Search Results (505)

Oil and gas development is characterized by high technical complexity, strong interdisciplinarity, long investment cycles, and significant uncertainty. To meet the need for quick evaluation of overseas oilfield projects with limited data and experience, this study develops an analogy indicator system and tests multiple machine-learning algorithms on two analogy tasks to identify the optimal method. Using an initial set of basic indicators and a database of 1436 oilfield samples, a combined subjective–objective weighting strategy that integrates statistical methods with expert judgment is used to select, classify, and assign weights to the indicators. This process results in 26 key indicators for practical analogy analysis. Single-indicator and whole-asset analogy experiments are then performed with five standard machine-learning algorithms—support vector machine (SVM), random forest (RF), backpropagation neural network (BP), k-nearest neighbor (KNN), and decision tree (DT). Results show that SVM achieves classification accuracies of 86% and 95% in medium-high permeability sandstone oilfields, respectively, greatly surpassing other methods. These results demonstrate the effectiveness of the proposed indicator system and methodology, providing efficient and objective technical support for evaluating and making decisions on overseas oilfield development projects. Full article

Amidst the escalating demand for high-precision structural health monitoring in large-scale engineering applications, carbon fiber-reinforced polymer fiber Bragg grating (CFRP-FBG) sensors have emerged as a pivotal technology for fatigue life evaluation, owing to their exceptional sensitivity and intrinsic immunity to electromagnetic interference. A time-series predictive architecture based on long short-term memory (LSTM) networks is developed in this work to facilitate intelligent fatigue life assessment of structures subjected to complex cyclic loading by capturing and modeling critical spectral characteristics of CFRP-FBG sensors, specifically the side-mode suppression ratio and main-lobe peak-to-valley ratio. To enhance model robustness and generalization, Principal Component Analysis (PCA) was employed to isolate the most salient spectral features, followed by data preprocessing via normalization and model optimization through the integration of the Adam optimizer and Dropout regularization strategy. Relative to conventional Backpropagation (BP) neural networks, the LSTM model demonstrated a substantial improvement in predicting the side-mode suppression ratio, achieving a 61.62% reduction in mean squared error (MSE) and a 34.99% decrease in root mean squared error (RMSE), thereby markedly enhancing robustness to outliers and ensuring greater overall prediction stability. In predicting the peak-to-valley ratio, the model attained a notable 24.9% decrease in mean absolute error (MAE) and a 21.2% reduction in root mean squared error (RMSE), thereby substantially curtailing localized inaccuracies. The forecasted confidence intervals were correspondingly narrower and exhibited diminished fluctuation, highlighting the LSTM architecture’s enhanced proficiency in capturing nonlinear dynamics and modeling temporal dependencies. The proposed method manifests considerable practical engineering relevance and delivers resilient intelligent assistance for the seamless implementation of CFRP-FBG sensor technology in structural health monitoring and fatigue life prognostics. Full article

The translocation of cadmium (Cd) in the soil-rice system is complicated; therefore, most of the soil-plant models of Cd have not been extensively studied. Hence, we studied the back-propagation artificial neural network model (BP-ANN) and multiple regression model (MLR) to predict the cadmium (Cd) content in rice grain and soil through testing soil parameters. In this study, 486 pairs of rice grains and corresponding soil samples of 456 vectors were used for training + validation, and 30 vectors were collected from the southwestern karst area of Guangxi Province as a test data set. In this study, the Cd content in rice was successfully predicted by using the factors soil available cadmium (A_Cd), total soil cadmium (T_Cd), soil organic matter (SOM), and pH, which have a more significant impact on rice, as the main prediction variables. Root mean square error (RMSE), Relative Percent Difference (RPD), and correlation coefficient (R²) were used to assess the models. The R², RPD, and RMSE values for R_Cd medium obtained by the MLR model with pH, T_Cd, and A_Cd as entered variables were 0.551, 2.398, and 0.049, respectively. The R² and RMSE values for R_Cd medium obtained by the BP-ANN model with pH, T_Cd, and A_Cd as entered variables were 0.6846, 2.778, and 0.104, respectively. Therefore, it was concluded that BP-ANN was useful in predicting R_Cd and had better performance than MLR. Full article

