Shahab Poshtkouhi

A General Approach for Quantifying the Benefit of Distributed Power Electronics for Fine Grained MPPT in Photovoltaic Applications Using 3-D Modeling

This paper deals with photovoltaic power installations in urban environments. A general simulation method is developed to quantify the total energy yield for photovoltaic (PV) installation sites exploiting different levels of Distributed Maximum Power Point Tracking (DMPPT) granularity. The process includes 3-D modeling, shading evaluation of the installation site, and irradiance calculations on the PV surfaces on an hourly basis throughout the year. Three leading microconverter topologies are analyzed and the cost/performance tradeoff is discussed for panel-level DMPPT. The energy yield evaluation technique is confirmed by means of several miniature PV acquisition units for frequent irradiance and temperature measurements in the installation site. The yearly energy yield benefit is shown to be highly dependent on the relative shading in the three installation sites. It is found that the energy yield benefit easily outweighs the power electronics costs in two of the three installations for panel-level DMPPT. The analysis method can be used by PV installers and system designers for accurate energy yield prediction, as well as power electronics engineers who need to bound the cost of their design based on the net energy benefit of the installed PV system.

DC-DC converter for high granularity, sub-string MPPT in photovoltaic applications using a virtual-parallel connection

This paper focuses on fine-grained, sub-string MPPT in photovoltaic applications. An accurate energy forecasting analysis using 3D modeling of an urban PV installation site is used to show that performing MPPT at the sub-string level inside the panel can result in a substantial increase in the yearly energy yield. The virtual-parallel concept is used to exploit the robustness of parallel-connected PV modules to irradiance mismatches, while maintaining the wiring advantages of the physical series configuration. A power-efficient 300 kHz dc-dc converter prototype with three asymmetrical phases is implemented to demonstrate the benefits of DMPPT with substring voltage equalization within the panel. A variable interleaving scheme is used to reduce the output voltage ripple during shading conditions, which increases the output capacitor lifetime and improves the system reliability. The prototype is measured with a single PV panel under a variety of shading test cases to confirm the power benefit of the two auxiliary phases, which reaches up to 30%.

Analysis of distributed peak power tracking in photovoltaic systems

It has been demonstrated that performing localized maximum peak power tracking (MPPT) on each photovoltaic (PV) panel, instead of using a single MPPT controller across the PV string can substantially increase the total harvested power, since each panel experiences unique illumination and temperature conditions. In this work, the effect of the dc-dc converter efficiency on the power savings from distributed MPPT (DMPPT) is analyzed for a wide range of test cases and different PV panel parameters. The benefit of DMPPT for a practical system is shown to be up to 25% for a standard deviation of σ= 0.2. A set of modular hardware-based PV panel emulators (ePVs) is presented. The ePVs can be programmed to match the unique i/v curves of real panels under various conditions and can therefore be used to optimize future DMPPT systems.

Multi-input single-inductor dc-dc converter for MPPT in parallel-connected photovoltaic applications

This paper focuses on photovoltaic systems with multiple parallel-connected panels. It is shown that distributed MPPT must be performed on each panel to maintain maximum power harvesting in partial shading conditions. This is especially true for PV systems made with panels having different electrical parameters. The multi-input, single-output (MISO) dc-dc converter provides a low-cost implementation of distributed MPPT for solar applications. A controller with digital peak and valley current control is used to operate the MISO converter in pseudo-CCM mode. A digital input current estimation algorithm based on the inductor current is proposed to iteratively reach DMPPT for each input, while eliminating the need for several current sensors in the system. The overall power benefit from the MISO converter ranges from 7 % to 43 % in the experimental MISO buck prototype. The proposed low-cost DMPPT solution and control algorithm provide very promising power savings compared to the conventional MPPT approach. The novel solution allows PV systems to be easily expanded without being restricted to panels from a single manufacturer.

A dual-active-bridge based bi-directional micro-inverter with integrated short-term Li-Ion ultra-capacitor storage and active power smoothing for modular PV systems

This work targets modular nanogrids for remote locations, where photovoltaic modules can be gradually introduced to grow the renewable energy capacity at minimal capital cost, while reducing diesel fuel consumption. Todays grid-tied micro-inverters provide a modular solution for ac power generation in nanogrids but battery storage remains centralized, requiring an additional ac-dc converter. The main contribution of this work is a new micro-inverter platform and control scheme with bidirectional power flow between the nanogrid, the photovoltaic module and integrated short-term storage, using new high energy-density Lithium-Ion Capacitor technology. A real-time power smoothing algorithm is also proposed and the performance of the new 100 W micro-inverter is experimentally verified under various closed-loop dynamic conditions.

