Design And Research On Temperature Control System Of High Volume High Temperature Process

Today’s industry is developing rapidly, and more and more goods are made of metal materials. As a result, the market for metal materials processing technology also has a higher demand. High temperature process gas furnace is one of the most commonly used equipment in the field of metal heat processing. In the process of heat treatment, the accuracy and stability of temperature control affect the processing quality and the processing difficulty. This determines the temperature control system to the high precision, high real-time and other direction of development. In the field of locomotive manufacturing, some locomotive components are too large, and the domestic high temperature processing furnace on the market can not meet the requirements of the processing technology on the volume.This thesis presents a design of temperature control system for large volume high temperature process atmosphere furnace. The effective heating capacity of the project is expected to reach 3000mm*2000mm*2000mm, and it is divided into 32 temperature zones for temperature acquisition. In order to improve the control precision and the heating speed of the system, the distributed control strategy is adopted. This collaborative control strategy on the one hand, the real-time synchronization of hardware requirements are higher, on the other hand, the operation of the module can not be too high.The design takes the multi chip BF-518 as the core, the K thermocouple as the temperature sensor, AD7606 as the analog digital conversion unit, with the cold junction compensation, the filter circuit of the conditioning amplifier circuit, to achieve accurate data collection. Multi chip DSP is composed of a distributed system, which combines the IEEE1588 Ethernet protocol to achieve the real-time synchronization of the nanosecond. Thanks to the unique function multiplication calculator, DSP DMA controller, VDK kernel optimization, after the implementation of the project code optimization can achieve the system operation work load is less than or equal to 50%. PID neural network strong decoupling, no overshoot and other characteristics of the system makes the system temperature, temperature control accuracy. The learning function of neural network makes the workload of control algorithm design greatly reduced. After verification test, the performance parameters meet the design requirements.