Journal of Engineering Research and Reports

This study explores the impact of sodium silicate on wood ash blended cement concrete, which contributes significantly to CO₂ emissions. This research holds great potential for the construction industry by aiding in the reduction of greenhouse gas emissions, conserving natural resources, and promoting waste minimization. The specific research objectives are Characterizing sand, granite, and wood ash, Producing concrete mixes with wood ash and sodium silicate (Na₂SiO₃) replacements of 0.5%, 1.0%, 1.5%, and 2.0%, and determine their compressive strength, workability, density, and water absorption, and Developing a Long Short Term Memory (LSTM) deep learning model to determine optimal proportions of cement, w/c ratio, wood ash, and sodium silicate to achieve target compressive strength, workability, water absorption, and density in concrete mixes. Concrete specimens were tested for compressive strength, workability, density, and water absorption. A Long Short Term Memory (LSTM) machine learning model is significant because it is designed to capture historical information of time series data and is suitable for predicting long-term nonlinear series. The LSTM predicted compressive strength based on inputs of cement, water/cement ratio, sodium silicate, and wood ash. Results indicated that sodium silicate significantly enhanced early compressive strength and workability, making this blend a viable option for sustainable construction where early strength is essential.