Semiparametric Approach to Optimal Sensor Location Design for a Photovoltaic Power Plant

• Activity Number: 187 - Contributed Poster Presentations: Section on Nonparametric Statistics
• Type: Contributed
• Date/Time: Monday, July 29, 2019 : 10:30 AM to 12:20 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Jane L Harvill* and Justin R Sims and Nalini Ravishanker and Clifford W. Hansen
• Companies: Baylor University and University of Tennessee at Martin and University of Connecticut and Sandia National Laboratories
• Keywords: Nonlinear time series; Vector time series; Functional coefficient autoregressive model; Photovoltaics

Abstract

Assessment of a utility scale photovoltaic (PV) power plant's potential performance is a critical aspect in the initial plant design and construction, and accurate monitoring of plant efficiency is crucial to profitable plant operation. Both assessment and monitoring rely on measurement of irradiance at the plant's location. These measurements are typically made using pyranometers which provide temporally dense, but spatially sparse data. Because plant output is directly related to total irradiance over the plant's footprint, a natural question is, ``What is the optimal number and layout of sensors for measuring and predicting solar irradiance?'' We propose a sensor design algorithm in an attempt to answer this question. The algorithm makes use of vector functional coefficient autoregressive (VFCAR) models to determine optimal sensor designs. To illustrate utility, we apply the algorithm to spatially and temporally dense irradiance data collected from a 1.2 MW PV plant located in Lanai, Hawaii.