XLXumei Lin

Xumei Lin

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Xumei Lin has a total of 23 co-authors. Here are the top co-authors by number of shared publications:

  • Shijie Yu

    School of Information and Control Engineering, Qingdao University of Technology, Qingdao 266520, China

    4 shared publications
  • Peng Wang

    School of Information and Control Engineering, Qingdao University of Technology, Qingdao 266520, China

    3 shared publications
  • Chenglong Yin

    School of Information and Control Engineering, Qingdao University of Technology, China

    2 shared publications
  • Guanghui Zhu

    Department of Information and Control Engineering , Qingdao University of Technology , Qingdao , China

    2 shared publications
  • Penggang Wang

    Department of Civil Engineering , Qingdao University of Technology , Qingdao , China

    2 shared publications
  • Shiyuan Wang

    School of Information and Control Engineering Qingdao University of Technology Qingdao China

    2 shared publications

Latest Publications

  • Quantitative detection of local corrosive conditions in reinforced concrete using an embedded, structurally optimized electromagnetic sensor

    by Xumei Lin, Jiahui Shen, Chi Ni, Jinqian Guo, Wang Penggang

    Purpose: This study aims to propose an embedded, structurally optimized electromagnetic sensor for quantitatively detecting local corrosive conditions in reinforced concrete by correlating its magnetic response with the corrosion-induced mass loss of the sensor steel shell. Design/methodology/approach: A square hollow steel shell sensor integrating a permanent magnet and Hall elements was designed as an embedded detection component. The sensor forms a stable, short-range and repeatable magnetic response channel, allowing local corrosion severity to be quantified from magnetic-field variations caused by sensor steel shell corrosion. Two shell corrosion models were established to reveal the corrosion-response mechanism, and three sensor-structure models were compared to optimize the shell cavity and sensitive detection region. An equivalent magnetic circuit model was developed to relate magnetic flux density to shell mass loss, and accelerated corrosion tests were conducted for validation. Findings: The proposed sensor showed strong correlations between magnetic response and the mass loss ratio of the sensor steel shell. The coefficients of determination were 0.9972 and 0.9932 under uniform and nonuniform corrosion, respectively, supporting quantitative assessment of local corrosion severity under controlled accelerated-corrosion conditions. Research limitations/implications: This study was validated through modeling and accelerated corrosion tests. Further long-term field studies under different concrete environments and service conditions are needed to verify the sensor performance in practical reinforced concrete structures. Practical implications: The proposed embedded electromagnetic sensor provides a feasible method for local and quantitative detection of corrosive conditions in reinforced concrete. It may support early warning of potential reinforcement-corrosion risks and durability assessment of reinforced concrete structures. Social implications: Reliable detection of locally corrosive conditions can help improve the safety, durability and maintenance efficiency of reinforced concrete infrastructure by providing early warning of potential reinforcement-corrosion risks and helping extend service life. Originality/value: This work introduces an embedded electromagnetic sensor for quantitative characterization of sensor steel shell corrosion. The sensor steel shell acts as a corrosion-sensitive element whose mass loss reflects local corrosive conditions in reinforced concrete.

  • Backstepping sliding mode control based on dual-path extreme learning machine and disturbance observer for robotic manipulator trajectory tracking

    by Xumei Lin, Chenglong Yin, Zhiyuan Zhang, Runze Xu

    To achieve high-precision trajectory tracking for robotic manipulators subject to unknown nonlinearities and external disturbances, a backstepping sliding mode control strategy incorporating a dual-path extreme learning machine (DP-ELM) and a disturbance observer is proposed. The DP-ELM approximates the system’s unknown nonlinearities via two parallel paths: a main path driven by joint angle errors to capture dominant nonlinearities, and an auxiliary path utilizing angular velocity errors to compensate for unmodeled high-frequency residuals. To enhance approximation capability and computational efficiency, a multi-scale feature extraction mechanism and a Top-K dynamic sparse activation strategy are integrated into the network. Furthermore, an exponentially convergent disturbance observer is designed to estimate external disturbances online. Both numerical simulations in Simulink and experiments on a physical platform of a 3-degree-of-freedom (3-DOF) robotic manipulator demonstrate that the proposed scheme yields superior tracking accuracy, rapid convergence, and strong robustness against disturbances compared to existing methods.

