Competition on EC in the Energy Domain: SG Applications – winners announcement
Vicky and Rasul have recently won a Bronze medal in IEEE CEC/GECCO 2021 Competition on “Evolutionary Computation in the Energy Domain: Smart Grid Applications”. The competition has two independent tracks, each tackling an optimisation problem in the energy domain. Participants of the competition could use different evolutionary algorithms for each track (if they participate in both). The algorithm designed by our team, Presto (Population REgeneration STar-guided Optimization), won 3rd place in track one and 4th place in track two. There were 20 teams in the competition. Presto has been inspired from Particle Swarm Optimization and Differential Evolution algorithms for continuous optimisation. Compared to these algorithms, Presto provides more efficient control over diversification by utilizing (statistically) independent feasible solutions throughout the algorithm. In musical terminology, presto is a tempo – it means fast.
A summary of the problems in the competition:
Track 1) Bi-level optimization of bidding strategies in Local Energy Markets (LEM)
A complex bi-level problem in which competitive agents in the upper-level try to maximize their profits, which is based on the price determined in the lower-level problem (i.e., the clearing price in the LEM), thus resulting in a strong interdependence of bi-level decisions.
Track 2) Flexibility management of home appliances to support a Distribution System Operator (DSO) flexibility requests
A complex problem seeking to efficiently match a flexibility request from the DSO, in which the households participating in the DR program are paid remuneration according to their preferences and the shifting/real-time modification of their baseline profiles.