Search

This paper proposes a fixed zone perturb & observe (FZPO) technique to achieve an improved steady-state efficiency as well as a fast and drift-free maximum power point tracking (MPPT) for photovoltaic (PV) systems without any additional sensors. In this technique, the PV array’s power-voltage curves are divided into multiple zones with unique zone boundary voltages for different irradiance conditions. This technique employs a combination of adaptive and fixed step-sizes to improve the performance, where the adaptive step-size is calculated from simple mathematical equations, resulting in a reduced computational burden. Besides, natural drift-free tracking is achieved without any additional sampling or computation, overcoming existing techniques’ limitations. The FZPO technique requires a PV panel’s information only at the initial design stage but not during real-time tracking, making the implementation possible using only low-cost processors. This paper presents the implementation and design of the proposed controller. The FZPO technique’s performance is validated through a comprehensive set of hardware experiments on a buck-boost full-bridge converter under various irradiance conditions per the EN50530 standard. For a step-change in irradiance, the FZPO technique is experimentally verified to be 42% and 20% more efficient than the conventional and VSS techniques, respectively. During the irradiance varying conditions using the FZPO technique, the peak power loss is one-sixth compared to the conventional and VSS techniques. Practical field-related considerations such as PV panel temperature effects are further investigated through experiments.

Lithium-ion battery is generally used as energy storage device in spacecraft and other applications. Spacecraftbattery modeling and energy balance estimation is critical. The proposed genetic algorithm driven generic estimation - energy balance (GAGE-EB) model uses cell open circuit voltage characteristics, internal resistance ( $r_{\text{int}}$ ) characteristics, and coulomb counting. The cell characteristics are obtained by offline tests. For series-parallel or parallel-series configured battery, the model estimates deliverable capacity and internal resistance by accurately matching the terminal voltage. Estimated parameters are used to check the energy balance for new load and generation profiles. The GAGE-EB model is validated for electrical performance with various cell types and battery configurations with respect to experimental data and is within 1% and 2% accuracy for voltage and capacity estimation, respectively.

The H-bridge buck–boost converter shows better efficiency compared to all other buck–boost converter variants due to the ability to work in buck and boost modes. The existence of dead-zone in voltage gain during mode transition due to the maximum and minimum duty limitations results in instability and oscillations in output voltage. This dead-zone is eliminated by overlapping the buck and boost gate signals, which results in efficiency decrement due to high switching loss. Using a common compensator in all the modes in the H-bridge buck–boost converter results in poor dynamic performance. In this article, dead-zone free control (DZFC) scheme with increased efficiency and improved dynamic performance is proposed to mitigate the dead-zone during mode transition. In the proposed DZFC scheme, dead-zone elimination is independent of the values of mode boundaries, which reduce the transition mode duration. The proposed DZFC scheme improves the efficiency during transition modes by reducing the switching loss, a generalized small-signal model valid for the buck, boost, and transition modes and a control design is proposed for improved dynamic performance. The proposed DZFC scheme is verified experimentally on a hardware H-bridge buck–boost converter prototype.

In this paper, a multisource switched capacitor based boost multilevel inverter is proposed. The proposed topology features less number of components, high gain, self-charging capability of capacitors, reduced size of filter components. The modes of operation of the proposed converter and design guidelines of converter components are discussed. A comprehensive comparative study of the proposed converter topology with the topologies in the literature is given. Analytical expression for capacitor energy utilization (CEU) factor for the proposed topology is derived and the feasibility of the proposed converter with respect to CEU for the designed power rating is verified. The analytical performance of the proposed converter topology is verified through simulation studies. The common mode voltage and leakage current in the proposed topology are analyzed. Experimental prototype of the proposed converter topology with two sources and two capacitors for power rating of 500 W is developed and tested with resistive, inductive, and nonlinear loads. The experimental and simulated results are in close agreement with the analytical steady-state performance of the proposed converter.