  • Attention enhanced EfficientNet for concrete structure crack classification with generative adversarial network augmented data

    by Xumei Lin, Shiyuan Wang, Jiahui Shen

    Cracks are an important indicator of the decline in the load‐bearing capacity of buildings. Therefore, it is of great significance to detect and classify cracks in reinforced concrete (RC) building exterior walls. Accurately and automatically classifying cracks remains challenging due to the highly irregular nature of crack images, irregular lighting conditions, and background texture noise. An EfficientNet network model combined with a HiLo attention mechanism was proposed to achieve precise identification and classification of RC exterior wall cracks. Firstly, existing crack datasets were categorized, and a combination of classical data augmentation and conditional generative adversarial networks was used to augment the data, improving the model's generalization ability under different imaging conditions and mitigating the adverse effects of the unbalanced dataset. Furthermore, crack images were divided into high‐frequency (Hi‐Fi) and low‐frequency (Lo‐Fi) feature maps by applying the HiLo attention mechanism. Hi‐Fi module suppressed background noise and captured the edge details of cracks by retaining relatively high‐resolution feature maps. Lo‐Fi module extracts global features of cracks through window segmentation and average pooling operations, thereby enhancing the classification capability of model. The experimental results showed that the HiLo‐EfficientNet model achieved the best image classification performance with an overall classification accuracy of 91.63% compared to the original and mainstream deep learning models.


  • Assessment of reinforced concrete corrosion degree based on the quantum particle swarm optimised-generative adversarial network

    by Xumei Lin, Shijie Yu, Peng Wang, Shiyuan Wang

    Reinforced concrete corrosion inspection methods based on deep learning have been widely used in the engineering field to monitor the service status of reinforced concrete. However, in engineering practice, it is difficult to obtain a large amount of reinforced concrete corrosion data of different types, which greatly hinders the improvement of the accuracy of neural network models in predicting corrosion conditions. The classic generative adversarial network (GAN) model gives poor model quality for datasets with small amounts of data and high concentration. This paper proposes an improved generative adversarial network approach to optimise reinforced concrete corrosion data. Firstly, a quantum particle swarm optimisation (QPSO) algorithm is used to improve the generative adversarial network. Then, existing corrosion characteristic data is used to train the improved generative adversarial network until the ideal equilibrium state is reached. Next, the feature data generated by the generator are fused with the original data and the fused data are input into several common machine learning models for training. Experimental results show that compared with other conventional results obtained by directly inputting corrosion data into a neural network model for training, the improved method makes full use of multi-source signal data and achieves better classification performance.

  • Distributed formation control for leader‐follower quadrotor unmanned aerial vehicles system based on practical sliding mode and event‐triggered mechanisms

    by Xumei Lin, Qihang Lin, Chenglong Yin, Chengcheng Peng

    This article investigates the distributed formation trajectory tracking problem of a homogeneous quadrotor unmanned aerial vehicles (UAVs) system using a leader‐follower consensus protocol in three‐dimensional space. To enable the follower UAVs in the swarm to track a single leader UAV and form a desired formation under undirected topology, a six‐degree‐of‐freedom dynamic model with bounded time‐varying disturbances is established, which is divided into a tracking subsystem and a stabilizing subsystem. A practical sliding mode control (PSMC) based event‐triggered algorithm is proposed to achieve distributed formation control for a leader‐follower quadrotor UAVs system and reduce the network transmission resources. To address the potential issue of inter‐UAVs collisions during formation, an artificial potential field (APF) method is introduced in the tracking subsystem, and an disturbance observer is introduced in the stabilizing subsystem to estimate the unknown external disturbances. The boundedness of signals in the closed‐loop system is demonstrated using Lyapunov stability theory, and the exclusion of Zeno behavior can be ensured by having a positive lower bound on the inter‐event time between two consecutive triggering events. Finally, the effectiveness of the proposed event‐triggered algorithm is validated through simulation examples.


  • Study on influencing factors and regularity analysis of eddy current method applied to detection of reinforcement corrosion

    by Xumei Lin, Shijie Yu, Peng Wang, Shiyuan Wang

    Rebar corrosion is one of the main reasons for the deterioration of concrete structures. In order to ensure the safe operation of reinforced concrete structures, it is necessary to detect the corrosion rate of rebars in time. This article adopts the eddy current method based on electromagnetic induction to detect steel corrosion, studies the mechanism of alternating current (AC) impedance change induced by steel corrosion in coils, and establishes a mathematical model between steel corrosion and coil AC impedance. Meanwhile, the quantitative relationship between excitation signal frequency, concrete thickness, coil parameters, and coil impedance is studied. Through experiments, it was found that in the early stages of steel corrosion, there is a linear relationship between the scanning peak value of the coil impedance and the mass loss rate of the steel bars, and the slope of the fitting curve decays exponentially with the increase of the concrete thickness. The method used in this paper has an absolute error of less than 0.01 and a relative error of less than 8 % in detecting the mass loss rate of rebars at the initial stage of steel corrosion. When the concrete thickness is less than 20 mm, the relative error is lower than 4 %.


