面向超宽带微波吸收的连续参数空间寻优：一种序列感知深度学习与进化优化方法

引用本文: 狄肖创, 袁牧山, 卢俊宇, 阳申全, 周翠清, 陈洋, 邹华维. 面向超宽带微波吸收的连续参数空间寻优：一种序列感知深度学习与进化优化方法[J]. 物理化学学报, 2026, 42(9): 100270. doi: 10.1016/j.actphy.2026.100270 shu

Citation: Xiaochuang Di, Mushan Yuan, Junyu Lu, Shenquan Yang, Cuiqing Zhou, Yang Chen, Huawei Zou. Navigating the continuous parameter space for ultra-broadband microwave absorption: a sequence-aware deep learning and evolutionary optimization approach[J]. Acta Physico-Chimica Sinica, 2026, 42(9): 100270. doi: 10.1016/j.actphy.2026.100270 shu

面向超宽带微波吸收的连续参数空间寻优：一种序列感知深度学习与进化优化方法

四川大学高分子材料工程国家重点实验室, 高分子研究所, 四川成都 610065

通讯作者: Email: cy3262276@163.com (陈洋); hwzou@163.com (邹华维)

摘要: 电磁污染的快速蔓延促使需要尽快开发具备超宽带能力的高性能微波吸收材料。然而，传统的试错设计范式受限于制备过程的路径依赖性，即特定的浸渍历史严格决定了最终的梯度分布和阻抗匹配。为此，本研究提出一种序列感知的逆向设计框架，将长短期记忆(LSTM)神经网络与遗传算法(GA)相结合。基于多步浸渍制备的聚氨酯/碳纳米管(PU/CNT)泡沫工艺—性能数据库，构建高保真LSTM代理模型以表征浸渍历程中的时间依赖关系，并预测随频率变化的复介电常数。随后，GA基于该预测模型在设计空间中寻优，确定可实现精确三层梯度构型的最优浸渍路径。优化后的泡沫通过介电损耗的阶梯式递增实现性能提升，获得−24.2 dB的平均反射损耗(RL)，并实现覆盖2–18 GHz的超宽有效吸收带宽(EAB)。本研究表明，利用包含历程信息的数据驱动策略可显著加速材料发现，为先进功能复合材料的智能设计提供可扩展范式。

关键词:

English

Navigating the continuous parameter space for ultra-broadband microwave absorption: a sequence-aware deep learning and evolutionary optimization approach

The State Key Lab of Polymer Materials Engineering, Polymer Research Institution, Sichuan University, Chengdu 610065, Sichuan Province, China

Corresponding author: Email: cy3262276@163.com (Yang Chen); hwzou@163.com (Huawei Zou)

Received Date: 13 January 2026
Accepted Date: 27 February 2026
Revised Date: 12 February 2026
Available Online: 15 September 2026

Abstract: The rapid proliferation of electromagnetic (EM) pollution necessitates the urgent development of high-performance microwave absorption (MA) materials with ultra-broadband capabilities. However, conventional trial-and-error design paradigms are constrained by the path-dependent nature of the fabrication process, where the specific impregnation history strictly governs the final gradient distribution and impedance matching. To address this, this study proposes a sequence-aware inverse design framework that integrates a Long Short-Term Memory (LSTM) neural network with a Genetic Algorithm (GA). Leveraging a process-property database derived from multi-step impregnated polyurethane/carbon nanotube (PU/CNT) foams, a high-fidelity LSTM surrogate model is developed to decode the complex temporal dependencies within the impregnation history and accurately predict frequency-dependent complex permittivity. Subsequently, the GA utilizes this predictive model to navigate the design space, identifying an optimal impregnation pathway that yields a precise three-layer gradient configuration. The resulting optimized foam, characterized by a rational stepwise increase in dielectric loss, achieves an exceptional average reflection loss (RL) of −24.2 dB and an ultra-wide effective absorption bandwidth (EAB) covering the full 2–18 GHz range. This work demonstrates the efficacy of history-based data-driven strategies in accelerating material discovery, offering a scalable paradigm for the intelligent design of advanced functional composites.

