Search

The grid forming inverter (GFMI) is a promising solution for the challenges introduced to the electrical grid by the high penetration of the inverter-based resources (IBRs). In this paper, GFMI with a virtual oscillator based on Andronov-Hopf oscillator dynamics control is modeled and tested in the phasor domain using DIgSILENT environment. To further investigate the inverter performance and operation, it is implemented in the IEEE 13 node distribution feeder. The results show that the inverter has a superior performance and the ability to stabilize frequency and voltage during disturbances in both grid-connected and islanded modes.

Today, renewable energy and microgrids are a major part of the electrical grid. The power sharing between different distributed energy resources (DER) is one of the main concerns about microgrids. Droop control is widely used to achieve accurate power sharing. The droop control depends on local voltage and frequency measurement. To provide better droop control performance, especially with the low voltage cables this paper proposes, is to provide droop control with measurements from the point of common coupling (PCC) using a communication system. In the case of a failure or a delay in the communication system, a trained radial basis function neural network (RBFNN) would be used to recover the PCC measurement using DER local measurement. The microgrid simulations were conducted using DIgSILENT software. The RBFNN training and testing were built using MATLAB®.

We propose a stochastic mixed-integer linear optimization model to size the storage system capacity of a solar system installed by electric power consumers at the distribution grid. Battery storage devices are chosen as the investment technology for storage capacity. Since multiple battery options are commercially available, the proposed model can accommodate multiple options to select the optimal option(s) and it's (their) optimal system capacity. The model is suitable for a solar net metering program, and its stochastic variables are the solar radiation and the consumer's power load. A numerical example with various experiments is used to conduct model's parametric analyses and assess the value of its stochastic solution. Experiment results demonstrated the model capability for selecting the best battery option when multiple options are available as well as indicated that solar consumers can decrease their total cost by investing in storage capacity and that the optimal number of battery units identified by the model is invariant for a large number of stochastic scenarios.

This paper proposes an alternative approach to finding the unknown switching angles of three-phase five-level cascade H-bridge multilevel inverters using the selective harmonic elimination PWM technique from the three-phase perspective over a wide range of modulation indices to improve Total Harmonic Distortion (THD). A mathematical model is introduced to find the pre-calculated angles using a three-phase modulation perspective in the interval (0, $\pi$/3) to eliminate selective harmonics on single-phase topologies that are used to construct three-phase inverters in high-power application. Looking from the three-phase perspective in the interval (0, $\pi$/3) to find the unknown angles improves the output in comparison with the interval (0, $\pi$/2) when looking from the single-phase perspective. Both approaches eliminate the selected harmonics; however, the proposed approach decreases the output voltage THD by 25% and increases the minimum value of the fundamental component by 125%. Further, finding the angles using the proposed method guarantees the desired shape of the voltage signal, further guaranteeing a minimum amplitude for the output generated voltage at low modulation indices. Simulation and experimental results are presented to validate the proposed method.

In this paper, a new set of algorithms has been developed to detect low charge fault for single compressor non-critically charged refrigeration system for both frozen and perishable goods. A mix of field units' data and experimental data were used to build the detection algorithms. Multiple detection algorithms were devised based on the operation modes. The system operates in multiple distinctive modes. Each mode has a defining characteristic. The used data includes several healthy and predefined low charge defected systems which gave us a prediction of the loss of charge percentage in different operational modes. This prediction provides us with a good knowledge of the system performance and operation.

A novel adaptive battery system with two battery sources is proposed to provide multiple output levels and increased benefits when integrated with multi-level inverters by improving the batteries' state of health (SOH) and state of charge (SOC). Applications of this topology include dynamic hybrid battery/inverter and photovoltaic/inverter systems. The system has been modeled and validated using Matlab software. Result show the benefits of the prosed system over existing approaches in overall operation and performance of the system.