Scholarlink

<p>Since the introduction of the Street Lighting National Programme (SLNP) in India in 2015, municipalities and local bodies have been upgrading and revamping existing discharge lamp-based street lighting systems with light-emitting diode (LED)-based ones for enhancing energy efficiency. In recent years, artificial neural networks (ANNs) have emerged as a potent machine learning tool for predictive modelling across major engineering disciplines. This study applied feedforward ANNs for street lighting planning for single-sided, opposite, and staggered layouts of LED luminaires of common power ratings (35–116&#xa0;W). Extensive photometric simulations were performed, and the generated datasets were utilized to train, validate, and test ANN models for the prediction of average illuminance, overall uniformity of illuminance, and installation energy efficiency. All the ANN models demonstrated good performance, and the error margin for the prediction of pertinent photometric and energy efficiency parameters was satisfactory. This machine learning-based approach can assist municipal engineers in street lighting project planning and implementation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Street Lighting Planning with Feedforward Artificial Neural Networks

  • 期刊论文

  • 发表日期：

    2026年4月23日

查看全文
  • Sourin Bhattacharya,
  • Parthasarathi Satvaya

摘要

Since the introduction of the Street Lighting National Programme (SLNP) in India in 2015, municipalities and local bodies have been upgrading and revamping existing discharge lamp-based street lighting systems with light-emitting diode (LED)-based ones for enhancing energy efficiency. In recent years, artificial neural networks (ANNs) have emerged as a potent machine learning tool for predictive modelling across major engineering disciplines. This study applied feedforward ANNs for street lighting planning for single-sided, opposite, and staggered layouts of LED luminaires of common power ratings (35–116 W). Extensive photometric simulations were performed, and the generated datasets were utilized to train, validate, and test ANN models for the prediction of average illuminance, overall uniformity of illuminance, and installation energy efficiency. All the ANN models demonstrated good performance, and the error margin for the prediction of pertinent photometric and energy efficiency parameters was satisfactory. This machine learning-based approach can assist municipal engineers in street lighting project planning and implementation.