首页 > AI前沿 > Power Setpoint Tracking for Large-Scale EV Charging: Transformer-Constrained Charging Policy Learning with Self-Tuning Constraint Prices

Power Setpoint Tracking for Large-Scale EV Charging: Transformer-Constrained Charging Policy Learning with Self-Tuning Constraint Prices

arXiv机器学习 2026-11-19 19:51 16 阅读 查看原文
用户需要我作为一个专业的新闻排版工程师,将提供的纯文本新闻正文排版成语义化的HTML片段。 需要遵循以下规则: 1. 只输出HTML片段本身,不要Markdown围栏、不要解释。 2. 每个自然段落用`

`包裹,长文用`

`/`

`划分小节。 3. 列表内容用`
  • `,引用用`
    `,代码用`
    `,重要句子可用``。
    4. 保留原文所有信息,不得增删改内容。
    5. 不要输出``/``/``等文档级标签。
    
    分析原文内容:
    原文是一段关于电力设定点跟踪(PST)和变压器约束充电策略学习(TC-CPL)的技术新闻/摘要。
    可以分为几个小节:
    1. 背景与问题定义(PST的定义和模型)
    2. TC-CPL方法介绍
    3. 实验设置与结果(600端口区域)
    4. 实验结果(42-kW变压器)
    5. 对比与自调优结果
    
    原文段落划分:
    段落1:In power setpoint tracking (PST), an operator follows a purchased charging profile without overloading shared transformers or undercharging vehicles. PST is cast as a constrained Markov decision process that minimizes tracking error subject to overload and mean user-satisfaction constraints.
    段落2:Transformer-Constrained Charging Policy Learning (TC-CPL) extends soft actor-critic with proportional-integral Lagrange-multiplier updates from measured violations, a residual on the incumbent dispatcher that is warm-started from its demonstrations, and a transformer-aware encoder shared by all ports. It observes no departure times or states of charge.
    段落3:Simulations (five seeds, 100 paired days) compare it with the incumbent, Round-Robin. On a 600-port, 122-substation district, raw overload falls from 3.59 to 0.0017 kWh/day (overloaded seed-days from 77% to 0.4%), and absolute tracking error falls by 27%. Delivered energy rises by 16%, and satisfaction rises from 0.959 to 0.989.
    段落4:On a 42-kW transformer with household load, overload falls from 1.86 to 0.0014 kWh/day and satisfaction rises from 0.974 to 0.985, but absolute tracking error is 22% higher.
    段落5:Against hand-set fixed multipliers, self-tuning is statistically indistinguishable on the district. On the single transformer, it raises satisfaction by 0.011 at similar overload, but a short sweep found a better-serving fixed pair. The overload multipliers stay below their cap.
    
    可以使用 `

    ` 或 `

    ` 来划分小节,让结构更清晰。 例如:

    电力设定点跟踪与模型构建

    In power setpoint tracking...

    TC-CPL 方法

    Transformer-Constrained Charging Policy Learning...

    区域仿真实验结果

    Simulations (five seeds, 100 paired days) compare it with the incumbent, Round-Robin. On a 600-port, 122-substation district, raw overload falls from 3.59 to 0.0017 kWh/day (overloaded seed-days from 77% to 0.4%), and absolute tracking error falls by 27%. Delivered energy rises by 16%, and satisfaction rises from 0.959 to 0.989.

    单变压器仿真实验结果

    On a 42-kW transformer with household load, overload falls from 1.86 to 0.0014 kWh/day and satisfaction rises from 0.974 to 0.985, but absolute tracking error is 22% higher.

    自调优与固定乘子对比

    Against hand-set fixed multipliers, self-tuning is statistically indistinguishable on the district. On the single transformer, it raises satisfaction by 0.011 at similar overload, but a short sweep found a better-serving fixed pair. The overload multipliers stay below their cap.

    检查规则: - 只输出HTML片段本身,无Markdown,无解释:是。 - 每个自然段落用`

    `包裹,长文用`

    `/`

    `划分小节:是。 - 列表、引用、代码等:这里没有明显的列表和引用,代码也不需要。重要句子用了``。 - 保留原文所有信息,不得增删改内容:检查文本是否完全一致。 - 段落1:In power setpoint tracking (PST), an operator follows a purchased charging profile without overloading shared transformers or undercharging vehicles. PST is cast as a constrained Markov decision process that minimizes tracking error subject to overload and mean user-satisfaction constraints. (一致) - 段落2:Transformer-Constrained Charging Policy Learning (TC-CPL) extends soft actor-critic with proportional-integral Lagrange-multiplier updates from measured violations, a residual on the incumbent dispatcher that is warm-started from its demonstrations, and a transformer-aware encoder shared by all ports. It observes no departure times or states of charge. (一致) - 段落3:Simulations (five seeds, 100 paired days) compare it with the incumbent, Round-Robin. On a 600-port, 122-substation district, raw overload falls from 3.59 to 0.0017 kWh/day (overloaded seed-days from 77% to 0.4%), and absolute tracking error falls by 27%. Delivered energy rises by 16%, and satisfaction rises from 0.959 to 0.989. (一致) - 段落4:On a 42-kW transformer with household load, overload falls from 1.86 to 0.0014 kWh/day and satisfaction rises from 0.974 to 0.985, but absolute tracking error is 22% higher. (一致) - 段落5:Against hand-set fixed multipliers, self-tuning is statistically indistinguishable on the district. On the single transformer, it raises satisfaction by 0.011 at similar overload, but a short sweep found a better-serving fixed pair. The overload multipliers stay below their cap. (一致) - 不要输出``/``/``等文档级标签:是。 微调标题,使其符合新闻排版风格,且不改变原意(标题是新增的,但规则说“长文用

    /

    划分小节”,所以新增小节标题是允许且符合要求的,只要正文内容不增删改)。 或者,为了绝对遵守“不得增删改内容”,我是否应该不添加任何原文没有的标题文字? 规则说“长文用

    /

    划分小节”,这通常意味着允许添加小节标题来组织结构。如果严格不增删改内容,可能指的是正文文本。为了安全,小节标题可以使用原文中的关键短语,或者简单的概括。 比如:

    背景与模型

    TC-CPL 方法

    区域仿真结果

    单变压器仿真结果

    对比分析

    或者,直接用原文的句子作为标题?不,标题通常是提炼的。只要正文一字不差即可。 让我们使用概括性小标题:

    电力设定点跟踪与模型

    TC-CPL 方法框架

    区域级仿真实验

    单变压器仿真实验

    自调优与固定乘