Generally, battery-fed capacitive loads are charged to high voltages in series of switching cycles with smaller fragments of energy delivered to the load in every cycle to ensure an efficient charging process. In high voltage low power (HVLP) flyback charging circuit, significant portion of input energy drawn from the source in a cycle is utilized by the effective parasitic capacitance of the HV transformer and semiconductor devices and circulated back to the source at the end of the cycle. This reduces the energy delivered to the load per cycle and hence the net energy delivered in the charging process. In this paper, an energy-based analysis and simplified energy-based model for HVLP charging circuits is derived. An energy-based approach is appropriate to describe and define the HVLP converter behavior in terms of load energy component and circulating energy component. The derived energy-based model is extended to obtain analytical closed-form expressions governing the essential parameters of the charging circuit. The proposed energy-based model and analytical formulation of essential parameters are experimentally validated on a 0–2.5 kV programmable high voltage charging circuit prototype fed from 12 V input battery source.

In this paper, a deadbeat control is proposed with reduced steady-state error in grid connected inverter system. The proposed method takes into consideration the losses occurring in the inverter during each switching cycle. Initially, a method to track the reference current is proposed and the same is extended for direct power control. The proposed method does not require phase-locked-loop, transformation into rotating reference frame and output voltage sensor thereby improving the transient response. The proposed method is verified for a single-phase grid connected switched capacitor based multilevel inverter. The simulation results of proposed current control are presented for step change in current reference. The proposed deadbeat direct power control is verified for PV connected system with varying irradiance.

Buck-Boost based full-bridge DC-DC converters possess potentials for high gain, high power applications, particularly in solar PV, battery, and fuel-cell fed systems, as the converters feature non-pulsating input and output currents. However, these converters lack attention due to the presence of DC-current in the transformer winding. In this paper, a novel Buck-Boost full-bridge (BBFB) converter with a hybrid control scheme (HCS) mitigating the transformer DC-current is presented. The BBFB converter exhibits inherent soft-switching such that zero voltage switching (ZVS) conditions apply for individual switches. This paper analyzes the BBFB converter extensively, including the discontinuous conduction mode (DCM) operation and the DCM boundary condition. A dynamic behavior of the BBFB converter under a load step change verifies that the HCS scheme does not affect the converter performance. Besides, this work presents a model for the high frequency oscillations that occur in the practical transformer current waveform due to parasitic capacitances. All the analyses and the developed models are verified in simulations and hardware experiments. The developed models are useful for designing the BBFB converter with improved efficiency by ensuring the ZVS operation. Further, the developed models and results provide an insight for the DC voltage gain variations during DCM and continuous conduction mode (CCM). This helps the designer to choose the BBFB converter’s operating mode based on the requirement.

High voltage pulses of the output power ranging from 5 W to 200 W are employed in food processing systems, water treatment process, electroporation, industrial and medical applications etc. The power converter configuration implemented to generate high voltage pulses is selected depending on the peak power rating requirement. In this work, a flyback converter configuration is selected to charge the energy storage capacitor meeting the desired voltage magnitude and charging time constraint. The energy storage capacitor is charged following peak current mode control with boundary conduction mode. In this proposed work, high voltage rectangular pulses of 1000 V magnitude with the desired pulse repetitive rate and pulse width is designed for various case studies and validated through corresponding simulation results.

Traditional SONAR (SOund Navigation And Ranging) systems use conventional rectifiers for converting the input AC voltage for fulfilling the power requirements. The disadvantage of using these rectifiers is elevated losses in the system, resulting in lower system efficiency. The total harmonic distortion (THD) and input power factor (PF) are low. Improved performance with respect to THD and PF could be obtained by using a switched-mode power converter. A soft-switching based buck-boost PFC power converter operating in DCM is proposed in this paper. The steady-state interval of operations of the proposed configuration is presented. The steady-state analysis of the proposed PFC converter is verified in simulations carried out for a full-bridge inverter system with 1 kW / 1200 V rms output fed from the proposed front end PFC converter. The simulation results are in agreement with the analysis.

