8 May 2023 in IEEE Transactions on Instrumentation and Measurement
Coal mine location-based services are vital fundamental facilities for achieving intelligence. Ultra-wide band (UWB) is widely used for underground mine positioning because of its powerful anti-interference capability. However, UWB suffers from range errors caused by first-arrival time (FAT) delay, non-line of sight (NLOS) as well as trilateral positioning deformation in narrow and long tunnels. Inertial Measurement Unit (IMU) sensor can effective solve the interference issue by tracing moving targets without external data, while it also needs precise initial alignment and may cause accumulative error. To efficiently increase positioning accuracy in underground tunnels, Motion Inertia Estimated Adaptive Kalman Filtering (MIAKF) is proposed combining the advantages of IMU and UWB. Firstly, heading memory is designed to store target motion historical state and estimate its forth coming state. Secondly, an dynamically updated inertial confidence parameter pair is introduced as the error weights of two sensors to measure the inertia motion deviation. Then, by converting the vector differences into angles and parameterizing these angles, the bias of each sensor can be measured. Finally, Kalman filtering error matrix is automatically adjusted by parameterized angles to obtain accurate position. Extensive experiments in factory and real scenarios show MIAKF can achieve 19% higher accuracy than benchmark methods.
12 July 2023 in Journal of Sound and Vibration
20 February 2025 in Expert Systems with Applications
10 February 2020 in IEEE Access
In this paper, a modified double-rope winding hoisting system (MDWHS) is proposed for ultra-deep mines. Nevertheless, the MDWHS has wire rope tension imbalance issues due to lifting inconsistency of two wire ropes. A combined control algorithm is designed for wire rope tension active control considering the nonlinear uncertainties subjected to the MDWHS, where a low gain state observer (LGSO) and a robust nonlinear adaptive controller are employed via backstepping theory. Firstly, to investigate the wire rope tension active control method, the dynamics model of the MDWHS is established. Then, the LGSO is employed to estimate unmeasured system state variables and to optimize perturbed system state variables. Finally, a robust nonlinear control method is employed to acquire prospective control input. The combined control algorithm can ensure the desired transient and final tracking accuracy, which is essential for tracking control of wire rope tension active adjustment. The stability of the overall closed-loop system with the presented control algorithm can be verified referring to the Lyapunov theory. Experimental results conducted on the established double-rope winding hoisting simulation experimental system demonstrate that the proposed active control method exhibit more advantageous performance on wire rope tension balance control compared with the conventional Proportion Integration (PI) and adaptive backstepping controller (ABC).
2 June 2025 in IEEE Transactions on Instrumentation and Measurement
In industrial applications, the non-stationary nature of long time series data from bearing operations poses a significant challenge due to data drift, influenced by varying operating conditions and environments. To tackle this issue, we propose a novel fault diagnosis model leveraging contrastive learning. This approach utilizes domain-differentiated contrastive feature learning to construct positive and negative sample pairs, fully capturing the commonalities within the same fault type and the differences between different fault types, thereby enhancing the model’s robustness against interference. Moreover, the model employs a patch-based transformer to capture dependencies within local subsequences, reducing computational complexity while maintaining the ability to abstract comprehensive signal representations. Additionally, the integration of multi-head fourier attention allows simultaneous analysis of time-domain and frequency-domain characteristics, enriching the feature extraction process. Our method is validated through comparative, parameter analysis, and ablation studies on datasets, demonstrating its effectiveness and potential for improving fault diagnosis accuracy in bearing systems, thereby reducing downtime and enhancing operational safety.
27 April 2026 in IEEE Transactions on Instrumentation and Measurement
To address the challenges of bearing fault diagnosis under variable operating conditions and limited labeled data in the target domain, we propose a domain adaptation model based on momentum contrastive learning. The model first uses momentum contrastive learning to extract robust contextual representations, and then applies nearest-neighbor contrastive learning to align these representations between source and target domains. To further ensure spatial structure consistency across domains, we incorporate maximum mean discrepancy and domain-adversarial learning, effectively mitigating the impact of diverse working conditions and data scarcity. In addition, we introduce a balanced dynamic adaptive threshold to improve the model’s resistance to noise, and employ wavelet-based weight initialization to enhance its extrapolation capability. Extensive experiments, including comparative studies, parameter analysis, and ablation studies under various noise levels, are conducted on datasets to validate our model’s effectiveness. This work significantly advances the practical application of bearing fault diagnosis in industrial settings by enhancing noise resistance and robustness.