ZZZhili Zhang

Zhili Zhang

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Zhili Zhang has a total of 21 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Obstacle-Resilient Topology Optimization for Industrial IoT via DRL-Driven Potential Field

    by Yi Ding, Zhili Zhang, Huayin Zhao, Liudi Wang, Peng Xue, Wentao Cui, Enqing Dong

    In industrial Internet of Things (IIoT), the nonline-of-sightconstraints caused by ubiquitous obstacles pose severe challenges to wireless communication. Existing clustering strategies suffer from problems of relying on idealized geometric prior information, leading to obstacle-ignorant structures, and easily causing topological discontinuities. These problems result in a severe mismatch between network topology and convergent traffic, thereby compromising both network connectivity and energy balance. To address these problems, a topology-adaptive clustering (TAC) framework based on deep reinforcement learning (DRL)-driven potential field is proposed. TAC reformulates the clustering problem as a generalized Voronoi tessellation constrained by physical potential fields. By utilizing a DRL agent to dynamically tune potential parameters within a nonconvex solution space, the framework achieves a posteriori topological awareness and adaptive obstacle avoidance. Furthermore, a generic mechanism is introduced to derive a theoretical closed-form solution for traffic load via graph theory. This graph-theoretic mechanism constructs a deterministic reward function decoupled from specific routing protocols, ensuring precise alignment between the clustering topology and intrinsic traffic demands to guarantee energy balance across various downstream protocols. Simulation results demonstrate that TAC adaptively generates obstacle-resilient topological structures, outperforming existing baseline protocols in terms of network connectivity and lifespan, offering a general, obstacle-resilient, and energy-efficient topology optimization framework for IIoTs.

  • An Energy Efficiency Optimization Scheme for Uniform Line-Distributed Wireless Geophone Networks

    by Yi Ding, Zhili Zhang, Huayin Zhao, Jinyuan He, Liudi Wang, Jin Kang, Peng Xue, Wentao Cui, Enqing Dong

    This article focuses on energy efficiency (EE) optimization for large-scale uniform line-distributed wireless geophone networks (WGNs). To address the issues of unbalanced subnetwork load and excessive energy consumption in traditional WGNs, we propose an EE optimization scheme—Markov-chain-based clustering and farthest vector forwarding (MCBC–FVF). For intracluster energy optimization, a state transition probability model based on Markov chain (MC) is constructed for cluster head (CH) election. An energy-aware objective function with a spatial bias term is designed to reduce and balance energy consumption. For intercluster energy optimization, a farthest vector forwarding (FVF) mechanism is introduced to mitigate communication failure and low packet delivery ratio (PDR) caused by excessive distances between CHs. It also helps reduce redundant traffic and suppresses path inflation. Compared with MH-LEACH, LEACH-C, and MMRP, simulation results based on IEEE 802.15.4 demonstrate that the proposed MCBC–FVF scheme improves the first node death (FND) time by 38.68%, 22.62%, and 17.20%, respectively, while reducing intercluster average energy consumption by 32.73%, 18.41%, and 43.38%, respectively. These results indicate that MCBC–FVF not only significantly prolongs network lifetime but also provides a novel integration of probabilistic modeling and topologyaware forwarding, offering a practical and effective solution for energy-constrained WGNs.

  • A Low Frequency Response Extension Method with Low Noise Floor, High Sensitivity and High Temperature Stability for Geophone

    by Zhili Zhang, Guoqing Zhang, Peng Xue, Lingquan Lv, Yi Ding, Wentao Cui, Enqing Dong

    In the fields of microseismic monitoring and deep earth exploration, low-frequency signal detection faces problems such as weak vibration signals, large ambient temperature changes, and complex ambient noise. A novel comprehensive low-frequency response extension method for a moving coil geophone is proposed to implement a high-sensitivity low-frequency vibration sensor with low noise floor for measuring seismic wave velocity. Based on exploring the principle of increasing the damping ratio of geophone to expand the linear variation range of acceleration with frequency, a model in which both damping and gain vary with temperature is established by introducing the temperature-dependent coil resistance parameter. Developed on the typical negative impedance conversion method, a damping adjustment circuit with temperature compensation, gain adjustment and differential noise suppression is designed comprehensively, thereby a new scheme of low-frequency response extension based on the force balance principle is proposed. To further suppress the inherent 1/f noise of the operational amplifier in the low frequency band, an acceleration-velocity conversion scheme based on a dual operational amplifier narrowband filter is designed to flatten the velocity-frequency curve while suppressing low-frequency noise. Through standard vibration table and field comparative tests, the proposed comprehensive low-frequency extension method not only extends the natural frequency of the conventional moving coil detector from 5Hz to 0.25Hz, but also has low noise, high gain (sensitivity of 420V/(m·s⁻¹)) and high-temperature stability.


