Contents lists available at ScienceDirect Sustainable Energy Technologies and Assessments journal homepage: www.elsevier.com/locate/seta Modelling impacts of utility-scale photovoltaic systems variability using the wavelet variability model for smart grid operations Michael Emmanuel ⁎ , Ramesh Rayudu, Ian Welch Smart Power & Renewable Energy Systems Group, School of Engineering and Computer Science, Victoria University of Wellington, Wellington 6140, New Zealand ARTICLE INFO Keywords: Cloud-speed coefficient Irradiance Photovoltaic Single point sensor Wavelet variability model ABSTRACT The increasing presence of large-scale photovoltaic (PV) systems in the distribution network requires a thorough interconnection study for effective planning and reliable grid operations. The proliferation of such systems now creates a critical need for their accurate modelling to enable planners and operators understand and fully characterize centralized PV variability with the ability to develop realistic projections of PV plant output variability. This article models impacts of variability and the locational value of such utility-scale (centralized) PV plants deployed close to distribution feeder source, midpoint and end using the wavelet variability model (WVM). This model is used to accurately simulate solar irradiance variability and the PV plant output taken into account its entire footprint and density, time series irradiance data from a single point sensor, and location- dependent cloud-speed coefficient. Also, since the variability observed from a single point irradiance sensor cannot provide the exact variability across the entire PV plant, this study uses a high-frequency solar irradiance data and geographic smoothing for accurate modelling of PV output variability. Further, impacts on the tap changer operation, voltage profile, load demand offset and line loading reduction on the IEEE-34 distribution test feeder are investigated. Introduction The on-going energy transition and gradual shift of distributed generation (DG) to renewable power production call for a critical analysis of the deployment scenarios and assessment of existing mod- elling techniques for these game-changing technologies [1,2]. As re- ported by International Renewable Energy Agency (IRENA), 2016 was a very remarkable and record breaking year for global renewable energy generation with 161 gigawatts (GW) capacity addition [3]. This rapid growth continues to underpin a very strong business case for these green power generation technologies coupled with their social-en- vironmental benefits such as job creation and down trending of greenhouse gas (GHG) emissions. Also, one of the major goals of the evolving smart grid as stipulated by IEEE Std 2030 is to allow increase in penetration of renewables with the capacity of serving as spinning reserves [4]. Other factors driving this evolution include electricity market deregulation, proposed legislation to increase taxation of GHG emissions and the need to have a more resilient grid [5,6]. However, the integration of these low-carbon sources of power (such as photo- voltaic (PV) systems) has the propensity to cause emergent behaviours in the traditional electric power system (EPS) due to its inherent variability, uncertainty and location-specificity [7–10]. The potential grid integration issues include voltage or power quality issues, increase in operation of voltage regulating devices and equipment overloading [11]. These impacts can manifest themselves locally (e.g., voltage and power quality problems) or on the grid system level (network balancing and potential interaction with other technologies such as energy storage systems) [12,13]. These effects suggest that the increasing presence of PV generation in the distribution network is becoming too large to be ignored for an effective active distribution system planning and reliable operation of the EPS. Therefore, to enable a graceful adoption of these technologies into the evolving distribution network, accurate modelling is pivotal for true impact assessment and the provision of mitigation measures. In recent times, various incentive programs have led to the increased deployment of PV systems and are currently being referred to as the most common DG system integrated with the EPS [1]. Conse- quently, there is a need for accurate modelling of solar system ag- gregate variability to enable utilities, distribution system planners and authorities having jurisdiction (AHJ) over the grid to help perform the required interconnection studies, fully characterize the impact of solar PV output variability on the reliable operation of the grid, and provide mitigation alternatives. Further, apart from changing atmospheric constituents and balance-of-system components, cloud movements have a more severe impact on PV output fluctuations at short durations https://doi.org/10.1016/j.seta.2018.12.011 Received 18 June 2018; Received in revised form 2 October 2018; Accepted 13 December 2018 ⁎ Corresponding author. E-mail addresses: michael.emmanuel@ecs.vuw.ac.nz, michaelikechi@gmail.com (M. Emmanuel). Sustainable Energy Technologies and Assessments 31 (2019) 292–305 2213-1388/ © 2018 Elsevier Ltd. All rights reserved. T