Data‐driven chance‐constrained dispatch for integrated power and natural gas systems considering wind power prediction errors

Stochastic wind power prediction errors hurt the normal operation of integrated and natural gas systems (IPGS). First, data-driven stochastic chance-constrained programming method is applied to deal with errors, its probability distribution accurately fitted by variational Bayesian Gaussian mixture model massive historical data. In addition, chance constraints tie-line reserve capacity turbine are built. Next, utilize more reasonably, operational characteristics optimal commitment power-to-hydrogen devices considered modelled in proposed strategy reflect actual situation IPGS. Then, original complicated dispatch problem converted into a tractable second-order cone via convex relaxation quantile-based analytical reformulation techniques. Finally, effectiveness validated numerical experiments based on modified IEEE 33-bus system 10-node micro hydrogen system.