Residential photovoltaic (PV) power systems convert sunlight into electricity for use in the home. The residence remains connected to the electric utility at all times. Therefore, if power demand exceeds what the PV system can produce, it’s simply drawn from the utility.
Increased use of these PV systems, distributed throughout the utility’s operating area, will ultimately affect the utility. Though they lighten the utility’s load, distributed installations must be monitored to ensure that they don’t degrade utility power quality (see “SEGIS Pushes Photovoltaics Into The Grid”).
During a utility outage, an advanced PV system with a backup battery can provide the required power. The PV array (Fig. 1) supplies dc to trickle charge the backup battery as well as drive the dc-ac inverter that supplies ac for the residence. During the night and at times when there’s insufficient sunlight, the home relies on the utility’s power. PV systems without a backup battery remain powerless during a utility outage.
A battery-backup system can keep “critical-load” circuits in the house operating during a utility outage. When an outage occurs, the PV system disconnects from the utility and uses the battery and inverter to power the home’s critical loads.
These critical-load circuits are wired from a subpanel separate from the rest of the electrical circuits. If the outage occurs during daylight hours, the PV array assists the battery in supplying the house loads. If the outage occurs at night, the battery and inverter supply the load.
The amount of time critical loads can operate depends on how much power they consume and the energy stored in the battery system. A typical backup-battery system may provide about 8 kWh with an eight-hour discharge rate. Therefore, it can handle a 1-kW load for eight hours. Average usage for a home, when running lights, TV, and a refrigerator, is 1 kW.
Typical System Components
To explain the complexity of distributed PV-inverter systems, it’s best to start with the components. First, the PV array employs multiple cells that produce dc power when subjected to solar radiation (Fig. 2). Today, these photovoltaics include monocrystalline silicon, polycrystalline silicon, amorphous silicon, cadmium telluride, and copper indium selenide/sulfide.
Because PV arrays require protection from the environment, they’re usually laminated behind a glass sheet. PV cells are electrically connected together to form PV modules, or solar panels.
Under ideal conditions, all solar cells are equally irradiated and work at the same current. However, some cells may occasionally be partially shaded. These shaded cells limit the current generated from other fully irradiated cells.
In extreme cases, current flow may be blocked because the cells are completely obscured. The shaded cells then behave like a load. The current generated from the fully irradiated cells produces overvoltages that can reach a panel’s breakdown threshold. These “hotspots” can cause shaded cells to overheat and, in some cases, permanently damage them due to current leakage. Bypass diodes connected in parallel with the string of cells prevent these hotspots.
Continue on next page
To this end, STMicroelectronics developed the SPV1001, which has the same functionality of a bypass diode, but with improved performance. It features low forward voltage drop and low reverse leakage current.
A properly controlled power MOS transistor charges a capacitor during OFF time and drives its gate during ON time with the charge previously stored in the capacitor. Properly set ON and OFF times lower the average voltage drop across the drain and source terminals, reducing power dissipation.
The inverter (Fig. 3) ties the generated ac into the utility grid for typical grid-tied residential applications. Residences and businesses with a grid-tied electrical system are usually permitted to sell their energy to the utility grid through a policy known as “net metering,” whereby the entity that owns the PV power source receives compensation from the utility for its net outflow of power. In the U.S., grid-interactive power systems are covered by specific provisions in the National Electric Code, which also mandates certain requirements for grid-interactive inverters.
Inverters take the dc power from the PV modules (typically 250 to 600 V dc) and invert it to ac so it can be fed into the grid. The inverter must also synchronize its frequency with that of the grid (e.g., 60 Hz) using a local oscillator and limit the voltage to no higher than the grid voltage.
Since typical inverters have a fixed unity power factor, their output voltage and current are in phase. In addition, the phase angle is within 1° of the ac power grid. The inverter usually has an on-board computer that senses the ac grid waveform and outputs a voltage synchronized with the grid.
Grid-tie inverters use a number of different technologies, from the newer high-frequency transformers to conventional low-frequency transformers, to convert dc directly to 120 or 240 V ac. Most of these electrical systems feed into the public utility grid, which requires a transformer for galvanic isolation between the dc and ac circuits.
An optimal grid-tie inverter produces a true sine wave with very low harmonic distortion. An inverter with high harmonic distortion could impact the power quality of nearby utility loads.
Residential inverters must provide dependable production of PV power, which means minimization of disruptive and costly inverter failures. Historically, though, inverters have been one of the least reliable components in solar-power generation systems because harsh environmental conditions stress electronic equipment.
