摘要
BP神经网络算法具有寻优效率不高、易发生早熟且最终求解精度不够等特点,针对以上问题,文章提出一种基于改进二进制人工蜂群算法(Improved Binary Artificial Bee Colony Algorithm)的BP神经网络并行集成学习算法(IBABC-BP).首先,文章构建以高斯变异函数作为概率映射函数的离散二进制人工蜂群算法(IBABC),分析证明了算法的有效性,并通过在4个Benchmark标准测试函数上证明了其寻优精度和收敛速度较其他4种改进人工蜂群算法均有大幅提高;其次,将改进的二进制人工蜂群算法(IBABC)用于训练BP神经网络.设计了IBABC-BP并行集成学习算法;最后,将IBABC-BP算法用于雾霾评估预测,以合肥地区的雾霾历史数据作为仿真数据.实验结果表明,IBABC-BP算法在寻优精度和收敛速度上较原始BP算法、人工蜂群ABC-BP算法、遗传GA-BP算法等算法有明显的提升,可以有效地提高雾霾评估预测的准确性.
BP neural net work algorithm has the characteristics of slow learning speed, falling into the local optimum easily and the inaccurate operating result, and so on, in order to solve these problems, a parallel ensemble learning algorithm based on Improved Binary Artificial Bee Colony Algorithm (IBABC) and BP neural net work is proposed. Firstly, a kind of improved Binary Artificial Bee Colony Algorithm which based on Gauss Variation Function as probability mapping function is proposed in this paper, then prove the effectiveness of the algorithm and verify the convergence speed and accuracy by four Benchmark functions. Secondly, the IBABC algorithm is used to train the BP Neural Network and construct a parallel integration learning algorithm of IBABC-BP. Finally, the IBABC-BP algorithm is applied to the haze assessment forecast, the experiment results based on haze data in Hefei indicate that the IBABC-BP algorithm is superior to BP algorithmABC-BP algorithm and GABP algorithm in terms of convergence speed and accuracy, the IBABC-BP algorithm can improve the accuracy of the haze assessment forecast efficiently.
作者
贾凯
倪志伟
李敬明
陆玉佳
朱旭辉
JIA Kai;NI Zhiwei;LI Jingming;LU Yujia;ZHU Xuhui(Key Laboratory of Process Optimization and Intelligent Decision-Making, School of Management, Hefei University of Technology, Hefei 230009;School of Management Science and Engineering, Anhui University of Finance & Economic, Bengbu 233030)
出处
《系统科学与数学》
CSCD
北大核心
2019年第3期477-494,共18页
Journal of Systems Science and Mathematical Sciences
基金
国家自然科学基金重大研究计划培育项目(91546108)
国家自然科学基金重大项目(91490725)
国家自然科学基金创新群体项目(71521001)
安徽省自然科学基金青年项目(1908085QG298)
中央高校基本科研业务费专项资金(JZ2019HGTA0053,JZ2019HGBZ0128)
过程优化与智能决策教育部重点实验室开放课题资助课题
关键词
改进二进制人工蜂群算法
BP神经网络
高斯变异函数
雾霾评估预测
Improved binary artificial bee colony algorithm
back propagation neural network
Gauss variation function
haze assessment forecast
