基于深度强化学习的海上风电机组状态维护与备件库存联合优化

维护与备件库存管理是海上风电运维的两个密不可分的关键环节。为提高海上风电机组设计寿命周期内的运维经济性,构建状态维护与备件库存联合优化策略。首先,将海上风电机组构建为由叶片、齿轮箱、电气、偏航、轮毂、制动、传动链、发电机8个子系统组成的系统,然后将各子系统的劣化过程构建为多状态马尔可夫随机过程,建立维护与备件库存的交互模型,其中包括被动维护时间与随机故障以及备件库存的关系、备件库存对维护活动的影响等。随后,设计子系统的劣化状态与备件库存状态、状态检修动作与备件订购的表征方法,并以此形成深度强化学习Dueling DQN的框架,通过对深度网络的迭代训练,求解海上风电机组的最优维护与备件订购决策序列。最后,以某海上风电场内的风电机组为例,验证所提联合优化方法的优越性,并讨论强化学习的探索率、风电场的可达率对运维成本的影响。     

3MW双馈式风电机组变桨控制策略优化研究

随着风电机组单机容量的不断增大,风电机组零部件承受的载荷越来越大,电网对机组输出功率的品质也提出了更高的要求。文章在变桨PID控制技术的基础上,通过增加塔架阻尼器、低通滤波器和陷波器,并在变桨时使用变增益双PID控制器,使风电机组在输出功率较为平滑的同时,降低关键部件的载荷。结合3 MW双馈式风电机组开发GH Bladed的外部控制器程序,并和广泛应用的PID控制器进行仿真比较。仿真结果表明,该研究提出的控制策略改善了风电机组输出功率的品质,降低了塔架等关键部件的载荷。     

考虑加快电网频率恢复的风电机组自适应虚拟惯性控制策略

传统风电机组虚拟惯性控制的惯性时间常数较大,虽抑制了频率变化率（Rate of Change of Frequency,ROCOF）,但也阻碍了频率的恢复。文章提出了一种风电机组自适应虚拟惯性控制,该控制在频率扰动初期能够提供适当的虚拟惯性支撑,同时在频率恢复期间加快频率恢复。首先,文章控制策略在频率跌落或者上升阶段,根据ROCOF的最大值,自适应计算出虚拟惯性时间常数。该值一直保持到ROCOF的极性发生变化,切换至根据ROCOF计算出的虚拟惯性时间常数,此时的ROCOF较小,虚拟惯性时间常数趋于零。故在频率恢复阶段,风电机组退出虚拟惯性支撑,从而帮助频率加快恢复。在PSCAD/EMTDC中搭建仿真算例,仿真结果,表明文章所提控制策略能够提供虚拟惯性支撑和加快扰动后频率恢复。     

基于改进下垂控制的双馈风电机组频率控制策略

为高效利用风机旋转动能,提高风电机组的频率支撑能力,文章在传统下垂控制策略的基础上,提出一种基于改进下垂控制的双馈风电机组频率控制策略。该策略通过结合扰动后电网频率动态特性,提出了可随系统频率变化率（RoCoF）变化的自适应下垂控制系数,在不同扰动下,有效提高风电机组频率支撑能力,改善系统最大频率偏差和最大频率变化率。基于EMTP-RV仿真软件搭建了四机两区域系统模型进行仿真,结果表明,改进后的下垂控制可有效应对不同扰动工况,提高了风电机组频率响应能力,进一步地改善了系统频率稳定性。     

海上风电机组多导管架拓扑优化方法

为实现海上风电机组多导管架结构概念设计,对某型5 MW多导管架结构进行模态分析、整机载荷计算与极限工况下的刚强度分析。根据受力特点,建立多导管架结构的多目标拓扑优化模型,通过最小尺寸约束抑制棋盘格现象并设置对称面约束,得到不同权因子下的拓扑优化结果。基于某一加权因子下拓扑优化结果,重新建立新型多导管架有限元模型,并进行载荷重分析。通过极限工况下的静动态分析结果对比可知,优化结构一阶固有频率略有提高,最大变形和应力均大幅降低。上述结果证明了提出的拓扑优化流程在海上风电机组多导管架设计中的可行性和优越性。     

基于综合经济效益目标的风电机组偏航控制策略研究

基于偏航系统发电量仿真计算和偏航轴承疲劳寿命量化分析,建立了一种偏航系统重启对风控制模型。该模型以风电场寿命周期综合经济效益为目标,以控制偏航系统启停为策略,能够合理地平衡发电量与偏航次数之间的关系。为寻求最优控制策略,采用粒子群-遗传混合优化算法(PSO-GA)对该控制模型进行优化。结果表明:该偏航系统重启对风控制模型可以达到预期的优化目标,对风电场综合经济效益的提高具有指导意义。     

