风力机涡流发生器斜向布局多目标优化OACHSSCDCSTPCD
Multi-objective optimization of oblique layout of wind turbine vortex generator
基于最优拉丁超立方试验设计法确定涡流发生器斜向布局参数和试验方案,耦合计算流体力学方法(CFD)和滑移网格技术计算风力机的推力和扭矩,并耦合反向传播(BP)神经网络与遗传算法,建立风力机涡流发生器斜向布局气动性能模型.通过5组测试算例,建立遗传算法优化BP神经网络的风力机涡流发生器斜向布局气动性能模型,气动性能模型预测值和模拟仿真值的误差与均方根均较小,表明气动模型的准确性较高.耦合多目标遗传算法,构建风力机涡流发生器斜向布局多目标优化方法,开展风力机涡流发生器优化.结果表明:相比原涡流发生器方案,优化后的涡流发生器诱导涡的最大涡量增大10.47%,风力机叶片截面流动分离得到有效抑制,风力机功率提升9.963%,而推力仅增大1.864%,风力机气动性能得到进一步提升.
Based on the optimal Latin hypercube test design method,the oblique layout parameters and test scheme for the vortex generator were determined.The thrust and torque of the wind turbine were calculated by coupling computational fluid dynamics(CFD)and sliding grid technique.An aerodynamic performance model for the vortex generator of the wind turbine in an oblique layout was established by coupling back propagation(BP)neural network with a genetic algorithm.Through five groups of test cases,the aerodynamic performance model was refined with a genetic algorithm optimized BP neural network.The model showed minimal error and root mean square value between the predicted and simulated aerodynamic performance,indicating its high accuracy.Coupled with multi-objective genetic algorithm,a multi-objective optimization method for the oblique layout of the wind turbine's vortex generators was constructed to facilitate the optimization of the vortex generators.The results showed that the optimized vortex generator,in comparison with the original scheme,increased the maximum vorticity of the induced vortex by 10.47%,effectively suppressing the flow separation at the wind turbine blade section.In addition,the power of the wind turbine was increased by 9.963%,while the thrust was only marginally increased by 1.864%.Thereby,the aerodynamic performance of the wind turbine was further improved.
雍天;谭剑锋;邢肖兵;杨宇霄;夏云松
南京工业大学机械与动力工程学院,江苏南京 211800
动力与电气工程
BP神经网络多目标遗传算法涡流发生器斜向布局气动性能风力机
BP neural networkmulti-objective genetic algorithmvortex generatoroblique layoutaerodynamic performancewind turbine
《南京工业大学学报(自然科学版)》 2024 (001)
65-73 / 9
国家自然科学基金(12172165);江苏省自然科学基金(BK20211259);旋翼空气动力学重点实验室开放课题(RAL20200302)
评论