基于智能预测的连续变刚度半主动控制结构的减震性能研究

Investigation of seismic mitigation performance of a continuously variable stiffness semi-active control structure based on intelligent prediction

摘要

结构振动控制技术是提升土木工程结构安全性的重要手段,其中半主动控制因兼具被动控制的可靠性与主动控制的高效性而备受关注。然而,传统半主动控制往往面临系统时滞效应导致的性能瓶颈。针对这一难题,本文提出了一种融合深度学习预测与模糊逻辑决策的Transformer-Fuzzy半主动变刚度控制算法。该算法利用Transformer模型在长序列时间特征提取上的优势,对结构的未来动力响应进行预测,并将预测结果引入模糊控制中,通过模糊推理实现对结构刚度的自适应调节,并引入遗传算法(genetic algorithm,GA)对模糊控制器参数进行全局寻优,有效补偿了控制时滞。为验证所提出控制算法的有效性,提出了伺服驱动的新型变刚度装置,开展了系统的振动台模型试验,并基于试验数据对数值仿真模型进行了校核与验证。研究对比了无控、被动控制以及基于Transformer-Fuzzy半主动控制结构在不同频谱特性地震波及不同激励强度下的动力响应。试验与模拟结果表明,所建数值模型能准确捕捉结构动力特性,提出的控制策略具有优异的鲁棒性与自适应能力,其对结构位移抑制效果显著优于传统被动控制工况,证明了该方法在结构半主动控制领域的有效性与应用前景,为后续工程化研究奠定了基础。

Abstract

Structural vibration control technology is a pivotal strategy for enhancing the safety of civil engineering structures. Among various strategies, semi-active control has garnered significant research attention for combining the reliability of passive control with the high efficiency of active control. However, traditional semi-active control systems frequently encounter performance bottlenecks induced by system time delays. To address this challenge, this study proposes a Transformer-Fuzzy semi-active variable stiffness control algorithm that integrates deep learning-based prediction with fuzzy logic decision-making. Leveraging the advantages of the Transformer model in extracting long-sequence temporal features, the algorithm accurately predicts future structural dynamic responses. These predictions serve as inputs for a fuzzy inference system to achieve adaptive stiffness regulation. Furthermore, a genetic algorithm (GA) is employed to globally optimize the fuzzy controller parameters, effectively compensating for control time delays. To validate the efficacy of the proposed algorithm, a novel servo-driven variable stiffness device was developed, and systematic shaking table tests were conducted. Subsequently, a numerical simulation model was calibrated and validated based on the experimental data. Comparative analyses were performed on the dynamic responses of uncontrolled, passively controlled, and Transformer-Fuzzy semi-actively controlled structures under seismic excitations with varying spectral characteristics and intensities. Experimental and simulation results demonstrate that the developed numerical model accurately captures the dynamic characteristics of the structure, and the proposed control strategy exhibits exceptional robustness and adaptability. Its displacement suppression performance is significantly superior to that of conventional passive control schemes, confirming the effectiveness and application potential of the proposed method in semi-active structural control and laying a foundation for subsequent engineering implementation.

关键词

半主动控制 / 变刚度结构 / Transformer响应预测 / 模糊控制 / 振动台试验

Key words

semi-active control / variable stiffness structure / Transformer-based response prediction / fuzzy control / shaking table test

参考文献

[1] 周锡元, 阎维明, 杨润林 . 建筑结构的隔震、减振和振动控制[J]. 建筑结构学报, 2002, 23(2): 2-12,26.

[2] Zhou Xiyuan, Yan Weiming, Yang Runlin . Seismic base isolation, energy dissipation and vibration control of building structures[J]. Journal of Building Structures, 2002, 23(2): 2-12,26. (in Chinese)

[3] Saed T E, Nikolakopoulos G, Jonasson J E, et al. A state-of-the-art review of structural control systems[J]. Journal of Vibration and Control, 2015, 21(5): 919-937.

[4] Kumar G, Kumar R, Kumar A . A review of the controllers for structural control[J]. Archives of Computational Methods in Engineering, 2023, 30(6): 3977-4000.

[5] Yamada K, Kobori T . Control algorithm for estimating future responses of active variable stiffness structure[J]. Earthquake Engineering & Structural Dynamics, 1995, 24(8): 1085-1099.

[6] 姜南, 李忠献 . 基于半主动控制的MR阻尼器控制力优化[J]. 地震工程与工程振动, 2008, 28(3): 138-144.

[7] Jiang Nan, Li Zhongxian . Semi-active control based optimization of control force for MR dampers[J]. Earthquake Engineering and Engineering Dynamics, 2008, 28(3): 138-144. (in Chinese)

[8] Kobori T . Technology development and forecast of dynamically intelligent building (DIB)[J]. Journal of Intelligent Material Systems and Structures, 1990, 1(4): 391-407.

[9] Nagarajaiah S, Sahasrabudhe S . Seismic response control of smart sliding isolated buildings using variable stiffness systems: an experimental and numerical study[J]. Earthquake Engineering & Structural Dynamics, 2006, 35(2): 177-197.

[10] 涂文戈, 邹银生 . 基于能量方程的建筑结构半主动控制[J]. 地震工程与工程振动, 2005, 25(3): 145-151.

[11] Tu Wenge, Zou Yinsheng . Semi-active control of buildings based on energy response[J]. Earthquake Engineering and Engineering Dynamics, 2005, 25(3): 145-151. (in Chinese)

[12] Guclu R . Sliding mode and PID control of a structural system against earthquake[J]. Mathematical and Computer Modelling, 2006, 44(1/2): 210-217.

