Search

This paper presents a method for the control design of five-level DCC converters based on mixed-integer optimization and machine learning. The resulting controller is computationally simple and can be easily implemented on low-resource control hardware using simple nested "if-else" statements. The optimization problem is recalled from previous work by modifying the cost function to further enhance the dynamic performance. Additionally, and in contrast to previous works, the online implementation accomplished in this paper allows the system to cover a wider range of operating points. For this, the optimization problem is solved offline for several operating conditions, and the results are gathered into a dataset to train classification and regression trees (CARTs), which are later used online. Due to the generalization capability of the CARTs, a more flexible and less resource-intensive implementation is achieved which is capable of operating at points outside the ones considered in the training dataset. The resulting control strategy is compared in simulation and experiments with several alternative approaches found in the literature. This approach can be extended to other power converter topologies, allowing the implementation of optimized modulations. • Modulation of five-level DCC as a mixed-integer optimization problem. • Achievement of wider range of operating points. • Classification and regression trees used for implementation of optimization problem. • Guidelines for robust implementation of such tress. • In-detail comparison with existing approaches both in simulation and experiments. [ABSTRACT FROM AUTHOR]

Abstract: This paper addresses the problems associated with the dc-link capacitor voltages of the three-level neutral-point-clamped power converter: the imbalance of the capacitor voltages as well as the presence of an ac-voltage low-frequency oscillation in the dc link of the converter. In order to cope with them, a mathematical analysis of the capacitor voltage difference dynamics, based on a direct average continuous model, is carried out, considering a singular perturbation approach. The analysis leads to a final expression where a sinusoidal disturbance appears explicitly. Consequently, the two problems can be handled together using the ordinary formulation of a problem of regulating the output of a system subject to sinusoidal disturbances, applying classical control theory to design the controller. In this way, the controller is designed including the disturbance estimate provided by a Luenberger observer to asymptotically cancel the disturbance, while also keeping the capacitor voltages balanced. Experiments for a synchronous three-level neutral-point-clamped converter prototype are carried out to evaluate the performance and usefulness of the converter working as a grid-connected inverter under the proposed control law. [Copyright &y& Elsevier]

Abstract: In this paper, a novel control strategy for a nonlinear boost inverter is proposed. The idea is based on generating an autonomous oscillator that does not need an external reference signal. This aim is achieved by using energy-shaping methodology with a suitable Hamiltonian function which defines the desired system behavior. A phase controller is added to the control law in order to achieve 180°-synchronization between both parts of the circuit as well as synchronize the voltage output with a pre-specified signal, e.g. synchronization with the electrical grid. An adaptive control is designed for dealing with the common problem of unknown load. In order to analyze the stability of the full system, singular perturbation approach is used. The resulting control is tested by means of simulations. [ABSTRACT FROM AUTHOR]

This article revisits the modeling of the grid-connected neutral point-clamped converter as a time-varying nonlinear hybrid dynamical system. As a result, an embedded control law that guarantees stability and fulfills the control objectives, namely dc-link voltage regulation, power-to-grid energy exchange and capacitor voltage balancing, is proposed. In addition, a rigorous study of stability is carried out without depending on an averaged model and considering the hybrid dynamical character of the system. Moreover, the hybrid control guarantees a minimum dwell time. This approach could be extended to other topologies or control laws in the search for rigorous stability guarantee The algorithm is evaluated and tested experimentally, verifying the feasibility of the approach. This process involves adapting the algorithm to be programmed in a FPGA as well as some further modifications to improve the steady-state performance. The experimental results are performed under several reference changes and perturbations to prove the feasibility of the proposed approach. [ABSTRACT FROM AUTHOR]