Search

This study establishes an innovative systematic optimization model for megacity flood mitigation by combining multiple Low Impact Development (LID) devices, taking into consideration a Benefit-Cost (B/C) Analysis. The contributions provided by this study include: (1) analysis of flooding consequences on a developing megacity; (2) development of a technical approach enabling an automatic and effective optimization process, linking with an embedded Storm Water Management Model (SWMM); and (3) proposal of adaptive solutions using a combined layout design scheme. Our investigation sets the Benefit/Cost ratio as the objective function to deal with flooding in all return periods (RPs). The decision variables correspond to the allocated areas and quantity of LID devices, including porous pavements, bioretention cells, infiltration trenches, rain barrels, vegetable swales, green roofs, and tree boxes. Under such layout, the flooding loss was simulated with SWMM, and the optimal solution was solved by employing a Simulated Annealing (SA) algorithm. Min-Sheng Community in Taipei is chosen as a study area for demonstrating the applicability of the developed model. Results show that the B/C ratio of identified optimal design can reach 1.448, with green roofs and bioretention cells as main devices, and rain barrels and porous pavements as secondary supplies. Regarding rainfalls in all RPs, the peak flows and delay of peak times at Fu-Yuan Pumping Station can decrease significantly in the range of 5.75–29.80% and 12.50–20%, respectively; and 9.52%–23.49% and 12.50%–37.5% at the subcatchments. The efficiency of flood detention is higher for low RPs than high RPs, while the time-delay ability is smaller.

Multi-objective reservoir operation considering the trade-off of discharge-desiltation-turbidity during typhoons and sediment concentration (SC) simulation modeling are the vital components for sustainable reservoir management. The purposes of this study were (1) to analyze the multi-layer release trade-offs between reservoir desiltation and intake turbidity of downstream purification plants and thus propose a superior conjunctive operation strategy and (2) to develop ANFIS-based (adaptive network-based fuzzy inference system) and RTRLNN-based (real-time recurrent learning neural networks) substitute SC simulation models. To this end, this study proposed a methodology to develop (1) a series of multi-phase and multi-layer sediment-flood conjunctive release modes and (2) a specialized SC numerical model for a combined reservoir–reach system. The conjunctive release modes involve (1) an optimization model where the decision variables are multi-phase reduction/scaling ratios and the timings to generate a superior total release hydrograph for flood control (Phase I: phase prior to flood arrival, Phase II/III: phase prior to/subsequent to peak flow) and (2) a combination method with physical limitations regarding separation of the singular hydrograph into multi-layer release hydrographs for sediment control. This study employed the featured signals obtained from statistical quartiles/sediment duration curve in mesh segmentation, and an iterative optimization model with a sediment unit response matrix and corresponding geophysical-based acceleration factors, for efficient parameter calibration. This research applied the developed methodology to the Shihmen Reservoir basin in Taiwan. The trade-off analytical results using Typhoons Sinlaku and Jangmi as case examples revealed that owing to gravity current and re-suspension effects, Phase I + II can de-silt safely without violating the intake's turbidity limitation before reservoir discharge reaches 2238 m3/s; however, Phase III can only de-silt after the release at spillway reaches 827 m3/s, and before reservoir discharge reaches 1924 m3/s, with corresponding maximum desiltation ratio being 0.221 and 0.323, respectively. Moreover, the model construction results demonstrated that the self-adaption/fuzzy inference of ANFIS can effectively simulate the SC hydrograph in an unsteady state for suspended load-dominated water bodies, and that the real-time recurrent deterministic routing of RTRLNN can accurately simulate that of a bedload-dominated flow regime.

