齿槽型直线发电机多工况优化设计和无传感控制研究

Slotted tubular generators are widely used in the applications of direct waver energy conversion systems, because of their high power density and high power factor. Since of the strong corrosion, strong impact and the uncertain activity of the wave, the optimization and high reliability control strategies, which are suitable for the generators operated in various conditions, become the key issues in the field. To this, in this project, the double-sided slotted tubular generator(DSSTG) is applied in the direct-drive wave energy conversion system, and a novel multi-condition optimization method and a nonlinear sensorless control strategy are proposed and highlighted. In details, a coupling field finite element method is used to analyze the dynamic characteristics of the generator. The nonlinear mathematical models are built based on the variation rules between the dimensions and the parameters. Based on the repetitive current compensation strategy and the corresponding energy consumption functions, a novel optimization method, which combines the topology and the control technology, is proposed. This method for multi-condition generators can realize the optimizations of the power density and the force ripples at the same time. With the optimized construction, multi-condition control strategies with high reliability for the generator are researched. By introducing the multi-condition nonlinear mathematical models of the generator into the sliding mode observer algorithm, a novel nonlinear sensorless control strategy for the direct-drive wave energy conversion systems is obtained. With the control strategy, a prototype and a measurement system of the DSSTG are then built to verify the viability and the accuracy of the proposed theories. The research results not only can effectively promote the development of the direct-drive wave energy conversion systems in China, but also can provide necessary theoretical basis and method reference for the research of optimization and control technologies of the electrical machine. This project has important theoretical and practical significance.

齿槽型圆筒直线发电机因其高功率密度和高功率因数，在波浪能直驱发电系统中得到广泛应用。海洋强腐蚀、强冲击的恶劣环境以及波浪能的随机性引发的高可靠性、多工况优化设计和控制策略是行业发展的关键问题。对此，本项目以双层气隙式齿槽型圆筒直线发电机为研究对象，重点对其在波浪能直驱发电应用中的多工况优化设计和无传感控制方法展开研究。拟基于多耦合场有限元分析，挖掘尺寸和参数间的非线性变化规律，构建其变参数非线性数学模型；通过不同拓扑下多工况电流补偿策略及能耗计算，研究可实现功率和推力波动双优化的基于电流补偿的发电机多工况全局优化新方法；基于发电机多工况非线性参数计算，构建非线性滑模观测器算法，实现多工况波浪能系统的无传感控制；最终搭建样机测控系统实现对理论和方法的验证。本项目研究成果不仅可推动我国波浪能发电业的发展，而且能够为电机的优化和控制研究提供理论基础和方法借鉴，具有重要的理论和实际意义。

本项目针对波浪能发电系统应用需求，提出了具有高效率高功率密度的新型双层气隙齿槽型直线电机拓扑结构，完成了齿槽直线电机多物理场建模、多工况特性分析、智能建模方法、全局优化设计和滑模控制新方法及实验方法研究。.针对双层气隙齿槽型直线电机高推力密度、内层散热难的问题，利用有限元建立了电磁—温度—应力场模型，完成了磁场分布、损耗计算、温升分布及应力分布计算，分析了不同电流负载及不同冷却条件下电机的性能变化，得到了电机政策运行条件。.针对双层齿槽型直线电机推力波动大的缺点，为削弱推力波动的同时保持电机的高功率密度，提出了切齿型和混合励磁分块型两种新型拓扑结构。研究成果证实切齿型拓扑与错极型结构相比，相同的功率密度下推力削弱能力提升了约20%；混合励磁分块型拓扑与现有正弦分块励磁结构相比，推力波动由38%降至5%，且推力均值提升了13%。.针对解析模型精确度不高的问题，为提高模型准确度的同时减少优化设计计算时间，提出了集成建模方法和迭代式小样本多目标优化算法。前者结合解析子域模型和机器学习补偿模型，将原解析模型误差由15%降低至1.5%，多目标优化计算时间缩短了80%；后者在对混合式分块励磁拓扑优化设计中，仅通过4次迭代，100个样本的有限元分析结果即完成了优化目标，极大的缩短了计算时间。.针对直线电机直驱系统的集成扰动问题，分别提出了谐波电流补偿式控制算法和抗扰动式滑模控制方法研究。前者针对逆变器死区时间造成的电流谐波引起的推力波动问题展开研究，基于谐波电流平均值的dq补偿算法，有效降低了5次、7次谐波电流，使推力波动降低了10%；后者将自抗扰算法与滑模控制方法相结合，利用粒子群算法进行参数优化后不仅消除了超调量，而且具有更快的响应速度、更小的稳态误差和抖振与更强的抗扰动性能。.该项目研究成果经实验证实切实有效，不仅为高功率密度低推力波动电机及控制系统的发展提供了新方法，而且在波浪能发电及电梯驱动等场合也具有重要的应用价值。