  • Back-stepping projective synchronization of fractional-order unified systems based on the lower triangular structure

    by Chenchen Peng, Anqing Yang, Xumei Lin, Ling Ren, Zihao Zhao

    The synchronization of fractional-order chaotic systems (FOCSs) plays an important role in modern control theory, the projective synchronization (PS) as a class of synchronization problems, also has huge applications and has attracted much attention. It is, however, shown in the obtained literature that the results on the PS of FOCSs either loss the rigorous theoretical demonstration or verify from the viewpoint of numerical simulations. How to derive a necessary and sufficient condition to guarantee the PS of complex FOCSs by a simple controller is still open. To this end, this article is concerned with the PS of fractional-order unified systems (FO-USs) that are important in FOCSs covering fractional-order Lorenz, Chen and Lü systems, where the controller is presented based on the lower triangular structure by use of the back-stepping technique. The necessary and sufficient criterion for the PS of FO-USs is proposed by solving an algebraic equation, and the controller for the PS of FO-USs is derived based on the lower triangular structure combined with back-stepping approach. Finally, the simulation results are reported to verify the correctness and efficiency of the obtained results.


  • A reinforced corrosion assessment method based on a new magnetic sensor and improved adaptive filtering

    by Xumei Lin, Guanghui Zhu, Shijie Yu, Peng Wang, Penggang Wang

    Corrosion assessment of reinforced concrete structures is the basis for subsequent corrosion repair and early warning of structural hazards. Aiming at the problem of accurately measuring the corrosion degree of reinforced concrete structures in coastal area, a new Magnetic Sensor (NMS) based on the theory of magnetic medium is designed in this paper. A functional model of the relationship between magnetic induction intensity change and reinforcement mass loss based on multi-point detection is constructed, and an improved adaptive filtering Least Mean Square (LMS) algorithm is proposed to realize damage identification and quantitative calculation of reinforcement corrosion comprehensively. Through numerical simulation experiments, the feasibility of designing the sensor is verified. The measured corrosion test of the sensor prototype shows that the calculated corrosion rates of different steel corrosion degrees are in good agreement with the actual values, with an average relative error of 1.64 %. The sensor provides a new method for corrosion monitoring of reinforced concrete structures.

  • Prediction of corrosion degree of reinforced concrete based on improved QPSO-neural network

    by Xumei Lin, Guanghui Zhu, Shijie Yu, Chuan Hu, Penggang Wang

    The corrosion assessment of reinforced concrete is an important problem. Most of the current detection methods are to perform corrosion evaluation on a single corrosion feature. The evaluation result depends on expert experience and existing prior knowledge information. In this paper, a neural network model based on improved quantum particle swarms optimisation (QPSO-NN) is proposed, which uses quantum particle swarms to optimise neural network (NN) weights and thresholds to predict the corrosion degree of reinforced concrete. The improved QPSO optimise the adjustment strategy of the Mean Best Position weight and improves the convergence speed of particles. The Simulated annealing algorithm (SA) is introduced in the iterative process of the particles to enhance the global optimisation capability of the particles. Under a variety of environmental parameters(pH, water/cement ratio, Chloride concentration), a variety of corrosion characteristic data are detected through the designed embedded acquisition system. The improved QPSO-NN corrosion model algorithm has better convergence speed by simulated analysis and has better accuracy of reinforcement corrosion assessment by experimental.


  • Open Access

    Neural‐network‐based robust terminal sliding‐mode control of quadrotor

    by Xumei Lin, Yulu Wang, Yunfei Liu

    A novel robust terminal sliding‐mode control (RTSMC) based on radial basis function (RBF) neural network is proposed firstly for controlling the attitude and position of the quadrotor, guaranteeing the system converged to stability point in a limited time. After establishing the nonlinear kinematics and dynamics models of the system, robust control is adopted in the RBF neural network terminal sliding‐mode controller such that the impact of external interference is reduced effectively. Resorting to the Lyapunov function, the asymptotically stable condition for the considered system is obtained, in which the convergence of the system is deduced in a finite time. Furthermore, simulation results are given to show the faster convergence speed and strong robustness for the considered system with RBF and RTSMC. Finally, the effectiveness and robustness of the developed control strategy is validated experimentally.