Data-driven virtual inertia LQR control for new energy systems

In order to improve the inertia support capability of new energy grid connected inverters for power systems, this paper mainly studies the design problem of an optimal quadratic regulation control algorithm based on the virtual inertia of DC capacitor energy. Based on the typical topology of new energy grid connected inverters, the dynamic foundation of virtual synchronous machines was quantitatively analyzed. On the basis of analyzing the dynamic characteristics, we further quantitatively obtained the mathematical relationship between inertia parameters and system frequency response characteristics. When considering load changes, the system frequency changes, which can lead to the charging and discharging process of DC capacitors. By designing a virtual inertia controller, the energy of the DC capacitor can be further released, providing inertia support for the system. It is worth noting that the design of the LQR control algorithm in this article is different from the traditional control algorithm design framework based on the Riccati equation. This article mainly proposes a data-driven LQR control algorithm design method. The algorithm design framework can be independent of system model parameters, and based on the input/output data of the system. The least squares method can be used to learn the optimal controller parameters through iterative algorithm design. In the simulation experiment section at the end of the paper, the effectiveness and advantages of the proposed controller design method were verified.