Data-driven control and learning systems

The 2017 Special Section on Data-Driven Control and Learning Systems of the IEEE Transactions on Industrial Electronics provides the latest advances both in theory and application. A novel mixed iterative ADP algorithm is proposed to solve the optimal battery energy management and control problem in smart residential microgrid systems. Based on the data of the load and electricity rate, two iterations called P-iteration and V-iteration are constructed. In another paper, an integrated model-data-based zero phase error tracking feedforward control (ZPETFC) strategy is proposed for high-precision motion systems with complex and non-minimum phase (NMP) dynamics. A data-assisted modeling method for motion control of a two-link robotic fish is presented to tackle the unavailability of the complex hydrodynamics thrust mechanism. A new circuit has been proposed for implementing a reduced-interval type-2 neural fuzzy system using weighted bound-set boundaries (RIT2NFS-WB) with online tuning ability. In another paper a novel optimal learning algorithm for partially unknown voltage-source inverters (VSIs) operating in parallel is presented. A novel data-driven framework using kernel partial least squares based on optimized preference matrix for fault diagnosis is presented. Compared with traditional methods, the proposed method can overcome the drawback of original features loss of the centralized mapped data in the feature subspace and can improve the accuracy of fault diagnosis.