基于 MISSA-BP 算法及有限状态反馈的结构振动控制研究

Structural vibration control based on MISSA-BP algorithm and finite state feedback

摘要

针对麻雀搜索算法搜索过程中易困入局部且收敛速度有提升空间，传统 BP 神经网络主动控制效果不稳定，全部楼层布置传感器经济性和实用性较差等问题，提出一种多策略改进的麻雀搜索算法（multi-strategy improved sparrow search algorithm，MISSA），并选择 5 个基准测试函数进行性能评估。将 MISSA 与 BP 神经网络结合设计了 MISSA-BP 算法，将传感器的布置数量和位置及对应的 BP 神经网络同步优化，并对 2 种地震波作用下的 20 层 Benchmark 模型进行振动控制。结果表明: MISSA 具有更好的寻优精度和速度，探索能力更强；MISSA-BP 算法比训练效果更优的 BP 神经网络控制效果更好，该算法仅需 7 个楼层传感器反馈的位移和速度信息就可达到与全状态反馈 LQR 控制接近的控制效果，且具有一定的泛化能力。

Abstract

In response to the problems of being easily trapped in local areas and having room for improvement in convergence speed during the search process of the sparrow search algorithm，the unstable active control effect of traditional BP neural networks，and the poor economy and practicality of arranging sensors on all floors，a multi-strategy improved sparrow search algorithm（MISSA）is proposed，and five benchmark test functions are selected for performance evaluation. Then the MISSA-BP algorithm was designed by combining MISSA with BP neural network. The number and position of sensors and their corresponding BP neural network were synchronously optimized，and vibration control was performed on the 20 layer Benchmark model under the action of two types of seismic waves. The results indicate that MISSA has superior optimization accuracy and speed，as well as stronger exploration ability. Additionally，the MISSA-BP algorithm exhibits better control performance than the BP neural network with improved training performance. This algorithm requires feedback solely on displacement and velocity information from seven floor sensors to achieve control effects similar to full state feedback LQR control，and it also demonstrates a certain degree of generalization ability.

关键词

有限状态反馈 / 改进麻雀搜索算法 / BP 神经网络 / 传感器

Key words

finite state feedback / improved sparrow search algorithm / BP neural network / sensor

参考文献

[1] Xue J K， Shen B. A novel swarm intelligence optimization approach:sparrow search algorithm[J]. Systems Science & Control Engineering, 2020， 8（1）: 22-34.

[2] 吕鑫, 慕晓冬, 张钧, 等. 混沌麻雀搜索优化算法[J]. 北京航空航天大学学报, 2021， 47（8）: 1712-1720.

[3] Lü Xin， Mu Xiaodong， Zhang Jun， et al. Chaos sparrow search optimization algorithm[J]. Journal of Beijing University of Aeronautics and Astronautics, 2021， 47（8）: 1712-1720（in Chinese）.

[4] 毛清华, 张强, 毛承成, 等. 混合正弦余弦算法和 Lévy 飞行的麻雀算法[J]. 山西大学学报（自然科学版）, 2021， 44（6）: 1086-1091.

[5] Mao Qinghua， Zhang Qiang， Mao Chengcheng， et al. Mixing sine and cosine algorithm with Lévy flying chaotic sparrow algorithm[J]. Journal of Shanxi University（Natural Science Edition）, 2021， 44（6）: 1086-1091（in Chinese）.

[6] Venanzi I. A review on adaptive methods for structural control[J]. The Open Civil Engineering Journal, 2016， 10: 653-667.

[7] 汪权, 韩强强, 王肖东, 等. 地震作用下高层建筑结构的分散神经网络振动控制研究[J]. 计算力学学报, 2019， 36（1）: 77-82.

[8] Wang Quan， Han Qiangqiang， Wang Xiaodong， et al. Decentralized neural networks vibration control of tall buildings under earthquakes[J]. Chinese Journal of Computational Mechanics, 2019， 36（1）: 77-82（in Chinese）.

