Stratigraphy Analysis of the Nahr Umr Formation in Zubair oil field, Southern Iraq

This hybrid deep-learning study focuses on pollutant concentration. It illuminates convolutional neural networks (CNN) and long short-term memory in hybrid deep learning methods (LSTM). CNNs are essential to deep learning, especially image processing. They are ideal for pollution concentration analysis because they extract complex data features. LSTM is another important tool for this study. LSTMs are recurrent neural networks (RNNs) that can process and store data sequences. Time-series data analysis, common in pollution concentration research, benefits from them. Understanding deep learning and hybrid learning's impact on pollutant concentration issues. It investigates a hybrid CNN-LSTM model that combines CNN feature extraction with LSTM sequence processing. This fusion lets the model make smart predictions from input data sequences. PCA is key to this investigation. PCA dimensionality reduction finds variables with significant relationships.

Technological Frontier on Hybrid Deep Learning Paradigm for Global Air Quality Intelligence

Priscila,

Celin Pappa,

Banu

et al. 2024

An Artificial Intelligence Technique in Industry 4.0 for Predicting the Settlement of Geocell-Reinforced Soil Foundations

Jeyanthi,

Venkatakrishnaiah,

Raju

2024

Civil and geotechnical engineering professionals face the challenge of settlement prediction to ensure the secure and long-lasting construction of a geocell-reinforced soil foundation (GRSF). In this study, a new adaptive method for forecasting geocell settlement has been developed. It is based on the adaptive artificial neural network (ANN) technique and elephant herding optimization (EHO). The goal is to reduce erosion on steep slopes, strengthen soft ground, and increase the carrying capacity of retaining structures, foundations, roadways, and railroads. The confinement effect, which occurs when the geocell disperses the loads across a larger area and enhances the soil's ability to sustain loads, makes the research novel. Numerical results from plate load tests on unreinforced and geocell-reinforced foundation beds have validated the proposed model.

Evaluation and Intelligent Modelling for Predicting the Amplitude of Footing Resting on Geocell-Based Weak Sand Bed Under Vibratory Load

Jeyanthi,

Venkatakrishnaiah,

Raju

2024

The main consideration while designing geo-structures to sustain vibration loads is the accuracy of the displacement amplitude estimation. The displacement amplitude of a footing on a geocell-reinforced bed exposed to vibration stress can be computed using sophisticated data. AI modelling has replaced many traditional methodologies. Thus, the current work introduces a hybrid paradigm called NFC-TSA, which stands for neuro-fuzzy controller and tunicate swarm algorithm. Comprehensive field vibration experiments provided the reliable database utilised to train and evaluate the model. To develop the model's precise prediction objective, displacement amplitude was used as an output index. Several parameters impacting the foundation bed, geocell reinforcement, and dynamic excitation were considered as input variables. Existing methods, ANN-EHO, JSA, MOA, RNN, and ANN-MGSA, were compared to the NFC-anticipated TSA's accuracy.