摘要
由于天津临港地区航道清淤产生了大面积的疏浚土,其含水率高、压缩性大,无法满足施工要求,所以一种大幅度提高疏浚土的抗压强度的固化技术成为解决问题的关键。针对疏浚土可塑性差、强度低、大面积放置占用土地的问题,为了达到废弃物资源化的目的,采用炉渣粉、粉煤灰代替大量的水泥作为胶凝材料,通过单掺、三掺试验,以无侧限抗压强度为指标,通过不同配比的正交试验筛选出最优的固化剂配比。建立神经网络预测模型,以正交试验数据为训练样本,以正交试验掺量范围内的试样为测试样本,对固化土的强度进行预测。结果显示:网络输出结果与试验测试结果有11%的误差,模型预测精度较高。研制的高强度固化剂不仅可以解决疏浚土的放置和处理问题,还能减少水泥的使用量,提高粉煤灰和炉渣粉的利用率,同时也为工程提供了一种再生资源,具有很大的发展前景。
A large area of dredged soil from the dredging of the waterway is there in the port-vicinity area of Tianjin.As the water content and compressibility of the dredged soil are high and then cannot meet the relevant construction requirement,a solidifying technology for largely increasing the compressive strength of it becomes the key link to solve the problem.Aiming at the problems from the dredged soil,i.e.poor plasticity,low strength and large area of land to be occupied,slag powder and fly ash are used to replace a large amount of cement for taking as the cementitious materials,and then an optimal mixing ratio for the solidifying agent is screened out by means of the tests of single-mixture and three-mixtures through the orthogonal experiments made on different mixing ratios by taking the unconfined compressive strength as the index,so as to achieve the objective of the resourcalization of waste.The strength of the solidified soil is predicted through establishing a neural network prediction model by taking the orthogonal experimental data as the training samples and the samples within the range of orthogonal experimental mixing amounts as the testing samples.The results show that the error of 11%is there between the network output result and the testing result from the experiment,of which the prediction accuracy of the model is higher.The developed high strength solidifying agent cannot only solve the problem of the placement and treatment of the dredged soil,but can also reduce the amount of the cement to be used,increase the utilization rates of both slag powder and fly ash,meanwhile,provide a renewable resource for the relevant construction project,thus has a great developing prospect.
作者
张伟
刘晓强
李顺群
张宇亭
王欢
刘双菊
ZHANG Wei;LIU Xiaoqiang;LI Shunqun;ZHANG Yuting;WANG Huan;LIU Shuangju(School of Civil Engineering,Tianjin Chengjian University,Tianjin 300384,China;Tianjin Research Institute for Water Transport Engineering,Tianjin 300456,China)
出处
《水利水电技术》
北大核心
2020年第4期20-26,共7页
Water Resources and Hydropower Engineering
基金
国家自然科学基金项目(51579123)
中央公益性基金项目(tks180211,tks190403)
天津市重点研发项目(16YFXTSF00280)
天津市自然科学基金项目(15JCQNJC43700)。
关键词
疏浚土
高强度
神经网络
炉渣粉
粉煤灰
变形
含水率
dredged soil
high strength
neural network
slag powder
fly ash
deformation
moisture content