2 results about "Quadratic nonlinearity" patented technology

This invention discloses a reactive power optimization method for distribution networks based on probabilistic power flow sensitivity analytical transformation, belonging to the field of power system optimization and dispatching technology. The method includes: considering the impact of source-load power uncertainty on distribution network dispatching, constructing a reactive power optimization model based on chance constraints, and using probabilistic power flow sensitivity to characterize the probability distributions of voltage and branch power random variables respectively; linearizing the quadratic nonlinear branch security constraints using a regular polygon approximation method to decouple the optimization variables from the random variables; establishing analytical expressions for voltage and branch security chance constraints through affine transformation of probabilistic power flow sensitivity; and analytically transforming the nonlinear chance constraints into deterministic constraints based on probability transformation theory, transforming the original problem into a mixed-integer second-order cone programming model for solution. This invention can fully tap the potential of photovoltaic reactive power regulation, significantly improve voltage stability and reduce network losses, while also having a computational efficiency advantage over the traditional sample mean method.

The invention relates to a complex terrain wind resource assessment method, and the method comprises the steps: carrying out the downscaling of obtained 3km WRF simulation data into a 100m refined meteorological element wind field through employing CALMET, and obtaining 100m WRF / CALMET simulation data; the 100m WRF / CALMET simulation data and the 3km WRF simulation data are respectively compared with the actually measured data of the weather station, and regional wind resource evaluation is carried out; aiming at a wind speed difference found in wind energy resource evaluation, correcting a wind speed deviation by respectively using an LSTM neural network and a VMD-LSTM model, and selecting an optimal wind speed deviation correction scheme; aiming at wind direction differences found in wind energy resource evaluation, a BP neural network and a multivariate quadratic nonlinear regression model are respectively used for correcting wind direction deviation, and an optimal wind speed deviation correction scheme is selected. Compared with the prior art, the method has the advantages that the simulation precision is effectively improved on the premise of keeping certain physical consistency, a theoretical basis and technical support are provided for wind energy development of a complex terrain area, and the target of carbon emission reduction in green development of the area is achieved.