桥梁承台深基坑钢板桩围堰施工成本预测分析

Study on construction cost prediction of steel sheet pile cofferdam for deep foundation pit excavation of bridge cap

制订桥梁承台深基坑开挖钢板桩围堰施工方案,并根据施工方案设计现场劳动力与材料等施工参数;利用鸡群算法优化极限学习机的权重与偏置,把经过优化后的权重和偏置输入极限学习机模型中,构建最好的极限学习机模型;在最佳极限学习机模型内输入施工参数,输出施工成本预测结果,完成桥梁承台深基坑开挖钢板桩围堰施工成本预测。以贵州省重点工程——乌当(羊昌)至长顺高速公路工程TJ-8标段羊昌河大桥为研究对象,进行相关的试验测试。试验结果证明:该方法可有效预测桥梁承台深基坑开挖钢板桩围堰施工时,钢板桩围堰打入阶段与基坑清淤及支撑阶段等不同施工阶段人工费与材料费等成本,成本预测的残差与相对误差均较低。

Develop a construction plan for the excavation of steel sheet piles and cofferdams for deep foundation pits on bridge platforms, and design construction parameters such as on-site labor and materials based on the construction plan;Using the chicken swarm algorithm to optimize the weights and biases of the extreme learning machine, input the optimized weights and biases into the extreme learning machine model, and construct the best extreme learning machine model;Input construction parameters into the optimal limit learning machine model, output construction cost prediction results, and complete the cost prediction of steel sheet pile cofferdam construction for deep foundation pit excavation of bridge bearing platforms. Relevant experimental tests were conducted on a certain bridge as the research object. Experimental results have shown that this method can effectively predict the labor and material costs during the construction of steel sheet pile cofferdam for deep foundation pit excavation of bridge bearing platforms, as well as the labor and material costs during different construction stages such as the steel sheet pile cofferdam driving stage and the foundation pit dredging and support stage. The residual and relative errors in cost prediction are relatively low.