A technique is developed that combines a new closed-form analytical form of the lumped capacitance model (LCM) and thermography to predict the internal heat generation of a lithium-ion pouch cell. A series of tests is conducted to validate the new technique using a polyimide film heater mounted as a vertical plate in front of an infrared camera. A programmable power supply is used to subject the heater to three types of pre-defined heat profiles. The technique performs very well in reconstructing each known heat profile. The technique is then used to predict the rate of heat generation of a 14.5 Ah lithium iron phosphate cell under different rates of discharge. Compared to the heat rate measured by a calorimeter, the model predicts an average heat rate within 10% for moderate C-rates and lower. For C-rates higher than 2C, a simple correction algorithm is applied due to cell temperature gradients becoming more pronounced. (C) 2015 Elsevier B.V. All rights reserved.