EFFECT OF ELECTRICAL DISCHARGE PATTERN ON SPARK IGNITED FLAME KERNEL DEVELOPMENT

Fuel-lean combustion and exhaust gas recirculation (EGR) in spark ignition engines improve engine efficiency and reduce emission. However, flame initiation becomes more difficult in lean and dilute fuel-air mixture with traditional spark discharge. This research proposal will first provide an intensive review on topics related to spark ignition including properties of electrical discharge, flame kernel behavior and spark ignition modeling and simulation. Focus will be laid on electrical discharge pattern effect as it is showing prospect in extending ignition limits in SI engines. An experimental setup has been built with an optically accessible constant volume combustion vessel. Multiple imaging techniques as well as spectroscopy will be applied. By varying spark discharge patterns, preliminary test results are available on consequent flame kernel development. In addition to experimental investigation of spark plasma and flame kernel development, spark ignition modeling with detailed description of plasma channel is also proposed for this study.

Characterization of water-soluble organic compounds in ambient aerosol using ultrahigh-resolution elctrospray ionization fourier transform ion cyclotron resonance mass spectrometry.

Atmospheric aerosol water-soluble organic compounds (WSOC) exist in a complex mixture of thousands of organic compounds which may have a significant influence on the climate-relevant properties of the atmospheric aerosol. To understand the potential influences, the ambient aerosol was collected at a nonurban mountainous site near Steamboat Springs, CO. The WSOC fraction was analyzed using positive and negative electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry. Approximately 2400 and 4000 molecular formulas were identified from the detected positive and negative ions, respectively. The formulas contained carbon (C), hydrogen (H), oxygen (O), nitrogen (N), and sulfur (S) atoms over the mass range of 100-800 Da in both ionization modes. The number range of double bond equivalents (DBE), the mean O:C, H:C, and oxidation state of carbon for the positive ions were 0 – 18, 0.25 ± 0.15, 1.39 ± 0.29, and -0.89 ± 0.23, respectively. Comparatively, the negative ion values were 0 – 14, 0.53 ± 0.20, 1.48 ± 0.30, and -0.41 ± 0.45, respectively. Overall, the positive ion molecular formulas were less oxygenated than negative ions as seen with the lower O:C and OSc values. Molecular formulas of the positive ions classified as aliphatic, olefinic, and aromatic compound classes based on the aromaticity index values. Aliphatic compounds were the CHNO and CHO formulas that had mean DBE values of about 5 and 3, respectively. However, a majority of the CHOS, CHNOS, and CHS formulas were defined as olefinic compounds and had mean DBE values of about 12, 13, and 10, respectively. Overall, more than half of the assigned molecular formulas contained sulfur and were olefinic to aromatic compounds with a DBE range of 7-18. Source of the unsaturated sulfur containing compounds is currently unknown. Several nitrogen containing compounds were in common with the field and laboratory studies of the biomass burning aerosol and aged secondary organic aerosol products of the limonene ozonolysis.

SCALABLE APPROXIMATION OF KERNEL FUZZY C-MEANS

Virtually every sector of business and industry that uses computing, including financial analysis, search engines, and electronic commerce, incorporate Big Data analysis into their business model. Sophisticated clustering algorithms are popular for deducing the nature of data by assigning labels to unlabeled data. We address two main challenges in Big Data. First, by definition, the volume of Big Data is too large to be loaded into a computer’s memory (this volume changes based on the computer used or available, but there is always a data set that is too large for any computer). Second, in real-time applications, the velocity of new incoming data prevents historical data from being stored and future data from being accessed. Therefore, we propose our Streaming Kernel Fuzzy c-Means (stKFCM) algorithm, which reduces both computational complexity and space complexity significantly. The proposed stKFCM only requires O(n2) memory where n is the (predetermined) size of a data subset (or data chunk) at each time step, which makes this algorithm truly scalable (as n can be chosen based on the available memory). Furthermore, only 2n2 elements of the full N × N (where N >> n) kernel matrix need to be calculated at each time-step, thus reducing both the computation time in producing the kernel elements and also the complexity of the FCM algorithm. Empirical results show that stKFCM, even with relatively very small n, can provide clustering performance as accurately as kernel fuzzy c-means run on the entire data set while achieving a significant speedup.