灰局勢決策對粒子群優法因子在類水筒模式上之應用

本研究以水筒模式行為機制的概念為出發點，對於都市不透水層比例較高的情形下，將原有的四段水筒所組成的基本型式變換成地表上及地面下二種機制的組合體，地表上採用類似水筒的概念建立降雨後產生逕流的行為機制，在產生漫地流後進入地面下的下水道系統中模擬，遂成為新的降雨-逕流模式，因架構在水筒模式的基礎上，故取名為類水筒模式（Para-Tank Model, PTM）。另本研究採用近代受自然界生物群體行為啟發衍生的隨機優化算法-粒子群優法（Particle Swarm Optimization, PSO）尋找入滲及窪蓄之貯蓄高（H1）、下水道系統之貯蓄高（H2）、地形積淹特性流出率（λ1）及下水道承載能力流出率（λ2）等4個參數。在最佳化演算過程中速度方程式有三個因子，即加速常數(Acceleration Constants) c1和c2及慣性權重(Inertia Weight) w等，本文以灰局勢決策（Grey Situation Decision-Making, GSDM）將前述三個因子作為事件，採0.2、0.5、0.8的值為其對策，形成27組局勢集，利用均方根誤差（Root Mean Squared Error, RMSE）、效益係數（Coefficient of Efficiency, CE）、總體積誤差百分比（Percent Error of Total Volume, VER）及流量誤差平方值（Squared Value of Flow Error）等四種目標決策指標做統一測度效果的分析，比較綜合效果測度後，以c1=0.8、c2=0.2、w=0.5時綜合效果測度最高為最佳的決策，成為PSO在尋找類水筒模式參數值時的因子最佳值。

關鍵字

類水筒模式；粒子群優法；灰色系統；灰局勢決策；最佳化

並列摘要

The relationship between rainfall and runoff has been the most important part of hydrological analysis. It is easier to get rainfall than to get runoff; therefore, the methods of calibration analysis regarding their relationship were actively developed in previous research to simulate the runoff mechanisms with rainfall data. In this study, the concept of tank model mechanism serves as a starting point to convert the original basic type of four tank sections into a combination of aboveground and underground mechanisms to address a higher proportion of impermeable layer in cities. A concept similar to a tank is used to establish the operational mechanisms produced after rainfall on the surface. It is simulated in the underground sewer system after overland flow, and therefore, a new rainfall–runoff model is established, called the Para-Tank Model (PTM). The particle swarm optimization (PSO) is employed to calculate the parameter values, including infiltration and depression head, sewer system head, terrain flooding feature outflow rat and sewer carrying capacity outflow rate, required by PTM. We also investigate three factors of the acceleration equation, i.e., acceleration constants c1 and c2 and inertia weight w, which are then used as events in PSO for parameter optimization in PTM during rainfall–runoff simulation. With Grey Situation Decision-Making, the values of 0.2, 0.5, and 0.8 are respectively used to create 27 groups of situation sets using the indices of the four objectives, root mean squared error, coefficient of efficiency, percent error of total volume, and squared value of flow error, in order to analyze the systematic effectiveness. After comparing the comprehensive effect measures, an optimal decision is reached when the combined effectiveness was at the highest when c1 = 0.8, c2 = 0.2, and w = 0.5 and becomes the optimal parameter value for the PTM.

並列關鍵字

Para-tank model ； particle swarm optimization ； gray system theory ； gray situation decision-making ； optimization

參考文獻

[63] Tang, L., Landers, R. G., and Balakrishnan, S. N., 2008, “Parallel Turning Process Parameter Optimization Based on A Novel Heuristic Approach,” Journal of Manufacturing Science and Engineering, Transactions of the ASME, Vol. 30, No. 3, pp. 354-368.

[10] Nelder, J. A., and Mead, R., 1965, “A Simplex Method for Function Minimization,” The Computer Journal, Vol. 7, No. 4, pp. 308-313.

[12] Powell, M. J. D., 1964, “An Efficient Method for Finding the Minimum of a Function of Several Variables without Calculating Derivatives,” The Computer Journal, Vol. 7, No. 2, pp. 155-162.

[18] Duan, Q., Sorooshian, S., and Gupta, V.K., 1992, “Effective and Efficient Global Optimization for Conceptual Rainfall-Runoff Models,” Water Resources Research, Vol. 28, No. 4, pp. 1015-1061.

[19] Sorooshian, S., Duan, Q., and Gupta, V.K., 1993, “Calibration of Rainfall-Runoff Models: Global Optimization to the Sacramento Soil Moisture Accounting Models,” Water Resources Research, Vol. 29, No. 4, pp. 1085-1194.

國際替代計量

灰局勢決策對粒子群優法因子在類水筒模式上之應用

全文下載

主題瀏覽