Neural Network Predict The Flowing Property Of Self-compacting Concrete And Experiment On Frame Member

SCC(self-compacting concrete) is belonged to the category of the high performance concrete and it is one of the main directions development on the concrete science from now on. Currently, the research and application on the SCC in our country is still placed in the development stage, and it is very important to go deep in to carry on experiment research about SCC's flowing property and beam.The overall calculation method is adopted to prepare two strength grades (C30, C40) of SCC whose coarse aggregate sizes are different (coarse aggregate sizes is 5-10mm and 5~20mm) and the strength grade (C40) of self-compacting light-aggregate concrete (SCLC) whose coarse aggregate sizes is 5~20mm. For finding the best work function of SCC, SCC is requested to have favorable filling performance, rebar clearance passing performance and cohesiveness, and prevented from bleeding and segregation under the premise of assure that SCC's flowing property due to by dint of rheology theory and analyse mechanism on concrete's vibrationless moulding and compact. This solves contradiction about fluidity capability and segregation capability. Through SCC's tests on flowing property, filling performance, passing ability and segregation are done by adopting slump test tube, V funnel and L fluidity instrument, the results of SCC's flowing property are gotten under the case whose water gel ratio and coarse aggregate are changed.Due to the diversity and complexity of the SCC ingredients which influence each other, it is difficult to describe the relationship between the workability of fresh and a single mixing effect factor. In the paper, the properties of fresh concrete are predicted by using neural network that can be used to carry out nonlinear regression with multi-factor. It shows that the artificial neural network can be used as a new approach to predict the flowing property of SCC. The artificial neural network's forecasting result can be gained when data that are dosage of ingredient by calculating in the mixing ratio design principles are input into trained network. The mixing ratio is adjusted according to the forecasting result until it meet request. This workload on the mixing ratio can be eased consumedly.According to the nerve network and the mixing ratio experiment, the pouring experiments about the fresh of SCC transit the steel bar handrail are finished. The three-point loading is done on the concrete beams that are formed by SCC.