基于混沌粒子群算法的神经网络的财务管理预警优化模型  

A Model of Financial Management Early Warning Optimization Based on Neural Network and Chaos Particle Swarm Optimization

作　　者：张静 ZHANG Jing(School of Economics and Management,Shaanxi Post and Telecommunication College,Xianyang 712000,China)

机构地区：[1]陕西邮电职业技术学院,经济管理学院,陕西咸阳712000

出　　处：《微型电脑应用》2023年第2期20-23,共4页Microcomputer Applications

基　　金：陕西省中华职业教育社(ZJS202136)。

摘　　要：以避免企业发生经济损失为出发点,提出基于混沌粒子群算法的神经网络的财务管理预警优化模型,提升了预警结果的可靠性。由于神经网络的财务管理预警模型存在局部极值和随机选择参数现象,导致预测结果存在较大随机性,将具备遍历特性的混沌优化算法引入至标准粒子群算法中,赋予每个粒子遍历特性,控制粒子的聚集程度,避免局部极值的发生,获取全局最佳值,保证模型的预警结果。测试结果表明:优化后模型可获取全局遍历最佳解;在迭代次数为50次时完成收敛,且适应度值为0.000 5;全局寻优性能良好,可零误差完成财务管理预警,具备极佳的预警可靠性。In order to avoid the economic loss of enterprises, this paper puts forward an optimization model of financial management early warning based on chaos particle swarm optimization and neural network, and improves the reliability of early warning results. Due to the phenomenon of local extremes and random selection of parameter amplitudes in the financial management early warning model of neural network, the prediction results of traditional methods are quite random. The chaotic optimization algorithm with traversal characteristics is introduced into the standard particle swarm algorithm to give each particle a traversal features, control the aggregation degree of particles, avoid the occurrence of local extreme values, obtain the global best value, and ensure the early warning results of the model. The test results show that the optimized model can obtain the best global traversal solution;the convergence is completed when the number of iteration is 50, and the fitness value is 0.0005;the global optimization performance is good. The financial management early warning can be completed with zero error, and it has excellent performance of early warning reliability.

关 键 词：混沌粒子群 优化 神经网络 财务管理 预警模型 遍历特性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]