Automatic calibration for urban rainfall - runoff models | |
| Author | Nittaya Wangwongwiroj |
| Call Number | AIT Diss. no. WM-02-03 |
| Subject(s) | Runoff Calibration Rain and rainfall |
| Note | A dissertation submitted in pmtial fulfillment of the requirements for the degree of Doctor of Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Dissertation ; no. WM-02-03 |
| Abstract | In this study, an automatic calibration procedure applicable for urban rainfall-runoff modelling has been developed and evaluated. The basis of the calibration scheme is the MOUSE software package developed by DHI. The shuffled complex evolution (SCE) method developed at the University of Arizona was applied as an optimisation algorithm. The data used in the study consist of synthetic rainfall/runoff data together with measured time series data from JOnkiiping catchment in Sweden and the data from two catchments in Bangkok, Thailand. In addition, the surface runoff data from Japan are also applied to ensure that the developed procedure is generally applicable. Manual calibration procedure for the tluee surface runoff models in MOUSE namely the time/area model, kinematic wave model and linear reservoir model was elaborated using real sewer data from urban catchments in Sweden and Thailand. The manual calibration results support the idea that manual calibration is time consuming and subjective. Hence, there is a need to develop an automatic calibration scheme to come up with these problems. The developed MOUSE-SCE scheme was evaluated in detail based on the time/area model. Three calibration tests have been carried out using different objectives; i.e. singlecalibration to fit water balance, single-calibration to fit RMSE, and multi-objective calibration to fit both water balance and RMSE. For synthetic data study, it is possible to locate the global optimum solutions. However, it was found that multiple optima occur due to strong parameter interaction and they are equally good to fit with the same optimnm objective function. For real sewer data study, it could not easily obtain converging calibration results. The difficulties in getting optimum solutions are not only dependent on the setup data required for automatic calibration, but they also depend on the occurrence of dry weather flow data that must be subtracted from combined sewer flows before calibration. Two additional algorithms are proposed for MOUSE-SCE automatic calibration scheme, i.e. the new time/area curves and the data processing for subtracting of dry weather flow and inconsistent data. In the first algorithm, a new set of equations representing the time/area curves applicable to general shapes of catchments is proposed. The algorithm is powerful in improving the calibration results. The second proposed algorithm of data processing was tested using synthetic data. The calibration results demonstrate the applicability of this approach. Finally, the peak flow values obtained from calibration scheme was used to the design of a sediment basin to ascertain the applicability of the automatic calibration results for engineering practical aspects. |
| Year | 2002 |
| Corresponding Series Added Entry | Asian Institute of Technology. Dissertation ; no. WM-02-03 |
| Type | Dissertation |
| School | School of Engineering and Technology |
| Department | Department of Civil and Infrastucture Engineering (DCIE) |
| Academic Program/FoS | Water Engineering and Management (WM) |
| Chairperson(s) | Mark, Ole; |
| Examination Committee(s) | Gupta, Ashim Das;Luketina, David Andrew;Nguyen Thi Kim Oanh;Larsen, Torben; |
| Scholarship Donor(s) | King Mongkut's University of Technology Thonburi; |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 2002 |