Accurate forecasts of electrical substations are mandatory for the efficiency of the Advanced Distribution Automation functions in distribution systems. The paper describes the design of a class of machine-learning models, namely neural networks, for the load forecasts of medium-voltage/low-voltage substations. We focus on the methodology of neural network model design in order to obtain a model that has the best achievable predictive ability given the available data. Variable selection and model selection are applied to electrical load forecasts to ensure an optimal generalization capacity of the neural network model. Real measurements collected in French distribution systems are used to validate our study. The results show that the neural network-based models outperform the time series models and that the design methodology guarantees the best generalization ability of the neural network model for the load forecasting purpose based on the same data.Index Terms-Model design, machine learning, neural network, short-term load forecast, variable selection, virtual leave-one-out.
This article presents a project designed to increase the monetary value of photovoltaic (PV) solar production for residential applications. To contribute to developing new functionalities for this type of PV system and an efficient control system for optimising its operation, this article explains how the proposed system could contract to provide ancillary services, particularly the supply of active power services. This provision of service by a PV-based system for domestic applications, not currently available, has prompted a market design proposal related to the distribution system. The mathematical model for calculating the system's optimal operation (sources, load and exchanges of power with the grid) results in a linear mix integer optimisation problem in which the objective is to maximise the profits achieved by taking part in the electricity market. Our approach is illustrated in a case study. PV producers could gain by taking part in the markets for balancing power or ancillary services despite the negative impact on profit of several types of uncertainty, notably the intermittent nature of the PV source.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.