A Fitting Method of the Temperature Characteristic Curve of Sensor

To improve effectively temperature compensation characteristic of sensor, a neural network model fitting the temperature characteristic curve was proposed. The convergence of the neural network algorithm was proposed and proved. The theory criterion to select learning rate was provided by the convergence theorem. The simulating example of the sensitivity-temperature characteristic curve of sensor was given. The result showed that the temperature characteristic fitting curve of sensor using the neural network algorithm was very both smooth and accurate. The fitting precision was up to 10
^-6
Therefore, the method of curve fitting based on the neural network algorithm is effective.