Streamline segment analysis of turbulent premixed flames

Abstract

In this study the streamline segment analysis has been carried out for turbulent premixed flames in order to analyze the statistical properties of the flame surface, which are associated with the nearby extremal points. The analysis is based on a DNS database of freely propagating turbulent premixed flames with a wide variation of turbulence and combustion parameters, spanning different combustion regimes. Typically a streamline segment can be characterized by its curve length l and the difference of the velocity magnitude at two extremal points Du. It has been found that an expression for the probability density function (PDF) of l, originally developed for non-reacting turbulent flows, remains valid for turbulent premixed flames in the thin reaction zones regime at high values of turbulent Reynolds number. However, the agreement between the theoretical prediction and the PDF of l obtained from DNS data worsens with decreasing turbulent Reynolds number, as well as in the corrugated flamelets regime. The joint PDFs between l and Du and between l and Dqu have been analyzed in detail, and it has been found that turbulence intensity, heat release parameter and the regime of combustion have significant effects on these joint PDFs. Detailed explanations have been provided for the observed differences in the topological behaviors of the streamline segment statistics in response to the combustion kinematics and dynamics.

DNS assessment of relation between mean reaction and scalar dissipation rates in the flamelet regime of premixed turbulent combustion

“…Nevertheless, the range of Re t ≈ 100 reached in the DNS by Nishiki et al [17,18] is typical even for recent simulations. For instance, among eight contributions that aimed at the DNS of premixed turbulent flames and were published in the proceedings of the recent 34th Combustion Symposium [12,[29][30][31][32][33][34][35], Re t was substantially larger than 100 in only two papers [29,34], with the combustion chemistry being reduced to a single reaction in five papers [12,30,32,33,35].…”

A direct numerical simulation study of vorticity transformation in weakly turbulent premixed flames

“…Accordingly, the range of Re t ≈ 100 reached in the DNS by Nishiki et al 28,36 is typical even for recent simulations. For instance, among eight papers that aimed at DNS of premixed turbulent flames and were published in the proceedings of the latest 34th Combustion Symposium, 45,[47][48][49][50][51][52][53] Re t was substantially larger than 100 only in two papers, 45,52 with the combustion chemistry being reduced to a single reaction in five papers. 47,[49][50][51]53 We may also note that the DNS data by Nishiki et al 28,36 were analyzed in several recent papers.…”

“…For instance, among eight papers that aimed at DNS of premixed turbulent flames and were published in the proceedings of the latest 34th Combustion Symposium, 45,[47][48][49][50][51][52][53] Re t was substantially larger than 100 only in two papers, 45,52 with the combustion chemistry being reduced to a single reaction in five papers. 47,[49][50][51]53 We may also note that the DNS data by Nishiki et al 28,36 were analyzed in several recent papers. [40][41][42] Nevertheless, the DNS by Nishiki et al 28,36 suffered from a low ratio of the transverse width yz = 4 mm of the computational domain to the integral length scale L = 3.5 mm of turbulence.…”

“…Nevertheless, as far as the goals of the present study are concerned, this limitation should not be overestimated, because (i) the governing phenomena, i.e., the baroclinic torque and dilatation, are localized to flame fronts, which are much thinner than the width yz , and (ii) small-scale effects are commonly less sensitive to the size of the computational domain. On the other hand, a low ratio of yz /L allowed Nishiki et al 28,36 to reach sufficiently large ratios of L/δ L , whereas the vast majority of recent DNSs 44,45,[47][48][49][50][51][52][53] …”

“…While the former case was selected in vast majority of recent DNS studies, 44,45,[47][48][49][50][51][52][53] the latter case addressed, in particular, by Nishiki et al 28,36 is also worth investigating. To conclude this section, let us compare the present simulations with previous DNS research [31][32][33][34] into vorticity transformation in premixed turbulent flames.…”