Key words:

Microwave absorption
/ Inverse design
/ Long short-term memory (LSTM) networks
/ Genetic algorithm
/ Gradient structures

1. [1]
  H. Yan, S. Xuan, X. Fan, Y. Shan, X. Xu, X. Yao, Compos. Sci. Technol. 248 (2024) 110471, https://doi.org/10.1016/j.compscitech.2024.110471. doi: 10.1016/j.compscitech.2024.110471
2. [2]
  Z. Liu, R. Zhang, S. Wang, W. Zhao, G. Yu, L. Wu, Compos. Sci. Technol. 240 (2023) 110094, https://doi.org/10.1016/j.compscitech.2023.110094. doi: 10.1016/j.compscitech.2023.110094
3. [3]
  X. Du, F. Yan, M. Cheng, H. Li, C. Peng, Y. Liu, D. Liu, D. Lan, G. Wu, Z. Jia, Int. J. Miner. Metall. Mater. (2025), https://doi.org/10.1007/s12613-025-3317-1. doi: 10.1007/s12613-025-3317-1
4. [4]
  Z. Zhao, Y. Qing, L. Kong, H. Xu, X. Fan, J. Yun, et al., Adv. Mater. 36 (4) (2024) e2304182, https://doi.org/10.1002/adma.202304182. doi: 10.1002/adma.202304182
5. [5]
  G. Qin, Y. Liu, Y. Yan, Z. Cheng, G. Ma, K. Zhang, X. Huang, Compos. Part B Eng. 298 (2025) 112397, https://doi.org/10.1016/j.compositesb.2025.112397. doi: 10.1016/j.compositesb.2025.112397
6. [6]
  X. Wang, Y. Yuan, L. Wang, X. Sun, T. Zhang, C. Liu, Y. Li, J. Adv. Ceram. 14 (2025) 112298, https://doi.org/10.26599/JAC.2025.9221172. doi: 10.26599/JAC.2025.9221172
7. [7]
  T. Zhang, J. Qiu, S. Wang, Y. Juan, J. Li, W. Wang, Adv. Funct. Mater. 35 (2025) 2414601, https://doi.org/10.1002/adfm.202521010. doi: 10.1002/adfm.202521010
8. [8]
  G. Qin, Y. Liu, Y. Yan, G. Ma, B. Gao, Z. Cheng, T. Chen, Adv. Funct. Mater. 35 (2025) e19796, https://doi.org/10.1002/adfm.202519796. doi: 10.1002/adfm.202519796
9. [9]
  K. Zhang, Y. Liu, X. Li, X. Wang, J. Liu, X. Liu, Adv. Mater. 37 (35) (2025) 2506386, https://doi.org/10.1002/adma.202506386. doi: 10.1002/adma.202506386
10. [10]
  R. Guo, S. Li, J. Li, L. Wang, W. Jiang, Adv. Colloid Interface Sci. 344 (2025) 103612, https://doi.org/10.1016/j.cis.2025.103612. doi: 10.1016/j.cis.2025.103612
11. [11]
  Q. Jiang, Y. Qiao, C. Xiang, A. Uddin, L. Wu, F. Qin, Adv. Compos. Hybrid Mater. 5 (4) (2022) 3190, https://doi.org/10.1007/s42114-021-00394-y. doi: 10.1007/s42114-021-00394-y
12. [12]
  Y. Liu, J. Zhou, C. Li, H. Zhang, Y. Wang, Y. Yan, et al., Nat. Commun. 16 (1) (2025) 202, https://doi.org/10.1038/s41467-024-55776-9. doi: 10.1038/s41467-024-55776-9
13. [13]
  C. Li, L. Liang, B. Zhang, Y. Yang, G. Ji, Nano-Micro Lett. 17 (1) (2024) 40, https://doi.org/10.1007/s40820-024-01549-4. doi: 10.1007/s40820-024-01549-4
14. [14]
  Z. Liu, B. Wang, S. Wei, W. Huang, Y. Wang, Y. Liang, L. Guo, X. Wang, R. Shi, W. Fu, et al., Nano Mater. Sci. (2025), https://doi.org/10.1016/j.nanoms.2025.09.001. doi: 10.1016/j.nanoms.2025.09.001
15. [15]
  Y. Ai, R. Xing, N. Ren, R. Huang, M. Cui, R. Su, J. Kong, Matter 8 (9) (2025) 102289, https://doi.org/10.1016/j.matt.2025.102289. doi: 10.1016/j.matt.2025.102289
16. [16]
  J. Zhou, X. Huang, D. Lan, Z. Jia, G. Wu, Carbon 248 (2026) 121143, https://doi.org/10.1016/j.carbon.2025.121143. doi: 10.1016/j.carbon.2025.121143
17. [17]
