A method for predicting audible noise of transmission lines based on optimized bp neural network
A BP neural network and transmission line technology, which is applied in the field of audible noise prediction of transmission lines, can solve the problems of large prediction error, lack of comprehensive coverage, and prediction error of audible noise of high voltage level lines.
Image
Examples
Embodiment 1
[0095] Such as Figure 1-2 As shown, the inventive method of this example is: firstly obtain the factors that have an influence on the audible noise Y of the transmission line as input data, including: voltage X 1 , wire diameter X 2 , wire section X 3 , split number X 4 , Splitting distance X 5 , wire-to-ground distance X 6 , The distance between the wire and the measuring point X 7 , temperature X 8 , Humidity X 9 , wind speed X 10 , air pressure X 11 , altitude X 12 , background noise X 13 .
[0096] The input data contains 13 neurons, and the order of magnitude differs greatly. In order to ensure the equal status of each factor and speed up the convergence speed, the normalized preprocessing method is used to preprocess the input data and normalize the data to [-1,1 ] interval.
[0097] The ant colony algorithm assumes that there are N parameters in the network, which includes all weights and thresholds. First, sort these parameters, for parameter P i (1≤i≤N...
PUM
Login to View More Abstract
Description
Claims
Application Information
- IPC
- G06N3/08; G06Q10/04
- Inventors
- 刘蕊; 马潇