Register for free to
view the full innovation profile.
AI self-learning models of gas demand forecasting for flexible power generation plants, integrating real-time data from Electricity Grid as inputs to forecasts.
About
Our vision is to deploy our expertise in the utility industry and data modelling, our software platform that incorporates Data Analytics and Artificial Intelligence with modelling, forecasting and optimisation capabilities in order develop mathematical /AI models for gas demand forecasting needs of the “flexible power generation” sites for use in near-real-time. These models could use the relevant site information and the real-time data from the Electricity Grid as inputs to produce the gas demand forecasts for these flexible power plants. With these forecasts WWU would be able to more accurately optimise and schedule their Gas networks, thereby reducing costs and risks whilst ensuring security of supply.
Key Benefits
This innovation applies to many industries however in the context of this proposal it has been aimed at the energy market and specifically for real-time gas demand forecast of the “flexible Power generation” plants. The benefits universally apply: Accurate real time and more granular forecasts provide the ability to timely meeting of the demand; Better optimisation of resources Reduced costs and reduced risks Improved efficiency Better insight and improved planning Social satisfaction and sharing resources Improves the environment
Applications
We propose to use our decades of expertise in the energy forecasting to analyse the challenge and to deploy our Nominator platform to develop one or more new mathematical /AI forecasting models of gas demand. These would enable WWU to incorporate gas demand forecasts for the smaller flexible power generation sites into their current gas scheduling operations. Such models would in real-time capture the flow of electricity to use as a driver, amongst other drivers, to forecast the flexible gas requirements of these power plants. Our Expertise proposal covered our forecasting models, assuming WWU currently has a satisfactory forecasting software solution capable of hosting or implementing our proposed forecasting models. In this our second, Innovation proposal, we introduce our Nominator software platform, which could house and implement our models if WWU wants to consider an alternative solution to the one it has currently implemented. In our experience with electricity balancing, flexible generation is under tight control, not least by its owner/operators, by NGC (ESO), and by the local DNOs. As the move towards embedded renewable generation and ‘smart grids’ gathers momentum, power surges and price volatility are on the increase, so reliable gas generation that can ramp up/down at very short notice will be much in demand. That may not help the gas industry, who will have to accommodate rapid fluctuations in highly-localised gas demand to fuel this generation. The preceding paragraph is the starting point for our solution. Although notification of the sudden operation of flexible gas generation may come as a surprise to the gas supply and gas distribution company(s), it certainly won’t be a surprise to the electricity DNO and NGC ESO. Some of these power stations operate directly in the NGC ESO balancing mechanism, meaning they are expected to respond instantaneously to commands issued by NGC ESO, and are wired in to NGC’s real-time control systems accordingly. At first glance – and subject to confirmation by WWU – we would not expect these despatch instructions from NGC ESO to be immediately relayed to the gas supply and distribution companies, hence the element of surprise to gas companies when these plant start running.