A relatively new option for optimizing solar system efficiency and reliability is use of micro-inverters that connect to each individual solar panel. Equipping each panel with its own micro-inverter allows the system to accommodate its changing load and atmospheric condition, which provides optimal conversion efficiency for both the individual panels and the entire system.
Micro-inverter architectures also enable simpler wiring, which translates into lower installation costs. By making consumers' solar power systems more efficient, the time taken for the system to “pay back” on the initial investment for the technology shortens.
Power inverters are the critical electronic component in a solar power system. In commercial applications, these components interface with the photovoltaic (PV) panel, the batteries that store the charge and the local power distribution system or the public utility grid. Fig. 1 shows a typical solar inverter, which takes a very low voltage from the dc output of a PV array and converts it into some combination of dc battery voltages, ac line voltages and distribution grid voltages.
In a typical solar energy-harvesting system, multiple solar panels are connected in parallel to a single inverter that converts the variable dc output of multiple PV cells into a clean sinusoidal 50-Hz or 60-Hz voltage source.
Furthermore, it should be noted that in Fig.1, the microcontroller (MCU) block, TMS320C2000 or MSP430 microcontrollers typically includes such critical on-chip peripherals as pulse-width-modulation (PWM) modules and A/D converters.
The primary design goal is to maximize conversion efficiency. This is a sophisticated, iterative process that involves algorithms, known as maximum power point tracking algorithms (MPPTs), and the real-time controllers that execute them.
MAXIMIZING POWER CONVERSION
An inverter that does not use an MPPT algorithm simply connects the modules directly to the battery, forcing them to operate at battery voltage. Almost invariably, battery voltage is not the ideal value for harvesting the maximum solar energy available.
Fig. 2 illustrates the traditional current/voltage characteristic for a typical 75-W module and 25°C cell temperature. The dotted line plots voltage (PV VOLTS) against power (PV WATTS). The solid line plots voltage against current (PV AMPS). As shown in Fig. 2, at 12 V, the output power is approximately 53 W. In other words, by forcing the PV modules to operate at 12 V, power is limited to approximately 53 W.
With an MPPT algorithm implemented, the situation changes dramatically. In this example, the voltage at which the module achieves maximum power is 17 V. So, the role of the MPPT algorithm is to operate the module at 17 V, thereby extracting the full 75 W regardless of battery voltage.
A high-efficiency dc-dc power converter converts the 17-V module voltage at the controller input to battery voltage at the output. Because the dc-dc converter steps the 17 V down to 12 V, the battery charge current for the MPPT-enabled system in this example would be:
(VMODULE / VBATTERY) × IMODULE, or, (17 V/12 V) × 4.45 A = 6.30 A.
Assuming 100% conversion efficiency in the dc-dc converter, there would be an increase in charge current of 1.85 A, or 42%, would be achieved.
Although this example presumes the inverter is handling the energy from a single solar panel, conventional systems typically have multiple panels connected to a single inverter. This topology has both advantages and disadvantages, depending on the application.
There are three main types of MPPT algorithms: perturb-and-observe, incremental conductance and constant voltage. The first two methods are often referred to as “hill climbing” methods, because they depend on the fact that on the left side of the MPP, the curve is rising (dP/dV > 0) while on the right side of the MPP the curve is falling (dP/dV < 0).
The perturb-and-observe (P&O) method is the most common. The algorithm perturbs the operating voltage in a given direction and samples dP/dV. If dP/dV is positive, then the algorithm knows it adjusted the voltage in the direction toward the MPP. It keeps adjusting the voltage in that direction until dP/dV is negative.
P&O algorithms are easy to implement, but they sometimes result in oscillations around the MPP in steady-state operation. They also have slow response times and can even track in the wrong direction under rapidly changing atmospheric conditions.
The incremental conductance (INC) method uses the PV array's incremental conductance dI/dV to compute the sign of dP/dV. INC tracks rapidly changing irradiance conditions more accurately than P&O. However, like P&O, it can produce oscillations and be confused by rapidly changing atmospheric conditions. Another disadvantage is that its increased complexity increases computational time and slows down the sampling frequency.