大型海上风电机组变桨距PID控制策略研究

文章针对大型海上风机的叶片具有较大惯性的问题,分析了传统变桨距PI控制器的不足,从而设计出带有超前环节的变桨距PID控制器,其能改善因惯性较大而导致的不良控制效果,并通过典型工况的仿真计算及性能分析,表明PID控制器的整体性能优于PI控制器,不仅具有稳定风轮转速、减轻风机振动的优点,还降低了风机的极限载荷水平,从而在保证安全性的基础上降低了机组成本。     

计及疲劳损伤评价的大型风电机组降载控制优化

针对目前大型风电机组疲劳载荷日益严重的问题,进行了降载荷控制优化的研究:首先采用美国国家可再生能源实验室(NREL)的5 MW风电机组搭建FAST和Matlab的联合控制模型;在此基础上设计了一种在额定风速以下提前进行变桨控制的优化方案,并依据疲劳损伤与发电量综合评价选取最优提前变桨区间;然后在额定风速以上结合功率和载荷两方面控制目标,先分风速区间优化独立变桨控制器PI参数,再拟合得到独立变桨控制的变增益调度优化函数;最后加入载荷反馈回路,在变增益独立变桨策略的基础上进一步降低叶根挥舞弯矩峰值。结果表明:提前变桨策略能有效降低额定风速附近的载荷与功率波动,加入叶片载荷反馈回路的变增益独立变桨控制相比统一变桨控制能显著减小叶片疲劳损伤。     

数据驱动的风电机组偏航参数适应性优化

为了增强风电机组偏航系统自适应水平,提升风能利用率,提出一种基于K近邻聚类（KNN）算法风电机组偏航控制参数优化方法。为准确描述风向变化,建立改进Weibull概率分布建立风向评估模型,即以风向波动的幅值（A）和波动持续时间（T）作为风况的数据标签来描述风向。对比风电机组不同偏航参数下的运行数据确定聚类中心（已知风况下的最佳偏航参数）,通过基于KNN算法的风电机组偏航控制参数优化模型,得到不同风况下风电机组最佳的偏航参数。通过对风电机组运行数据进行算例分析表明,该方法高风速时可提升风电机组发电效率,并在低风速时减少偏航启动次数。     

基于布谷鸟算法的风电机组装配序列优化

装配序列规划（ASP）是风电机组制造的重要技术。ASP是一个组合优化问题,风电机组最优装配序列的搜索空间和计算量均很大。提出基于布谷鸟算法的风电机组装配序列优化方法。首先,从三维装配体模型中提取多种装配约束信息并表示成装配约束矩阵,以降低最优装配序列的搜索空间;继而构造装配序列目标函数,建立装配序列规划模型,方便算法计算出最优装配序列;对离散布谷鸟算法（DCA）改进,求解ASP模型,获得风电机组产品的最优装配序列。最后通过实验验证了ASP模型的有效性和DCA的优良性能。     

基于风电场机组大部件运行状态的年度检修策略

风电场机组年度检修是保证所有风电机组正常运行的重要手段,合理的检修方案可以降低风电场运行维护成本。为确定风电场的最优年度检修方案,文章提出了一种基于大部件运行状态的风电场年度检修策略。首先,综合考虑大部件的工作役龄、维修历史及当前运行状态,以威布尔比例强度模型计算部件的故障概率;然后,定义检修改善因子,描述年度检修对部件故障概率的降低程度,建立考虑故障停机时间、故障维修费用及检修成本的年度检修策略优化模型;最后,通过算例分析验证该策略的正确性及有效性。     

双馈风电机组的高电压穿越控制策略

分析了广义电网电压骤升(含对称、不对称骤升)故障下双馈风电机组电磁暂态过渡过程,评估了影响机组高电压穿越运行的关键因素,并梳理了电压骤升、骤降故障诱发双馈机组脱网机理的异同。提出了一种基于瞬态灭磁控制和无功电流支持的双馈机组高电压穿越控制方案。仿真和实验结果表明,所述方案能够显著加快故障电网条件下双馈感应电机定子磁链中直流、负序分量的衰减,进而快速抑制机组电磁转矩和母线电压的波动;同时能够满足并网规范对机组无功电流输出的要求,实现机组的故障穿越运行。     

Multivariable control strategy for variable speed, variable pitch wind turbines

Reliable and powerful control strategies are needed for wind energy conversion systems to achieve maximum performance. A new control strategy for a variable speed, variable pitch wind turbine is proposed in this paper for the above-rated power operating condition. This multivariable control strategy is realized by combining a nonlinear dynamic state feedback torque control strategy with a linear control strategy for blade pitch angle. A comparison with existing strategies, PID and LQG controllers, is performed. The proposed approach results in better power regulation. The new control strategy has been validated using an aeroelastic wind turbine simulator developed by NREL for a high turbulence wind condition.     

