A Combined Persistence and Physical Approach for Ultra-Short-Term Photovoltaic Power Forecasting Using Distributed Sensors

Yakov Malinkovich, Moshe Sitbon, Simon Lineykin, Kfir Jack Dagan, Dmitry Baimel

Research output: Contribution to journalArticlepeer-review

Abstract

This paper presents a novel method for forecasting the impact of cloud cover on photovoltaic (PV) fields in the nowcasting term, utilizing PV panels as sensors in a combination of physical and persistence models and integrating energy storage system control. The proposed approach entails simulating a power network consisting of a 22 kV renewable energy source and energy storage, enabling the evaluation of network behavior in comparison to the national grid. To optimize computational efficiency, the authors develop an equivalent model of the PV + energy storage module, accurately simulating system behavior while accounting for weather conditions, particularly cloud cover. Moreover, the authors introduce a control system model capable of responding effectively to network dynamics and providing comprehensive control of the energy storage system using PID controllers. Precise power forecasting is essential for maintaining power continuity, managing overall power-system ramp rates, and ensuring grid stability. The adaptability of our method to integrate with solar fencing systems serves as a testament to its innovative nature and its potential to contribute significantly to the renewable energy field. The authors also assess various scenarios against the grid to determine their impact on grid stability. The research findings indicate that the integration of energy storage and the proposed forecasting method, which combines physical and persistence models, offers a promising solution for effectively managing grid stability.

Original languageEnglish
Article number2866
JournalSensors
Volume24
Issue number9
DOIs
StatePublished - May 2024

Keywords

  • cloud cover
  • energy storage system
  • nowcasting
  • photovoltaic (PV) fields

All Science Journal Classification (ASJC) codes

  • Analytical Chemistry
  • Information Systems
  • Atomic and Molecular Physics, and Optics
  • Biochemistry
  • Instrumentation
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'A Combined Persistence and Physical Approach for Ultra-Short-Term Photovoltaic Power Forecasting Using Distributed Sensors'. Together they form a unique fingerprint.

Cite this