The monitoring and control of combustion systems co-firing coal and biomass is a critical consideration when aiming to increase the proportion of biomass being combusted. This is because it is likely that the combustion will become increasingly unstable as the biomass proportion increases. In order to develop a flame monitoring and control system, flame signal data sets were collected from combustion measurements taken on a 500kW pilot scale combustion test facility. The sensors used were photodiodes with sensitivities in the UV, visible and IR wavelengths. The analysis of these data, identified flame features that can be related to operational parameters such as flame stability, excess air level, NOx and CO emissions. These features were then applied in the development of an intelligent flame monitoring and optimisation system for individual burners based on these low cost sensors. The testing of the monitoring and control system on a pilot scale burner and at full scale are described in this paper.