Tracking systems are included in more than one-third of new photovoltaic developments in Europe, but less than 10% in the US. If properly controlled, they can capture more energy from each panel than fixed racks – up to 40% more in most parts of North America – so selecting the best hardware and the right control algorithm is critical to realizing optimum energy production and reliability.
Key requirements for control systems are cost, reliability, and energy productivity. Energy productivity can be defined as the number of kilowatt-hours produced by each panel (or each kilowatt of panels). Photovoltaic panels carry power ratings – typically 200 watts to 400 watts per panel – based on standard testing methods1. Control systems that enable effective tracking of the sun can produce significantly more energy, as shown in Figure 1.
This article considers flat-plate photovoltaic arrays, which should be controlled to within about 1 degree of the optimum orientation east-west and north-south. Concentrated PV technology requires much higher precision tracking.
Hardware for control systems includes printed circuits, connectors, weatherproof enclosures, and cables – all commercial parts or custom components made with common and inexpensive technologies. Thus cost considerations are generally less critical than reliability and energy productivity.
For large systems, hardware costs are typically less than $0.04 per peak watt, compared with $2.00 to $5.00 for all system costs combined. Systems less than 10 kilowatts may have control hardware costs about $0.10 per watt.
Similarly, control systems for photovoltaic tracking systems can be held to high standards of reliability, because they can be protected from extreme weather and generally do not carry high current. Care is taken to use best practices, such as de-rating of components, and sealing of junction boxes against weather. Cables which are exposed to sunlight must be chosen for life under ultraviolet exposure, or shrouded with uv-resistant material.
Types of Mounting Systems
The primary difference among control systems is in energy production, and it depends largely on the type of mechanical motion. The National Renewable Energy Laboratory’s PVWatts database2 defines the following array types:
Horizontal fixed racks: panels mounted on fixed racks and held horizontal at all times. This type does not track the sun and hence requires no tracking control system. Tilted fixed racks: panels mounted on fixed racks which are tilted toward the south, typically at an angle approximately equal to the latitude of the site. This arrangement is typically about 20% more productive than fixed horizontal racks, and requires no tracking control system. Horizontal one-axis tracking systems: panels tracking the sun east-west through the day but fixed at a tilt of 0 degrees. This arrangement is typically 40% more productive than fixed horizontal racks, and requires a control system for the east-west motion. Tilted one-axis tracking systems: panels tilted up toward the south (in the northern hemisphere), typically at an angle approximately equal to the latitude, and tracking the sun east-west through the day. Systems with this configuration typically produce about 60% more energy than horizontal fixed racks, and have control requirements the same as horizontal one-axis tracking systems. Two-axis tracking systems: panels track the sun east-west through the day and north-south through the seasons, producing typically 70% more energy than horizontal fixed racks (Figure 2). The control system is more complex than one-axis systems but typically not much more expensive. a. Upright-pole two-axis tracking systems typically pivot in the azimuth direction, requiring a rotational control moving the panels from east to west through the day and a linear or rotational control adjusting the tilt of the panels through the seasons. b. Rail tracking systems use linear controls to adjust the orientation of the panels in both the east-west direction and the north-south direction.
Two strategies are employed to determine the position of the sun. The first employs sensors and iterative adjustment of the array to find the orientation that produces the maximum power. This strategy has the advantage of simplicity, but can be disrupted by the presence of clouds or other shadows. Once the position of the sun is lost, it may not be found again, and the array can remain poorly oriented.
A second strategy is based on the location (latitude and longitude) of the array and date/time data. Location data can be established within a microprocessor at the time of installation, and the microprocessor can include a clock. With these data, and a well-chosen solar position algorithm, the microprocessor can determine the position of the sun, even if it is obscured by clouds or other obstacles. The array can be positioned accurately regardless of weather, so when the sun comes out the array will already be correctly positioned.
Both methods are commonly employed, with the latter being generally considered more reliable.
An additional input to some controllers comes from an anemometer (wind speed sensor). When wind speed exceeds a set limit, the panels may be “stowed” – oriented so as to minimize drag against the wind and reduce risk of damage to the array. The stow position may be horizontal in regions with no snow, or tilted to the south in snowy regions to allow snow to slide off.
Tracking systems may include motors, actuators, and/or hydraulics. The control system must be configured accordingly, typically using encoders.
When panels are close together, they may partially shade each other early and late in the day. Simple algorithms provide for “back-tracking” – panels are kept at the “noon” position during those hours, so they continue to produce energy and do not shade each other (Figure 3). This feature reduces total energy produced, typically by 1% to 2%, but allows closer spacing of panels which increases the amount of energy produced per unit array area.
Tracking control systems typically gather little information; energy production is more commonly logged by the inverters that convert DC power to AC power. Data gathered by the tracking control system is primarily for evaluating its function and providing alerts when the system is not working properly.
On advanced networked solar tracking control systems, power and energy data can be logged by the solar tracker controller. In these systems, the data can be directly correlated with the tracker movements to verify correct and optimum operation.
Energy lost due to system down-time may be many times the energy consumed for operation of the tracking system, and of much more value than the cost of the control hardware. Thus, the primary considerations for control system selection are reliability and energy production.
Energy production can be accurately estimated by analysis of the control system algorithm together with insolation data from NREL or another proven source. Reliability can be evaluated using data for the hardware components and analysis of the control algorithm to understand potential failure modes during operation.
Tracking systems can increase energy production dramatically, reducing the number and cost of the panels and other system components needed to meet a given load. Control systems of high reliability and low relative cost are available to optimize the function of tracking systems and minimize the cost per unit of energy produced.
1. http://www.fsec.ucf.edu/en/publications/ pdf/fsec-gp-68-01.pdf
2. http://www.nrel.gov/rredc/pvwatts/ changing_parameters.html