| Summary: | Thesis (M.Eng.)--Memorial University of Newfoundland, 1998. Engineering and Applied Science Bibliography: leaves 166-173 Factors affecting initial dilution and bacterial concentration at an area near the outfall discharge, e.g. sewage flow rate, seawater currents and bacterial decay, are highly variable. Because of this, a probabilistic approach for ocean outfall design and analysis is essential in predicting the performance of the outfall and in reflecting the probabilistic nature of the initial dilution and bacterial concentration. -- The intention of this thesis is to develop and apply a design procedure using a probabilistic method to calculate initial dilution and bacterial concentration at a location of interest. The scope of the study is directed at design and analysis for a horizontal buoyant round jet in a density unstratified seawater environment. Uncertainty of five parameters of design, i.e. sewage flow rate, tidal height, seawater currents, decay parameter, and bacterial concentration in the sewage before discharge into seawater are taken into account in this study. -- A comparison of the probabilistic approach with the deterministic approach shows that the probabilistic approach may provide a full range of possible values of the parameters of interest other than a fixed value. Associated probability values for the parameters of interest can also be obtained using the probabilistic methods. The procedure for outfall design using a probabilistic approach is straight forward, and may work in practice because the analysis of an existing outfall (the Spaniard's Bay Outfall, Newfoundland, Canada) has resulted in good agreement with field data. -- Comparison among the various probabilistic methods studied shows that all methods generally give the same answers for the case of initial dilution, except for a small probability of failure which is typically less 4 %. It is found that First Order Second Moment ( FOSM), Improved Mean-First Order Second Moment (IM-FOSM) and Advanced First Order Second Moment (AFOSM) with assumed normal parameters work well for use in analysis of initial dilution. In practice, the use of FOSM is recommended for its simplicity. -- For the case of bacterial concentration, FOSM gives poor results because the performance function in this case is complex and non linear, but AFOSM with assumed non-normal parameters is recommended. Monte Carlo Simulations (MCS) may also be used if a fast computer and software are available. It should be remembered that the choice of the probability method should consider the problem under investigation as well as the cost and facilities available.
|
|---|