  M. Zhang, T. Shang, Y. Wang, Y. Song, Chem. Eng. J. 523 (2025) 168856, https://doi.org/10.1016/j.cej.2025.168856. doi: 10.1016/j.cej.2025.168856
18. [18]
  X. Luo, H. Xie, Y. Ma, D. Lan, G. Wu, Z. Jia, Int. J. Miner. Metall. Mater. (2025), https://doi.org/10.1007/s12613-025-3252-1. doi: 10.1007/s12613-025-3252-1
19. [19]
  H. Wu, X. Ren, W. Hu, Y. Tang, H. Yin, H. Fan, H. Yuan, C. Wang, Y. Xin, Carbon 233 (2025) 119883, https://doi.org/10.1016/j.carbon.2024.119883. doi: 10.1016/j.carbon.2024.119883
20. [20]
  X. Zuo, C. Zhou, T. Yuan, N. Wen, C. Sun, H. Zhang, Y. Zhang, J. Tian, C. Li, Z. Fan, et al., Chem. Eng. J. 494 (2024) 152866, https://doi.org/10.1016/j.cej.2024.152866. doi: 10.1016/j.cej.2024.152866
21. [21]
  M. Han, Z. Jia, D. Lan, Z. Gao, G. Wu, Chin. J. Chem. (2026), https://doi.org/10.1002/cjoc.70494. doi: 10.1002/cjoc.70494
22. [22]
  Y. Huang, Q. Fan, J. Chen, L. Li, M. Chen, L. Tang, D. Fang, Compos. Sci. Technol. 191 (2020) 108066, https://doi.org/10.1016/j.compscitech.2020.108066. doi: 10.1016/j.compscitech.2020.108066
23. [23]
  X. Ma, Y. Duan, L. Huang, H. Lei, X. Yang, Compos. Part B Eng. 233 (2022) 109659, https://doi.org/10.1016/j.compositesb.2022.109659. doi: 10.1016/j.compositesb.2022.109659
24. [24]
  X. Li, J. Liu, Z. Jia, D. Lan, D. Ai, Z. Gao, F. Bai, G. Wu, J. Mater. Sci. Technol. 268 (2026) 41, https://doi.org/10.1016/j.jmst.2025.12.046. doi: 10.1016/j.jmst.2025.12.046
25. [25]
  R. Fan, D. Li, W. Liao, T. Liu, X. Li, T. Tang, et al., Compos. Part B Eng. 287 (2024) 111880, https://doi.org/10.1016/j.compositesb.2024.111880. doi: 10.1016/j.compositesb.2024.111880
26. [26]
  L. Liang, Y. Lin, Y. Huang, M. Chen, Compos. Part A Appl. Sci. Manuf. 160 (2022) 107069, https://doi.org/10.1016/j.compositesa.2022.107069. doi: 10.1016/j.compositesa.2022.107069
27. [27]
  Y. Lin, X. Yang, Y. Huang, M. Chen, Compos. Struct. 272 (2021) 114235, https://doi.org/10.1016/j.compstruct.2021.114235. doi: 10.1016/j.compstruct.2021.114235
28. [28]
  Y. Huang, X. Yuan, Y. Yuan, J. Liu, X. Fan, L. Zhao, Z. Cheng, X. Li, G. Wu, Carbon 144 (2019) 449, https://doi.org/10.1016/j.carbon.2018.11.052. doi: 10.1016/j.carbon.2018.11.052
29. [29]
  Y. Huang, D. Wu, M. Chen, K. Zhang, D. Fang, Carbon 177 (2021) 79, https://doi.org/10.1016/j.carbon.2021.02.066. doi: 10.1016/j.carbon.2021.02.066
30. [30]
  X. Ren, Z. Jia, Z. Gao, S. Zhang, Y. Zhang, D. Lan, G. Wu, Adv. Funct. Mater. 36 (2026) e24264, https://doi.org/10.1002/adfm.202524264. doi: 10.1002/adfm.202524264
31. [31]
  Q. An, D. Li, W. Liao, T. Liu, D. Joralmon, X. Li, J. Zhao, Adv. Mater. 35 (2023) e2300659, https://doi.org/10.1002/adma.202300659. doi: 10.1002/adma.202300659
32. [32]
  W. Jiang, S. Xu, C. Lv, D. Lan, S. Zhang, Z. Gao, Z. Jia, G. Wu, Carbon 245 (2025) 120784, https://doi.org/10.1016/j.carbon.2025.120784. doi: 10.1016/j.carbon.2025.120784
33. [33]
  D. Li, X. Zheng, H. Gu, X. Xuan, B. Liu, H. Feng, M. Chen, Nano Mater. Sci. 6 (4) (2024) 456, https://doi.org/10.1016/j.nanoms.2023.09.005. doi: 10.1016/j.nanoms.2023.09.005
34. [34]
