Apart from optimisation techniques involved in solving load flow equations, and the occasional optimal placement of distribution line regulators, modern optimisation techniques are rarely used in distribution planning. This may have been appropriate in the past when networks were simpler, but in the modern and emerging power system uncertainties are multiplying. These uncertainties consist in varying uptake scenarios of photovoltaics, electric vehicles and customer based battery systems, and also in variations of demand management uptake rates, project execution timing and fundamental economic activity – to name a few. These complexities compound to ensure that it is very likely, given finite planning time and resources, that optimum least cost solutions and work plans are not discovered.

This paper will demonstrate a new methodology for optimised planning to solve this problem. It will first present a recently created optimised demand management program called Optimised Incremental Pricing (OIP)1 which is currently being implemented within Energy Queensland. This example is a specific case of a more generalised optimisation solution which will be outlined and demonstrated. This solution will consist of a risk-optimisation approach, where solutions, constraints, loads / generation and consequences are all represented stochastically. In other words, these fundamental quantities will be represented as random variables rather than singular projected values which don’t describe inherent uncertainty.

Using Monte Carlo analysis, these variables will be balanced and optimised to determine the least risk (and therefore least cost) solution program over all possible future histories, therefore discovering the most robust solutions in the presence of risk and uncertainty. This system will also enable the balancing between high certainty traditional augmentation and less certain customer partnered solutions (i.e. demand management initiatives).