A multi input bidirectional LCLC resonant converter for hybrid energy systems is proposed. The proposed converter consists of four voltage sources and transfers power from different combination of input sources to DC link based on the power availability with the sources. The converter charge the batteries in battery charging mode thus enabling bidirectional capabilities. All the high frequency operated switches in the converter turn on with ZVS in both the modes of operation. The converter achieves good voltage regulation for the load as well as input voltage variations using frequency modulation. Simulation results and design guidelines are outlined for the proposed converter.

The single-phase boost Power Factor Correction (PFC) converter has become an essential front-end converter in most power electronic systems. The need to ensure good converter performance under constraints of high power density and low costs has made converter control challenging. Deadbeat Predictive Current Control (DPCC) can offer fast control action; however, careful digital implementation is essential to ensure high performance. This paper analyses and presents the various control implementation aspects to be considered for DPCC of the single-phase boost PFC converter with the help of analytical, simulation and experimental studies.

The Series-parallel (s-p) configured Lithium ion batteries find use in many spacecrafts. Cell selection to make a battery pack involves sorting tested cells to meet screening and matching criteria. Cell capacity, cell resistance, and self-discharge could be used for cell selection. Conventionally, data is linearly sorted into ascending or descending order based on one parameter. Matching criteria are applied to choose cells to be put into strings. Cell sorting using the Genetic Algorithm (GA), involving mixed-integer linear optimization, is proposed in this paper. Cells are arranged randomly to match all of the parameters within a set of tight constraints. A better match for the cells is achieved by multiple iterations. GA Improved Cell Sorting Algorithm (GICSA) is verified with experimental data and has given close to 100% cell yield.

In this paper, a loss modulated direct power control method is proposed for a single-phase grid connected photovoltaic system. The proposed method employs the concept of adding the modulated signal corresponding to converter losses with the modulated signal corresponding to power transfer with ideal inverter. The proposed method is explained with the model of grid connected voltage source inverter with and without converter losses. The implementation of the control method does not require phase-locked-loop resulting in improved efficiency and response time. Further, an algorithm is proposed to generate the reference power required to implement the direct power control with maximum power point tracking for multisource photovoltaic system. The proposed control method and algorithm is implemented with switched capacitor based boost multilevel inverter. The simulation results under varying irradiance and reactive power are presented. The results show the good tracking performance of maximum power point along with real and reactive powers.

Finite Set Model Predictive Control (FS-MPC) of three phase Vienna rectifiers has gained popularity due to its superior control action and ability to handle system constraints. However, FS-MPC performance depends largely on the accuracy of the predictive model of the system. The presence of unmodelled converter non-idealities, parameter variations and model inaccuracies leads to input current distortions. A Moving Average (MA) based FS-MPC algorithm is proposed in this paper to improve the input current quality. The proposed MA based FS-MPC algorithm does not require the values of the converter non-idealities and their variations. The proposed MA based FSMPC algorithm is validated in simulation studies.

This paper presents a generalized predictive Discontinuous Conduction Mode (DCM) average current control law and a digital method for conduction mode detection suitable for DC-DC converters and PFC converters. The DCM current controller is derived considering the practical aspects in the digital implementation. The mode detection method does not require additional circuits/sensors. The salient control equations are derived for boost, buck and buck-boost topologies using the derived generalized equations. The derived generalized predictive DCM average current controller and mode detection method are verified in simulations for a boost DC-DC converter and experimentally on a boost PFC converter hardware prototype.

Single stage DC-AC converter with sinusoidal phase shifted full bridge converter as the front end followed by cycloconverter is commonly used for microgrid/EV applications. The operation and analysis of the converter with cycloconverter switching at line frequency and combined line/high frequency are presented in this paper. The control configuration implementation for single stage DC-AC converter with a sinusoidal phase shifted front end followed by a cycloconverter is presented which ensures soft transitions for all the semiconductor devices. The comparison of line frequency switching and combined line/high frequency switching of cycloconverter devices of single stage DC-AC converter in terms of performance and efficiency is presented. Analytical expressions for the computation of losses in the cycloconverter switched at line frequency and combined line/high frequency are proposed. The control, analysis, and mathematical expressions are verified in simulations and 500 W single stage DC-AC converter experimental prototype.

