Suo Bin

Suo Bin


Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
16 March 2022 in IEEE Access
Verified
Journal Article
Open access
YYYing Yan
Ying Yan
SBSuo Bin
Suo Bin
ZLZiwei Li
Ziwei Li
Failure mode and effects analysis (FMEA) is a widely used, powerful tool to identify and assess potential failure modes in products and to make products more reliable. Due to the complexity of products and lack of knowledge, FMEA involves many uncertainties in practice. In previous studies, numerous modified FMEA methods based on fuzzy logic and Dempster-Shafer (D-S) evidence theory have been employed to address these uncertainties. These studies focus on how to handle uncertainties and to identify a more reliable prioritization of risk priority numbers (RPNs). However, studies have not sufficiently examined how many uncertainties are present in resulting RPNs. To better model and process various types of uncertainties in FMEA, two new area-based metrics are constructed in this paper. One is the interval area metric (IAM), which is used in RPN representation. The other is the dimensionless uncertainty metric (DUM), which is used to measure how many uncertainties there are in RPN. IAM is used to rank the risks in failure modes, and DUM is used to rank the uncertainties in failure modes. Then, an expert system is presented to qualitatively evaluate the DUM, which can help FMEA users intuitively judge whether further investigation should be performed to alleviate the epistemic uncertainties in each failure mode. Finally, a practical risk evaluation case regarding the grinding wheel system of a numerically controlled (NC) machine is provided to demonstrate the application and effectiveness of the proposed FMEA. The case study shows that the calculation programs of IAM and DUM do not require any assumptions or need to address conflict among experts. In addition, proposed method can not only give a more accurate rating of each failure mode, but also help designers intuitively see the uncertainty grade of each RPN, which is useful to help them understand FMEA results.

15 October 2025 in Measurement Science and Technology
Verified
Journal Article
Closed access
YQYang Qi
Yang Qi
LZLing Zhao
Ling Zhao
AGAo Gu
Ao Gu
BWBin Wu
Bin Wu
SBSuo Bin
Suo Bin
Rolling bearings, as one of the most vital components in rotating machinery, are frequently exposed to severe noise interference during operation, posing a significant challenge for accurate and rapid fault identification. To address this issue, this study proposes a novel fault diagnosis framework termed transformer–Kolmogorov Arnold networks (TKANs), which integrates the global feature extraction capability of the Transformer with the non-linear noise suppression advantage of the KAN linear layer. In the proposed TKAN model, raw vibration signals are first segmented into structured samples to fully preserve temporal dynamics. A four-layer Transformer module is then employed to extract high-dimensional representations from the input data, leveraging multi-head self-attention to enhance discriminative feature learning across different subspaces. To improve robustness under noisy conditions, a KAN linear layer with B-spline activation is incorporated in place of traditional linear mappings, effectively smoothing the feature space and attenuating noise-induced fluctuations. Extensive experiments are conducted on two widely used benchmark datasets—Case Western Reserve University and Xi’an Jiaotong University—to evaluate the performance of TKAN in both clean and noisy environments. Comparative results against five representative deep learning models (multilayer perceptron, convolutional neural network (CNN), KAN, LSTM–KAN, and CNN–KAN) demonstrate that TKAN achieves superior performance across multiple evaluation metrics (accuracy, precision, recall, and F1-score). Furthermore, under various levels of Gaussian, uniform, and impulse noise, TKAN consistently maintains high classification accuracy, underscoring its strong noise resilience and diagnostic robustness. This study provides a novel approach for fault diagnosis of bearings in noisy environments, offering significant practical and research value.

2 April 2014 in International Journal of Nonlinear Sciences and Numerical Simulation
Verified
Journal Article
Closed access
SBSuo Bin
Suo Bin
CZChao Zeng
Chao Zeng
YCYong-Sheng Cheng
Yong-Sheng Cheng
JLJun Li
Jun Li
For the reliability evaluation of a complex system, the aleatory uncertainty and epistemic uncertainty always exist at the same time because of the inadequate experimental data and the incomplete information. To reduce the computational cost of the reliability evaluation, an effective method based on evidence theory was developed. In this method, the mean of the belief measure and the plausibility measure was taken as the approximation of the system reliability. The discretization methods for uncertain parameters were discussed in the cases of only aleatory uncertainty was involved in the system, and mixed aleatory and epistemic uncertainties were involved, respectively. Algorithms of the evaluations of belief and plausibility functions were proposed for monotonic and non-monotonic systems. Four numerical examples under different conditions were studied. Simulation results showed that, the proposed method was much more effective than Monte Carlo method without sacrificing the accuracy of the resulting reliability. It was a general method suitable for various systems with different types of information.

19 March 2025 in Materials
Verified
Journal Article
Open access
PDF file
WZWei Zheng
Wei Zheng
SBSuo Bin
Suo Bin
CZChunping Zhou
Chunping Zhou
YLYueqi Li
Yueqi Li
Feng ZhangFeng Zhang
Feng Zhang
This paper presents research into the failure modes and mechanisms of active electromagnetic functional structures. An active electromagnetic functional structure is usually achieved by loading active devices on a frequency-selective surface and designing an active frequency-selective surface (AFSS) structure with adjustable transmission and cutoff characteristics through feeding. This has shown significant advantages in electromagnetic countermeasures for fighter jets and has become one of the key technologies for the future development of such equipment in the context of integration and intelligence. The active electromagnetic functional structure is different from the traditional composite material structure because it includes active functional devices to realize the electromagnetic function but also includes a structural component to bear the complex environment. The failure modes are also more complex, including both structural failure and functional degradation or even failure caused by device failure, and the coupling relationship between different failure modes is complex. However, the failure modes of active electromagnetic functional structures in complex environments have received little research attention, and the corresponding fault modes and mechanisms are poorly understood, thereby limiting their practical applications. Thus, in the current study, environmental stress analysis is performed to verify the main failure modes and mechanisms of active electromagnetic functional structures through fault excitation tests. Importantly, the obtained results clarify the stress sensitivity of the structure and provide effective support for its engineering applications.

24 August 2026 in Engineering Analysis with Boundary Elements
Verified
Journal Article
Closed access
LZLing Zhao
Ling Zhao
HZHui Zhang
Hui Zhang
YQYang Qi
Yang Qi
SBSuo Bin
Suo Bin

14 April 2022 in Mathematical Problems in Engineering
Verified
Journal Article
Open access
PDF file
YYYing Yan
Ying Yan
SBSuo Bin
Suo Bin
From the perspective of D-S evidence theory and area measurement, a risk-based comprehensive decision-making method that considers both the expected utility and the uncertainty of the scheme is proposed under the interval uncertainty environment of attribute values. The upper and lower bounds of the synthetic probability distribution of attributes values in different natural states are constructed based on the belief measure and plausibility measure. Based on the area measurement, a method for calculating the expected utility of each scheme is proposed. To reflect the influence of the uncertainty in the evaluation value of each scheme attribute on the final decision result, two indexes are defined: the evaluation uncertainty of attributes (EUA) and the uncertainty of the expected utility of scheme (UEU). Finally, considering the expected value of the expected utility and its uncertainty, three decision methods, namely, risk-neutral, risk-averse, and risk-preference, are constructed. An example is considered to show that the proposed method is effective and practical, and the uncertainty of the expected utility has a significant impact on the result of risky decisions. The new method can solve the problems of existing methods that overlook the impact of epistemic uncertainty on the decision-making process.