为减轻水利工程对河流生态系统造成的不利影响,针对重要珍稀水生动物中华鲟资源的有效保护及可持续发展问题,本文在葛洲坝下游中华鲟产卵场水动力数学模型及产卵适合度评价模型的基础上,利用BP人工神经网络的学习能力构建水库调度与产卵场产卵适合度关系模型,并将关系模型嵌入葛洲坝现行的水库调度模型,获得中华鲟产卵期(10—11月)优化中华鲟产卵生境的水库生态调度模型.生态调度模型以中华鲟产卵场产卵适合度及水电厂发电量最大为双目标函数,以同期水库需满足的航运效益、水资源利用效益及水轮机运行效益为主要约束条件,通过比较相同入库流量下生态调度和现行调度的水库综合效益,得出在中华鲟产卵期,优化中华鲟产卵生境的生态调度在仅损失0.15%发电量的同时,使坝下中华鲟产卵场产卵适合度增加39%,使中华鲟获得更适宜的水动力环境,增加产卵概率,有效保护中华鲟资源.
In order to mitigate the adverse effects of hydraulic engineering on river ecosystems,and for the effective protection and sustainable development of Chinese sturgeon, a relational model between the reser?voir operation and the suitability of Chinese sturgeon spawning ground has been built up by BP artificial neural network,on the baois of the hydrodynamic model and spawning fitness evaluation model. By the rela?tional model embedded in the existing Gezhouba reservoir operation model,this paper gets the eco-schedul?ing model for the optimization of Chinese sturgeon spawning habitats. The target functions of the ecological scheduling model compose of the largest spawning fitness of Chinese sturgeon spawning ground and the larg?est electricity generation of the hydroelectric power plant,and the main constraints compose of the naviga?tion benefits, water resource use efficiency and turbine operational efficiency that should be balanced dur?ing the reservoir operation. By comparing the com