[13] Chakraborty S, Sensy K, Ji X D . Control of a building system with sudden nonlinearity using the sliding mode technique[J]. Journal of Vibration and Control, 2025, 31(15/16): 3339-3356.

[14] 汪志昊, 陈政清 . 基于自供电磁流变阻尼器的隔震高架桥半主动控制[J]. 地震工程与工程振动, 2010, 30(1): 126-133.

[15] Wang Zhihao, Chen Zhengqing . Semi-active control of isolated elevated highway bridge with self-powered MR damper[J]. Earthquake Engineering and Engineering Dynamics, 2010, 30(1): 126-133. (in Chinese)

[16] Akbari M, Zand J P, Falborski T, et al. Advanced seismic control strategies for smart base isolation buildings utilizing active tendon and MR dampers[J]. Engineering Structures, 2024, 318: 118756.

[17] 孟庆利, 张敏政 . 基于模式识别的模糊半主动控制策略[J]. 地震工程与工程振动, 2005, 25(6): 152-158.

[18] Meng Qingli, Zhang Minzheng . Fuzzy control strategy based on mode identification used in semi-active control[J]. Earthquake Engineering and Engineering Dynamics, 2005, 25(6): 152-158. (in Chinese)

[19] Braz Manuel T, Barros R C . Semi-active fuzzy based control system for vibration reduction of a SDOF structure under seismic excitation[J]. Smart Structures and Systems, 2018, 21(4): 389-395.

[20] 杨润林, 闫维明, 周锡元 . 结构半主动控制研究中存在的若干问题[J]. 建筑结构学报, 2007, 28(4): 64-75.

[21] Yang Runlin, Yan Weiming, Zhou Xiyuan . Key problems in semiactive structural control[J]. Journal of Building Structures, 2007, 28(4): 64-75. (in Chinese)

[22] Smith O J M . Closer control of loops with dead time[J]. Chemical Engineering Progress, 1957, 53(5): 217-219.

[23] Kalman R E . A new approach to linear filtering and prediction problems[J]. Journal of Basic Engineering, 1960, 82(1): 35-45.

[24] Yang J N, Pan S, Huang H . An adaptive extended Kalman filter for structural damage identifications II: unknown inputs[J]. Structural Control and Health Monitoring, 2007, 14(3): 497-521.

[25] 郑史雄, 雷川鹤, 贾宏宇, 等 . 人工智能技术在桥梁抗震领域的应用综述[J]. 地震工程与工程振动, 2023, 43(4): 1-13.

[26] Zheng Shixiong, Lei Chuanhe, Jia Hongyu, et al. Application overview of artificial intelligence technology in bridge seismic field[J]. Earthquake Engineering and Engineering Dynamics, 2023, 43(4): 1-13. (in Chinese)

[27] Kim H S, Kim U . Development of a control algorithm for a semi-active mid-story isolation system using reinforcement learning[J]. Applied Sciences, 2023, 13(4): 2053.

[28] Pang Hui, Liu Fan, Xu Zeren . Variable universe fuzzy control for vehicle semi-active suspension system with MR damper combining fuzzy neural network and particle swarm optimization[J]. Neurocomputing, 2018, 306: 130-140.

[29] 高经纬, 涂建维, 刘康生, 等 . 基于GA-LSTM的高层建筑结构地震响应的分散控制研究[J]. 振动与冲击, 2021, 40(10): 114-122.

[30] Gao Jingwei, Tu Jianwei, Liu Kangsheng, et al. Decentralized control for the seismic response of high-rise building structures based on GA-LSTM[J]. Journal of Vibration and Shock, 2021, 40(10): 114-122. (in Chinese)

[31] Zhang Ruiyang, Chen Zhao, Chen Su, et al. Deep long short-term memory networks for nonlinear structural seismic response prediction[J]. Computers & Structures, 2019, 220: 55-68.

[32] Wu Yongxin, Yin Zhanpeng, Zhang Houle, et al. Prediction of nonlinear seismic response of underground structures in single-and multi-layered soil profiles using a deep gated recurrent network[J]. Soil Dynamics and Earthquake Engineering, 2023, 168: 107852.

[33] Zhang Gao, He Chang, Zhang Jin, et al. Prediction of bridge structure response and resilience assessment under main-aftershock: LSTM-Transformer model based on adaptive learning rate framework[J]. Engineering Structures, 2025, 345: 121449.

[34] Zhang Qingyu, Guo Maozi, Zhao Lingling, et al. Transformer-based structural seismic response prediction[J]. Structures, 2024, 61: 105929.

[35] Zhang Fanxing, Lv Yang, Li Bingchen . Semi-active control of MRE base isolation frame structure utilizing online learning deep neural network fuzzy control[J]. International Journal of Structural Stability and Dynamics, 2025: 2650282.

[36] Zhang Han, Wang Liangkun, Shi Weixing . Structural translational-torsional coupling response control using Transformer-based bidirectional semi-active TMD[J]. Journal of Building Engineering, 2025, 114: 114438.

[37] Nagarajaiah S, Narasimhan S . Seismic control of smart base isolated buildings with new semiactive variable damper[J]. Earthquake Engineering & Structural Dynamics, 2007, 36(6): 729-749.

[38] Ancheta T D, Darragh R B, Stewart J P, et al. NGA-West2 database[J]. Earthquake Spectra, 2014, 30(3): 989-1005.

[39] Zhang Han, Wang Liangkun, Shi Weixing . Seismic control of adaptive variable stiffness intelligent structures using fuzzy control strategy combined with LSTM[J]. Journal of Building Engineering, 2023, 78: 107549.

基金资助

国家自然科学基金项目(52308526)