  • A Novel Inverse Filtering-Negative Feedback Low-Frequency Extension Method for Seismic Geophone

    by Zhili Zhang, Yi Ding, Peng Xue, Huayin Zhao, Guoqing Zhang, Wentao Cui, Enqing Dong

    To address the challenge of low-frequency seismic signal measurement in ultra-deep oil and gas exploration, an inverse filtering-negative feedback low-frequency extension (IF-NF-LFE) method is proposed based on the negative feedback closed-loop (NFCL) model. An adjustable negative feedback circuit (ANFC) is firstly designed by incorporating tunable resistors. For the NFCL model comprising the ANFC and geophone electromagnetic dynamics model, the system transfer function is derived through the force-balance principle, followed by analysis of system consistency performance and frequency response characteristics. Theoretically cascading an ideal inverse filter can ensure voltage-velocity linearity but exhibit complex circuit and notable low-frequency inherent noise. Consequently, an approximate transfer function is constructed via the least-squares algorithm to asymptotically approach the ideal inverse filter, and the stability of the IF-NF-LFE system is analyzed. Finally, we implement the extension circuit, and propose a dual-frequency calibration strategy to optimize the consistency of sensitivity and cutoff frequency. Using the proposed method, a low-frequency extension circuit is designed for a 5 Hz moving-coil geophone, and its performance (sensitivity of 200 V/(m·s⁻¹) ±5%, corner frequency extended to 0.2 Hz ±5%, consistency and noise floor) is validated through standard vibration table testing. Furthermore, a small batch of compact and rugged integrated low-frequency geophones are manufactured, with multiple comparative verification tests performed under field conditions.

  • PKA2-Net: Prior Knowledge-Based Active Attention Network for Accurate Pneumonia Diagnosis on Chest X-Ray Images

    by Yu Fu, Peng Xue, Zhili Zhang, Enqing Dong

    To accurately diagnose pneumonia patients on a limited annotated chest X-ray image dataset, a prior knowledge-based active attention network (PKA 2 -Net) was constructed. The PKA 2 -Net uses improved ResNet as the backbone network and consists of residual blocks, novel subject enhancement and background suppression (SEBS) blocks and candidate template generators, where template generators are designed to generate candidate templates for characterizing the importance of different spatial locations in feature maps. The core of PKA 2 -Net is SEBS block, which is proposed based on the prior knowledge that highlighting distinctive features and suppressing irrelevant features can improve the recognition effect. The purpose of SEBS block is to generate active attention features without any high-level features and enhance the ability of the model to localize lung lesions. In SEBS block, first, a series of candidate templates T with different spatial energy distributions are generated and the controllability of the energy distribution in T enables active attention features to maintain the continuity and integrity of the feature space distributions. Second, Top- n templates are selected from T according to certain learning rules, which are then operated by a convolution layer for generating supervision information that can guide the inputs of SEBS block to form active attention features. We evaluated the PKA 2 -Net on the binary classification problem of identifying pneumonia and healthy controls on a dataset containing 5856 chest X-ray images (ChestXRay2017), the results showed that our method can achieve 97.63% accuracy and 0.9872 sensitivity.


  • Semi-Supervised Adversarial Learning for Improving the Diagnosis of Pulmonary Nodules

    by Yu Fu, Peng Xue, Taohui Xiao, Zhili Zhang, Youren Zhang, Enqing Dong

    Achieving the pathological type diagnosis of pulmonary nodules on chest CT is a critical step in the early detection of lung cancer and treatment of patients. Based on a small and unbalanced self-constructed dataset, we achieved intelligent diagnosis of five pathological types including adenocarcinoma, squamous cell carcinoma, small cell carcinoma, inflammatory and other benign diseases for the first time. In order to reduce the dependence of deep convolutional neural network (DCNN) on a large amount of training data, a reverse adversarial classification network (RACN) was proposed based on semi-supervised learning, which consists of a reverse generative adversarial network (RGAN) for unsupervised regression and a supervised classification network (CN). In RGAN, five specific normal distributions P with different means and variances were assigned to represent the five pathological types, and then a special regression task was designed by mapping pulmonary nodules to the random sampling Z of P. The input of generator in RGAN is set to 3D nodule volume data, the inputs of discriminator are set to Z and the output of generator. The regression task enables RGAN to extract specific features, which will be deeply integrate into CN to improve the classification performance. Experiments showed that the average sensitivity of RACN in detecting malignant nodules was 0.6525, where the sensitivity of adenocarcinoma, small cell carcinoma and squamous cell carcinoma was 0.8426, 0.5604 and 0.5543. Besides, the RACN can achieve 93.21% accuracy for diagnosing malignant nodules on the public LIDC-IDRI dataset, obtaining the state-of-the-art results.