变幅扑翼摘要修改实录

学术写作 | 从实验流水账到贡献导向

Posted by 陈陈 on September 28, 2026

编者按:本文记录一次会议论文摘要的修改过程,对象是课题组学生投往某国际会议的一篇扑翼飞行器论文——A Variable-amplitude Flight Strategy Study for Flapping Wing Aircraft Vehicle。

一、摘要原稿:

下面是学生的原始初稿:

This study presents Aquila-S, a single-drivetrain flapping-wing platform for regulating aerodynamic output and comparing electrical input power under equivalent mean-force constraints. A Pixhawk–STM32G4 architecture supports continuous full-amplitude flapping, low-frequency variable-amplitude flapping, phase-locked gliding, and periodic flap-gliding. Source-code co-simulation evaluated mode scheduling and phase control, while fixed-bench tests assessed aerodynamic forces and electrical input power. Variable-amplitude flapping provided adjustable low-output operation, but repeated drivetrain reversal limited the attainable frequency. Static screening identified θ = −10.0° as the target glide phase for subsequent flap-glide tests. These tests used 3 m/s inflow, a 30° body installation angle, and 3 Hz active flapping. At a 30% glide fraction, cycle periods of 4 and 6 s satisfied the prescribed mean-force tolerances. Relative to continuous flapping, these conditions reduced mean electrical input power by 20.1% and 19.8%, respectively. Force-matched flap-glide operation offers a promising approach to reducing mean electrical input power.

逐句诊断:

句 问题
1 single-drivetrain 不是领域标准术语;Aquila-S 项目代号在摘要中不传递信息;核心结论被埋到最后一句
2 四种模式列举尚可,但缺乏与后文的衔接线索
3 Source-code co-simulation 含义模糊,实际是软件程序在环仿真
4 传动反转限制频率不是什么优点,在摘要中突出强调没有必要
5 Static screening 同样含义模糊,读者只能推测其代表的意思
6–9 参数堆砌(3 m/s、30°、3 Hz、4 s、6 s),具体数字 20.1%、19.8% 淹没在细节中
10 结论笼统,Force-matched 术语突兀,没有顺畅地收尾

二、结构重组:

学术摘要的第一句就应该告诉读者:你做了什么,最核心的结论是什么。原稿把”扑-滑结合可降低功率”埋在最后一句,改后将定性结论(can significantly reduce)前置到首句。

后续按“平台→控制→验证→调节机制→实验→量化结果→意义”逻辑递进展开。每句话为下一句话搭桥:第二句用 The prototype 回指首句;第三句用 then 顺承验证步骤;第四句先声明调节能力,为实验做铺垫;第五句定量测试,承接上句。

三、做减法:

以下信息被删去:

  • single-drivetrain:非标准术语,不是本文贡献,读者不关心。
  • Aquila-S 项目代号:摘要空间有限,代号不传递信息,正文再提。
  • Static screening identified θ = −10.0°:这个细节不重要且难以理解。摘要不应被旁支分散注意力。
  • 具体工况参数:原稿列了 3 m/s、30°、3 Hz、4 s、6 s。摘要不是实验报告——读者若对条件感兴趣会查正文。最终仅保留”30% 滑翔占比”和”约 20% 节能”两个关键数字。
  • 传动反转限制频率:负面信息,更不需要在摘要体现。

删掉”自己觉得重要”但读者并不关心的细节,是学术写作的第一道门槛。

四、修改稿:

This study develops a variable-amplitude flapping-wing prototype and demonstrates that flap-glide operation can significantly reduce mean electrical input power. The prototype employs a Pixhawk–STM32G4 control architecture that implements four flight modes: continuous full-amplitude flapping, low-frequency variable-amplitude flapping, phase-locked gliding, and periodic flap-gliding. A Software-in-the-loop (SITL) co-simulation platform is then established to validate the mode scheduling and phase control accuracy. Variable-amplitude flapping adjusts aerodynamic output and mean electrical power via amplitude and frequency modulation. Wind-wall-based fixed-bench tests quantify aerodynamic forces and electrical power under equivalent mean-force constraints. For the selected operating conditions, a 30% glide fraction reduces mean electrical input power by approximately 20% relative to continuous flapping. These results establish flap-glide scheduling as a promising strategy for power-efficient flapping-wing flight.

逐句功能:

句 功能 要点
1 核心贡献 significantly reduce 定性先行,不堆砌数字
2 平台能力 prototype 回指首句,冒号引出四种模式
3 验证手段 then 顺承,明确验证对象是模式调度与相位控制
4 调节机制 声明变幅扑动能力,为实验做铺垫
5 实验方法 风墙台架定量测试
6 关键结果 仅留两个核心数字(30%、~20%),其余工况不列
7 意义提升 收尾点明扑-滑调度是高效飞行策略

五、一些体会和建议

  1. 摘要不是缩小版的全文。 它是一份自给自足的贡献声明——读者不用读正文也应该能说出你做了什么、得到了什么。把所有细节塞进去不等于”信息丰富”,等于”没有重点”。
  2. 做减法的判断力比做加法的勤奋更重要。 写摘要前的第一件事不是打开正文复制粘贴——而是问自己:如果只能留三个信息点,会是哪三个。非关键信息都删减。
  3. 句子之间要有逻辑连接。 写完一段后,一定要自己多读几遍。如果每句话都彼此独立,只是一股脑地把意思说完了而没有关联和逻辑,就需要重写。多用连接词让句子”流动”起来。

做实验、跑仿真、分析数据不是科研的全部,如何展示自己做的工作同样重要。我们要学会从读者的角度来思考。对于一篇摘要,绝大多数读者并不关心实验细节,而只关心三件事:你做了什么、得到了什么、为什么重要。