging systems, can also be combined into visual, animated interpretations to facilitate rapid qualitative understanding of experimental results. Finally, digital finite impulse response filters, cycle time reference signals, and floating average reference lines can all be utilized to eliminate noise and extract equivalence ratio data from passive OH emission and active hydrocarbon absorption sensors. By preprocessing these complex signal files with automated routines, the cost-effectiveness of data intensive research programs can be maintained and key results can be successfully extracted from a myriad of noisy, machine-readable raw data.