Assessment of solar photovoltaic potentials on urban noise barriers using street-view imagery

• Published in: Renewable energy Vol. 168; pp. 181 - 194
• Main Authors: Zhong, Teng, Zhang, Kai, Chen, Min, Wang, Yijie, Zhu, Rui, Zhang, Zhixin, Zhou, Zixuan, Qian, Zhen, Lv, Guonian, Yan, Jinyue
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.05.2021
• Subjects: Acoustic noise measurement; Computational framework; Deep learning; Estimation results; Learning-based methods; Machine learning; Object detection; Photovoltaic cells; Photovoltaic noise barrier (PVNB); Photovoltaic panels; Renewable energy source; Solar cell arrays; Solar energy; Solar photovoltaic system; Solar photovoltaics; Solar power generation; Solar radiation assessment; Street-view images; Urban growth; Street-view images; Photovoltaic noise barrier (PVNB); Solar radiation assessment; Object detection; Machine learning
• ISSN: 0960-1481; 1879-0682; 1879-0682
• DOI: 10.1016/j.renene.2020.12.044

Solar energy captured by solar photovoltaic (PV) systems has great potential to meet the high demand for renewable energy sources in urban areas. A photovoltaic noise barrier (PVNB) system, which integrates a PV system with a noise barrier, is a promising source for harvesting solar energy to overcome the problem of having limited land available for solar panel installations. When estimating the solar PV potential at the city scale, it is difficult to identify sites for installing solar panels. A computational framework is proposed for estimating the solar PV potential of PVNB systems based on both existing and planned noise barrier sites. The proposed computational framework can identify suitable sites for installing photovoltaic panels. A deep learning-based method is used to detect existing noise barrier sites from massive street-view images. The planned noise barrier sites are identified with urban policies. Based on the existing and planned sites of noise barriers in Nanjing, the annual solar PV potentials in 2019 are 29,137 MW h and 113,052 MW h, respectively. The estimation results show that the potential PVNB systems based on the existing and planned noise barrier in 2019 have the potential installed capacity of 14.26 MW and 57.24 MW, with corresponding potential annual power generation of 4662 MW h and 18,088 MW h, respectively. •A photovoltaic noise barrier system is promising source for harvesting solar energy.•Detect existing noise barrier from street-view images with 96.22% accuracy.•Solar PV potentials of existing and planned noise barriers are estimated.•Installed capacity of proposed PVNB systems reach 14.26 MW and 57.24 MW.•Potential annual power generation are 4662 MW h and 18,088 MW h.