Design and fault diagnosis of DCS sintering furnace’s temperature control system for edge computing

Abstract

Under the background of modern industrial processing and production, the sintering furnace’s temperature control system is researched to achieve intelligent smelting and reduce energy consumption. First, the specific application and implementation of edge computing in industrial processing and production are analyzed. The industrial processing and production intelligent equipment based on edge computing includes the equipment layer, the edge layer, and the cloud platform layer. This architecture improves the operating efficiency of the intelligent control system. Then, the sintering furnace in the metallurgical industry is taken as an example. The sintering furnace connects powder material particles at high temperatures; thus, the core temperature control system is investigated. Under the actual sintering furnace engineering design, the Distributed Control System (DCS) is used as the basis of sintering furnace temperature control, and the Programmable Logic Controller (PLC) is adopted to reduce the electrical wiring and switch contacts. The hardware circuit of DCS is designed; on this basis, an embedded operating system with excellent performance is transplanted according to functional requirements. The final DCS-based temperature control system is applied to actual monitoring. The real-time temperature of the upper, middle, and lower currents of 1# sintering furnace at a particular point is measured to be 56.95°C, 56.58°C, and 57.2°C, respectively. The real-time temperature of the upper, middle, and lower currents of 2# sintering furnaces at a particular point is measured to be 144.7°C, 143.8°C, and 144.0°C, respectively. Overall, the temperature control deviation of the three currents of the two sintering furnaces stays in the controllable range. An expert system based on fuzzy logic in the fault diagnosis system can comprehensively predict the situation of the sintering furnaces. The prediction results of the sintering furnace’s faults are closer to the actual situation compared with the fault diagnosis method based on the Backpropagation (BP) neural network. The designed system makes up for the shortcomings of the sintering furnace’s traditional temperature control systems and can control the temperature of the sintering furnace intelligently and scientifically. Besides, it can diagnose equipment faults timely and efficiently, thereby improving the sintering efficiency.

Ensemble learning based transmission line fault classification using phasor measurement unit (PMU) data with explainable AI (XAI)

“…In order to automatically identify and pinpoint symmetrical and asymmetrical faults, as well as high-impedance faults (HIFs), in a distribution system, Jibin et al [39] According to recent studies illustrated in Table 1, ML techniques [40][41][42][43] are quite accurate in categorizing various fault types. Real-time labeled datasets [37] of power system disruptions are rarely present, hence synthetic data has played a vital role in these fields.…”

Logic adaptive optimization model for dynamic temperature compensation of heating furnace based on proportion‐integral‐derivative position algorithm

“…In many cases, it is not necessary to control these three parameters at the same time, and the whole temperature control system can be well controlled by adjusting one or two of the parameters. In view of the fact that the response speed of the control system is too slow, the problem can be solved by adjusting the parameter k in the control system [12][13][14]. In the process of constructing the optimal temperature control model, setting the gas flow of the heating furnace, and adopting PID positional algorithm modelling to analyze and optimize the control rules to avoid large controlled deviation, in order to further improve the system response speed, expert rule compensation is required.…”

Research on Fault Diagnosis of Switchgear Based on Temperature Cloud Image Technology

“…In this paper, the metal armoured open switch cabinet is selected to establish a multi-physical field co-simulation model. Since the model of the switch cabinet is relatively complex and the establishment of the finite element software is relatively cumbersome, the model is established in SolidWorks software and imported into Fluent for analysis [3]. An important reason for the heating of switchgear is excessive load.…”