The majority of present marine finfish production is conducted in flexible net cages which can deform when they are subjected to water movements generated by currents. The ability to monitor net deformation is important for performing cage operations and evaluation of fish health and welfare under changing environment. This paper presents a new method for real-time monitoring of net cage deformations that is based on an integrated approach where positioning sensor data is incorporated into a numerical model. An underwater positioning system was deployed at a full-scale fish farm site, with three acoustic sensors mounted on a cage measuring positions of the net at different depths. A novel numerical model with an adaptive current field was used to simulate net cage deformations, where the magnitude and direction of the current could be adapted by continuously assessing deviations between the simulated and the measured positions of the net. This method was found to accurately predict the pre-defined current velocity profiles in a set of simulated experiments. In the field experiment, a good agreement was also obtained between the simulated positions of the net and the acoustic sensor data. The integrated approach was shown to be well suited for in-situ real-time monitoring of net cage deformations by using a significantly reduced number of sensors.