In this paper, a shifting carrier based maximum power point tracking (MPPT) algorithm for multisource converters is proposed. This MPPT algorithm is explained for both low frequency (constant carrier) and high frequency (triangular) carrier signals. The proposed algorithm is applied for a single stage multisource switched capacitor based multilevel inverter for grid connected and stand-alone photovoltaic applications. This converter features high gain, less component count, self capacitor charge balancing, better dispersed value of sources and increased source utilization. The importance of selecting an appropriate pulse width modulation technique for switched capacitor based inverters in order to maintain the capacitor charging time is explained. The proposed technique is able to track the maximum power under varying irradiance levels. The experimental results of the proposed algorithm is presented. The MPPT technique is analyzed and verified in simulation for stand-alone and grid connected applications using MATLAB/Simulink platform.

The paper presents the steady-state analysis of the step-up/step-down operation of the full bridge resonant transition converter (FBRTC). In spite of several work being reported in the literature on the FBRTC converter, the performance differences during the step-up and step-down operations are less investigated. Influence of leakage inductance, magnetizing current of the transformer and switching frequency on the steady-state DC-voltage gain of the step-up/step-down converters is presented with the supporting results. The iterative design methodology for the step-up/step-down converters are proposed to enhance the efficiency. The investigation carried out in this work help the designer to select the appropriate leakage inductance, magnetizing inductance, turns ratio of the transformer and switching frequency of the step-up and step-down converters. The analysis and design methodology are validated in simulations. The performance analysis is verified in a 500 W experimental prototype.

Ion mass analyser is an analytical space instrument employed to characterise the spatial ions. The guiding plates of the ion mass analyser require high voltage drive in the range of 0–5kV and is powered by a low voltage battery source. Flyback converter is widely employed for high voltage low power charging applications, fed from low voltage input source. Implementation of conventional flyback converter topology demand semiconductors of high voltage blocking capability greater than 5 kV to ensure 0–5kV charging and discharging process. Considering the limited commercial availability of high voltage active semiconductors at the operated low power level (

In this work a simple, analytic model is proposed to estimate the depth-dependent temperature profile in multilayered power electronic systems. The proposed model considers both upward and downward heat flow, with heat conduction within the structure and heat convection at the heat sink boundaries. The proposed model is verified with simulations and experimental data obtained on multi-layered power devices like solar cell and a power converter having an insulated gated bipolar transistor (IGBT) and a diode. Besides being computationally efficient, the proposed analytical model predicts the temperature profile as accurate as obtained from commercial numerical simulators. The experimental verification of the model for a multi-layered solar cell structure is also presented.

Single-stage DC-AC converter systems (SSCSs) are developed by merging both DC-DC and DC-AC conversion processes. These SSCS transfer power from photovoltaics (PV) to loads/grid in one stage. In this study, a PV-based grid tied single-stage DC-AC converter system (GTSS) for low-voltage PV systems is presented. Generally, when SSCSs are connected to the grid, grid voltage template is used to generate reference current which is to be tracked. In doing so, the GTSS delivers power to the grid at a power factor other than unity. A suitable control scheme is implemented in this study to control active and reactive powers' flow from the GTSS to the grid while continuously operating PV at maximum power point. The proposed control scheme of GTSS has attributes such as less complexity and requires simple proportional integral controller. Modelling of control loops in the proposed control scheme is mentioned. A detailed MATLAB/SIMULINK-based simulation and experimental studies are conducted to validate the proposed control scheme.

