The development prospects of solar energy technology

Solar power is becoming a viable alternative energy source as energy costs continue to rise. Germany was the world's largest solar market until 2007, driven by government legislation that actively encouraged the use of renewable energy (e.g. the Renewable Energy Act "Energieeinspeisungsgesetz"). Now, other countries have surpassed Germany, with Spain, for example, leading the world in the number of new solar power plants in 2008, and Italy, France and the United States expecting significant growth in installed solar power capacity . The various incentives have driven demand higher, which in turn has stimulated capacity growth. However, the recent global economic crisis and the sudden withdrawal of Spanish incentives for the solar market in 2008 have resulted in an oversupply of solar chips , leading to a 40%-50% drop in prices. This has brought photovoltaic technology closer to the so-called "grid parity" goal, where the cost of solar power generation is equivalent to the current market price of electricity. Germany is expected to achieve grid parity in 2015.

Solar modules generate a DC voltage , and solar inverters convert this DC power into AC power, which is then fed into the grid. This article will explore the latest trends in solar inverter design.

One important trend is the adoption of higher powers. Today, solar power plants with peak power generation exceeding 100kW are becoming more common, and this trend also exists in smaller-scale power generation systems: the average power increases from 5kWp to 10kWp.

The boost + H-bridge topology is one of the most commonly used topologies for solar inverters. It is a two-stage non-isolated topology. The first stage is the boost stage, which is used to step up the variable output voltage of the module (e.g. 100V – 500V) to a larger intermediate voltage, which must be greater than the actual peak mains voltage (e.g. 230V x sqrt(2), or >325V). The boost stage also has an important role, that is, in order to maximize efficiency, the solar module must operate to produce as much power as possible, and the power curve of the solar module can be obtained by multiplying the output current by the output voltage value. There is a maximum point in the power characteristic, which is called the "maximum power point" or MPP, and this exact location varies with factors such as module type, temperature and sunlight shadow.

Using a software technique called maximum power point tracking, or MPPT, along with a customized algorithm, the inverter’s input stage tracks this maximum power point.

The second stage of the inverter converts the constant intermediate voltage into a 50Hz AC voltage, which is then fed into the mains. This output is synchronized with the phase and frequency of the mains. This stage, being connected to the mains, must meet certain safety standards even in fault conditions. In addition, there is a new draft of VDE 0126-1-1 related to the Low Voltage Directive, which requires solar inverters to actively support the mains supply network even in the event of power quality degradation, in order to minimize the risk of more widespread power outages. Within the existing regulatory constraints, it is possible to design an inverter that can shut down in real time in the event of a power outage to achieve self-protection. However, when solar inverters become popular and account for a significant share of total power generation, it is possible to cause a larger mains power outage if the connected solar inverters are directly shut down at the first power outage, because the inverters will be shut down one by one and quickly reduce the power in the grid. The new draft directive therefore aims to improve the stability and power quality of the mains distribution network at the expense of only making the inverter output stage slightly more complex.

Solar inverters must be reliable to minimize maintenance and downtime costs. They must also be efficient to maximize power generation. Solar inverter designers also put considerable effort into maximizing efficiency.

There are many ways to improve the efficiency of a boost inverter. Since boost inverters can operate in either continuous conduction mode or boundary conduction mode (CCM or BCM), different optimization schemes are derived. In CCM mode, a major contributor to losses is the reverse recovery current of the boost diode ; in this case, silicon carbide diodes or Fairchild Semiconductor 's Stealth diodes are generally used. Solar inverters are more commonly operated in BCM mode, and although CCM is usually recommended for this power level, the reason for using BCM mode is that the forward voltage of the diode is much lower in BCM mode. In addition, BCM mode also has much higher EMI filter and boost inductor ripple current. Here, good high-frequency inductor design is a solution.

A new approach is to use two interleaved boost stages instead of one. This way, the current through each inductor and each switch can be halved. In addition, with interleaving, the ripple current on one stage can compensate for the ripple current on another stage, thus removing the input ripple current over a wide operating input range. Controls such as the FAN9612 interleaved BCM PFC can easily meet the requirements of the solar boost stage.

There are two options for the boost switch in the inverter: IGBT or MOSFET . For input stages that require a rated switching voltage of more than 600V, 1200V IGBT fast switches such as FGL40N120AND are often used. For input stages that only require a rated voltage of 600V/650V, MOSFETs are selected.

Designers of output H-bridge stages have traditionally used 600V/650V MOSFETs, but new draft specifications requiring four-quadrant operation of the output stage have rekindled interest in IGBTs in this area. MOSFETs have built-in body diodes, but their switching performance is poor compared to the combined package diodes used in IGBTs. New field-stop IGBTs can switch voltages at 10V/ns, which is a significant improvement over the conduction losses of older products. The integrated diode has excellent soft recovery performance, which helps reduce EMI caused by high di/dt above 500A/us. For 16kHz-25kHz switching, IGBTs such as Fairchild Semiconductor's FGH60N60UFD are recommended.

Another trend in solar inverter design is to expand the input voltage range, which results in a reduction in input current at the same power level, or an increase in power level at the same input current. When the input voltage is higher, IGBTs with higher rated voltages (in the 1200V range) are required , resulting in greater losses. One way to solve this problem is to use a three-level inverter.

Using two electrolytic capacitors in series , the high input voltage can be split in two, connecting the middle point to the neutral line, and then a 600V switch can be used. The three-level inverter can convert between three levels: +Vbus, 0V and –Vbus. In addition to being more efficient than the solution built with 1200V switches, the three-level inverter has the advantage of greatly reducing the output inductance.

For unity power factor, the function of the three-level inverter can be explained as follows. During the positive half-wave Q5 is always on, Q6 and Q4 are always off. Q3 and D3 form a step-down converter, producing an output sine wave voltage. If only unity power factor is required, Q5 and Q6 can be designed to switch at 50Hz, using very slow IGBTs with very low Vce (saturation voltage), such as FGH30N60LSD. If lower power factor is required, Q5 and Q6 must be operated at the switching frequency for a short period of time. The diodes of Q3 and Q4 should be fast soft recovery diodes. Q3 and Q4 can be arranged as fast recovery MOSFETs , such as FGL100N50F, or fast IGBTs, such as FGH60N60SFD.

Based on the above analysis, the three-level inverter topology can achieve an efficiency of more than 98%, and therefore may become the mainstream structure of non-isolated inverters with power levels above 5kWp.