Summary This study applies an Adaptive Network-based Fuzzy Inference System (ANFIS) and a Real-Time Recurrent Learning Neural Network (RTRLNN) with an optimized reservoir release hydrograph using Mixed Integer Linear Programming (MILP) from historical typhoon events to develop a multi-phase intelligent real-time reservoir operation model for flood control. The flood control process is divided into three stages: (1) before flood (Stage I); (2) before peak flow (Stage II); and (3) after peak flow (Stage III). The models are then constructed with either three phase modules (ANFIS-3P and RTRLNN-3P) or two phase (Stage I + II and Stage III) modules (ANFIS-2P and RTRLNN-2P). The multi-phase modules are developed with consideration of the difference in operational decision mechanisms, decision information, release functions, and targets between each flood control stage to solve the problem of time-consuming computation and difficult system integration of MILP. In addition, the model inputs include the coupled short lead time and total reservoir inflow forecast information that are developed using radar- and satellite-based meteorological monitoring techniques, forecasted typhoon tracks, meteorological image similarity analysis, ANFIS and RTRLNN. This study uses the Tseng-Wen Reservoir basin as the study area, and the model results showed that RTRLNN outperformed ANFIS in the simulated outcomes from the optimized hydrographs. This study also applies the models to Typhoons Kalmaegi and Morakot to compare the simulations to historical operations. From the operation results, the RTRLNN-3P model is better than RTRLNN-2P and historical operations. Further, because the RTRLNN-3P model combines the innovative multi-phase module with monitored and forecasted decision information, the operation can simultaneously, effectively and automatically achieve the dual goals of flood detention at peak flow periods and water supply at the end of a typhoon event.

In this study, we apply artificial intelligence techniques to the development of two real-time pumping station operation models, namely, a historical and an optimized adaptive network-based fuzzy inference system (ANFIS-His and ANFIS-Opt, respectively). The functions of these two models are the determination of the real-time operation criteria of various pumping machines for controlling flood in an urban drainage system during periods when the drainage gate is closed. The ANFIS-His is constructed from an adaptive network-based fuzzy inference system (ANFIS) using historical operation records. The ANFIS-Opt is constructed from an ANFIS using the best operation series, which are optimized by a tabu search of historical flood events. We use the Chung-Kong drainage basin, New Taipei City, Taiwan, as the study area. The operational comparison variables are the highest water level (WL) and the absolute difference between the final WL and target WL of a pumping front-pool. The results show that the ANFIS-Opt is better than the ANFIS-His and historical operation models, based on the operation simulations of two flood events using the two operation models.

Summary In this study, we estimate small-scale temporal variations in groundwater pumpage and recharge in alluvial fan topography. We study these variations on a monthly scale using groundwater storage hydrograph and isotope analysis along with area rainfall hydrographs and data of historical irrigation pumpage variations. We consider multiple irrigation practices whose effect was measured only on an annual basis in mostly previous studies. In previous studies, pumpage for non-irrigation or irrigation purposes was regarded as known but illegal pumping was not accounted for. Also, recharge sources include rainfall, rivers, boundary inflow, and groundwater-irrigation, which have not been individually accounted for in previous studies. Pumpage can be estimated from the pumping rate and the number of pumping days. In alluvial fan topography, the actual pumping rate, including that because of illegal wells, can be deduced from the recession slope of the groundwater hydrograph. The annual pumping rate of non-irrigation pumping is assumed to be constant, and can be estimated as the slope of the groundwater hydrograph over consecutive non-rainy days during the non-irrigation periods of the dry season. For cases where there are multiple irrigation practices, the temporal distribution of the irrigation pumping rate can be estimated from the slope of the groundwater hydrograph and also from records provided by the Irrigation Association. In this study, we regard recharge as the difference between inflow and loss. Different types of inflow can be inferred from the storage hydrograph in rainy or non-rainy days, and the loss can then be obtained by using the groundwater balance equation. The isotope mass balance equation is then used to quantify various sources of recharge. In this study, we apply the proposed methodology to the Cho-Shui River alluvial fan, Taiwan. The results show that the proposed methodology can overcome previous restrictions of groundwater hydrograph and isotope analyses to effectively and rationally estimate monthly variations in pumpage and recharge, thus making it possible to research fine flow mechanisms within a multi-layer groundwater system.