Aerodynamic optimization of shrouded wind turbines

An axisymmetric Reynolds averaged Navier–Stokes solver is used along with an actuator disk model for the analysis of shrouded wind turbine flowfields. Following this, an efficient blade design technique that maximizes sectional power production is developed. These two techniques are incorporated into an optimization framework that seeks to design the geometry of the shroud and rotor to extract maximum power under thrust constraints. The optimal solution is also evaluated using a full three‐dimensional Reynolds averaged Navier–Stokes solver, suggesting the viability of the design. The predicted optimal designs yield power augmentations well in excess of the Betz limit, even if the normalization of the power coefficient is performed with respect to the maximum shroud area. Copyright © 2016 John Wiley & Sons, Ltd.     

Dynamic wind estimation based control for small wind turbines

We introduce a novel scheme for small wind turbines that gives dynamic estimation of wind speed from rotor angular velocity measurements. The estimation proceeds in two different dynamic observers, one giving a valid estimate for higher Tip Speed Ratios (TSRs) and which we call the Upper Wind Estimator (UWE) and the other called the Lower Wind Estimator (LWE) valid for lower TSRs. The meaning of “higher” and “lower”, and the precise regions of validity, are quantified. We further propose a coordinated control scheme using the UWE. Simulations are presented showing closed-loop performance of the turbine and the estimators both in the optimal TSR regulation condition, and the dynamic power-shedding condition caused by a wind gust. An analytic analysis of closed-loop stability and of the convergence and bias properties of the estimator is provided. Empirical data showing performance on a real turbine is also presented.     

Dynamic control of wind turbines

The paper presents an intelligent wind turbine control system based on models integrating the following three approaches: data mining, model predictive control, and evolutionary computation. To enhance the control strategy of the intelligent system, a multi-objective model is proposed. The model involves five different objectives with different weights controlling the wind turbine performance. These weights are adjusted in response to the variable wind conditions and operational requirements. Three control factors, wind speed, turbulence intensity, and electricity demand are considered in eight computational scenarios. The performance of each scenario is illustrated with numerical results.     

基于风电机组无功裕度预测的风电场无功分层控制策略

针对风电电压波动的问题,文章基于风电机组无功裕度预测,提出了一种风电场无功分层控制策略.该策略首先以并网点电压偏差和线路有功损耗最小为目标,使用二次规划算法在线实时求解最优并网电压,进而求解风电场无功参考值;其次,采用EWT-LSSVM预测算法进行风电功率预测,并提出预测功率校正方法实时修正预测功率,精确求解风电机组的...     

Aerofoil optimization for vertical‐axis wind turbines

An expression for the aerodynamic optimization of aerofoils for 2D lift driven vertical‐axis wind turbines is derived as a function of lift slope and drag. As lift slope is proportional to aerofoil thickness, the aerodynamic optimum is found in thick aerofoils, which are also structurally advantageous. Using a genetic optimization algorithm, the objective function is used to generate aerofoils whose performance in a vertical‐axis wind turbine is calculated using a potential flow solution of the induction field and 2D polars calculated with XFOIL. The results demonstrate power and structural gains. This approach can lead to reductions in rotor mass due to the thicker and thus stiffer aerofoils, without compromising aerodynamic performance. Copyright © 2014 John Wiley & Sons, Ltd.     

Adaptive tower damping control for offshore wind turbines

This paper deals with the problem of wind turbine tower damping control design and implementation in situations where the support structure parameters vary from their nominal design values. Such situations can, in practice, occur for onshore and especially offshore wind turbines and are attributed to aging, turbine installation, scour or marine sand dunes phenomena and biofouling. Practical experience of wind turbine manufacturing industry has shown that such effects are most easily quantified in terms of the first natural frequency of the turbine support structure. The paper brings forward a study regarding the amount to which nominal tower damping controller performance is affected by changes in the turbine natural frequency. Subsequently, an adaptive tower damping control loop is designed using linear parameter‐varying control synthesis; the proposed tower damping controller depends on this varying parameter which is assumed throughout the study to be readily available. An investigation of the fatigue load reduction performance in comparison with the original tower damping control approach is given for a generic three‐bladed horizontal‐axis wind turbine. Copyright © 2016 John Wiley & Sons, Ltd.