为了快速准确的预测基坑开挖对周围建筑物沉降的影响,本文提出一种结合多智能体粒子寻求LSSVM(最小二乘支持向量机)模型参数的算法,提高了LSSVM算法的预测精度.采用该算法对昆明市某基坑开挖过程中周围建筑物的沉降进行预测,并与其他预测方法进行对比,结果表明该算法具有收敛速度快、预测精度高等特点. In order to quickly and accurately predict the effect of excavation on the settlement of surrounding buildings, an algorithm to find the LSSVM (least squares support vector machine) model parameters by combining multi-agent particles is proposed in this paper, which improves the prediction accuracy of LSSVM algorithm. The algorithm predicts the settlement of surrounding buildings during the excavation of a foundation pit in Kunming and compares it with other forecasting methods. The results show that the proposed algorithm has the characteristics of fast convergence and high prediction accuracy.