1 AIT Asian Institute of Technology

Optimal risk-based design of a flood control system

AuthorTuantan Kitpaisalsakul
Call NumberAIT-Diss. no. WM-96-02
Subject(s)Flood control--Mathematical models

NoteA dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Engineering, School of Civil Engineering
PublisherAsian Institute of Technology
Series StatementDissertation ; no. WM-96-02
AbstractRisk-based design is a design approach developed for quantifying the probability of failure and the expected annual damage cost of a structural system which is subject to the variation of actual load and available resistance. The approach considers the variation of both load and resistance and the interaction between load and resistance of the structure in the estimation of probability of failure. Moreover, the expected annual damage cost can be estimated by analyzing the variation of load and resistance along with the consequence function of load and resistance relating to damage cost of failure. The solution will be the optimal capacity of the structural system such as the flood control structure with maximum net benefit for a design return period of flood protection failure. The concept of optimal risk-based design is applicable not only to a single structure but also the combination of multiple structures such as a flood control system. In this study, an optimal risk-based design procedure of a flood control system is developed. An interesting contribution is that the design approach is a computational framework using four relevant techniques, namely flood flow simulation analysis, coincident flood frequency analysis, load-resistance analysis and optimization of risk-based design. In this design approach, the four techniques are linked together as a computational sequence for determining the optimal design of a flood control system. Large floodings in Bangkok and its suburban areas are mostly due to failure of existing river dikes to prevent intrusion of flood water from the Chao Phraya river such as the river flood overtopping the river dike into the protected area. Other floodings due to dike leakage and dike gate closure failure are also possible. It is considered that in addition to the variation of the river flood level due to various factors, the resistance of the river dikes is also variable due to combined natural and man made effects. The river flood level is excessively high during the simultaneous occurrence of the flood inflow from the north and high tide at the river mouth. As it is impossible to properly solve such floodings by a single flood control component, it becomes necessary to apply a flood control system to tackle the overall causes of flooding. Therefore, this is interesting to apply an optimal risk-based design to a flood control system to solve flood problems for Bangkok. In this study, an optimal risk-based design of a flood control system is applied to determine the optimal capacity of a flood control system based on maximum net benefit and benefit/cost ratio with a design return period of flood protection failure for Bangkok. In this study, a proposed Chao Phraya river flood control system for Bangkok recommended by the Asian Institute of Technology (AIT) and the Thai-Austrian Consortium (TAC) (1986) is considered. It is consisted of a diversion dam at Pak Kret (70 Km) upstream of Bangkok, a 52 km long diversion channel from the diversion dam to the sea with a diversion sea gate at the end of the diversion channel, a sea barrier with a pumping station at the river mouth. The diversion scheme will divert excessive portion of flood flow through the diversion channel and release the remaining flow through the diversion dam to Bangkok. The sea barrier will prevent sea water intrusion into the river. The pumping station is a supplementary component of the sea barrier to pump flood water to the sea when the sea barrier is closed. This proposed flood control system will jointly utilize the existing river storage between the diversion dam and the sea barrier to absorb flood volume during closure of the sea barrier. By this flow regulation, the river water level at Bangkok can be controlled effectively during flood events. The design approach is consisted of four steps of analysis. Firstly, the flood flow simulation analysis is carried out to determine the flooding relationships of river flood levels at Bangkok and at two flood boundary stations at Bangsai and Fort Chula for various capacities of the flood control system. It considers the flood inflow from Bangsai which is mainly diverted along the diversion channel from Pak Kret to the diversion sea gate and partly released through the diversion dam at Pak Kret through the Chao Phraya river and the sea barrier to the sea. The overdike flow into flood plains in Bangkok is considered. Secondly, the coincident flood frequency analysis is undertaken to determine the probability density functions of the natural and regulated flood levels at Bangkok for a mean dike crest level of 1.5 m.MSL. Thirdly, the load-resistance analysis is done to determine the risk of flooding due to failure of existing river dike and expected annual damage cost. Both load and resistance are considered as variables. The floodings due to failure of existing river dike due to flood overtopping the river dike, dike leakage and dike gate closure failure are considered. The respective resistance are the dike crest level, the dike strength and the dike gate closure success while the common load is the river flood level. Fourthly, the optimization of risk-based design is done to determine the optimal size of diversion channel and capacity of pumping station at the sea barrier based on maximum net benefit and benefit/cost ratio for a design return period of flood protection failure (flooding). The flood control system with diversion channel basewidth ranging from 0 to 80 m and pumping capacity ranging from 0 to 2,000 m3 /s are considered. As the result of load-resistance analysis, it is found that the proposed flood control system can increase the return period of flood protection failure at Bangkok and reduce the expected annual damage cost considerably. When there is no flood control system, the return period of flood protection failure is 2 years and the expected annual damage cost is 5,000 Million Baht. With the optimal risk-based design, it is found that the proposed flood control system is remarkably profitable. The project period is considered for 50 years (1996-2045), with initial 10 years for construction and subsequent 40 years for flood protection. The project cost and benefit are estimated at present value of the year 1995. It is found that the optimal flood control system is consisted of a diversion dam, a 40 m basewidth diversion channel with a diversion sea gate and a sea barrier without pumping station. This scheme can protect flooding in Bangkok and its suburban area for a return period of 270 years with maximum net benefit and benefit/cost ratio. The project cost is 74 billion Baht while the net benefit is as high as 200 billion Baht with the benefit/cost ratio of 3.7 and the economic rate of return of 10.2%. If excluding the cost of improvement of city drainage system amounting to 43 Billion Baht, the cost will decrease to about 31 Billion Baht. By including other indirect benefits arising from this flood protection scheme in addition to the considered benefit from reduction of potential flood damage in the protected area, the profitability will be much larger. This flood control system shows both high flood protection and large profitability, therefore justifying viability for project investment.
Year1996
Corresponding Series Added EntryAsian Institute of Technology. Dissertation ; no. WM-96-02
TypeDissertation
SchoolSchool of Civil Engineering
DepartmentOther Field of Studies (No Department)
Academic Program/FoSWater Engineering and Management (WM)
Chairperson(s)Tawatchai Tingsanchali;
Examination Committee(s)Huynh Ngoc Phien;Okitsugu, Fujiwara; Plate-Ing, Erich J.;
Scholarship Donor(s)Royal Thai Government;
DegreeThesis (Ph.D.) - Asian Institute of Technology, 1996


Usage Metrics
View Detail0
Read PDF0
Download PDF0