The third method — the constant voltage method — makes use of the fact that, generally speaking, the ratio of VMPP/VOC ≈ 0.76. The problem with this method arises from the fact that it requires momentarily setting the PV array current to 0 to measure the array's open circuit voltage. The array's operating voltage is then set to 76% of this measured value. But during the time the array is disconnected, the available energy is wasted. It has also been found that while 76% of the open circuit voltage is a very good approximation, it does not always coincide with the MPP.
Because there is not a single MPPT algorithm that successfully addresses all common use scenarios, many designers go the extra step of having the system assess environmental conditions and select the algorithm that has the best fit. In fact, many MPPT algorithms are available, and it is not uncommon for solar panel manufacturers to provide their own.
Executing MPPT algorithms can be a tall order for inexpensive controllers because, in addition to the control functions that are the MCU's normal routine, algorithms require these controllers to have high-performance computing power. Advanced 32-bit real-time microcontrollers, such as those found in Texas Instruments' C2000 platform, are suitable for a wide range of solar applications.
Using a single inverter has several advantages, the two most prominent of which are simplicity and cost. Using MPPT algorithms and other techniques increases the efficiency of single-inverter systems, but only to a point. The downsides of a single-inverter topology can be significant, depending on the application. Most striking is the reliability issue: If just one inverter malfunctions, all of the energy being generated by all the panels is lost until the inverter is fixed or replaced.
Even when it is operating flawlessly, a single-inverter topology can have a negative impact on system efficiency. In most instances, each solar panel has different control requirements to reach maximum efficiency. Factors determining efficiency from panel to panel include manufacturing variations in its component PV cells, variations in ambient temperature, and a different degree of irradiance (raw power received from the sun) due to shadows and orientation.
Overall system conversion efficiency can be given an additional boost by using a micro-inverter for each individual solar panel instead of using a single inverter for the entire system. The micro-inverter topology's primary benefit is that energy continues to be converted even when one inverter malfunctions.
Other benefits to a micro-inverter approach include the ability to adjust conversion parameters on each panel using a high-resolution PWM. Because clouds, shadows and shade can vary the output of individual panels, equipping each panel with its own micro-inverter allows the system to accommodate changing loads. This provides optimal conversion efficiency for the individual panels as well as the entire system.
Micro-inverter architectures require a dedicated MCU for each panel to manage energy conversion. However, these additional MCUs can also be used to improve both system and panel monitoring.
Large solar panel farms, for example, benefit from inter-panel communication to help keep loads in balance and allow system administrators to plan in advance how much power will be available — and what to do with it. To take advantage of the benefits of system monitoring, however, the MCU must integrate on-chip communication peripherals (CAN, SPI, UART, etc.) in order to simplify interfacing with other micro-inverters in the solar array.
In many applications, using a micro-inverter topology can significantly improve overall system efficiency. At the panel level, a 30% increase in efficiency can be expected. But because applications vary widely, a percentage for an “average” system-level improvement has little meaning.
When evaluating the value of micro-inverters for an application, several aspects of the topology should be considered.
In small installations, the panels are likely to experience basically the same irradiance, temperature and shadowing conditions. As a result, micro-inverters may offer only a small efficiency advantage.
Having the panels operate at different voltages to maximize each panel's efficiency requires that each output voltage be normalized to the storage battery voltage with a dc-dc converter. To minimize manufacturing cost, the dc-dc converter and the inverter could be designed into a single module. The dc-ac converter for local line power or to tie into the distribution grid could also be part of the module.
The solar panels have to communicate with each other, which can add wires and complexity. This is another argument for creating a module that includes the inverter, the dc-dc converter and solar panel.
Each inverter's MCU must still be powerful enough to run multiple MPPT algorithms to accommodate different operating conditions.
Having multiple MCUs can increase the overall system bill of materials cost.
Cost is a concern whenever an architectural change is considered. To meet system price targets, having a controller for each panel means the silicon must be cost competitive, in a relatively small form factor, but still capable of handling the full range of control, communication and computational tasks simultaneously.
