基于骨架理论的机电特种设备启动故障检测系统设计

Design of start-up fault detection system for electromechanical special equipment based on skeleton theory

摘要:

针对当前机电特种设备启动故障检测方法缺少故障检测指标提取过程，导致系统准确率低、召回率高的问题，设计基于骨架理论的机电特种设备启动故障检测系统。设计二极管组成整流电流采集电路模块消除开关电弧，通过吸收太阳光直接加热或间接产生能量提供热量，构成太阳能光伏电池板或组件，以L298N为主驱动芯片，避免损坏稳压片，使用5V外部电源供电，采用大容量滤波电容，设计自由保护二极管的L298N电机驱动模块，利用有载调容变压器监测变压器低压侧的电压、电流，判断当前负载电流，完成系统硬件设计，重新定义离散化获得离散域骨架，基于骨架理论，提取节点散度、骨架节点、骨架曲率特征故障检测指标，由此设计故障检测流程，完成系统软件设计。实验结果表明，该系统的召回率较低，能够有效提高故障检测准确率。

Abstract:

In view of the lack of fault detection index extraction process in current mechanical and electrical special equipment start-up fault detection methods, resulting in low system accuracy and high recall rate, a start-up fault detection system based on skeleton theory is designed. The rectifier current acquisition circuit module composed of diodes is designed to eliminate the switching arc. The solar photovoltaic panel or module is composed of L298N as the main driver chip to avoid damaging the voltage regulator. The L298N motor driver module is designed with 5V external power supply and large capacity filter capacitor on load tap changer is used to monitor the voltage and current at the low-voltage side of the transformer, judge the current load current, complete the hardware design of the system, redefine the discretization to obtain the discrete domain skeleton. Based on the skeleton theory, the fault detection indexes of node divergence, skeleton node and skeleton curvature are extracted, and the fault detection process is designed, and the system software design is completed. The experimental results show that the recall rate of the system is low, which can effectively improve the accuracy of fault detection.