Predicting consistency of concrete in transit

摘要

Predefined parameters of influence, on which the concrete's consistency depends, are fed into a neural network which produces a response in the form of a prediction of the consistency when the concrete is delivered to the construction point. For training the neural network, the influencing parameters are derived from the production of the fresh concrete and the consistency of concrete on delivery. Some of the following parameters are used: the wts. of the binding agent, aggregate, additives, additional material and water; quality characteristics, e.g. granularity, solidity, manufacturer, added material types; process parameters, esp. moisture content, temp. and density of added water; electrical power consumption of the mixer drive; ambient temp.; transportation time.