Energy and Climate
Energy and ClimateRenewable SystemsRenewable EnergySolar EnergyPhotovoltaicsPV Systems ReliabilityInverter Reliability ProgramGoal 2: Development of Prognostics and Health Management

Goal 2: Development of Prognostics and Health Management

Degradation Signature Characterization

The data from the characterization of parts before and after stress testing will be analyzed for useful fingerprints that indicate degradation and remaining lifetime. One approach is to apply the characterization results to simulations to look for corresponding changes in behavioral parameters of the inverters. Figure 1 describes one potential behavioral parameter, voltage ripple [1]. Component degradation data are combined with system data and simulations to understand what components and signatures most strongly affect the inverter system. This process will uncover preferred indicators that are easy to measure with little additional instrumentation to inverters under production, minimizing the complexity and cost of PHM implementation.

Figure 1: SPICE simulation of voltage ripple on an inverter circuits. With additional sensor and further characterization, traits such as voltage ripple can serve as degradation indicators.

[1] J. D. Flicker, M. Marinella, R. Kaplar, and J. Granata, “PV inverter performance and reliability: what is the role of the bus capacitor?,” presented at the IEEE Photovoltaics Specialists Conference, 2012. For more information, contact Jennifer Granata.

Validation with Field Data

The laboratory results of the previous sections will be compared with field data gathered as part of a long-term study. Field data is composed of results from testing facilities, such as the Distributed Energy Technologies Laboratory (DETL) shown in Figure 1, as well as other collaborating sites and regional test centers. Figure 2 shows an example of temperature field data from an inverter instrumented with thermocouples [1]. The comparison will provide confidence that the experimental findings from the laboratory are applicable to real-life scenarios.

Figure 1: Photograph of the Distributed Energy Technologies Laboratory (DETL) at Sandia National Laboratories as one of the sources of field data.

 

Figure 2: An inverter instrumented with thermocouples provides temperature field data at a component-level resolution.

[1] N. R. Sorensen, E. V. Thomas, M. A. Quintana, S. Barkaszi, and A. Rosenthal, “Thermal study of inverter components,” presented at the IEEE Photovoltaics Specialists Conference, 2012. For more information, contact Jennifer Granata.

Predictive Modeling

The previously-described research results will form the foundation of predictive models such as ones used to generate the failure probability curve shown in Figure 1 [1]. These models will enable development of a PHM system that can report the remaining useful life of a component or subsystem to the operator. The system can also warn the operator about an impending failure and recommend replacement. The resulting tool will enable effective condition-based maintenance, minimizing the number of replacement parts and the amount of inverter-related downtime, ultimately lowering the lifetime cost of the PV system.

Reliability models for three control boards in inverters located in the Southeast based on field measurements applied to an Arrhenius formulation.

Figure 1: Reliability models for three control boards in inverters located in the Southeast based on field measurements applied to an Arrhenius formulation.

[1] N. R. Sorensen, E. V. Thomas, M. A. Quintana, S. Barkaszi, and A. Rosenthal, “Thermal study of inverter components,” presented at the IEEE Photovoltaics Specialists Conference, 2012.

Comments are closed.



Feedback

Feedback