The Application Of The Embedded Operating System At Control Of Furnace Temperature

This paper mainly researches how to transplant the Embedded Real-Time Operating System (μC/OS-Ⅱ) to 32 bit Microprocessor Unit, in order to control the temperature of an electrical resistance furnace. An electrical resistance furnace is a nonlinear system, changing with time and distributed parameter. It is difficult to express its characteristic using precise mathematic model. A normal PID controller is used in industry, due to its simple structure and being realized easyly. But a traditional PID controller cannot obtain a good effect on real-time controlling, owing to uncertain controlled object and environment. This paper not only adopts the adaptive Neuron Network PID arithmetic for the purpose of distinguish system parameters, but also brings forward Expert System control ruler in order to overcome the disadvantage of slow speed constringency of Neuron Network on the basis of Neuron Network PID arithmetic.Because the adaptive Neuron Network PID arithmetic need caculate many parameters and a normal slow speed CPU cannot process program on real time, Samsung S3C44BOX CPU is selected. The Samsung S3C44BOX CPU is 32/16 bit RISC Microprocessor and uses ARM7TDMI core. Its maximum CPU clock frequency is 75Mhz. The S3C44BOX is used in fields of cheap price handle devices and industry applications. The Embedded Real-Time Operating System μC/OS-Ⅱ is a preemptive multitask opening system, which source codes are opened.In the paper, firstly the hardware system of intelligently controlling temperature is designed. Secondly, on the basis of the special structure of S3C44BOX Microprocessor, through modifying μC/OS-Ⅱ and programming key codes using ARM assemble and C language, the Embedded Real-Time Operating System μC/OS-Ⅱ is transplanted to S3C44BOX. Last, according to Neuron Network PID arithmetic and Expert System control rulers, the single electrical resistance furnace is controlled on real time.