Haiwei Mu

Reservoir evaluation is essential in oil and gas exploration, influencing development decisions. Traditional classification methods are often limited by small sample sizes and low accuracy, restricting their effectiveness. To address this, we propose an intelligent classification method, GA-XGBoost, which integrates Genetic Algorithm (GA) optimization with Extreme Gradient Boosting (XGBoost) to enhance the classification accuracy in small-sample scenarios. The lithological, physical, and lithofacies characteristics of tight sandstone reservoirs are analyzed, and key evaluation parameters—including the mineral composition, porosity, permeability, oil saturation, and logging data (GR, SP, CAL, DEN, AC, LLS)—are selected. After data normalization, the GA-XGBoost model is developed and compared with SVM, XGBoost, and AdaBoost models. The experimental results demonstrate that GA-XGBoost achieves an 88.8% classification precision, outperforming traditional algorithms in both efficiency and accuracy. This study advances experiments on and the standardization of intelligent reservoir evaluations, providing a more reliable classification approach for tight sandstone reservoirs. Additionally, it contributes to the integration of geological exploration and computational intelligence, offering new insights into the application of machine learning in geosciences.

Manipulating the light scattering direction and enhancing directivity are important research areas in integrated nanophotonic devices. Herein, a novel, to the best of our knowledge, nanoantenna composed of hollow silicon nanoblocks is designed to allow directional emission manipulation. In this device, forward scattering is enhanced and backward scattering is restrained substantially in the visible region. Owing to electric dipole resonance and magnetic dipole resonance in this nanoantenna, Kerker’s type conditions are satisfied, and the directionality of forward scattering G F B reaches 44.6 dB, indicating good characteristics in manipulating the light scattering direction.

A highly sensitive refractive index (RI) sensor based on the surface plasmon resonance (SPR) phenomenon is proposed and demonstrated. The sensor is composed of a quasi D-shape ten-fold photonic quasi-crystal fiber (PQF) and indium tin oxide (ITO) film. The quasi D-shape and eccentric core structure strengthen coupling between the core mode and surface plasmon polariton (SPP) mode to enhance the sensing sensitivity, while the mechanical strength of the PQF is preserved. The properties of the sensor are determined numerically and optimized by the finite element method. In the RI range of 1.40–1.44, the maximum wavelength sensitivity and average wavelength sensitivity are 62,000 nm/RIU and 26,500 nm/RIU, respectively. Meanwhile, the loss spectra show smaller full-width at half-maximum and maximum FOM of 542.5 RIU⁻¹. This novel sensor which boasts ultra-high-sensitivity and excellent FOM has large potential in the determination of liquid RIs.

An optical nanoantenna consisting of a Au-dielectric core-shell nanocube dimer with switchable directionality was designed and described. Our theoretical model and numerical simulation showed that switching between forward and backward directions can be achieved with balanced gain and loss, using a single element by changing the coefficient κ in the core, which can be defined by the relative phase of the polarizability. The optical response indicated a remarkable dependence on the coefficient κ in the core as well as frequency. The location of the electric field enhancement was specified by the different coefficient κ and, furthermore, the chained optical nanoantenna and coupled electric dipole emitted to the optical nanoantenna played significant roles in unidirectional scattering. This simple method to calculate the feasibility of unidirectional and switchable scattering provides an effective strategy to explore the functionalities of nanophotonic devices.