This paper shows a new algorithm in the natural abc coordinates to calculate the current reference in a shunt active power filter. It modifies the Fryze's algorithm based on classical power theories for obtaining an optimal estimation in the current reference when the voltage in the utility is distorted. The aim of the method is twofold; first, to mitigate the harmonic components and second, to improve the power factor in the system blocking the reactive components. The feasibility of the algorithm in real-time applications has been tested with different waveforms and the results are compared by means of the total harmonic distortion in the grid current and the system power factor before and after the compensation. In addition, the results obtained in the experimental setup show the correct performance of the algorithm and demonstrate its easy implementation in industrial systems.

The electric energy demand is dramatically growing worldwide and demand reduction emerges as an outstanding strategy; it implies detailed information about the electricity consumption, namely load disaggregation. Typical automatic methods for load disaggregation require high hardware efforts to install one sensor per appliance, whereas Non-intrusive Load Monitoring (NILM) systems diminish the hardware efforts through signal processing and mathematical modeling. One approach to NILM systems is to model the load signatures via artificial intelligence. This paper proposes to employ S-Transform for the feature extraction stage and Support Vector Machines for the pattern recognition problem. Several experiments are presented and the results of the feature extraction with S-Transform and Wavelet Packet Transform are compared. Thus promising feature vectors based on S-Transform are presented with similar or superior performance than the approach based on Wavelet Packet Transform.

Multi-functional photovoltaic (PV) inverters incorporate ancillary services to enhance power quality and mitigate stability issues in distribution networks. These next-generation PV inverters will achieve a higher utilization of the inverter’s rated capacity, improving the cost-effectiveness of PV technology. However, the power required to perform ancillary services, such as load compensation capabilities, could exceed the inverter’s capacity, risking the components’ integrity. Therefore, multi-functional control algorithms must limit the power capacity according to the system’s nominal currents. Despite this, most control proposals do not address this issue when load balancing capabilities are included for PV inverter control. This paper proposes a flexible control strategy for three-phase multi-functional PV inverters, considering load balancing functionalities while keeping the inverter currents within safe operating limits. The proposal introduces two control parameters whose variation results in different load compensation capabilities. These parameters can be adapted dynamically according to the inverter rated capacity not used for active power injection and the load compensation requirements. Additionally, a control algorithm is proposed to limit the inverter current according to the nominal values supported by the device. This algorithm also allows setting compensation objectives following a priority scheme in which the injection of the PV active power is prioritized over the load compensation functionalities. Reactive power compensation and load balancing functionalities are also considered at a lower level of hierarchical priority. The proposal was evaluated through experimental tests on a multi-functional PV inverter prototype under various operational conditions. The experimental results show an excellent control strategy performance, achieving the control objectives under unbalanced load conditions.

This paper presents a new formulation of the voltage sag state estimation problem based on compressive sensing theory. The growing economic losses for voltage sags have led to a search for new mathematical methods for voltage sags diagnosis. Several studies have been focused on optimization problems based on techniques that are inaccurate when faults with large impedance are considered. To overcome these limitations, we proposed a ℓ1-based voltage sag state estimator (ℓ1-VSSE). A limited number of voltage sag meters, sensing matrices using residual voltages per unit, and the solution of a ℓ1-regularized least square problem (LSP) for each voltage sag detected, are novel characteristics of the proposed estimator. The ℓ1-VSSE has been validated by using the IEEE30-bus power system and the IEEE69-bus distribution system. The outcomes validate the efficient performance in the voltage sags estimation and its robustness to the fault types, fault impedance, and meshed or radial systems.