In this research, we extend the Gaussian-Polynomial Approximation Model with operation temperature as the parameter. A fourth-order temperature polynomial is used to describe the nonlinear feature of batteries discharge behavior for a fix current load under various temperatures. As a result, in the curve of percentage of remain capacity versus voltage, the major differences between the observed discharge curve and the curve generated from our temperature model is happened only in the beginning of the discharge. In the middle and the later part of the discharging, our model fits very well to the observed discharge curve. The differences are within 0.1 volts. As to the percentage of remaining battery capacity, the differences are within 3 percents. To the users, the important concern is how much the capacity remains to be used, rather than how much capacity they have used. Thus, the understanding of the later part of the discharge curve is much more important than that of the earlier part. Although the high order polynomial looks odd and complicated, the getting more popular microprocessor implementation of fuel gauge doesn’t seem to be a problem.
