中文摘要：

针对现有关于企业绩效管理评价的方法存在一定的缺陷,不能满足企业发展过程对制度评价的准确度要求的问题,通过利用AHP方法在复杂的评价指标体系中筛选出8个代表性的指标作为绩效评价的评价指标,随后对此采用拓扑结构的BP神经网络方法,构建评价模型,设计算法程序,结合计算机技术进行网络训练,得到该算法的仿真结果：算法的误差在训练步数为100时就已经达到了期望结果,而且预测数据的相关系数达到了0.98747,这表明基于拓扑结构的BP神经网络算法在企业绩效管理评价工作中具有一定的可行性与较高的准确性。

英文摘要：

In order to solve the problems existing in the evaluation methods of enterprise performance management,which can not meet the requirements of the enterprise development process to evaluate the accuracy of the system,eight representative indexes are selected as evaluation indexes of performance evaluation by using AHP method in the complex evaluation index system,and then the BP neural network method of topological structure is used to construct the evaluation model and the algorithm program is designed,the computer technology is combined to carry on the network training,and the simulation result of this algorithm is gotten：the error of the algorithm has reached the expected results when the training step is100,and the correlation coefficient of forecast data is 0.98747,it shows that the BP neural network algorithm based on topological structure in enterprise performance management evaluation has certain feasibility and high accuracy.