Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


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


Latest Publications

  • Open Access

    An Adaptive Fuzzy Recurrent Stochastic Configuration Network for Food Quality Prediction with Limited Samples

    by Shiqi Hu, Ningyi Sun, Yingying Chen, Cunsong Wang, Le Wang

    Accurate food quality prediction remains challenging when labeled samples are limited, and the relationships between measurable characteristics and target quality attributes are nonlinear and heterogeneous. To address this problem, this study proposes an adaptive fuzzy recurrent stochastic configuration network that integrates adaptive fuzzy rule-number selection, rule-coupled recurrent subreservoirs, global-residual-guided stochastic configuration, and regularized analytical output-weight learning. The proposed framework was evaluated on the Mackey–Glass benchmark and two practical food quality prediction tasks involving litchi soluble sugar content and wine quality. On the Mackey–Glass benchmark, it achieved a root mean square error of 4.47×10−4, representing a 37.8% reduction relative to the conventional fuzzy recurrent stochastic configuration network. For litchi sugar prediction, it achieved a root mean square error of 0.7200, a mean absolute error of 0.5660, a mean absolute percentage error of 2.632%, and a coefficient of determination of 0.9352, with the root mean square error reduced by approximately 10.2% relative to the conventional fuzzy recurrent stochastic configuration network. For white and red wine, the corresponding root mean square errors were 0.1829 and 0.1655, with coefficients of determination of 0.9513 and 0.9677, respectively. These results indicate that the proposed framework provides favorable predictive performance under the evaluated limited-sample settings and predefined data partitions, supporting its applicability to nonlinear food quality prediction with limited labeled data.


  • Short-term traffic flow prediction method based on ATT-CNN-BiLSTM hybrid network

    by Shiqi Hu, Ningyi Sun, Congguo Ma, Hongbiao Zhou, Wen Shi, Le Wang

    Short-term traffic flow forecasting is a fundamental component of intelligent transportation systems (ITS), enabling proactive traffic management and congestion mitigation. However, traditional statistical methods often struggle to capture the nonlinear and time-varying characteristics of traffic flow, while single deep learning models may fail to learn complementary temporal patterns across multiple scales. To address these limitations, this paper proposes a hybrid forecasting framework, attention mechanism-convolutional neural network-bidirectional long short-term memory (ATT-CNN-BiLSTM), which integrates one-dimensional convolutional neural networks for local temporal feature extraction, a BiLSTM network for modeling long-range bidirectional dependencies, and a squeeze-and-excitation attention mechanism for adaptive feature recalibration. Extensive experiments conducted on real-world traffic datasets demonstrate that the proposed model achieves stable and competitive prediction performance across datasets collected from different years. In addition, ablation studies verify the effectiveness of the attention mechanism and network configuration in improving overall forecasting accuracy. The proposed framework therefore provides an effective and practical solution for short-term traffic flow forecasting in ITS applications.


  • Kernel-space oversampling and adaptive ensemble learning for robust imbalanced data classification

    by Le Wang, Ningyi Sun, Shiqi Hu, Hongbiao Zhou, Han Chen

    This paper proposes a robust fuzzy kernel-based possibilistic clustering-means adaptive weighted random forest (FKAW-RF) framework for imbalanced data classification that integrates kernel-space oversampling with adaptive ensemble learning. The method first maps the original dataset into a high-dimensional kernel space, where a fuzzy kernel-based possibilistic clustering-means (FKPCM) algorithm is employed to identify and remove boundary samples, thereby reducing the influence of noise and class overlap. Within this kernel space, a kernel-conventional synthetic minority oversampling technique (KSMOTE) strategy is applied to perform high-quality oversampling of minority class instances, and the generated samples are projected back into the input space through preimage estimation. To further enhance classification robustness, an adaptive weighted random forest (AW-RF) is constructed as the ensemble classifier, in which underperforming or redundant decision trees are pruned based on feature correlation analysis, and the remaining trees are adaptively weighted using dual penalty factors reflecting both local and global error rates. Extensive experiments on UCI, KEEL, and Tennessee-eastman (TE) benchmark datasets demonstrate that the proposed kernel-space oversampling and adaptive ensemble learning approach consistently outperforms conventional synthetic minority oversampling technique (SMOTE) variants and state-of-the-art classifiers across multiple evaluation metrics, including accuracy, precision, recall, and G-mean, confirming its effectiveness and robustness for challenging imbalanced classification tasks.


