Enhancing the efficiency of photovoltaic systems
Among today’s various renewable energy sources, solar energy is by far the most prevalent. To harness this source, arrays or solar panels are formed by linking solar modules that contain semiconducting material which absorbs photons of sunlight. The photons energize electrons in the semiconducting material, freeing them from their atoms. This, in turn, creates direct current (DC) that must be converted to alternating current (AC). Photovoltaic (PV) technology is the mechanism used to implement this solar-to-electrical conversion. Unfortunately, it generally can only achieve efficiency of roughly 19 percent. The only way to maximize the use of harvested solar energy while minimizing module and system size is to achieve efficiency of greater than 95 percent.
There are two types of PV systems. The first type is configured as an off-grid or standalone system that operates independently of the electric utility grid. The second type of PV system can be integrated with the utility grid, which enables energy to be shared between the PV system and the grid. One benefit of this approach is that surplus power can be sold back to the utility.
Regardless of which approach is taken, each PV system uses similar components, including PV modules, a cooling system, an energy storage system or battery bank, the load, a utility grid interface, and a PV inverter system (see Figure 1). While these components vary depending on functional and operational requirements, the PV inverter system is the heart of any implementation. It performs all DC-to-AC conversion, power protection, monitoring, and control functions.
Figure 1: Typical PV energy system
Click on image to enlarge
Designing a PV inverter system
The two primary sub-components in a PV inverter systems typically are the controller used to implement system management tasks and control algorithms, and the AC-to-DC conversion circuit.
The controller is used for tasks including grid and system monitoring, system synchronization with utility power for grid-connected systems, and output power quality monitoring. The controller also performs protective functions for safety and compliance with various standards and regulations. Other key functions include data logging, firmware updates, and communications with the system operator, as well as battery charging control for standalone systems, and smart metering used for grid-connected PV systems. One other important controller responsibility is the execution of control and energy management algorithms. In addition to being very computationally demanding, these tasks can also impact power efficiency.
The DC-to-AC conversion circuit also plays an important role, handling all the tasks related to converting raw DC power from the panels into clean AC power that is consistent with the utility grid’s voltage and power quality requirements. To accomplish this, the circuit uses a set of switching power devices such as metal oxide semiconductor field effect transistors (MOSFETs) or insulated gate bipolar transistors (IGBTs). The inverter circuit also includes active filtering circuitry to reduce the distortion caused by harmonics resulting from high-frequency switching.
Choosing the right configuration
Designers have a choice of several possible PV conversion circuit configurations. The choice depends on a number of factors, including the number of power processing stages, the type of power decoupling, the types of intra-stage interconnections, and the type of grid interface.
There are also important considerations related to power levels. Micro inverters are typically integrated in the PV module for power levels up to 400 watts (W), whereas string inverters are used for power levels up to 10 killowatts (Kw). For power levels between 5 and 50 kW, multi-string inverters are generally the best choice. For higher power levels, central inverters should be used.
Optimizing efficiency
Solar energy harvesting demands that systems achieve extremely high levels of efficiency. There are two key factors that have the biggest influence on conversion circuit efficiency. The first is topology, and the second is the type of components – including semiconductor switching devices, magnetic elements and capacitors, to name a few — and their operational characteristics.
Among the various inverter topologies that have been implemented, two have achieved higher efficiencies for grid-connected centralized inverters than alternatives. The first topology is called the Highly Efficient and Reliable Inverter Concept (HERIC). The HERIC topology uses an extra switch and diode pairs at the output, which reduces losses by decoupling the output inductor from the input capacitor. The second, more complex, topology is the multilevel inverter (see Figure 2). Multilevel inverters feature half the voltage stress in each switch as compared to a HERIC topology. The multilevel inverter approach uses much lower voltage, which yields higher efficiency and lower device costs. An additional benefit of multilevel inverters is that the size of the electromagnetic interference (EMI) level and output filter (for cleaning the harmonics) can be reduced, which reduces overall system cost.
Once the topology is chosen, designers must select the components that will be used. In addition to an inverter, a PV system can also have a DC-to-DC conversion stage for maintaining a constant and controlled input voltage level at the inverter, and decoupling the control of voltage and power. A drawback of using a DC-to-DC conversion stage is that it can negatively impact system efficiency. To mitigate this potential hit on efficiency, designers can employ a number of techniques. One is to use silicon carbide (SiC) power transistors, which offer several advantages over traditional silicon or even gallium arsenide (GaAs) solutions, allowing for much greater power handling and higher switching rates. A number of projects are in the works to develop utility-grade devices with an eye toward creating solid-state power transformers and high-power inverters for wind and solar farms.