The BP (Back-Propagation) neural network model (hereafter referred to as the BP model) often gets stuck in local optima when predicting China’s silicon wafer price, which hurts the accuracy of the forecasts. This study addresses the issue by enhancing the BP model. It integrates the principles of genetic algorithm (GA) with particle swarm optimization (PSO) to develop a new model called the GA-PSO-BP. This study also considers the material price from both the supply and demand sides of the photovoltaic industry. These prices are important factors in China’s silicon wafer price prediction. This research indicates that improving the BP model by integrating GA allows for a broader exploration of potential solution spaces. This approach helps to prevent local minima and identify the optimal solution. The BP model converges more quickly by using PSO for weight initialization. Additionally, the method by which particles share information decreases the probability of being confined to local optima. The upgraded GA-PSO-BP model demonstrates improved generalization capabilities and makes more accurate predictions. The MAE (Mean Absolute Error) value of the GA-PSO-BP model is 31.01% lower than those of the standalone BP model and also falls by 19.36% and 16.28% relative to the GA-BP and PSO-BP models, respectively. The smaller the value, the closer the prediction result of the model is to the actual value. This model has proven effective and superior in China’s silicon wafer price prediction. This capability makes it an essential resource for market analysis and decision-making within the silicon wafer industry. Full article

To meet the requirements of the high spatiotemporal three-dimensional (3D) airflow field within the glide path corridor during carrier-based aircraft/unmanned aerial vehicles (UAVs) landings, this paper proposes a prediction method for high spatiotemporal resolution 3D ship airwake along the glide path by integrating computational fluid dynamics (CFD), backpropagation (BP) neural network, and Doppler wind lidar (DWL). Firstly, taking the conceptual design aircraft carrier model as the research object, CFD numerical simulations of the ship airwake within the glide path region are carried out using the Poly-Hexcore grid and the detached eddy simulation (DES)/the Reynolds-averaged Navier–Stokes (RANS) turbulence models. Then, using the high spatial resolution ship airwake along the glide path obtained from steady RANS computations under different inflow conditions as a sample dataset, the BP neural network prediction models were trained and optimized. Along the ideal glide path within 200 m behind the stern, the correlation coefficients between the predicted results of the BP neural network and the headwind, crosswind, and vertical wind of the testing samples exceeded 0.95, 0.91, and 0.82, respectively. Finally, using the inflow speed and direction with high temporal resolution from the bow direction obtained by the shipborne DWL as input, the BP prediction models can achieve accurate prediction of the 3D ship airwake along the glide path with high spatiotemporal resolution (3 m, 3 Hz). Full article

This paper presents the flow field reconstruction and prediction of powder fuel transport systems based on representative feature extraction from scattering images using deep learning techniques. A laboratory-built powder fuel supply system was used to conduct scattering spectroscopy experiments on boron-based fuel under various flow rate conditions. Based on the acquired scattering images, a prediction and reconstruction method was developed using a deep network framework composed of a Stacked Autoencoder (SAE), a Backpropagation Neural Network (BP), and a Long Short-Term Memory (LSTM) model. The proposed framework enables accurate classification and prediction of the dynamic evolution of flow structures based on learned representations from scattering images. Experimental results show that the feature vectors extracted by the SAE form clearly separable clusters in the latent space, leading to high classification accuracy under varying flow conditions. In the prediction task, the feature vectors predicted by the LSTM exhibit strong agreement with ground truth, with average mean square error, mean absolute error, and r-square values of 0.0027, 0.0398, and 0.9897, respectively. Furthermore, the reconstructed images offer a visual representation of the changing flow field, validating the model’s effectiveness in structure-level recovery. These results suggest that the proposed method provides reliable support for future real-time prediction of powder fuel mass flow rates based on optical sensing and imaging techniques. Full article

This study proposes a novel LK-BP-AFPSO model for the nondestructive evaluation of wood mechanical properties, combining a backpropagation neural network (BP) with adaptive fractional-order particle swarm optimization (AFPSO) and Liang–Kleeman (LK) information flow theory. The model accurately predicts four key mechanical properties—longitudinal tensile strength (SPG), modulus of elasticity (MOE), bending strength (MOR), and longitudinal compressive strength (CSP)—using only nondestructive physical features. Tested across diverse wood types (fast-growing YKS, red-heart CSH/XXH, and iron-heart XXT), the framework demonstrates strong generalizability, achieving an average prediction accuracy (R²) of 0.986 and reducing mean absolute error (MAE) by 23.7% compared to conventional methods. A critical innovation is the integration of LK causal analysis, which quantifies feature–target relationships via information flow metrics, effectively eliminating 29.5% of spurious correlations inherent in traditional feature selection (e.g., PCA). Experimental results confirm the model’s robustness, particularly for heartwood variants, while its adaptive fractional-order optimization accelerates convergence by 2.1× relative to standard PSO. This work provides a reliable, interpretable tool for wood quality assessment, with direct implications for grading systems and processing optimization in the forestry industry. Full article

