基于人工神经网络的锅炉原煤水分预测分析

Prediction and Analysis of Boiler Raw Coal Moisture Based on Artificial Neural Network

燃煤水分与煤种品质密切相关,准确的燃煤水分在线监测是判定煤种类型和煤质的重要手段,对电厂配煤掺烧、燃烧优化调整有着非常重要的意义。对传统的基于热平衡原理原煤水分在线监测和基于人工神经网络原煤水分预测方法进行对比,采用人工神经网络方法具有很强的非线性映射能力,训练样本的预测水分与实测水分的相对误差小,完全可以满足生产应用需求。

Coal moisture is closely related to coal quality. Accurate online monitoring of coal moisture is an important means to determine coal type and coal quality, which is of great significance to coal blending and combustion optimization in power plants. The traditional online monitoring method of raw coal moisture based on the principle of thermal balance is compared with the method of raw coal moisture prediction based on artificial neural network. The method of artificial neural network has strong nonlinear mapping ability, and the relative error between the predicted moisture of training samples and the measured moisture is small, which can fully meet the needs of practical application.