HLHaibo Liang

Haibo Liang

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Haibo Liang has a total of 149 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Study on an online calibration method for rheological parameter measurement of waterbased drilling fluids based on dual-tube differential pressure measurement

    by Hai Yang, Junjie Chang, Dunyu Jiang, Jianlong Wang, Haibo Liang

    In the double straight-pipe differential-pressure measurement system, diaphragm-pump pulsation and mechanical vibration contaminate the differential-pressure (DP) signals, thereby limiting the accuracy of on-line measurement of drilling-fluid rheology. To address this problem, an online calibration method integrating improved slope-based method time-varying filter-based empirical mode decomposition (ISBM-TVF-EMD) with maximum-likelihood estimation based on the cloud-adaptive particle swarm optimization zero-mean K -distribution is proposed. The ISBM-TVF-EMD method suppresses endpoint effects and mode mixing and extracts reliable differential-pressure information from the signals acquired from the two measuring pipes. The corrected pressure-drop and flow-rate data are then used to calculate the wall shear stress, wall shear rate, and rheological parameters. Experiments were conducted using ten water-based drilling fluids with different viscosities, with measurements obtained using a Fann rotational viscometer serving as reference values. Before calibration, the maximum relative errors of the measured rheological parameters ranged from 139% to 242%. After calibration, the maximum relative errors of apparent viscosity, plastic viscosity, and yield point were reduced to 2.58%, 2.89%, and 2.41%, respectively. The results demonstrate that the proposed method significantly improves the accuracy of rheological-parameter measurement based on double straight-pipe differential-pressure measurement and provides a feasible approach for continuous and automated on-line measurement of drilling-fluid rheology.

  • Open Access

    The Hybrid DBSCAN-Transformer Framework for High-Precision Phase Fraction Measurement in Low-Energy Gamma Flowmeter

    by Yibo Huang, Mingyang Liu, Lijing Fan, Yulin Liang, Qingjing Lin, Shihan Zhang, Haibo Liang, Lianzheng Zhang

    Multiphase flow metering is widely employed in the oil and gas industry, particularly for measuring gas-liquid-solid multiphase flow at drilling outlets. Low-energy gamma flowmeters offer relatively high metering accuracy, with phase fraction errors for gas, liquid, and solid typically within ±10%. However, in practical applications, fluid viscosity often causes substances to adhere to the photon detector, leading to measurement deviations that can reach 18% or more. To overcome this limitation, this paper proposes a hybrid Density-Based Spatial Clustering of Applications with Noise (DBSCAN)-Transformer regression framework, referred to as D-Transformer. DBSCAN removes isolated abnormal detector responses before overlapping time-series windows are generated, while the Transformer captures temporal dependencies among operating variables, raw phase-fraction readings, and multi-energy photon counts. Under experiment-wise five-fold evaluation, D-Transformer obtains R2 values of 0.982, 0.985, and 0.981 and RMSE values of 0.0134, 0.0122, and 0.0138 for the gas, liquid, and solid phase fractions, respectively. Component ablations and baseline comparisons show that the complete framework outperforms the no-ResNet, no-DBSCAN, CNN-GRU-Attention, CNN-LSTM, ridge-regression, and uncorrected-flowmeter alternatives.

  • Research on Precision Improvement Method for Oil-in-Water Concentration Based on Conductive Efficiency of Oil-Water Biphasic Channels

    by Weijie Zhang, Haibo Liang, Lei Zhao, Hai Yang

    The treatment and discharge of petroleum wastewater are critical for ensuring ecological environmental safety and the secure reinjection of produced water, where oil content directly affects the porous media characteristics of geological formations and the operational lifespan of water treatment equipment. However, conventional detection methods suffer from limitations including poor timeliness, complicated operational procedures, and susceptibility to secondary pollution. Consequently, low-cost, rapid-analysis, and user-friendly oil-in-water sensors have been increasingly deployed for in-situ measurement of water quality parameters. Nevertheless, under complex water quality conditions, temperature fluctuations not only alter the refractive index of oil-water two-phase mixtures but also directly influence the fluorescence quantum yield of optical sensors, thereby introducing measurement errors. To address this challenge, this paper proposes a novel method for enhancing oil content measurement accuracy based on oil-water two-phase conductivity efficiency. This method integrates temperature compensation based on Support Vector Regression (SVR) and corrects the molar refraction through the Maxwell model and Lorentz-Lorenz equation to achieve physical constraints for oil content determination. Experimental results demonstrate that under tests with various oil substances, the average error of oil content measurement is reduced from approximately 29.9% to about 2.45%, indicating that the proposed method effectively eliminates the interference of water temperature variations.