[9] 汪权, 王文, 韩新节, 等. 基于神经网络算法的建筑结构振动分散控制研究[J]. 计算力学学报, 2021， 38（5）: 580-585.

[10] Wang Quan， Wang Wen， Han Xinjie， et al. Decentralized control of building structure based on neural network algorithm[J]. Chinese Journal of Computational Mechanics, 2021， 38（5）: 580-585（in Chinese）.

[11] 潘兆东, 谭平, 刘良坤, 等. 基于自适应 RBF 神经网络算法的建筑结构递阶分散控制研究[J]. 土木工程学报, 2018， 51（1）: 51-57.

[12] Pan Zhaodong， Tan Ping， Liu Liangkun， et al. Hierarchical decentralized control of building structure based on adaptive RBF neural network algorithm[J]. China Civil Engineering Journal, 2018， 51（1）: 51-57（in Chinese）.

[13] 王志伟, 葛楠, 李春伟. 基于 BP 神经网络算法的结构振动模态模糊控制[J]. 山东大学学报（工学版）, 2020， 50（5）: 13-19.

[14] Wang Zhiwei， Ge Nan， Li Chunwei. Fuzzy control of structure vibration mode based on BP neural network algorithm[J]. Journal of Shandong University（Engineering Science）, 2020， 50（5）: 13-19（in Chinese）.

[15] 王文. 基于有限状态反馈遗传 BP 算法的高层结构振动控制研究[D]. 合肥: 合肥工业大学， 2022.

[16] 宋立钦, 陈文杰, 陈伟海, 等. 基于混合策略的麻雀搜索算法改进及应用[J]. 北京航空航天大学学报, 2023， 49（8）: 2187-2199.

[17] Song Liqin， Chen Wenjie， Chen Weihai， et al. Improvement and application of hybrid strategy-based sparrow search algorithm[J]. Journal of Beijing University of Aeronautics and Astronautics, 2023， 49（8）: 2187-2199（in Chinese）.

[18] Mirjalili S. SCA:a sine cosine algorithm for solving optimization problems[J]. Knowledge-Based Systems, 2016， 96: 120-133.

[19] Mirjalili S. Moth-flame optimization algorithm: a novel nature-inspired heuristic paradigm[J]. Knowledge-Based Systems, 2015， 89: 228-249.

[20] 张永, 陈锋. 一种改进的鲸鱼优化算法[J]. 计算机工程, 2018， 44（3）: 208-213.

[21] Zhang Yong， Chen Feng. A modified whale optimization algorithm[J]. Computer Engineering, 2018， 44（3）: 208-213（in Chinese）.

[22] Mirjalili S， Lewis A. The whale optimization algorithm[J]. Advances in Engineering Software, 2016， 95: 51-67.

[23] 欧进萍. 结构振动控制:主动、半主动和智能控制[M]. 北京: 科学出版社， 2003.

[24] 刘中宪, 乔一丁, 朱朔, 等. 基于 IBEM 和 QMESSA 的饱和场地地下空洞反演[J]. 应用力学学报, 2026， 43（3）: 665-674.

[25] Liu Zhongxian， Qiao Yiding， Zhu Shuo， et al. Inversion of subsurface cavity in saturated sites based on IBEM and QMESSA[J]. Chinese Journal of Applied Mechanics, 2026， 43（3）: 665-674（in Chinese）.

[26] 毛晓敏, 张慧华, 纪晓磊, 等. 基于 XFEM 与 BP 神经网络的裂纹智能识别[J]. 应用力学学报, 2022， 39（6）: 1158-1167.

[27] Mao Xiaomin， Zhang Huihua， Ji Xiaolei， et al. Intelligent crack identification based on XFEM and BP neural network[J]. Chinese Journal of Applied Mechanics, 2022， 39（6）: 1158-1167（in Chinese）.

基金资助

国家自然科学基金资助项目(52378298)