Jinghuan Ma

Honglin Wen, Pierre Pinson, Jinghuan Ma et al.

We present a data-driven approach for probabilistic wind power forecasting based on conditional normalizing flow (CNF). In contrast with the existing, this approach is distribution-free (as for non-parametric and quantile-based approaches) and can directly yield continuous probability densities, hence avoiding quantile crossing. It relies on a base distribution and a set of bijective mappings. Both the shape parameters of the base distribution and the bijective mappings are approximated with neural networks. Spline-based conditional normalizing flow is considered owing to its non-affine characteristics. Over the training phase, the model sequentially maps input examples onto samples of base distribution, given the conditional contexts, where parameters are estimated through maximum likelihood. To issue probabilistic forecasts, one eventually maps samples of the base distribution into samples of a desired distribution. Case studies based on open datasets validate the effectiveness of the proposed model, and allows us to discuss its advantages and caveats with respect to the state of the art.

Jinghuan Ma

Local area packetized-power network (LAPPN) provides flexible local power dispatching in the future Energy Internet. With interconnections among multiple LAPPNs, power dispatching can be further extended to intra- and inter-LAPPN power interchanges. It becomes a significant issue to schedule the two kinds of power interchanges as, from a system perspective high utilization of available scheduling time slots and low overall transmission loss should be guaranteed, and from a subscriber perspective a high scheduled ratio of transmission requests with a fair transmission sequence in terms of transmission urgency are expected. To this end, we propose a cooperative power dispatching framework for connected LAPPNs, including subscriber matching and two-layer power transmission scheduling. The former matches subscribers from different LAPPNs, considering both subscriber preferences and power transmission loss. The latter coordinates the intra- and inter-LAPPN power packet transmission to maximize the amount of energy delivered with a guaranteed fairness on user urgency. Simulation results of a two-LAPPN system are provided, which demonstrate that the proposed framework can achieve effective and efficient power dispatching in terms of the mentioned concerns, and reveal facts on ideal system capacity and how to manipulate the proportions of the two kinds of transmissions according to network status.