Improved Estimation Method for the Capacitor Voltage in Modular Multilevel Converters Using Distributed Neural Network Observer

Abstract

The Modular Multilevel Converter (MMC) has emerged as a key component in HVDC systems due to its ability toefficiently transmit large amounts of power over long distances. In such systems, accurate estimation of the MMCcapacitor voltage is of utmost importance for ensuring optimal system performance, stability, and reliability. Traditional methods for voltage estimation may face limitations in accuracy and robustness, prompting the needfor innovative approaches. In this paper, we propose a novel distributed neural network observer specificallydesigned for MMC capacitor voltage estimation. Our observer harnesses the power of a multi-layer neural networkarchitecture, which enables the observer to learn and adapt to the complex dynamics of the MMC system. Byutilizing a distributed approach, we deploy multiple observers, each with its own set of neural network layers, tocollectively estimate the capacitor voltage. This distributed configuration enhances the accuracy and robustness ofthe voltage estimation process. A crucial aspect of our observer’s performance lies in the meticulous initializationof random weights within the neural network. This initialization process ensures that the observer starts with asolid foundation for efficient learning and accurate voltage estimation. The observer iteratively updates its weightsbased on the observed voltage and current values, continuously improving its estimation accuracy over time. Thevalidity of proposed algorithm is verified by the result of estimated voltage at each observer in capacitor of MMC.