  M. Shi, Z. Jia, D. Lan, Z. Gao, S. Zhang, G. Wu, Adv. Funct. Mater. 36 (2026) e28665, https://doi.org/10.1002/adfm.202528665. doi: 10.1002/adfm.202528665
35. [35]
  J. Yang, Y. Guo, X. Li, Y. Qiu, X. Wang, J. Zhou, B. Li, L. Zhang, Y. Bian, X. Xiao, et al., Adv. Mater. 38 (6) (2026) e17422, https://doi.org/10.1002/adma.202517422. doi: 10.1002/adma.202517422
36. [36]
  J. Liang, F. Ye, Q. Song, Y. Cao, C. Xiao, Y. Qin, Y. Lin, B. Huang, X. Wang, C. Li, Compos. Part B Eng. 297 (2025) 112298, https://doi.org/10.1016/j.compositesb.2025.112298. doi: 10.1016/j.compositesb.2025.112298
37. [37]
  M. Feng, K. Zhang, J. Xiao, B. Liu, H. Cheng, Y. Li, Z. Zhao, B. Liang, Compos. Part B Eng. 263 (2023) 110862, https://doi.org/10.1016/j.compositesb.2023.110862. doi: 10.1016/j.compositesb.2023.110862
38. [38]
  Y. Huang, D. Wu, K. Zhang, H. Yang, W. Dong, M. Chen, D. Fang, Compos. Sci. Technol. 213 (2021) 108898, https://doi.org/10.1016/j.compscitech.2021.108898. doi: 10.1016/j.compscitech.2021.108898
39. [39]
  C. Wang, Z. Lv, M. P. Mohan, Z. Cui, Z. Liu, Y. Jiang, J. Li, C. Wang, S. Pan, M. F. Karim, et al., Adv. Mater. 33 (41) (2021) e2102131, https://doi.org/10.1002/adma.202102131. doi: 10.1002/adma.202102131
40. [40]
  L. Liang, X. Yang, C. Li, R. Yu, B. Zhang, Y. Yang, G. Ji, Adv. Mater. 36 (24) (2024) e2313939, https://doi.org/10.1002/adma.202313939. doi: 10.1002/adma.202313939
41. [41]
  N. Yang, H. Gao, S. Ju, Z. Zhang, Y. Zhao, X. -W. Guo, S. Xing, J. Liu, J. Zhang, K. Duan, et al., Compos. Part B Eng. 307 (2025) 112882, https://doi.org/10.1016/j.compositesb.2025.112882. doi: 10.1016/j.compositesb.2025.112882
42. [42]
  R. Zhang, Y. Yuan, X. Wang, X. Sun, S. Wang, Z. Yang, Y. Ma, E. Zhang, Y. Li, Chem. Eng. J. 511 (2025) 161634, https://doi.org/10.1016/j.cej.2025.161634. doi: 10.1016/j.cej.2025.161634
43. [43]
  T. Zou, J. Shi, M. Wang, C. Lin, Adv. Funct. Mater. (2025) e12788, https://doi.org/10.1002/adfm.202512788. doi: 10.1002/adfm.202512788
44. [44]
  Y. Pan, K. Yu, D. Lan, Z. Zhang, Z. Chen, Carbon 245 (2025) 120824, https://doi.org/10.1016/j.carbon.2025.120824. doi: 10.1016/j.carbon.2025.120824
45. [45]
  S. Song, B. Zheng, L. Chen, H. Shu, D. Gao, D. Lan, T. Li, X. Liu, Y. Ma, J. Energy Storage 134 (2025) 118282, https://doi.org/10.1016/j.est.2025.118282. doi: 10.1016/j.est.2025.118282
46. [46]
  B. Zeng, F. Zhang, K. Zhao, M. Ahmad, J. Wu, L. Zhang, D. Lan, B. Zhang, J. Mater. Sci. Technol. 251 (2026) 193, https://doi.org/10.1016/j.jmst.2025.07.008. doi: 10.1016/j.jmst.2025.07.008
47. [47]
  B. Zhou, L. Tang, H. Cheng, M. Ahmad, J. Wu, L. Zhang, D. Lan, B. Zhang, Nano Res. 18 (9) (2025) 94907739, https://doi.org/10.26599/NR.2025.94907739. doi: 10.26599/NR.2025.94907739
48. [48]
  S. Xiong, L. Cai, Y. Zhang, Y. Ma, D. Lan, G. Chen, C. Dong, H. Guan, Rare Met. 44 (2025) 7720, https://doi.org/10.1007/s12598-025-03439-z. doi: 10.1007/s12598-025-03439-z
49. [49]
  X. Meng, J. Li, S. Zhang, D. Lan, M. Yu, T. Long, C. Wang, Adv. Fiber Mater. 7 (3) (2024) 736, https://doi.org/10.1007/s42765-024-00501-w. doi: 10.1007/s42765-024-00501-w

