论文部分内容阅读
提出了一个考虑农业需水量变化的遗传算法优化模型,用于水库优化调度。为了将农业需水量的不确定性纳入水库调度优化中,考虑了由入流量、月初水库蓄水量和需水量的不同组合作为自变量的不同的线性方程形式。通过传统的和模糊的回归分析得到优化调度策略的回归系数,在模糊回归分析中应用了对称和非对称模糊隶属函数。根据伊朗中部扎因代河水库的长期调度模拟结果评价调度策略的效率。评价标准包括可靠性、回弹性、总脆弱性和最大月脆弱性,以及统计学中的相关系数、效率系数和标准差,结果表明,采用入流量、蓄水量和需水量作为自变量的模糊线性回归方程,并采用非对称模糊隶属函数确定回归方程系数,在满足变化的需水量方面具有最好的长期性能。
A genetic algorithm optimization model considering the change of agricultural water demand is proposed for reservoir optimal dispatch. In order to incorporate the uncertainty of agricultural water demand into the reservoir scheduling optimization, different forms of linear equations with different combinations of inflow, early-stage reservoir storage and water demand as independent variables are considered. Through the traditional and fuzzy regression analysis, the regression coefficients of the optimal scheduling strategy are obtained, and the symmetrical and asymmetric fuzzy membership functions are applied to the fuzzy regression analysis. The efficiency of the dispatching strategy is evaluated according to the long-term dispatch simulation results of Zainhihe reservoir in central Iran. The evaluation criteria include reliability, resilience, total vulnerability and maximum monthly vulnerability, as well as statistical correlation coefficients, efficiency coefficients and standard deviations. The results show that the use of inflows, water storage and water demand as ambiguities as independent variables Linear regression equation, and use the asymmetric fuzzy membership function to determine the coefficient of regression equation, which has the best long-term performance in meeting the changing demand of water.