Recent deep learning models, including GPT-4, have achieved remarkable performance using the back-propagation (BP) algorithm. However, the mechanism of BP is fundamentally different from how the human brain processes learning. To address this discrepancy, the Forward-Forward (FF) algorithm was introduced. Although FF enables deep learning without backward passes, it suffers from instability, dependence on artificial input construction, and limited generalizability. To overcome these challenges, we propose Local Back-Propagation (LBP), a method that integrates layer-wise unsupervised learning with standard inputs and conventional loss functions. Specifically, LBP demonstrates high training stability and competitive accuracy, significantly outperforming FF-based training methods. Moreover, LBP reduces memory usage by up to 48% compared to convolutional neural networks trained with back-propagation, making it particularly suitable for resource-constrained environments such as federated learning. These results suggest that LBP is a promising biologically inspired training method for decentralized deep learning. Full article

Aiming at the problem of a full-width horizontal-axis roadheader being prone to diverge from the preset trajectory of the tunnel, a deviation correction control method based on particle swarm optimization–backpropagation (PSO-BP) neural network proportional–integral–derivative (PID) control is proposed. The track error model of the walking system and the transfer function model of the deviation correction control are established. The PSO-BP PID controller is designed; the beginning weights of BP are enhanced by the PSO, and the BP receives the optimal weights to instinctively adapt the PID parameters. An experiment on deviation correction control of the roadheader was carried out. The experimental results indicate that the maximum steady-state error of PSO-BP PID for deflection angle and angular velocity is reduced by 41.03% and 44.93%, respectively, compared with BP PID, and the average rise time for deflection angle and angular velocity is reduced by 75.76%. Full article

Accurate and reliable detection of coal mine gases is the key to ensuring the safe service of coal mine production. Fourier Transform Infrared (FTIR) spectroscopy, due to its high sensitivity, non-destructive nature, and potential for online monitoring, has emerged as a key technique in gas detection. However, the complex underground environment often causes baseline drift in IR spectra. Furthermore, the variety of gas species and uneven distribution of concentrations make it difficult to achieve precise and reliable online analysis using existing quantitative methods. This paper aims to perform a quantitative analysis of coal mine gases by FTIR. It utilized the adaptive smoothness parameter penalized least squares method to correct the drifted spectra. Subsequently, based on the infrared spectral distribution characteristics of coal mine gases, they could be classified into gases with mutually distinct absorption peaks and gases with overlapping absorption peaks. For gases with distinct absorption peaks, three spectral lines, including the absorption peak and its adjacent troughs, were selected for quantitative analysis. Spline fitting, polynomial fitting, and other curve fitting methods are used to establish a functional relationship between characteristic parameters and gas concentration. For gases with overlapping absorption peaks, a wavelength selection method bassed on the impact values of variables and population analysis was applied to select variables from the spectral data. The selected variables were then used as input features for building a model with a backpropagation (BP) neural network. Finally, the proposed method was validated using standard gases. Experimental results show detection limits of 0.5 ppm for CH₄, 1 ppm for C₂H₆, 0.5 ppm for C₃H₈, 0.5 ppm for n-C₄H₁₀, 0.5 ppm for i-C₄H₁₀, 0.5 ppm for C₂H₄, 0.2 ppm for C₂H₂, 0.5 ppm for C₃H₆, 1 ppm for CO, 0.5 ppm for CO₂, and 0.1 ppm for SF₆, with quantification limits below 10 ppm for all gases. Experimental results show that the absolute error is less than 0.3% of the full scale (F.S.) and the relative error is within 10%. These results demonstrate that the proposed infrared spectral quantitative analysis method can effectively analyze mine gases and achieve good predictive performance. Full article

Climate change, resource scarcity, and ecological degradation have become critical bottlenecks constraining socio-economic development. Basin cities serve as key nodes in China’s ecological security pattern, playing indispensable roles in ecological civilization construction. This study established an evaluation index system spanning five dimensions to assess the effectiveness of ecological civilization construction. This study employs the entropy-weighted Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) and Back-Propagation (BP) neural network methods to evaluate the level of ecological civilization construction in the Yellow River Basin from 2010 to 2022, to analyze its indicator weights, and to explore the spatio-temporal evolution characteristics of each city. The results demonstrate the following: (1) Although the ecological civilization construction level of cities in the Yellow River Basin shows a steady improvement, significant regional development disparities persist. (2) The upper reaches are primarily constrained by ecological fragility and economic underdevelopment. The middle reaches exhibit significant internal divergence, with provincial capitals leading yet demonstrating limited spillover effects on neighboring areas. The lower reaches face intense anthropogenic pressures, necessitating greater economic–ecological coordination. (3) Among the dimensions considered, Territorial Space and Eco-environmental Protection emerged as the two most influential dimensions contributing to performance differences. According to the ecological civilization construction performance and changing characteristics of the 48 cities, this study proposes differentiated optimization measures and coordinated development pathways to advance the implementation of the national strategy for ecological protection and high-quality development in the Yellow River Basin. Full article

