Accurate photovoltaic power forecasting is very important for daily dispatching planning of power grid. A short-term photovoltaic power forecasting model based on hybrid improved multivariate universe optimization(HIMVO) algorithm and support vector machine(SVM) optimization is proposed in this paper. Firstly

in order to get rid of the shortcomings that the conventional MVO algorithm easily falls into local optimization

the tent mapbased chaotic sequence is used to participate in population initialization. Then

in the updating of the position vector of the MVO algorithm

a non-linear inertia weight reduction strategy is introduced

and the differential evolution(DE) algorithm is added to carry out global search

and the HIMVO algorithm is proposed to optimize the parameters of SVM. Finally

simulation experiments are carried out in three different weather types using the proposed method. The prediction results are compared with that of the SVM and MVO-SVM methods

which verifies that the HIMVOSVM method can effectively improve the short-term photovoltaic power prediction accuracy.