Adding Granularity to PV Monitoring
Inside this Article
As a provider of energy monitoring solutions, DECK Monitoring works daily with developers and investors for commercial and utility-scale PV systems. On larger projects, stakeholders typically want varying degrees of visibility into system performance beyond cumulative energy generation. Based on past experience or contract terms, project partners often know exactly what monitoring package is required and specify the level of monitoring granularity they need to achieve project goals. Nonetheless, I often find myself fielding questions such as: What level of monitoring is appropriate for my project? Why monitor at the string or subarray level? Does it make economic sense to add this degree of monitoring granularity? Here I break down the answers to these and other questions that may come up when deciding what type of monitoring solution to specify.
Let us assume that using a revenue-grade meter to monitor overall PV system generation is a given. Most incentive programs require the use of a revenue-grade third-party meter to measure the system’s total production. In commercial or utility-scale projects, this is usually the foundation upon which other monitoring components are added. Granular monitoring of solar arrays is most commonly achieved using three different methods: inverter data, subarray monitoring and string-level monitoring.
Inverter data. The first and most common method of gathering system performance and status information is a direct data connection to the inverters. This data connection is often referred to as inverter specific or inverter direct. The inverter is typically considered to be the most common point of failure on-site. Because of the integral nature of the inverters to system production and their relatively high failure rate compared to other installed components, monitoring the inverters is usually the first additional set of data points a customer requests. Inverter data provide key information such as visibility into the performance of each inverter’s array, the power lost when converting from dc to ac and valuable fault code information for remote troubleshooting. This monitoring approach is usually the least expensive since there are little or no additional hardware costs. It also provides a good deal of useful information.
Subarray monitoring. Although monitoring at the inverter level provides data on the performance of the system by each inverter’s array, this level of monitoring is not specific enough to identify issues on the string or module level on larger systems. To monitor performance and operation on a more granular level, project developers may specify subarray or zone monitoring. This approach allows you to break apart each inverter’s array into any number of smaller metered arrays and provide greater visibility than inverter monitoring alone.
For example, by isolating the performance of individual combiner boxes, you can use the monitoring system to remotely identify problem strings and areas of the array with cleaning or shading issues—without costly site visits and analysis.
String-level monitoring. Some developers go one step further and monitor the system on a string level. String-level monitoring is usually achieved by specifying monitoring or smart combiner boxes that measure each string (or pair of strings) independently. This method of monitoring offers the highest degree of visibility into commercial or utility-scale projects and allows you to identify underperforming strings in which modules are down, need cleaning or maintenance, or are subject to excessive shading.
Increasing Performance Ratios
While granular PV monitoring solutions can give you a highly detailed look into the performance of a system, the associated hardware and software come at a price. To determine the level of monitoring that is appropriate for your project, you need to understand why such solutions were developed and are in increasingly high demand.
A standard measurement for system performance is known as the performance ratio. The performance ratio refers to the relationship between actual yield and the target yield. Many designers expect a well-designed system to have a performance ratio of 0.77. In practice, performance ratios on installed solar farms have a wide variation. If you look at the 15 largest solar projects in the online Photovoltaic Power Systems Program database of International Energy Agency solar farms worldwide, you will see that the average performance ratio of installed systems is 0.66. To take another example, a subset of solar farms installed and maintained by the Tucson Power Company maintained a performance ratio of 0.79. There is a 13% difference in the annual performance ratio between these two sets of data.
What causes these solar farms to have such different performance ratios? Losses in performance are often due to issues on-site that require ongoing maintenance to optimize energy harvest. Historical data from these projects and others show that solar farms need unscheduled maintenance far beyond the initial installation period and that drops in the performance ratio are usually related to lack of maintenance. Solar monitoring solutions offer enhanced visibility and enable timely maintenance and optimization of the system’s performance. Through remote monitoring, system owners can identify issues before they affect the performance ratio of the project for a substantial and costly amount of time. The Tucson Power Company, for example, found a need for unscheduled site maintenance approximately every 7 months to maintain a high performance ratio.