Dynamic prediction of energy delivery capacity of power networks: Unlocking the value of real-time measurements

The paper focuses on advances in short-term prediction (1-4 Hours) of dynamic line rating as an example of what can be achieved by the combination of advanced network sensors and the latest machine learning, data-mining tools. Combining these tools has allowed us to achieve reliable and usable predictions that allow the network operators to switch from a static approach to a manageable dynamic one that significantly increases asset utilisation without reducing security of supply.