拉西瓦水电站泄流雾化影响范围研究-水利水电工程专业论文.docxVIP

摘 摘 要 拉西瓦水电站地处高山峡谷，坝高、泄洪落差大、下游河道狭窄，加之左岸 坝肩下游存在一个稳定性较差的高边坡II 4变形体，而采用的挑流消能泄洪方式 所引起的水流雾化问题又比较突出。预测拉西瓦水电站泄洪雾化对下游环境的影 响范围和程度，为制定下游防护工程措施提供可靠的依据，使泄洪引起的雾化不 危及高边坡稳定，保证电站的安全运行，有着重要意义。 本文在泄流雾化机理研究的基础上，以BP神经网络为基础，研究建立了拉 西瓦水电站挑流泄洪雾化神经网络模型，并用网络模型预测出拉西瓦水电站不同 水位和宣泄流量下的雾化影响范围，经与用力学分析模型计算预测的雾化影响范 围对比分析得到： (1)当泄洪高速水流从挑坎挑出时，水舌在空中掺气和水舌入水时溅水而 形成雾化水流。雾化水流是一种复杂的水气两相流，泄流雾化主要由水舌落水喷 溅引起。 (2)水电站的泄流雾化是宣泄流量、水位差、泄流孔口形式等多因素相互 作用的结果，具有明显的非线性输入、输出关系。人工神经元网络是一种模拟非 线性输入、输出关系的有效工具，因此，通过建立泄流雾化的人工神经网络模型 预测雾化影响范围是可行的； (3)用BP神经网络模型预测的拉西瓦水电站泄流雾化影响范围与力学分 析模型的计算结果基本一致，同力学分析模型相比具有计算量小、方便快捷的特 点。为今后深入研究泄流雾化的影响提供了一条新的方法和途径； (4)根据预测结果，提出减轻拉西瓦水电站泄流雾化影响的防护建议和措 施。 关键词：拉西瓦水电站神经网络模型泄流雾化预测 ABSTRACTLaxiwa ABSTRACT Laxiwa hydropower station is located in the place with hi．gh mountains and deep valleys。Because the dam is highthe water head is large and the riverbed of downstream is nalTow,the atomization problem aroused by flood discharge is serious。 Predicting the influence of Laxiwa hydropower station flood discharge atomization for the environment of downstream，can offer reliable basis for preparing the protection project measure of downstream，make the atomization that flood discharge arouses not dangerous to the stabilize of high side slope，and guarantee the safe operation of power station．So it is much valuable to predict the flood discharge atomization． On the basis of researching flood discharge atomization mechanisms this paper takes BP neural network method as foundation，established the Laxiwa hydropower station trajectory flood discharge atomization neural network model．With the neural network model the Laxiwa hydropower station atomization influence range has been predicted under the different water head and flow rate．By contrast with the physical analysis model about the atomization influence range，the following conclusion can be dawn。 a。When high-velocity flow caused by flood discharging deflects from deflecting bucket，atomized water flow forms in both the way of nappe earation in the air and splashing。Atom