Event Title
Location
London
Event Website
http://www.csce2016.ca/
Description
Ultra-lightweight concrete (ULWC) has potential applications for floating structures and architectural elements because of its dry density coming in at under 1000 kg/m3. The objective was to develop an artificial neural network (ANN) to aid the ULWC designer according to his needs. Boundary conditions were set for each material and 13 constraints based on the water binder ratio, density, air content, binder and aggregate content. The ANN predicted the compressive strength with a comfortable margin of error, with the gap encountered being attributed to variability in workability. Precise constraints and boundary conditions are needed to ensure a lower variability in workability. The ANN, coupled with a genetic algorithm, can generate millions of mixes for a given compressive strength in a short amount of time. The designer is able to choose mixes according to additional needs, such as the carbon footprint, absolute density, polymer content, cost, etc.
Included in
Civil Engineering Commons, Construction Engineering and Management Commons, Structural Engineering Commons
MAT-701: PREDICTING THE COMPRESSIVE STRENGTH OF ULTRA-LIGHTWEIGHT CONCRETE BY AN ARTIFICIAL NEURAL NETWORK
London
Ultra-lightweight concrete (ULWC) has potential applications for floating structures and architectural elements because of its dry density coming in at under 1000 kg/m3. The objective was to develop an artificial neural network (ANN) to aid the ULWC designer according to his needs. Boundary conditions were set for each material and 13 constraints based on the water binder ratio, density, air content, binder and aggregate content. The ANN predicted the compressive strength with a comfortable margin of error, with the gap encountered being attributed to variability in workability. Precise constraints and boundary conditions are needed to ensure a lower variability in workability. The ANN, coupled with a genetic algorithm, can generate millions of mixes for a given compressive strength in a short amount of time. The designer is able to choose mixes according to additional needs, such as the carbon footprint, absolute density, polymer content, cost, etc.
https://ir.lib.uwo.ca/csce2016/London/Materials/1
