Anisotropy-based Sensorless Control at Excessive Load

通过在电力驱动中避免位置传感器，除其他外，可以节省成本和空间，并实现更高的可靠性。低速范围内的无传感器控制方法利用所谓的电感各向异性进行位置估计。由于磁饱和，各向异性通常与电流/负载有关。这导致低速下无传感器控制的负载限制，超过此限制，基于各向异性的方法在闭环操作中变得不稳定。负载限制的确切值主要取决于机器设计，因此在设计过程中可能受到影响或优化。但为此，在设计过程中就能够正确地计算出载荷限值是至关重要的。这类方法在文献中偶有提出，采用各向异性比和各向异性角作为稳定性指标。然而，由于实际上这两个标准都与不稳定性没有直接联系，因此这些预测会导致不正确的优化结果。这一应用是基于对不稳定性问题的第一个完整的理论推导，其中各向异性驱替率被证明是关键的度量。简化形式（2D）的理论已经得到了模拟的支持，现在必须进行实验验证。理论（3D）的扩展将提高载荷极限预测的准确性，并将在项目中进行实验验证。在对三维预测规则进行验证后，将其集成到机器设计过程中，以便制造针对各向异性负载限制进行优化的机器。本文将最终证明，在多大程度上可以增加无传感器过载能力，以及即使迄今为止被排除在外的应用，如电动汽车，是否可以在没有编码器的情况下实现。

By avoiding position sensors in electric drives, it is possible, inter alia, to save cost and space, and achieve improved reliability. Methods for sensorless control in the lower speed range employ the so called anisotropy of the inductance for the position estimation. Due to magnetic saturation, the anisotropy is generally current-/load-dependent. This results in a load limit for sensorless control at low speeds, above which anisotropy-based method become unstable in a closed loop opration.The exact value of the load limit mainly depends on the machine design and could therefore be influenced or optimized in the design process. But for this it is essential to be able to correctly calculate the value of the load limit already in the design process. Such approaches have been proposed sporadically in the literature which use the anisotropy ratio and the anisotropy angle as stability indicators. However, since actually both criteria have no direct connection with the instability, those predictions lead to incorrect optimization results.This application is based on a first complete theoretical derivation of the instability problem, wherein the anisotropy-displdacement rate turned out to be the critical measure. The theory in simplified form (2D) has been supported by simulations and must now be experimentally verifified. An extension of the theory (3D) shall improve the accuracy of the prediction of the load limit and also will be verified experimentally in the project.After verification of the 3D prediction rule it will be integrated into the machine design process in order to manufacture machines optimized with respect to the anisotropy-load limit. It will herein finally turn out, to which extent the sensorless overload capacity can be increased and whether even hitherto excluded applications, such as electromobility, can be realized even without encoder.