论文部分内容阅读
丙烯腈收率是丙烯腈装置的关键指标,如何得到丙烯腈收率是厂家很关注的研究,将新型优化算法用于丙烯腈收率软测量建模是1种较好的尝试。将新型微粒群优化算法用于同样新型的文化算法种群空间的优化,设计文化微粒群优化算法。它由种群空间和信念空间2部分组成,在种群空间和信念空间分别采用各自算法并行演化,同时,2个空间又根据一定的协议相互联系。分别将该算法和基本微粒群算法用于一些常用测试函数的优化问题;结果表明,与基本微粒群算法相比,文化微粒群算法加强了全局搜索能力,更容易收敛于全局最优解。最后将文化微粒群优化算法用于优化神经网络,构成文化微粒群神经网络,并将其应用于丙烯腈收率软测量建模。结果表明,此模型精度高,应用前景广阔。
Acrylonitrile yield is a key indicator of acrylonitrile plant. How to get the yield of acrylonitrile is a great concern for the manufacturers. It is a good attempt to use the new optimization algorithm in the soft sensor modeling of acrylonitrile yield. The new particle swarm optimization algorithm is used to optimize the same new culture algorithm population space and the design of the cultural particle swarm optimization algorithm. It is composed of two parts of the population space and the belief space, and each uses the respective algorithm to evolve in parallel in the population space and the belief space. Meanwhile, the two spaces are connected with each other according to a certain protocol. The algorithm and the basic particle swarm optimization algorithm are respectively used to optimize some commonly used test functions. The results show that compared with the basic particle swarm optimization algorithm, the PSO enhances the global search ability and converges to the global optimal solution. Finally, the cultural particle swarm optimization algorithm is used to optimize the neural network to form a cultural particle swarm neural network, which is applied to the soft sensor modeling of acrylonitrile yield. The results show that the model has high precision and wide application prospect.