Temperature Control System Of Thermostatic Water Bath In Gravity-capillary Viscometer

Gravity-capillary viscometer is the instrument for testing viscosity of liquid, it is applied in many fields, such as petroleum, chemical industry, chemical fiber, food and medicine. Thermostatic water bath is the one key part of gravity-capillary viscometers, and its temperature control is the one key technology of viscometers. Under the supports of Changsha WeiTe Technique Development Company's Project ("Intelligent Viscometer"), some studies are performed.The conception of viscosity is presented, the relationship of viscosity and temperature is expounded, and the research status and development trend of temperature control method for thermostatic water bath of gravity-capillary viscometer is introduced.According to the thermostatic water bath's some characteristics, such as nonlinearity, great hysteresis, an complex intelligent control method is proposed to meet the temperature error of thermostatic water bath by national standards of viscosity testing ("method for measurement and test of the viscosity GB/T256-88"), which combines improved Bang-Bang control, self-organizing fuzzy control, human-simulated intelligent control and PWM control of DSP. When the temperature error between the desired temperature and actual temperature in the water tank is large, the improved Bang-Bang controller is employed to reduce the rise time of temperature, the self-organizing fuzzy controller is used to decrease the system overshoot when the temperature error is small, the human-simulated intelligent controller is applied when the temperature error is less than the steady-state error of fuzzy controller. All experimental results show that the thermostatic water bath with complex intelligent control method has satisfied performance, its overshoot is slight and steady-state error is less than 0.05℃, which is better than the Chinese National Standards for viscosity measurement.A nonlinear error compensation method for digital temperature sensor based on modified radial basis function neural networks (RBFNN) ensemble is proposed. Firstly some independent member networks based on RBFNN are founded to compensate sensor's error in respective range according to the digital temperature sensor's error characteristic and the generator of member network. Then the estimator of digital temperature sensor's output and the weight controller are used to obtain the output weight of RBFNN ensemble, and the final temperature result is gotten. The experiments show that the model based on RBFNN ensemble is effective to compensate the nonlinear error of temperature sensor.The conformation and operating principle of gravity-capillary viscometer is introduced, and the circuit is designed, which includes temperature data acquiring circuit, temperature control circuit, and man-machine interface circuit. The software of complex intelligent control and compensation for sensor's nonlinear error is given. The temperature control system has the function of heating and cooling, its range is 15-40℃, temperature resolution is 0.01℃, and its steady-error is less than 0.1℃.