Introduction
Sandia National Laboratories (SNL) has developed the capability to assess the performance, cost and reliability of fielded photovoltaic systems.  The tools and analysis techniques are documented in the referenced publications.  Sandia’s capability and experience provides an avenue to meet the requirements for independent verification of actual progress in systems technology for programs such as the Technical Pathway Partnerships of the DOE Solar America Initiative.

Capability
Sandia’s capability was developed to provide a historical perspective on system performance based on actual field operation.  An MS Access database has been created as a tool to collect system, component and O&M information as well as support data analyses.  These analyses are based on information captured and made available, including performance, initial costs and maintenance records, by partners who have systems installed in the field.  Figure 1 represents the input screen for entering system and component information.  In the top half of the screen system data are entered that are both technical and non-technical in nature.  Component information is entered in the bottom half of the screen by selecting a type of component in the system and then entering the appropriate information.  For the example shown, the inverter tab has been selected which allows component information at the sub-component level to be recorded.

Figure 1:  Screen Shot for Data Entry

Similar input screens exist for entering performance,  maintenance and failure information.  Performance information can be recorded as monthly or annual energy production.  For maintenance records, the system and components are identified, as are the details associated with the event, including the failure date, the cost needed to restore system operation and how the system failed.

Using these data, an analysis methodology has been developed to provide information on system and component initial costs, performance, reliability and O&M costs.  To describe system performance, PV energy parameters that have been established by the International Energy Agency (IEA) Photovoltaic Power Systems Program (IEC standard 61724) are used.  Three of the IEC standard 61724 system performance parameters – final yield, reference yield, and performance ratio – define the system field performance in terms of energy production, solar resource, and system losses.  These provide an easily understood method to not only compare system performance with other system options but also to permit system owners/customers to determine if system performance is meeting expectations.  Additional metrics have evolved as data have become available from the various applications.  Annual O&M costs as a percent of the initial system cost and an energy figure of merit (system cost per watt/30 year energy production) were early analyses implementations.  Additionally, charts of the number of O&M failures for a component as a percentage of the total number of failures, and likewise costs, are part of a standard output.  For some off-grid systems where O&M was provided on a quarterly basis, a running average on total O&M as a percent of capital cost was created by yearly quarters.  For other off-grid systems such as water pumping, the comparison of 30-year life cycle costs with a PV system and line extension were used instead of levelized energy cost.  In all cases the business model used in fielding the system was a guide for the goal of providing a practical view of the data to describe costs.

Top of page

Partnerships
Sandia’s interests are not different from those of energy providers who want to use photovoltaics for part of their energy needs, and want to make the best business decisions possible.  These energy providers want their dollars to translate into fielded systems that perform reliably with concomitant, manageable maintenance costs.  These businesses as well as DOE’s program can also become beneficiaries of Sandia’s work in system life cycle costs.  The references in this document illustrate the use of the database tool and the analysis techniques with partners that include investor owned utilities and small private businesses.  Studies of utility scale and residential grid-tied systems have been done with Arizona Public Service and Tucson Electric Power.  Additonally, a study of residential sized grid-tied systems has been completed with Conversation Services Group (Austin, TX).  Examples of the analysis capability are shown in the following text.

For a family of systems that have been installed over a number of years, the calculation of costs over time can be presented in current year dollars.  Figure 2 is an example of installed systems costs of residential grid-tied systems in Arizona expressed in 2004 dollars.  Similar charts for components (i.e., inverters, modules, etc) can be easily created.

Figure 2 – System Costs in 2004 Dollars

In addition to total system costs, the break-down of system costs can be captured and analyzed as a way to identify areas of high cost that need to be reduced. This cost break-down comes from a family of utility scale systems that are in operation in Arizona. The database allows the cost per watt to be calculated over a period of time or for the lifetime of the family of systems. Conversion to cost per watt AC is done using the calculated performance ratio.

Table 1:  System Cost

Performance of fielded PV systems can be described using the metric Final Yield. This metric can be expressed either in monthly or annual values. Final Yield is calculated by dividing the energy produced (kWhac) by the STC aggregate rating of the modules (kWstc). The figure below illustrates monthly Final Yield for a family of utility scale systems in Arizona.

In addition, Performance Ratio (PR) for a system represents the losses in the system in converting to ac energy and are . The PR for the same family of utility scale systems in Arizona is 0.79.

Figure 2 – System Costs in 2004 Dollars

Operation and maintenance events can be presented as a chart such as the one shown to the side. Here the unscheduled maintenance events by component are represented as a per cent of the total unscheduled maintenance cost. Over the operating history from mid-2001 through 2004, a total of 94 unscheduled maintenance events were recorded for this same family of utility scale systems in Arizona. The events are grouped by categories including data acquisition (DAS), inverters, junction boxes, modules (PV) and system level events. In a slightly different view of the data, the information can be presented as the cost of component failures as a percentage of the total cost of all failures.

Figure 4:  Unscheduled O&M by Component as Percent of Total O&M

Maintenance cost can be expressed as a percentage of total capital investment such as the table shown below.  Here the maintenance cost is calculated by year yielding an average annual maintenance cost of 0.16% of the initial system cost.

Top of page

Summary
Sandia has developed the tools and analytical capability to evaluate the performance, cost and reliability of fielded PV systems.  A flexible and comprehensive database has been created as a convenient method of documenting, analyzing and archiving this information.  The database tool allows performance metrics such as final yield and performance ratio to be easily calculated and displayed.  In addition, operation and maintenance costs can be calculated, failures modes and trends can be identified, and life cycle costs can be calculated for comparison with conventional energy sources or other solar technologies.  This capability provides a useful measurement tool to assess and evaluate progress for installed systems using both PV and CSP technologies.

References

Top of page