Research On Data Analysis Method For Concrete Performance Prediction

With the increasing progress of society and the rapid development of intelligent concrete(concrete)manufacturing,enterprises have increasingly high requirements for concrete quality control and cost reduction.The composition and mix proportion of concrete directly affect its quality and cost,and the mix proportion design needs to consider the performance indicators of concrete.Therefore,it is crucial to guide the mix proportion design by predicting the performance of concrete.Currently,most practitioners at home and abroad use some empirical formulas to predict performance,but empirical formulas are only derived based on limited experimental data,and their prediction results are not accurate.Moreover,each formula has a limited scope of application,which requires extremely high engineering experience for practitioners.Therefore,it is necessary to analyze and use them carefully in combination with actual situations.Based on the entire process of concrete design,production,testing,and analysis,this article uses data analysis methods to predict the performance of concrete,and introduces the architecture and practical application of this method in detail.At the same time,an intelligent manufacturing data analysis platform has been designed and implemented,specifically for predicting concrete performance.The main tasks completed in the thesis include:(1)Aiming at the needs of concrete manufacturing enterprises,a data analysis method for concrete performance prediction was proposed.The method in this paper makes full use of data acquisition systems and associated testing equipment,and takes the data flow direction as a guide to collect and uniformly store concrete intelligent manufacturing data,integrating multiple digital systems and testing equipment into a whole;At the same time,a prediction model was constructed using neural networks,and based on the collected data,accurate prediction of the 28 day strength and slump of concrete was achieved.(2)A data analysis platform for intelligent concrete manufacturing is designed.The platform is mainly divided into five layers: data layer,data collection layer,computing layer,data mining layer,and application layer;From the perspective of concrete practitioners,indepth functional analysis and detailed design of the platform were conducted,with main functions including data source management,data quality analysis,task scheduling,and so on.(3)An intelligent manufacturing data analysis platform for concrete performance prediction is implemented.Based on Hadoop distributed cluster technology,the platform is built and components are configured to achieve task scheduling and analysis.This platform can help users better utilize intelligent concrete manufacturing data for performance prediction and production analysis,providing better decision-making basis for concrete practitioners.The concrete intelligent manufacturing data analysis platform developed in this paper achieves unified storage and comprehensive analysis of multi-source data.The platform can not only help concrete manufacturing enterprises improve the quality of concrete manufacturing and reduce production costs,but also provide decision-making reference for concrete performance prediction,provide optimization suggestions for other aspects of intelligent concrete manufacturing,and have important significance for improving the intelligent concrete production.