扫一扫看文章

计量

PDF下载量: 0
文章访问数: 13
HTML全文浏览量: 2

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

面向超宽带微波吸收的连续参数空间寻优：一种序列感知深度学习与进化优化方法

通讯作者: Email: cy3262276@163.com (陈洋); hwzou@163.com (邹华维)

English

Navigating the continuous parameter space for ultra-broadband microwave absorption: a sequence-aware deep learning and evolutionary optimization approach

Corresponding author: Email: cy3262276@163.com (Yang Chen); hwzou@163.com (Huawei Zou)

CCS Chemistry：嵌段共聚物自组装成 “三明治”二维有机材料，实现纳米级压力传感功能

CCS Chemistry：颜徐州、鲍哲南：你的皮肤可能是一片SPMs

CCS Chemistry：工作高效不疲劳，只须让催化剂动起来

CCS Chemistry：静电组装导向聚合- 聚电解质纳米凝胶量化制备新策略

CCS Chemistry：金黄色葡萄球菌醛基脱氢酶是如何识别特异性底物的？

计量

通讯作者: 陈斌, bchen63@163.com

中国化学会秘书处

面向超宽带微波吸收的连续参数空间寻优：一种序列感知深度学习与进化优化方法

通讯作者: Email: cy3262276@163.com (陈洋); hwzou@163.com (邹华维)

English

Navigating the continuous parameter space for ultra-broadband microwave absorption: a sequence-aware deep learning and evolutionary optimization approach

Corresponding author: Email: cy3262276@163.com (Yang Chen); hwzou@163.com (Huawei Zou)

CCS Chemistry：嵌段共聚物自组装成 “三明治”二维有机材料，实现纳米级压力传感功能

CCS Chemistry：颜徐州、鲍哲南： 你的皮肤可能是一片SPMs

CCS Chemistry：工作高效不疲劳，只须让催化剂动起来

CCS Chemistry：静电组装导向聚合- 聚电解质纳米凝胶量化制备新策略

CCS Chemistry：金黄色葡萄球菌醛基脱氢酶是如何识别特异性底物的？

计量

文章相关

通讯作者: 陈斌, bchen63@163.com

中国化学会秘书处

CCS Chemistry：颜徐州、鲍哲南：你的皮肤可能是一片SPMs