Scholarlink

<p>Optimization of Electric Vehicles (EV) charging is traditionally operated by factors such as cost, grid constraints and operational efficiency. However, the human centric factors including perceived waiting time and psychological discomfort plays a significant role in shaping the user’s charging experience. So, this paper presents a Stress-Aware Particle Swarm Optimization (SAPSO) framework, which explicitly incorporates behavioral stress factors in the process of EV charging optimization and therefore, optimizes continuous-valued charging profiles over a 24-slot daily horizon across four charging stations subject to power constraints. The performance of SAPSO is evaluated by using 1,254 charging sessions and then compared against traditional PSO without stress modelling, uniform scheduling, and a greedy heuristic. Experiments shows that the proposed SAPSO reduces the mean normalized stress from 0.5156 to 0.4433, thereby, corresponding to a 14.03% improvement, outperforms traditional PSO, which achieves a 12.80% reduction and at the same time, SAPSO also maintains a comparable state-of-charge (SoC) gain of approximately 0.069. Although uniform and greedy strategy produces higher stress reductions of about 17–18%, they produce significantly reduced energy delivery, as seen in low SoC gains, and are therefore less practically feasible. Moreover, SAPSO maintains the mean charging price per kWh near the original charging pattern and yields the most consistent pricing results as shown by the lowest coefficient of variation amongst all the methods. Paired tests are used to establish the statistical significance of differences in post-optimization stresses, and the results indicate that SAPSO attains high human-centric performance as compared against the competing strategies (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(p \ll 0.001\)</EquationSource> </InlineEquation>). These findings justify the use of behavioral stress included in EV charging optimization have the potential to promote user experience and does not negatively affect essential operational efficiency.</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

Human-centric EV charging optimization using stress-aware particle swarm optimization

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月19日

查看全文
  • Faraz Masood,
  • Zara Fatima Siddiqui,
  • Arman Rasool Faridi,
  • Ahmad Wali Siddiqui,
  • Ali Haider Shamsan

摘要

Optimization of Electric Vehicles (EV) charging is traditionally operated by factors such as cost, grid constraints and operational efficiency. However, the human centric factors including perceived waiting time and psychological discomfort plays a significant role in shaping the user’s charging experience. So, this paper presents a Stress-Aware Particle Swarm Optimization (SAPSO) framework, which explicitly incorporates behavioral stress factors in the process of EV charging optimization and therefore, optimizes continuous-valued charging profiles over a 24-slot daily horizon across four charging stations subject to power constraints. The performance of SAPSO is evaluated by using 1,254 charging sessions and then compared against traditional PSO without stress modelling, uniform scheduling, and a greedy heuristic. Experiments shows that the proposed SAPSO reduces the mean normalized stress from 0.5156 to 0.4433, thereby, corresponding to a 14.03% improvement, outperforms traditional PSO, which achieves a 12.80% reduction and at the same time, SAPSO also maintains a comparable state-of-charge (SoC) gain of approximately 0.069. Although uniform and greedy strategy produces higher stress reductions of about 17–18%, they produce significantly reduced energy delivery, as seen in low SoC gains, and are therefore less practically feasible. Moreover, SAPSO maintains the mean charging price per kWh near the original charging pattern and yields the most consistent pricing results as shown by the lowest coefficient of variation amongst all the methods. Paired tests are used to establish the statistical significance of differences in post-optimization stresses, and the results indicate that SAPSO attains high human-centric performance as compared against the competing strategies ( \(p \ll 0.001\) ). These findings justify the use of behavioral stress included in EV charging optimization have the potential to promote user experience and does not negatively affect essential operational efficiency.