No abstract

Our results broaden the spectrum of diabetes phenotypes resulting from KCNJ11 mutations. They indicate testing for KCNJ11 mutations should be considered not only for neonatal diabetes but also for other forms of dominantly inherited diabetes with later onset, especially where these are associated with a low body mass index and low birth weight.

Abstract. In kernel-based machines, the integration of several kernels to build more flexible learning methods is a promising avenue for research. In particular, in Multiple Kernel Learning a compound kernel is build by learning a kernel that is the weighted mean of several sources. We show in this paper that the only feasible average for kernel learning is precisely the arithmetic average. We also show that three familiar means (the geometric, inverse root mean square and harmonic means) for positive real values actually generate valid kernels.

An increasing awareness of energy efficiency has led to the development of several improved converter topologies, semiconductor devices and control schemes for distributed energy resources, and, particularly, for microgrids. Recent advances in energy management systems (EMS) for microgrids have improved upon existing methods in several aspects, including prediction of power generated by photovoltaics (PV), and optimal management of electrical energy storage. However, the actual generated PV power may deviate from predictions for several reasons, such as rapid cloud changes or system faults. This paper contributes to the ongoing research on EMS control schemes by proposing a model predictive control (MPC) scheme that adapts to the difference between the actual and predicted output power of PV. The key benefit of this approach is its ability to rapidly adapt to varying operating conditions of the PV without increasing the computational burden of a typical MPC scheme. The feasibility of the scheme is demonstrated using simulations of 5 kW microgrid system compromising a 5 kW/400 Ah battery, 10 kW PV and 5 kW grid/load connection. The proposed scheme reduces variations in the state of charge (SOC) of a battery. The proposed scheme also reduces the energy taken from grid and this improvement in performance is a function of the difference between the actual and the predicted power.