DSP-based Intelligent Temperature Control System For Semiconductor Lasers

Semiconductor lasers with its small size, light weight, good color, high efficiencyhas been widely applied in the field of military defense, industrial measurement (suchas infrared gas detection), satellite communications, and medical. As thesemiconductor laser emission wavelength and intensity is affected by temperature.For example, the variation coefficient between the emission wavelength of a genericsemiconductor laser and temperature is about0.2-0.3nm/°C. Moreover, thetemperature is easy to make laser burned. We need to precisely control thetemperature of the semiconductor laser in order to achieve laser stability of a singlewavelength.According to the requirements of the laser normal working with precisiontemperature control, temperature overshoot and settling time, traditional PID(Proportional-Iintegral-Derivative) control algorithm can not meet the requirementsby experiment. So we design active temperature control system based on fuzzyadaptive PID algorithm of the laser with the TMS320LF2407digital signal processorcore control chip.Firstly, the hardware circuit of the laser temperature control system is designedto complete. It include three parts that is temperatue signal acquisition, temperatuesignal control and temperatue signal monitor. Temperature signal acquisition circuit iscomposed by encapsulated in the laser internal10K thermistor and externalprecision resistor and16bit high-precision serial A/D chip LTC1864. Temperatuesignal control part include TMS320LF2407, thermoelectric cooler driver moduleMAX1968and16bit high-precision D/A chip LTC1655. The principle of temperatuesignal monitor is that temperature value is collected through the MAX232serial chipto the computer, in order to monitor the temperature of the laser within the process ofchange. Finally, the temperature control system hardware is completed.Secondly, in order to theoretically verify the feasibility of fuzzy adaptive PIDalgorithm,we used Matlab software to simulate the fuzzy adaptive PID algorithm.First,we established the mathematical model of the laser internal temperature of theheat balance system. Then,we get first-order delay model based on theoretical analysisand experimental environment. we established laser temperature control system of PID algorithm model and fuzzy adaptive PID algorithm model by Simulink thecomponents of Matlab. Through we compared the traditional PID model with fuzzyadaptive PID model, the fuzzy adaptive PID algorithm is theoretically proved thefeasibility of the design process. It has excellent performance by comparing the fuzzyadaptive PID algorithm model than the traditional PID algorithm model in the controltemperature overshoot, precision temperature control, and time to reach stably.Finally, the design and experiment of taking TMS320LF2407as the hardware forcircuit’s controlling core, will prove that the actual measured results are consistentwith the simulation results from application of the Simulink. We focus on testing theperformance of the laser temperature control system based on fuzzy PID algorithm.The accuracy of temperature control system is±0.05℃, and the setting time is lessthan70s. In addition, the experiments on laser temperature control system verify thatcompared with conventional PID algorithm, the fuzzy adaptive PID algorithm canmake PID parameters self-regulating and the temperature overshoot is smaller, what’smore, the process of temperature rising is smoother, when the temperature of externalenvironment changes. So the temperature control system for semiconductor laserbased on fuzzy adaptive PID algorithm meets the requirements of the infrared gasdetection and other engineering applications.