Research On Concrete Production Data Management System And Compressive Strength Prediction Method

China is a major infrastructure construction country and is moving towards being a great infrastructure construction country,and concrete is the largest raw material used in infrastructure projects.In response to the current situation of inefficient information flow among the participating units in ready-mixed concrete production and the prevalence of blind production,this paper designs and implements a concrete production data management system to provide global visibility of the entire production process.There is also the problem of difficulty in determining the quality of concrete at the production site.This paper uses the concrete production data management system as a platform to carry out research on concrete compressive strength prediction model by deploying the system to obtain data in the mixing building of the HanjiangTo-Weihe River Walley Water Diversion Project,and provide information for judging the quality of concrete,the main research contents are as follows:(1)The records of the whole process of concrete production at the production site are mainly realized through paper forms.In view of the shortcomings such as inefficiency of paper forms in circulation between different units and difficulties in access,this paper,based on the analysis and arrangement of all form data,designs and implements the system to collect data at the site,and enters the contents of paper forms generated in real time into the system and saves them in electronic form,realizing the paper files electronic backup.In addition,in order to improve the efficiency of the system to collect information,the paper analyzes the various testing processes of concrete and its raw materials,and realizes the automatic acquisition of some testing indexes through two improved measures of sensors and image recognition.(2)The traditional concrete compressive strength experiments need to undergo a long curing process,and the curing time is usually 28 days,during which the concrete is often already in use at the construction site.The site can not find the substandard quality of concrete and discarded in a timely manner,and the subsequent reinforcement of the concrete structure is economically costly.To address this problem,this paper uses the historical data in the system database to first establish a concrete compressive strength prediction model in an offline learning mode,which can predict the compressive strength of concrete immediately after the concrete ratio is determined.In order to find the adaptability of the algorithm model to the system,this paper proposes to establish a concrete compressive strength prediction model driven by real-time data,and improve it based on hoeffding tree as the model algorithm.Through the comparison of results,it is verified that our model not only achieve high prediction accuracy,but also the online learning model algorithm in this paper has low complexity and low resource consumption,which is easy to deploy in the embedded system.(3)The visualization design and implement of the system is completed for the actual needs of the production site,and the site personnel can realize the real-time monitoring of the production site information through the system.In addition,the real-time data-driven compressive strength prediction model is combined with the system to test the functional implementation of each part of the system,and the test results prove the feasibility of the scheme in this paper.