燃煤火力发电厂BP神经网络轮机燃料流量智能控制技术探索  

Exploration of intelligent control technology of BP neural network turbine in coal-fired power plant

作　　者：吕木 池彬彬 LV Mu;CHI Binbin(Zhejiang Zheneng Lanxi Power Generation Co.,Ltd.,Jinhua,Zhejiang 321100,China)

机构地区：[1]浙江浙能兰溪发电有限责任公司,浙江金华321100

出　　处：《计算机应用文摘》2025年第3期101-103,108,共4页

摘　　要：当前燃煤火力发电厂控制策略普遍存在稳态误差较大的问题,为此对燃煤火力发电厂BP神经网络轮机燃料流量智能控制技术进行了研究。文章构建了一个混合型燃煤轮机模型,能够对燃煤轮机内部的关键参数(如燃煤轮机作业的机械参数)进行精确监控和计算。同时,应用BP神经网络对梯度计算与反向传播算法进行了训练,通过计算目标函数的梯度,可以了解网络参数对误差的影响程度,从而进行有效的权值调整。实验结果表明,控制阀可以快速开启且切换过程中没有出现抖动现象,实现了平稳的切换,符合预期效果。这一技术的成功应用不仅提高了燃煤火力发电厂的运行效率,也为类似系统的控制策略提供了新的思路和方法。In the current control strategy of coal-fired power plants,there is a problem of large steady-state error,so the exploration of BP neural network turbine fuel flow intelligent control technology for coal-fired power plants is studied.This paper constructs a hybrid coal turbine model,which can accurately monitor and calculate the key internal parameters of coal turbine,such as the mechanical parameters of coal turbine operation.At the same time,BP neural network is used to train the gradient calculation and backpropagation algorithm.By calculating the gradient of the objective function,the influence of network parameters on the error can be understood,and the weight can be adjusted effectively.The experimental results show that the control valve can be opened quickly and achieve a stable switching state without shaking during the switching process,which is in line with the expected effect.The successful application of this technology not only improves the operating efficiency of coal-fired power plants,but also provides a new idea and method for the control strategy of similar systems.

关 键 词：BP神经网络 火力发电厂 轮机 燃料 流量控制 

分 类 号：TK472[动力工程及工程热物理—动力机械及工程]