Ready-mix concrete strength prediction system
A machine learning-based system that uses multiple trained models to predict the strength of ready-made concrete by integrating blending, manufacturing, and quality information addresses the inaccuracies of existing methods, providing reliable strength predictions.
JP7653331B2Active Publication Date: 2025-03-28TAIHEIYO CEMENT CORP
8 Cites 0 Cited by
Patent Information
- Application Number
- JP2021139882
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-08-30
- Publication Date
- 2025-03-28
- Estimated Expiration
- 2041-08-30
AI Technical Summary
Technical Problem
Existing methods for predicting the compressive strength of ready-made concrete are inaccurate due to variations in the hydration reaction caused by differences in mixers and manufacturing environments, making it difficult to predict the strength of concrete over time.
Method used
A system utilizing multiple trained models generated by machine learning, which incorporates blending-related, manufacturing-related, and quality-related information to predict the strength of ready-made concrete with high accuracy.
Benefits of technology
The system achieves high accuracy in predicting the strength of ready-made concrete by considering various factors that influence the hydration reaction, thereby improving the reliability of strength predictions before quality inspections.
✦ Generated by Eureka AI based on patent content.
Abstract
To provide a ready-mixed concrete strength prediction system capable of highly accurately predicting ready-mixed concrete strength before quality inspection.SOLUTION: A ready-mixed concrete strength prediction system comprises a storage unit which stores a plurality of learned models, a data input unit which accepts an input of input data for prediction, and a data output unit which outputs data containing strength prediction information of the ready-mixed concrete. The plurality of learned models include a first learned model for deriving primary strength prediction information, a second learned model for deriving secondary strength prediction information, and a third learned model for deriving tertiary strength prediction information. Blending-related information, production-related information, and quality-related information of ready-mixed concrete to be predicted are entered as input data for prediction.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art