Dr. Espinosa-Neira, Eduardo

The interest on weak and micro-grid systems has grown up substantially last time, specially tied up to distributed power generation systems (DPGSs), isolated systems as aircraft, or islanding power systems. These kinds of grids are usually under significant variation in their quantities, specifically in their voltage amplitude and/or frequency. On this line, many studies about synchronization methods have been developed, which may work under variations on the frequency value, unbalanced voltage, and even with harmonic distortion. However, power converters connected to this class of systems are poorly documented-specifically controlled rectifiers. In fact, most of the controlled grid connected converters (GCCs) are defined to work in a fixed frequency and balanced system. This paper deals with a GCC connected to a variable-frequency and unbalanced voltage supply system control through a predictive algorithm with a fixed resolution sampling strategy. Furthermore, the current references are imposed in order to help the weak-grid source subjected to unbalancing, taking more power from the phase with highest voltage amplitude and relaxing the other phases. This issue makes to calculate every phase current reference independently and accordingly the voltage amplitudes keep the dc-link voltage in a desired value. The results show the feasibility of the proposed algorithm, where the performance is highlighted by simulated and experimental waveforms.

Multi-cell converters are widely used in medium-voltage AC drives. This equipment is based on power cells that operate with low-voltage-rating semiconductors and require an input multipulse transformer. This transformer cancels the low-frequency current harmonics generated by the three-phase diode-based rectifier. Unfortunately, this transformer is bulky, heavy, expensive, and does not extend the existing power cell (three-phase rectifier—Direct Current (DC) voltage-link—single-phase inverter) to the transformer. In this study, a harmonic cancelation method based on finite control set-model predictive control (FCS–MPC), extending the power cell’s modularity to the input transformer. On the other hand, it considers treating the two disadvantages of the FCS–MPC: High switching frequency and spread spectrum. The details were developed in theory and practice to obtain satisfactory experimental results.

The use of controlled power converters has been extended for high power applications, stacking off-the-shelve semiconductors, and allowing the implementation of, among others, AC drives for medium voltages of 2.3 kV to 13.8 kV. For AC drives based on power cells assembled with three-phase diode rectifiers and cascaded H-bridge inverters, a sophisticated input multipulse transformer is required to reduce the grid voltage, provide isolation among the power cells, and compensate for low-frequency current harmonics generated by the diode-based rectifiers. However, this input multipulse transformer is bulky, heavy, and expensive and must be designed according to the number of power cells, not allowing total modularity of the AC drives based on cascade H-bridges. This study proposes and evaluates a control strategy based on a finite control set-model predictive control that emulates the harmonic cancellation performed by an input multipulse transformer in a cascade H-bridge topology. Hence, the proposed method requires conventional input transformers and replaces the three-phase diode rectifiers. As a result, greater modularity than the conventional multicell converter and improved AC overall input current with a THD as low as 2% with a unitary displacement power factor are achieved. In this case, each power cell manages its own DC voltage using a nonlinear control strategy, ensuring stable system operation for passive and regenerative loads. The experimental tests demonstrated the correct performance of the proposed scheme.

The use of power converters has considerably grown up, in part, because refined control strategies have been recently proposed, including nonlinear schemes such as predictive control. This approach is used in this work considering a variable grid frequency environment in order to get an appropriate response for a wide ac mains frequency range. Indeed, in order to achieve appropriate, both dynamic and static, responses for all operating frequencies, the number of samples per period is kept constant and independent of the ac mains frequency. This allows a fixed resolution of the sensed voltages and/or currents, which is preferable if high-performance digital control schemes are required. However, imposing a constant number of samples per period requires a variable sampling time in systems that feature variable ac mains frequency. On the other hand, predictive control has been developed and well documented just for a constant sampling frequency. This work presents how to extend the predictive control algorithms for variable sampling time allowing high-performance waveforms and wider ac mains frequency range. Simulated and experimental results show the feasibility of the proposed control strategy which corroborates the mathematical and model analysis.