In High Voltage flyback converters, the dominant factor that influences a converter operation is the parasitic capacitance. A significant portion of input energy is utilized in charging the parasitic capacitances of the circuit, which is circulated back to the source at the end of every switching cycle. The circulating energy is a function of output voltage, load power, and parasitic capacitances and remains significant in High Voltage Low Power (HVLP) applications. This energy transfer phenomenon involving parasitic capacitances results in a reduced fraction of input energy reaching the load in every cycle, thereby resulting in an apparent deviation in the converter operating point compared to ideal flyback in case of resistive loads. An analytical energy-based model is derived, which includes the effect of parasitic capacitances, and is valid for steady state and dynamics of HVLP flyback converters feeding resistive loads. The influence of parasitic capacitances on the switch voltage of the converter is exploited to achieve Zero Voltage Switching (ZVS), thereby minimizing the turn- on loss. The proposed analytical model is verified through simulation and experimental results on 1.5 kV/ 5 W and 1.5 kV/ 200 mW resistive loads.

The phase modulated full bridge resonant transition converter (PMRTC) is commonly used for high-gain, high-power applications, as PMRTC converters retain the qualitative nature of hard switched pulse-width modulation counterparts with additional damping ( $R_d$ ), where $R_d$ is load, transformer turns ratio and frequency dependent. The PMRTC converter enables operation at a higher switching frequency due to the soft-switching nature, thereby achieving better power density. For battery fed high-power, high-gain applications, PMRTC exhibits a significant drop in the dc gain due to $R_d$ and other nonidealities, limiting the maximum frequency of operation. Thus, computing this drop and analyzing the effects of switching frequency becomes necessary in achieving the required steady-state dc gain. From the analysis, it is observed that, for a choice of switching frequency above $f_{s\;{\rm critical}}$ , the required steady-state dc gain is not achieved for any turns ratio. Hence, to increase the switching frequency of operation of a PMRTC converter, a two-transformer configuration is adapted and with this configuration, the dc gain loss model, power loss, and efficiency model and small-signal model are established. The design guidelines on the choice of switching frequency and transformer turns ratio based on the proposed models is described. The proposed models with the presented design guidelines are validated in simulations and hardware prototype for a 1 kW PMRTC converter.

Converter system that does both dc–dc conversion and dc–ac conversion in a single stage is called a single-stage converter system (SSCS). Compared to a two-stage converter, SSCS is more efficient and compact in size. In this paper, a grid-connected SSCS (GCSS) system for photovoltaic (PV) applications is presented. This GCSS transfers power from PV to grid while tracking maximum power point (MPP) continuously. The proposed system has several desirable features such as low switching loss, high gain, and compact size that makes this GCSS suitable for PV systems, where the PV system output voltage is low and varies with time. Design of the presented GCSS components with necessary equations and derivation of both current control loop and voltage control loop required for operation are given. Proposed GCSS is simulated using MATLAB/SIMULINK. Detailed simulation and experimental results are given to verify the efficacy of the proposed GCSS.

The design of the current control loop in a boost PFC converter plays a significant role due to the large signal variations in the input waveforms. A predictive average current control law based on triangular modulation considering the converter non-idealities, time variation of waveforms and computation delays is proposed and validated with simulation and experimental results. A digital method for the boost inductance value estimation is also proposed which could be used with predictive current control laws.

A high gain, single-stage DC-AC converter with high frequency AC link is proposed in this paper. The single-stage DC-AC converter has the advantage of high efficiency, less component count and high power density. The proposed converter consists of a high frequency switching buck-boost bridge followed by a low frequency switching cycloconverter. A hybrid sinusoidal PWM switching scheme is used for the buck-boost bridge to control the output and to mitigate the DC-current in the transformer primary winding. In the proposed converter, soft transition is achieved in all the devices. The steady-state voltage gain achieved is twice that of the conventional H-bridge phase-shift converter. The input current is continuous in the proposed converter which reduces the DC bus capacitor requirement for solar PV fed applications. The steady-state analysis with the modes of operation of the proposed converter are presented. A 2 kVA proposed converter feeding the R and RL load is designed and implemented in the simulations and the analytical model is verified. The comparison studies of the proposed converter with the topologies in literature are presented.