Probabilistic load flow with versatile non-Gaussian power injections

•A complete and well-defined probabilistic load flow procedure is presented.•Gaussian Mixture Model is a key point to model monitored injections.•Standard distributions are also taken into account to model non-monitored injections.•The proposal is presented in such level of detail that could be easily implementable as a planning tool.

A probabilistic load flow distinguished by the versatility in the way in which input data can be provided is presented. The main contribution of the proposal involves taking advantage of the data available as well as completing any missing information. This enables the proposal to be applied at any voltage level, even in medium-voltage networks where there is a glaring lack of systematic data collection. The use of the Gaussian Mixture Model is also a key feature of the proposed solution, and determinant in the final solution. Not only does the detailed and thorough analysis through numerous tests demonstrate the good performance of the proposed procedure, but it also confirms the accuracy of the results.