  • Research on the identification of formation water intrusion in oil-based drilling fluids and quantitative analysis methods

    by Hai Yang, Zhenhuai Wang, Zhi Qiu, Haibo Liang, Ren Wang, Jianlong Wang, Jialing Zou

    During drilling operations, formation water intrusion is a common and complex downhole condition, and failure to identify it promptly and accurately may lead to wellbore instability or even blowouts. Existing wellhead parameter-based monitoring methods are easily affected by operating conditions and often suffer from delayed parameter response. To address these limitations, this paper proposes an integrated multi-source data fusion framework for the identification, quantitative analysis, and operating-condition compensation of formation water intrusion in oil-based drilling fluids. First, the correlation between characteristic parameters and intrusion status is analyzed to support model construction. Then, a logistic regression-based multi-source data fusion model is established to achieve effective intrusion identification. In addition, by investigating the relationships among moisture content, outlet flow, and water influx, and introducing weighted compensation for conductivity, density, and other parameters, a quantitative water influx model is developed. In data simulation, the mean absolute error of the proposed model is 127.71 l, indicating good predictive accuracy. Furthermore, a compensation model for non-water-intrusion conditions is constructed, and the L-SHADE-XGBoost algorithm is used to classify operational scenarios and reduce interference from non-intrusion factors. The classification accuracy, precision, and recall all exceed 90%, and the compensated data agree more closely with actual measurements. Field verification further demonstrates that the proposed system can stably perform real-time monitoring and quantitative analysis of formation water intrusion. These results confirm its applicability and reliability and provide technical support for safe drilling operations in oil and gas wells.

  • Periodic Modulation Convolution Parse Self-Learning for fault feature enhancement in rolling bearing

    by Aiguo Li, Haibo Liang

    Periodic modulation transients caused by localized defects in rolling bearings are key indicators for condition monitoring and fault diagnosis. However, under complex operating conditions and strong noise interference, vibration signals often become highly nonstationary, with weakened modulation characteristics and fault features easily masked by broadband noise. To address these challenges, a Periodic Modulation Convolution Sparse Self-Learning (PM-CSSL) method is proposed for adaptive enhancement and accurate extraction of fault features. First, symplectic geometry mode decomposition (SGMD) is employed to denoise the signal while preserving periodic impulse information. Then, a parameterized atomic prototype based on the Laplace wavelet is constructed, and a candidate atom dictionary is generated using Latin hypercube sampling (LHS). The optimal atom is adaptively selected via a maximum correlation coefficient criterion and further periodized to form physically interpretable modulation atoms. Next, frequency-domain mapping and an energy concentration criterion are used to identify the optimal modulation atom. A convolutional sparse representation model is then established in the frequency domain to enhance modulation features and reconstruct the signal. Finally, envelope demodulation is applied, and the extracted frequencies are compared with theoretical fault frequencies for accurate fault identification. Experimental results demonstrate that the proposed method effectively enhances fault-related features and improves diagnostic accuracy under noisy conditions.

  • A Cascaded Learning Framework for Data-Driven Information Extraction from Water Quality Sensors

    by Weijie Zhang, Haibo Liang, Lin Yang, Hai Yang

    Traditional water quality detection is time-consuming and labor-intensive, limiting real-time monitoring. Optical water quality sensors are widely used in marine, aquaculture, oil extraction, and domestic water applications. However, in complex water environments such as oilfield produced water, their high sensitivity and low signal-to-noise ratio generate large defective datasets that fail to meet industrial standards. To address this, we propose an adaptive sliding-window cascaded learning framework based on global and local correlations. The framework dynamically adjusts window lengths to capture temporal dependencies, employs adaptive K-means clustering for multimodal distributions, and incorporates a recursive feature elimination–based attention mechanism to prioritize core variables. A data-driven module further constrains the optimized dataset, enabling dynamic updates and multi-category fusion to correct suspended solids measurements. Experimental results show that compared with linear fitting using raw datasets, the proposed TC-DD framework reduces the average coefficient of dispersion from 0.354 to 0.00214, increases R from 0.22 to 0.97, narrows the average confidence interval width by 74.0%, and improves suspended solids measurement accuracy by 12.6% under complex water conditions. This method extracts useful information from measured data more efficiently, providing reliable support for suspended solids concentration evaluation and operational guidance in complex water environments.