A state-of-the-art Machine Learning (ML) based approach, by modeling the behavior of Gallium Nitride (GaN) power electronic devices, is presented in this paper. Switching voltage and current waveforms of these novel devices are accurately predicted using the developed supervised ML algorithm. This was utilised to build a more generic black-box model for these devices. Moreover, long short-term memory unit (LSTM) and gated recurrent unit (GRU) device models have been proposed to make the approach more user friendly. The performance of the developed approach is verified using a set of simulations and experimental tests under 450 V, 10 A test conditions. Model results demonstrate an error rate of 0.03 and convergence speed of 3s with excellent stability. Compared to the existing models, the developed ML-based model produces more accurate results, converges faster and has a better stability. Additionally, the developed ML-based GaN model offers the ability to select the best fit available GaN model (Panasonic, GaN Systems, Transphorm etc.). It automatically configures them into a system that would optimally yield the desired power conversion. This enables a shorter learning curve for the power electronics community, which would lead to acceptance and faster adoption of these devices by the power electronics industry.

This paper proposes a three-port Zeta-KY dc-dc converter which is fed with hybrid sources like photovoltaic (PV) cells and batteries. The converter proposed here is a multi-input single-output (MISO) structure which harnesses the benefits of Zeta and KY converters. The combination of these converters is highly advantageous since the Zeta converter provides lesser output voltage ripples with high gain and the KY converter topology suits well for withstanding load transients. The KY converter used in this research work is subjected to a topological change to facilitate bidirectional power flow. The bidirectional flow is essential to save the excess power in PV source in batteries during low load conditions. This novel multiport topology with bidirectional facility is first of its kind and has not been discussed earlier in the research arena. In the proposed work, two control algorithms are developed and deployed: the first one ensures the maximum power extraction from the PV and the second one maintains constant dc bus voltage and manages bidirectional power flow. MATLAB Simulink and hardware prototype of the proposed system has been realized for a 72 V dc bus and a 500 W electric vehicular drive. The simulation and experimental results reveal that the proposed system is viable for medium power electric shuttle applications. The proposed system is subjected to various test cases and it is observed that the source and load intermittencies are catered very well by the proposed three port Zeta-KY converter. The developed multiport converter is feasible for renewable energy applications.

Induction heating (IH) is a process of heating the electrically conducting materials especially ferromagnetic materials with the help of electromagnetic induction through generating heat in an object by eddy currents. A well-entrenched way of IH is to design a heating system pertaining to the usage of ferromagnetic materials such as stainless steel, iron, etc., which restricts the end user’s choice of using utensils made of ferromagnetic only. This research article proposes a new scheme of induction heating that is equally effective for heating ferromagnetic and non-ferromagnetic materials such as aluminium and copper. This is achieved by having a competent IH system that embodies a series resonant inverter and controller where a competent flexible load modulation (FLM) is deployed. FLM facilitates change in operating frequency in accordance with the type of material chosen for heating. The recent attempts by researchers on all metal IH have not addressed much on the variable shapes and sizes of the material, whereas this research attempts to address that issue as well. The proposed induction heating system is verified for a 2 kW system and is compatible with both industrial and domestic applications.

Photovoltaic (PV) systems-based energy generation is relatively easy to install, even at a large scale, because it is scalable in size and is thus easy to transport. Harnessing maximum power is only possible if maximum power tracking (MPPT) functionality is available as part of the power converter control that interfaces the PV panels to the grid. Solar exposure covering all PV panels is unlikely to happen all the time, which is known as a partial shading (PS) phenomenon. As a result, depending on the MPPT algorithm adopted, it may fail to find a maximum global power peak, being locked into a local power peak. This research work discusses an alternative MPPT control technique inspired in the social group optimization (SGO) algorithm. SGO belongs to the meta-heuristic optimization techniques family. In this sense, the SGO method ability for solving global optimization problems is explored to find the global maximum power point (GMPP) under the presence of local MPPs. The introduced SGO–MPPT was subjected to different PS conditions and complex shading patterns. Then, its performance was compared to other global search MPPT techniques, which include particle swarm optimization (PSO), the dragon fly algorithm (DFO) and the artificial bee colony algorithm (ABC). The simulation outcomes for the SGO–MPPT characterization showed good results, namely rapid global power tracking in less than 0.2 s with reduced oscillation; the efficiency of solar energy harness was slightly above 99%.

The output of a photovoltaic array is reduced considerably when PV panels are shaded even partially. The impact of shading causes an appreciable loss in power delivery, since the PV panels are connected in series and parallel to contribute to the required voltage and power for the load. The prevailing research on mitigating the shading impact is mostly based on complex reconfiguration strategies where the PV panels are subjected to complex rewiring schemes. On the other hand, to disperse the shading many studies in the literature defend the physical rearrangement of the panels. The available intensive reconfiguration schemes, such as the series parallel (SP), bridge link (BL), honeycomb (HC), and total cross tied (TCT) schemes, try only to mitigate the shading impact and there is no scope for compensation; as a result, a loss of output power is inevitable. In the proposed research work, both the mitigation of and the compensation for the losses incurred due to shading are studied. In this work, an optimal reconfiguration scheme is adopted to reduce the shading impact and a power electronic circuit with a battery source is designed to compensate for the shading losses in all aspects. In the optimal reconfiguration scheme, a bifurcation strategy is adopted in each column and the electrical connections of the PV panels are interchanged such that the shading impact is dispersed. The power electronic circuit consists of a half-bridge buck converter with a battery source that injects the current required by a shaded column. This setup compensates for the shaded PV array’s power and improves the efficiency of the total system. The proposed scheme was implemented in a 3200 W system and subjected to various shading patterns, including single panel shading, corner shading, long and wide shading, and random shading. The proposed scheme was simulated in the MATLAB Simulink environment and compared with static 4 × 4 PV array configurations, including the series parallel (SP), bridge link (BL), honeycomb (HC), and total cross tied (TCT) configurations. The comparative performance was assessed in terms of mismatch power loss, fill factor, and efficiency. The proposed system is suitable for all shading patterns and was proved to be very efficient even in the worst shading, where 1353 W was saved.