基于仿射扰动反馈的风力发电机组多目标随机模型预测控制

Affine-disturbance-feedback-based Multi-objective Stochastic Model Predictive Control of Wind Turbine

风力发电的间歇性和随机不确定性是其规模化并网的主要障碍。为提高风电机组控制系统应对风能随机扰动的能力,该文提出一种基于仿射扰动反馈的风力发电机组多目标随机模型预测控制策略。以输出功率准确跟踪设定值和降低执行机构疲劳载荷为目标,结合风力发电机组运行工况和约束条件,构建基于期望目标函数和机会约束的随机优化模型,采用仿射扰动反馈和风速扰动的概率信息将其转换成确定性优化模型,求解得到当前时刻的最优控制律。以NREL5MW风电机组为研究对象进行算例验证,结果表明,通过选取合适的输出约束违背概率值,所提出的控制策略能够有效抑制风速的随机干扰,在减小执行机构动作幅度的基础上,实现机组输出功率对额定值的准确跟踪。

The intermittency and random uncertainty of wind power generation are the main obstacles to large-scale grid connection.In order to improve the ability of wind turbine control systems and deal with wind energy random disturbance,an affine-disturbance-feedback-based multi-objective stochastic model predictive control(ADF-MSMPC)strategy is proposed in this paper.Aiming at accurately tracking the power output set-point and reducing the actuators’fatigue loads,a stochastic optimization model consisting of expected objective function and chance constraints is constructed based on the turbine realistic operating condition and constraints.With the aid of affine disturbance feedback policy and wind speed probability information,the stochastic optimization model is then transformed into a tractable deterministic optimization model,and therefore the current optimal control law can be obtained.The benchmark NREL 5MW wind turbine is chosen as the research object in this paper,and the simulation results show that through selecting the appropriate probability value of output constraint violation,the proposed control strategy can effectively suppress the random wind speed disturbance,reduce the actuators’action amplitudes and realize the accurate tracking of rated power output.