At-side flood frequency analysis was carried out for 64 unregulated streamflow stations laocated in various regions of Thailand with each data set having at least 20 observations. Using the extreme value type 1 (EV1), general extreme value (GEV) and log-logistic (LLG) distributions as the parent models, along with commonly used methods of parameter estimation, it was found that the GEV is the most suitable distribution, and the method of probability weighted moments is the most desirable method for parameter estimation in this case

At-site flood frequency analysis for Thailand

Piecewise continuous quadratic interpolation

This article presents a computer procedure for modelling the mixing characteristic in a reactor. The basic approach employed in this study is to model the mixing pattern of a real reactor by using an equivalent reactor system consisting of several ideal continuous stirred tank reactors (CSTRs) placed in series. The concentration profile yielded by this CSTRs system is compared with the actual profile of the real reactor obtained experimentally. The concentration profile of the CSTRs system that is closest to the actual profile is then taken as the system that will adequately represent the mixing characteristic of the actual reactor. The proposed procedure is tested on an IBM personal computer

S. L. Ong

Papers

Data analysis via generalised continuous parabolic interpolation

Program DSSRSSR for the solution of nonlinear optimisation models

An efficient program for the design of a segmentation tank