【摘要】 精准可靠地预测锅炉NO_x排放量对电站锅炉低氮运行有着重要意义,为了提升模型的预测效果,提出一种基于鲸鱼优化算法-最小二乘支持向量机(WOA-LSSVM)的锅炉NO_x排放量预测建模方法。首先归一化处理初始样本数据,然后通过WOA算法对LSSVM中的核函数宽度和惩罚因子两个参数进行寻优求解,建立WOA-LSSVM黑箱模型,最终得到模型输出,同时将采用果蝇优化算法(FOA)、粒子群优化算法(PSO)优化参数建立的LSSVM预测模型和单一LSSVM预测模型作为对比研究。仿真结果表明,采用WOA优化的LSSVM模型在NO_x排放量预测方面明显优于其他选定模型,具有稳定且较高精度的仿真性能。更多还原

【Abstract】 Given the significance of accurate and reliable prediction of boiler NO_x emission to low-nitrogen operation of power plant boilers, this paper proposed a predictive modeling methods of NO_x emission based on the whale optimization algorithm-least squares support vector machine(WOA-LSSVM). The first step was to normalize the initial sample data. Then this paper built a WOA-LSSVM black box model and obtained the model output by optimizing the kernel function width and penalty factor in LSSVM by WOA algorithm. At the same time, this paper adopted fruit fly optimization algorithm(FOA) and particle swarm optimization algorithm(PSO) in optimizing parameter to establish LSSVM prediction model. And this paper compared the improved LSSVM prediction model with single LSSVM prediction model. The simulation results showed that the WOS-optimized LSSVM model is superior to other selected models in NO_x emission prediction, and that it features stable and high-precision simulation performance.更多还原