摘要
研究了电厂热工过程参数辨识问题。针对传统的蚁群算法在热工过程参数辨识中运行迭代时间长,而且存在容易出现早熟现象而陷入局部最优解的缺陷,提出了用人工免疫蚁群算法对电厂热工过程进行参数辨识的方法。将人工免疫的思想引入到传统的蚁群算法中,将特征信息作为疫苗注射给"蚂蚁",使"蚂蚁"具有免疫能力,新算法模型克服了传统蚁群算法的缺点。用新算法对电厂热工过程参数辨识的仿真结果表明,新算法有效避免了算法出现停滞的现象,提高了算法全局搜索能力和辨识的准确度。
The parameter identification of power plant thermal process is studied. The traditional ant colony algo- rithm is an algorithm of searching the global optimal solutions. This algorithm takes a long iterate time to run, easily encounters premature problem and traps into a local optimal solution. In order to reduce the search time, avoid trap- ping into a local optimal solution and increase its efficiency, the artificial immune system is introduced into the tradi- tional ant colony algorithm, and the system injects feature information as vaccine into ants and makes the ant have immunity. This method overcomes the shortcomings of traditional ant colony algorithm. The simulation for parameter identification of power plant thermal process shows that, new algorithm avoids the algorithm stagnation phenomenon, and improves the global search ability of the algorithm and the identification accuracy.
出处
《计算机仿真》
CSCD
北大核心
2015年第2期127-130,260,共5页
Computer Simulation
关键词
蚁群算法
全局最优解
人工免疫
疫苗
Ant colony algorithm
Global optimal solutions
Artificial immune
Vaccine