A second way to improve PV system efficiency is to use maximum power point tracking (MPPT) algorithms, which enables the PV system to better control the power inverter as it reacts to changes in operating conditions. A number of MPPT algorithms are available, each with advantages and disadvantages. These algorithms regulate the PV output for maximum power delivery, ensuring that the system maintains the optimum operating point. These algorithms also ensure that the inverter draws no more than the maximum PV array output power, thus preventing inverter collapse. There are two ways to deploy an MPPT algorithm – either in the main controller, or in the individual PV modules. The latter approach allows each to track independently, which is preferred in cases where the operating conditions for individual PV modules differ significantly. This finer control over power conversion can greatly enhance the efficiency of the conversion process in applications using a larger number of modules.
A third approach for improving efficiency is to employ pulse width modulation (PWM) technology to control power switching components in the inverter circuit during DC-to-AC conversion. A PWM algorithm is used to control the switching component’s states. This ensures that the time-average value of the voltage command is met. PWM algorithms can reduce losses in the inverter while optimizing the voltage utilization of the DC bus. They also offer the benefit of being well understood, and they can easily be implemented in either hardware or software.
One last technique for improving efficiency is to use power factor correction (PFC). Capacitive and inductive loads cause a poor power factor, which is the ratio of real power to reactive power where real power is useful and reactive power is wasted (the result of current and voltage being out of phase). With a power factor of one, the voltage and current are in phase, which provides maximum power. By actively correcting the power factor, designers can, in effect, improve system efficiency.
The FPGA option
Although PV inverters have traditionally been implemented using a variety of processors including microcontrollers and digital signal processors (DSPs), a third option is to include programmable logic in the solution. This is possible using field programmable gate array (FPGA) technology, which enables customized controllers to take advantage of ongoing cost, performance, flexibility and gate capacity improvements. FPGA technology has reached the point where it can now outperform microcontrollers, DSPs and ASICs at the same price range.
FPGA technology provides particularly compelling benefits when there is the demand for highly optimized solutions with special algorithmic functions such as PWM, MPPT and PFC implementation. In these cases, FPGAs provide a very low-cost hardware and software customization platform. To further improve flexibility, designers can opt for flash-based FPGAs that can be reprogrammed at any time, thus reducing development costs while permitting field upgrades and bug fixes.
Another option is the customizable system-on-chip (cSoC), which combines programmable logic with an embedded controller and configurable analog. cSoC technology offers many options for further enhancing the design and performance of a PV inverter system. The integration of board-level components into a single monolithic IC reduces cost and power dissipation. Plus, because there is no board-level wiring, there are shorter circuit delays. Eliminating the long wires that are otherwise needed to connect devices on the board also enables designers to avoid parasitic ringing, oscillations and other associated problems, as well.
cSoCs also give designers the flexibility to implement control functions in hardware, software, or a combination of both. Plus, the cSoC’s programmable analog can be used to monitor and evaluate operating conditions, giving end-users and utilities critically important visibility into potential failures before they occur so they can take pre-emptive actions. Finally, the highly parallel nature of the cSoC’s programmable fabric enables arithmetic co-processing. By using hardware acceleration techniques, cSoCs increase computational throughput. They can implement any required DSP functions that cannot be feasibly implemented in the embedded microcontroller. Whenever a signal processing function is required, the embedded microprocessor makes calls to a coprocessor that has been constructed in the cSoC’s programmable logic core. This offloads the microprocessor while delivering the needed throughput. System designers can choose from a number of off-the-shelf IP cores that greatly simplify the task of implementing DSP algorithms in hardware for PV system applications.
As demand grows for more efficient and reliable PV systems, system designers have a number of options to consider. The inclusion of programmable logic in today’s PV module solutions offers opportunities to improve system cost and performance ratios while adding valuable new features and capabilities.
About the author:
Rufino Olay is an Industrial Business Manager at Microsemi Corp., responsible for alternative-energy-focused designs and applications. His expertise includes business development, P&L responsibilities, NPI, and production launches for the FPGA-, PV-, and wireless-infrastructure industries. Olay holds a bachelor of science in electrical engineering from San Jose State University.
Visit: Microsemi Corporation
See related links:
PV industry looks to a bright future
Energy harvesting, wireless sensor networks & opportunities for industrial applications
Anti-aging elixir for solar cells
CZTSe thin-film solar cells results
Increasing energy yield of PV panels
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
