基于卷积神经网的柴油发电机组转速控制系统设计  

Design of Diesel Generator Speed Control System Based on Convolutional Neural Network

作　　者：张华彦 杜治虎 陈向 邸维正 ZHANG Huayan;DU Zhihu;CHEN Xiang;DI Weizheng(Shanghai Regeon Electric Co.,Ltd.,Shanghai 201612,China)

机构地区：[1]上海昱章电气股份有限公司,上海201612

出　　处：《自动化与仪表》2023年第6期23-28,共6页Automation & Instrumentation

基　　金：华北电力大学中央高校基金项目(2016MS125)。

摘　　要：柴油发电机组的转速波动会导致供电电压和频率波动,影响用电设备使用效能和电能质量。为达到有效实现转速控制的目的,该文设计了基于卷积神经网控制模型的柴油发电机组转速控制系统。系统硬件结构包括转速控制器、供油执行器和传感器。通过信号采集与预处理,获取柴油机各个部位控制量的变化范围;基于卷积神经网络实现一维信号向二维灰度图像的转换;建立柴油发电机组转速控制模型。经实验论证分析,该文系统执行器的静态特性稳定,与其它系统对比,动态性能更优。The speed fluctuation of diesel generator set will lead to the fluctuation of power supply voltage and frequency,and affect the use efficiency and power supply quality of electrical equipment.In order to achieve the purpose of effectively realizing speed control,the speed control system of diesel generator set based on convolutional neural network is designed.System hardware structure includes electronic speed controller,fuel supply actuator and sensor.Through signal acquisition and preprocessing,the variation range of the control parameter of each part of the diesel engine is obtained.Based on convolution neural network,the conversion from one-dimensional signal to two-dimensional gray image is realized.Establish the speed control model of diesel generator set.Through experimental demonstration and analysis,the static characteristics of the actuator of this system are stable and the dynamic performance is better than other systems.

关 键 词：卷积神经网 控制器 柴油发电机组 转速信号 转速控制系统 执行器 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]