Data-Driven Distributed Combined Primary and Secondary Control in Microgrids

Abstract

This paper presents a comprehensive data-driven distributed combined primary/secondary controller design method for microgrids. This method provides transient and steady-state performance including power-sharing and voltage and frequency restoration while guaranteeing stability for fixed communication delay. The measured data is directly used for controller design with no need for knowledge about the order or structure of the system and grid physical parameters. Moreover, no assumption is made on X/R-ratio of feeders. All the control specifications are formulated as frequency-domain constraints on the 2-norm of weighted sensitivity functions. Then, using a recently developed frequency-domain robust control design method, a distributed fixed-structure controller is synthesized in one step. The performance of the obtained controller is validated using Hardware-In-the-Loop (HIL) experiments. The results show considerable improvement in transient performance, while providing power-sharing and voltage and frequency restoration using distributed implementation. Index Terms-Data-driven controller design, frequency control, voltage control, power-sharing, microgrid control, distributed control.

Grid Forming Inverter Modeling, Control, and Applications

“…Content may change prior to final publication. [3], [21]- [27] Angle-based [28], [29] PSC [30], [31] VISMA [32], [33] Swing equation emulation [34]- [42] Augmented VSG control [4], [43]- [49] Synchronverter [50]- [55] Matching control [56]- [58] Virtual oscillator-based [59]- [68] H∞\H 2 -based [69], [70] ViSynC [71] Frequency shaping-based [72] (K Droop ) is…”

“…In the control design, the performance and stability specifications are defined as constraints on the ∞-norm of the sensitivity functions, and the optimization problem is solved iteratively until an optimal controller is reached. The robust fixed structure control design is extended to distributed control in [70]. The control design is based on an experimentally identified frequency response, and the order of the controller is not restricted, and it is up to the designer.…”

Safe Data-Driven Secondary Control of Distributed Energy Resources

“…Predictions made by the sensitivity-based prediction model ( 5) can become less accurate, as a result of large disturbances or lack of control excitation, causing the controller to take unsafe actions. In order to achieve a safer operation, we propose to incorporate the inferred knowledge about the modeling residuals (12) into the optimization (27) as follows:…”

“…The authors of [11] propose a data-driven voltage regulation approach based on the estimated topology and line parameters of radial power distribution systems. A number of works (see, e.g., [8], [12]) proposed data-driven approaches that use measurement data to directly perform controller synthesis without system identification. For example, the authors of [8] propose a data-driven frequency control method that uses the datasets generated by the linear quadratic regulator (LQR) to learn a static feedback gain viable for various amounts of inertia present in the system.…”

“…For example, the authors of [8] propose a data-driven frequency control method that uses the datasets generated by the linear quadratic regulator (LQR) to learn a static feedback gain viable for various amounts of inertia present in the system. The authors of [12] propose a data-driven H 2 optimization that directly uses measurements for synthesis of primary and secondary controllers. Another body of works (see, e.g., [7],…”