Thermal flow through materials can be modeled as an electronic circuit. This is pretty common:

The material's thermal resistance (1/conductivity) is modeled as a resistor, it's thermal capacitance (mass x specific heat) as a capacitor, heat flow in Watts as current, etc.

Does anyone know how to model a non-linear phase change material such as paraffin? These materials have a large heat of fusion, and so the temperature (modeled as voltage) will increase until the melting point is reached, at which point it will stop until the heat of fusion has been met. After this point both the thermal capacitance and the thermal conductivity will change. Having the properties change at the melting point is fairly straight forward, just use a voltage-controlled switch. It's the heat storage during the melting phase (or heat release during the freezing phase) that I'm having problems with.

 

If heat = current, then you need a current sink with a fixed capacity. It absorbs all current until the capacity is used up. I'm probably telling you what you already know.