Integrating the right mix of control peripherals on-chip as well as high analog integration are both essential elements to keeping system cost low. High-performance also is critical in order to execute the algorithms that have been developed to optimize efficiency at every step of the conversion, system monitoring and storage process.
The increased cost of using multiple MCUs can be mitigated by choosing an MCU that can handle most of the entire system's requirements, including ac-dc conversion, dc-dc conversion, and communication between panels in addition to the demands of the micro-inverter itself.
Taking a closer look at these high-level requirements is the best way to determine what functionally is required of the MCU. Load balancing control, for example, is required when the panels are connected in parallel. The MCU must be able to detect the load current and increase or decrease the output voltage by turning off the output MOSFET. This requires a fast on-chip ADC to sample the voltage and current.
There are no “cookie-cutter” designs for micro-inverters. This means designers must be smart and innovate with new tips and techniques, especially in inter-panel and inter-system communications. The preferred MCU should support a variety of protocols, including some that are not the usual suspects such as power line communications (PLCs) and controller area networks (CANs). Power line communications, in particular, could reduce system cost by eliminating wires dedicated to communication. However, this requires a high-performance PWM capability, fast ADC and high-performance CPU to be built into the MCU.
An unexpected but highly valued feature of an MCU intended for solar inverter applications is dual on-chip oscillators, which can be used for clock failure detection to enhance reliability. The ability to run two system clocks simultaneously can also be helpful in reducing problems when the solar panels are being installed.
Because so much innovation is destined to happen in solar micro-inverter design, perhaps the most important feature for an MCU is software programmability. This feature would allow the highest degree of flexibility in power circuit design and control.
Equipped with an advanced number-crunching core to handle algorithmic calculations efficiently plus a combination of on-chip peripherals for power conversion control, the C2000 microcontrollers already widely used in conventional solar panel inverter topologies. A less expensive option is the recently introduced Piccolo series of C2000 microcontrollers, feature architectural advancements and enhanced peripherals in package sizes starting at 38-pins to bring the benefits of 32-bit real-time control to applications like micro-inverters that demand lower overall system costs.
In addition, various members of the Piccolo MCU series have integrated dual on-chip 10-MHz oscillators for clock comparison, on-chip VREG with power-on reset and brown-out protection, multiple high-resolution 150-ps PWMs, a 12-bit, 4.6 mega-sample/s ADC and interfaces for the I2C (PMBus), CAN, SPI, and UART communications protocols. Fig. 3 shows a computer system configuration for use with a micro-inverter-based PV system.
Performance is a critical characteristic for micro-inverters. Although Piccolo devices are less expensive and have a smaller footprint than other C2000 MCU members, the devices feature advancements, such as a programmable, floating-point control law accelerator (CLA) designed to off-load complex high-speed control algorithms, freeing the CPU to handle I/O and feedback loop metrics for up to a five times performance increase in closed-loop applications.
One of solar power-based systems' shortcomings is conversion efficiency. Solar panels harvest approximately 1 mW of average power from each 100-mm2 PV cell. Typical efficiency is roughly 10%. The capacity factor PV sources (i.e., the ratio of average power produced to the amount of power that would be produced if the sun was always shining) is about 15% to 20%. There are a number of reasons for these results, including the vagaries of sunlight itself, which disappears altogether at night and is often reduced by shadows and weather conditions during the day.
PV conversion introduces additional variables into the efficiency equation, including the temperature of the solar panels and their theoretical peak efficiency. For design engineers, another problem is the fact that PV cells produce voltages that vary erratically around 0.5 V. This variation has serious implications when choosing a power conversion topology. It is possible, for example, for a poorly implemented power conversion technology to consume a high percentage of the harvested PV power.
To accommodate the fact that the sun does not shine 24 hours a day, solar-powered systems include batteries as well as the sophisticated electronics required to charge these batteries efficiently. When batteries are incorporated into the system, additional dc-dc conversion is necessary for battery charging, and battery management and monitoring are also required.
Many solar-powered systems also interface with the power grid, which requires phase synchronization and power factor correction. There are also several use scenarios that require sophisticated control. For example, fault projection must be built-in to guard against events such as brownouts and blackouts on the public electricity grid. These are only the top level issues that design engineers must take into account.