Most components of a solar PV power plant are directly exposed to the outside environment, subjecting them to temperature fluctuations and extremes, humidity, corrosive elements, and dust. The installation’s geographic location influences environmental stresses, too, and that also must be factored into any reliability analysis.
The accurate prediction of solar inverters’ component stresses and associated wear-out mechanisms wrought by natural cycles requires a complex time-dependent modeling approach. Because temperature cycling contributes to device wear-out, simpler constant hazard rate and mean time before failure (MTBF) calculations just aren’t accurate enough.
Continue on next page
Normally, grid-tied inverters will shut off if they don’t detect the presence of the utility grid. If, however, load circuits in the electrical system happen to resonate at the frequency of the utility grid, the inverter may be fooled into thinking that the grid is still active even after shutdown. This is called “islanding.”
Anti-islanding protection is built into inverters designed for grid-tie operation. It will inject small pulses that are slightly out of phase with the ac electrical system to cancel any stray resonances present when the grid shuts down.
Since 1999, the U.S. standard for anti-islanding protection has been UL 1741, harmonized with IEEE 1547. Any inverter listed to the UL 1741 standard may be connected to a utility grid without the need for additional anti-islanding equipment, anywhere in the U.S. or other countries that accept UL standards.
Most inverters can’t differentiate between a true utility outage (when an anti-islanding disconnect is required) and a grid disturbance or brownout situation during which the PV system could actually assist in supporting grid stability. Even for those inverters that can differentiate between these two different conditions in most situations, current regulations (IEEE 1547/UL 1741) typically require the inverter to disconnect from the grid. Finally, interactions between inverters from different manufacturers may result in false island detection and increased run-on times that compromise the safety of utility line personnel.
Maximum Power Point
Most solar grid-interactive inverters include a maximum power-point tracker (MPPT). It’s a high-efficiency dc-dc converter that presents an optimal electrical load to a solar panel or array and produces a voltage suitable for the load. As such, an MPPT enables the inverter to extract an optimal amount of power from the solar array by tracking the array’s maximum power point.
PV cells have a single operating point where the cell’s current and voltage values produce a maximum power output. These values correspond to a particular load resistance. The PV cell has an exponential relationship (Fig. 4) between current and voltage. The maximum power point (MPP) occurs at the knee of the curve, where the resistance is equal to the negative of the differential resistance (V/I = –dV/dI). MPPTs utilize a control circuit or logic to search for this point, allowing the converter to extract the maximum available cell power.
Traditional solar inverters perform MPP tracking for an entire array as a whole. In such systems the same current, dictated by the inverter, flows though all panels in the string. However, different panels have different IV curves, i.e. different MPPs (due to manufacturing tolerance, partial shading, etc.). Thus, in this architecture, some panels will perform below their MPP, resulting in the loss of energy.
Electric utilities want to monitor and control distributed residential PV generation systems, which would require the distributed PV systems to employ smart inverters. The best and most likely management scenario for distributed generation is a corresponding distributed hierarchy.
Continue on next page
Under this hierarchy, the utility would centrally monitor and control smart inverters, either directly or via another system, such as a plant controller. Direct management is likely with individual or master/slave configurations of commercial and residential inverters. A hierarchy of control is apt to exist with large-scale PV power generation facilities, as well as with a “virtual power plant” that aggregates multiple sources of distributed generation.
Although still evolving, the “intelligent” or smart inverter could include some of the essential requirements utilities need for exerting direct control:
• Automatic discovery with unique identification
• Power production monitoring
• Event data logging
• Time synchronization
• Remote on/off control
• Remote software updating
• Power-quality scheduling and control (including storage)
Additional capabilities may be needed for full control over the inverter and its prominent role in monitoring and managing other system components.
Forecasting can minimize problems caused by the adverse effects of irradiance transients, which otherwise create a barrier to effective short-term and long-term planning of distributed PV system integration. Rapid changes in cloud cover can produce intense power transients that shorten switch gear life and distort power quality and stability at the host facility and in the utility grid (see the table). The transients also make it difficult to effectively match PV power generation to electrical demand.
Integrating weather “awareness” into the inverter control system is the most promising approach toward reducing the adverse effects of cloud position, movement, and transparency. When coupled with the inverter’s ability to curtail output power using predefined ramping functions, it reduces duty factors on tap changers and other voltage-regulation equipment. This effect is more recognizable on feeders with higher penetrations of PV power generation.
As a future metric, the inverter’s ability to “foresee” cloud transients will allow for more seamless integration of energy storage capabilities. These enhancements lay the foundation that enables inverters to produce more stable power and avoid faults that can occur during and after cloud transients.