  • Open Access

    Adaptive fault‐tolerant attitude tracking control for hypersonic vehicle with unknown inertial matrix and states constraints

    by Le Wang, Yizhen Meng, Shiqi Hu, Zhiyu Peng, Wen Shi

    This paper proposes an adaptive fault‐tolerant control (FTC) method for hypersonic vehicle (HSV) with unexpected centroid shift, actuator fault, time‐varying full state constraints, and input saturation. The occurrence of unexpected centroid shift has three main effects on the HSV system, which are system uncertainties, eccentric moments, and variation of input matrix. In order to ensure the time‐varying state constraints, a novel attitude state constraint control strategy, to keep the safe flight of HSV, is technically proposed by a time‐varying state constraint function (TVSCF). A unified controller is designed to handle the time‐varying state constraints according to the proposed TVSCF. Then, the constrained HSV system can be transformed into a novel free‐constrained system based on the TVSCF. For the variation of system input matrix, input saturation and actuator fault, a special Nussbaum‐type function is designed to compensate for those time‐varying nonlinear terms. Additionally, the auxiliary systems is designed to compensate the constraint of system control inputs. Then, it is proved that the proposed control scheme can guarantee the boundedness of all closed‐loop signals based on the Lyapunov stability theory. At last, the simulation results are provided to demonstrate the effectiveness of the proposed fault‐tolerant control scheme.


  • Open Access

    Quality Control of Jinhua Ham from the Influence between Proteases Activities and Processing Parameters: A Review

    by Shiqi Hu, Xinglian Xu, Wangang Zhang, Chunbao Li, Guanghong Zhou

    Endogenous proteases are significant for Jinhua ham quality. Protein degradation affects the chemical traits, texture and the formation of flavor substances. Protease activities are affected by different process parameters, such as processing temperature, maturation time, salt content and the drying rate. They affect ham quality, which can be controlled by process parameters. The influences of key factors on Jinhua ham quality are briefly summarized, which can provide a theoretical basis for the selection of specific parameters in dry-cured ham processing. Furthermore, some suggestions are proposed for correcting and improving the flavor and textural defects of ham, yet the effectiveness depends on the operating conditions. The determination of enzyme activity is not real-time and unsupervised at the moment. Future research will focus on the determination of the actual endogenous protease activity and the quantitative relationship between the enzyme activity and main processing parameters.


  • Open Access

    Age Discrimination of Chinese Baijiu Based on Midinfrared Spectroscopy and Chemometrics

    by Shiqi Hu, Le Wang

    Baijiu is a traditional and popular Chinese liquor which is affected by the storage time. The longer the storage time of Baijiu is, the better its quality is. In this paper, the raw and mellow Baijiu samples from different storage time are discriminated accurately throughout midinfrared (MIR) spectroscopy and chemometrics. Firstly, changing regularities of the substances in Chinese Baijiu are discussed by gas chromatography-mass spectrometry (GC-MS) during the aging process. Then, infrared spectrums of Baijiu samples are processed by smoothing, multivariate baseline correction, and the first and second derivative processing, but no significant variation can be observed. Next, the spectral date pretreatment methods are constructively introduced, and principal component analysis (PCA) and discriminant analysis (DA) are developed for data analyses. The results show that the accuracy rates of samples by the DA method in calibration and validation sets are 91.7% and 100%, respectively. Consequently, an identification model based on support vector machine (SVM) and PCA is established combined with the grid search strategy and cross-validation methods to discriminate the age of Chinese Baijiu validly, where 100% classification accuracy rate is obtained in both training and test sets.