This study investigates how lighting conditions influence emotional and neural responses in a standardized, simulated museum environment. A multimodal evaluation framework combining subjective and objective measures was used. Thirty-two international students assessed their viewing experiences using 14 semantic differential descriptors, while real-time EEG signals were recorded via the EMOTIV EPOC X device. Spectral energy analyses of the α, β, and θ frequency bands were conducted, and a θα energy ratio combined with γ coefficients was used to model attention and comfort levels. The results indicated that high illuminance (300 lx) and high correlated color temperature (4000 K) significantly enhanced both attention and comfort. Art majors showed higher attention levels than engineering majors during short-term viewing. Among four regression models, the backpropagation (BP) neural network achieved the highest predictive accuracy (R² = 88.65%). These findings provide empirical support for designing culturally inclusive museum lighting and offer neuroscience-informed strategies for promoting the global dissemination of traditional Chinese culture, further supported by retrospective interview insights. Full article

High-standard farmland fertility enhancement projects can lead to the sustainable utilization of arable land resources. However, due to difficulties in project implementation and uncertainties in costs, resource allocation efficiency is constrained. To address these challenges, this study first analyzes the impact of geography and engineering characteristics on cost indicators and applies principal component analysis (PCA) to extract key influencing factors. A hybrid prediction model is then constructed by integrating the Northern Goshawk Optimization (NGO) algorithm with a Backpropagation Neural Network (BP). The NGO–BP model is compared with the RF, XGBoost, standard BP, and GA–BP models. Using data from China’s 2025 high-standard farmland fertility enhancement projects, empirical validation shows that the NGO–BP model achieves a maximum RMSE of only CNY 98.472 across soil conditioning, deep plowing, subsoiling, and fertilization projects—approximately 30.74% lower than those of other models. The maximum MAE is just CNY 88.487, a reduction of about 32.97%, and all R² values exceed 0.914, representing an improvement of roughly 5.83%. These results demonstrate that the NGO–BP model offers superior predictive accuracy and generalization ability compared to other approaches. The findings provide a robust theoretical foundation and technical support for agricultural resource management, the construction of projects, and project investment planning. Full article

Acid phosphatase (ACPase) is an essential industrial enzyme, but its production via recombinant bacterial fermentation is often limited by insufficient dissolved oxygen control. This study optimized the aerobic fermentation of the ACPase-producing recombinant bacterium Bacillus subtilis 168/pMA5-Acp by refining the bioreactor’s aerodynamic structure using computational fluid dynamics (CFD) simulations. This was combined with fermentation kinetics modeling to achieve precise process control. First, the gas distributor structure of the 5 L bioreactor was optimized using CFD simulation results. Optimal mass transfer conditions were identified through comprehensive analysis of K_La in different reactor regions (aeration ratio: 1.142 VVm, K_La = 264.2 h⁻¹). The simulation results showed that the optimized oxygen transfer efficiency increased 2.49 fold compared to the prototype. Second, the process control issue was addressed by developing a BP (backpropagation) neural network model to predict K_La under alternative media conditions. The prediction error was less than 5%, and the model was combined with the logistic equation to construct the bacterial growth kinetic model (R² > 0.99). The experiments demonstrated that using the optimized reactor with a molasses–urea medium (molasses 7.5 g/L; urea 15 g/L; K₂HPO₄ 1.2 g/L; MgSO₄·7H₂O 0.25 g/L) reduced production costs while maintaining enzyme activity (215.99 U/mL) and biomass (OD₆₀₀ = 101.67) by 90.03%. This study provides an efficient and cost-effective process solution for the industrial production of ACPase and a theoretical foundation for bioreactor design and scale-up. Full article

This study focuses on a blood pump system equipped with four radial active magnetic bearings (RAMBs). The finite element method (FEM) was employed to optimize the physical parameters of the system. Based on this optimization, two intelligent PID tuning strategies—particle swarm optimization (PSO) and backpropagation (BP) neural networks—were compared. First, a differential control model of a single-degree-of-freedom active magnetic bearing was developed, based on the topology and operating principles of the radial magnetic bearings. Then, magnetic circuit parameters were precisely identified through finite element simulation, enabling accurate optimization of the physical model. To enhance control accuracy, intelligent tuning strategies based on PSO and BP neural networks were applied, effectively addressing the limitations of conventional PID controllers, which often rely on empirical tuning and lack precision. Finally, simulation experiments were conducted to evaluate the optimization performance of PSO and BP neural networks in the magnetic bearing control system. The results demonstrate that the improved PSO algorithm offers significant advantages over both the BP neural network and traditional manual PID tuning. Specifically, it achieved a rise time of 0.0049 s, a settling time of 0.0079 s, and a steady-state error of 0.0013 mm. The improved PSO algorithm ensures system stability while delivering faster dynamic response and superior control accuracy. Full article