This paper deals with a crucial aspect in the control of grid-connected power converters, i.e., the detection of the fundamental-frequency positive-sequence component of the utility voltage under unbalanced and distorted conditions. Specifically, it proposes a positive-sequence detector based on a new decoupled double synchronous reference frame phase-locked loop (DDSRF-PLL), which completely eliminates the detection errors of conventional synchronous reference frame PLL's (SRF-PLL). This is achieved by transforming both positive- and negative-sequence components of the utility voltage into the double SRF, from which a decoupling network is developed in order to cleanly extract and separate the positive- and negative-sequence components. The resultant DDSRF-PLL conducts then to a fast, precise, and robust positive-sequence voltage detection even under unbalanced and distorted grid conditions. The paper presents a detailed description and derivation of the proposed detection method, together with an extensive evaluation using simulation and experimental results from a digital signal processor-based laboratory prototype in order to verify and validate the excellent performance achieved by the DDSRF-PLL

This paper deals with the control of doubly fed induction generators to ride through unbalanced voltage sags. A control strategy is proposed by choosing certain current reference values in the positive and negative sequences so that the torque and the DC voltage are kept stable during such unbalanced sags. Both rotor- and grid-side converters are considered, detailing the control scheme of each converter while considering the effect of the crowbar protection. The control strategy is validated by means of simulations.

This paper compares different cascaded and multilevel topologies to interface supercapacitors to a dc bus in regenerative braking applications. It is shown that the modular multilevel dc/dc converter (MMC) can benefit from both reduced voltage and increased frequency across the inductor to reduce its weight and volume when using phase shifting modulation. The proposed control method is able to balance supercapacitor voltage while providing precise output current control. The converter topology and control method are validated with simulations and experimental results.

In this paper, a high-accuracy all-digital resolverto-digital (R/D) converter is presented. The two basic components of a conventional tracking R/D converter, the phase detector and the loop filter, are software implemented by frequency-shifting techniques and a decoupled double synchronous reference frame-based phase-locked loop (DSRF-PLL). This PLL allows the simultaneous extraction of the angular position and speed of the rotatory resolver, even in the presence of gain and phase errors in the resolver. In order to increase accuracy and to minimize the time lag of the whole system, oversampling methods and downsampling finite-impulse response digital filters are introduced. Finally, DSP implementation issues, like the use of techniques to synchronize the resolver output signals with the excitation one, are discussed. Using these combined techniques and a standard DSP with a 12-bit analog-to-digital converter, resolutions of up to 14 bits can be achieved with a computation cost of about 13% of the total (100 MIPs). The paper presents the main techniques, simulation, and experimental results.

Electric Vehicles (EVs) have seen significant growth in sales recently and it is not clear how power systems will support the charging of a great number of vehicles. This paper proposes a methodology which allows the aggregated EV charging demand to be determined. The methodology applied to obtain the model is based on an agent-based approach to calculate the EV charging demand in a certain area. This model simulates each EV driver to consider its EV model characteristics, mobility needs, and charging processes required to reach its destination. This methodology also permits to consider social and economic variables. Furthermore, the model is stochastic, in order to consider the random pattern of some variables. The model is applied to Barcelona’s (Spain) mobility pattern and uses the 37-node IEEE test feeder adapted to common distribution grid characteristics from Barcelona. The corresponding grid impact is analyzed in terms of voltage drop and four charging strategies are compared. The case study indicates that the variability in scenarios without control is relevant, but not in scenarios with control. Moreover, the voltages do not reach the minimum voltage allowed, but the MV/LV substations could exceed their capacities. Finally, it is determined that all EVs can charge during the valley without any negative effect on the distribution grid. In conclusion, it is determined that the methodology presented allows the EV charging demand to be calculated, considering different variables, to obtain better accuracy in the results.