Flyback Mode for Improved Low-Power Efficiency in the Dual-Active-Bridge Converter for Bidirectional PV Microinverters With Integrated Storage

This paper targets photovoltaic microinverters (MIVs) with integrated battery storage. The dual-active-bridge (DAB) topology provides bidirectional power flow; however, it generally suffers from poor efficiency and limited regulation accuracy at low power. It is shown that by modifying one switch, the DAB converter can operate as a two-transistor flyback to resolve these two issues. In addition, the dc-link voltage in the two-stage MIV can be dynamically adjusted for optimal performance in DAB mode. The proposed dual-mode control scheme is demonstrated experimentally on a 100-W prototype, with up to 8% increase in converter efficiency at low power.

Battery and ultra-capacitor hybrid energy storage system and power management scheme for solar-powered Wireless Sensor Nodes

This paper presents a Wireless Sensor Node (WSN) architecture with solar power generation and a hybrid energy storage scheme. The WSN is composed of three key modules: Energy Harvesting, Energy Storage, and the Control/Processing unit. The harvesting module consists of a miniature 179 mW solar array and MPPT hardware. A rechargeable 350 mAh Lithium-Ion battery and an ultra-capacitor are used as the energy storage elements. The low-ESR ultra-capacitor efficiently supplies the load power, which can reach as high as 295 mW peak, while the battery provides high-density storage. These elements are interfaced through a digitally controlled bi-directional dc-dc converter, which efficiently regulates the power-flow in the WSN. Multiple sensors and circuitry are implemented to measure positional and environmental data, as well as receiving and transmitting data via RF communication. A long-term test of the WSN is conducted to demonstrate the effective system functionality.

A dual active bridge DC-DC converter with optimal DC-link voltage scaling and flyback mode for enhanced low-power operation in hybrid PV/storage systems

Todays PV micro-inverters (MIVs) provide a modular solution for generation, however the energy storage architectures remain centralized, requiring an additional bi-directional ac-dc converter, with complex cell balancing circuits. Distributing storage capacity within the smart PV panels allows power fluctuations to be locally buffered, while minimizing the need for additional power electronics and balancing circuits. The dual-active-bridge (DAB) topology, which is adopted in this paper, provides bi-directional power flow; however it generally suffers from poor efficiency at low power. It is shown that with a minor modification, the DAB can be operated as a two-transistor flyback converter for improved efficiency. In addition, the dc-link voltage can be dynamically adjusted for the best performance in DAB mode. The proposed control scheme is demonstrated on a 100 W prototype, with up to 8% increase in efficiency at low power.

Distributed power-management architecture for a low-profile concentrating-PV system

This paper presents a low-cost dc-dc distributed power-management architecture for a low-profile concentrating-photovoltaic (CPV) system. The proposed scheme uses inverting buck-boost converters connected across neighboring series-connected CPV cells to achieve a virtual-parallel connection and thus improve the tolerance of the overall system to parameter variations. Since the converters process the difference between corresponding cell currents and the string current, they have reduced power-handling and efficiency requirements. Current cascading is identified as the main challenge with this architecture. Statistical simulations are used to investigate the current distributions in the converters, and power benefits are analyzed using measured data from a six-cell CPV system. Experimental results from four- and three-cell systems using digitally controlled converters demonstrate clear power benefits in the presence of tracker misalignment and short-circuit current mismatches.

On-chip PLL-based methods for synchronizing active switches across the isolation boundary in Dc-Dc converters

Digital isolators are widely used in isolated power converters for transmitting high-frequency gating pulses. Opto-isolators generally suffer from low reliability, while emerging RF based isolators are expensive and power-hungry. In this work, two Phase-Locked-Loop (PLL) based synchronization schemes are introduced and demonstrated to operate a bidirectional Dual-Active-Bridge (DAB) dc-dc converter. The proposed Digital Isolator Sensing (DIS) scheme utilizes a miniature air-core pulse transformer for synchronization and data communication, while the Power Transformer Sensing (PTS) scheme eliminates the need for any digital isolator for synchronization by using the power transformer. The application of DIS scheme can be extended to any topology that requires active switches on both sides of the isolation boundary, whereas the PTS scheme is designed for the DAB converter. In both schemes, the reference clock is generated on the secondary-side controller, and sensed indirectly on the primary-side. A PLL with a 100 MHz delay line, precise 12-bit phase-shift control module, and high-frequency transceiver are implemented on-chip in a 0.18μm 80V BCD process, and are used to control an off-chip DAB power stage.