DIGITAL AGGREGATE: Analyzing Concrete Aggregate Physical Properties using Neural Network at Ready-Mix Plant
Presented By: Vivek Patel
Affiliation: Alter Biota- Canada
Description: The physical characteristics of aggregates, such as shape (circularity, roughness, roundness, etc.) and size (minimum diameter, maximum diameter, aspect ratio, etc.), have a significant impact on the fresh and hardened state properties of concrete. Although only aggregate particle size distributions are regularly tested at ready-mix concrete plants by sieving of representative samples, morphological properties are not commonly considered when proportioning concrete mixtures. Some characteristics like flatness, elongation, and texture indexes of aggregates are still manually measured, however, this happens infrequently for standards compliance reasons and is not used as a reference point when proportioning concrete mixtures. These additional morphological properties of aggregate, that fluctuate constantly, remain unknown during concrete production, requiring additional safety margin to ensure the mix-design meets the fresh and hardened state requirements. To address this limitation, an innovative technique known as DIGITAL AGGREGATE is proposed, which involves installing a computer vision system at the ready-mix plant to capture and analyze images of aggregates falling into the weighing bin for a batch of concrete in real-time. A convolutional neural network (CNN)-based model is trained on the image dataset to predict the segmentation mask areas of the aggregates. Statistical metrics are applied for evaluating the model's training and prediction performances. This novel inline aggregate data offers an enhanced opportunity for optimizing cement paste proportions in real-time from batch to batch, reducing the need for high safety margins and field adjustments to meet fresh state requirements.
Application of Evolutionary Algorithms for Optimization of Dense Packing of Concrete Aggregates
Presented By: Konstantin Sobolev
Affiliation: University of Wisconsin
Description: Sequential Packing Algorithm - SPA was developed to model the dense packing of large assemblies of particulate materials such as aggregate systems used for portland cement or asphalt concrete. The SPA performance was further enhanced using the artificial intelligence (AI) approach. The AI optimization based on genetic algorithms (GA) uses natural selection and genetics to estimate the fractal dimensions and porosity of Apollonian packing of spherical particles. Multi-cell packing procedures, fine adjustment of the algorithm’s parameters, as well as implementation of GA were demonstrated to be effective tools to optimize the computational resources, to speed-up the SPA and to pack a large number of spherical objects and also be applicable to packing of other geometrical shapes such as ellipsoids and fibers. The developed algorithm can be used to describe and visualize dense packings corresponding to concrete aggregates. Based on the simulation results, different particle size distributions of particulate materials and packing efforts are correlated to corresponding packing degree. These virtual packings agree well with the standard requirements and available research data.
Low-Embodied Carbon Concrete Enabled by Aggregate and Concrete Optimization
Presented By: Mohamadreza Moini
Affiliation: Princeton University
Description: Concrete production accounts for 8-9% of global anthropogenic CO2 emissions and 2-3% of energy demand. This study exhibits how optimizing aggregate gradation and packing can reduce cement use and associated carbon footprint by presenting theoretical particle packing criteria for enhancing binary and ternary aggregate blends from ready-mix plants. The optimization considers the experimental and theoretical (Modified Toufar model) aggregates packing degree (PD) and the overall gradation considering power curves (PC), alongside metrics like workability and coarseness factors, to evaluate the global warming potential and properties of the optimized concrete blends, compared against reference conventional ready mixtures (without optimized gradation). A good agreement between the theoretical PD obtained from the Modified Toufar model and experimental PD was found. These optimized mixtures showed lower compressive strength and reduced workability, attributed to a higher water-to-binder (w/b) ratio. Despite these drawbacks, the optimized mixtures meet the normal 28-day strength criteria of concrete, proving their adequacy for many applications. Furthermore, the mixtures with optimized aggregate exhibited a 25.17% decrease in global warming potential (GWP) over the non-optimized counterparts, thereby providing a facile approach to reducing cement consumption without losing performance.
Alkali-Aggregate Reaction in Concrete (AAR): Current Challenges and Research Needs
Presented By: Leandro Sanchez
Affiliation: University of Ottawa
Description: In 2024, the International Conference on Alkali-Aggregate Reaction in concrete (ICAAR) was held in Ottawa, ON, Canada. ICAAR-2024 was the 17th international conference on the topic; this conference normally takes place every 4 years and in 2024, the 50th anniversary of the event was celebrated. After so many years of in-depth research, numerous protocols, guidelines, and standards have been produced to help avoiding, mitigating, assessing, rehabilitating, and managing concrete infrastructure affected by AAR. However, there are still plenty of doubts and unknowns on both scientific aspects and practical procedures. The purpose of this work is to highlight my personal impressions after the 17th ICAAR, pointing out some of the current challenges and research needs and opportunities on the topic.