Wire icing existence decision-making tree model and method for judging wire icing existence and predicting wire icing duration
A decision tree, icing technology, applied in instrumentation, design optimization/simulation, calculation, etc., can solve problems such as line loss
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0080] A decision tree model for the presence or absence of icing on wires, comprising the following steps,
[0081] (1) Data preprocessing: Collect the meteorological monitoring data of all ice-covered stations in the area as the total sample Y, and classify and organize all parameters, including qualitative parameters and quantitative parameters, as follows. Table, where the qualitative parameter is the discrete variable Z i , (i=1,2,3,4,5,6), including six attributes of frost, fog, dew, rain, snow, and ice particles, the discrete variable Z i , (i=1,2,3,4,5,6) are defined as Z 1 (1, 0), Z 2 (1, 0), Z 3 (1, 0), Z 4 (1, 0), Z 5 (1, 0), Z 6 (1, 0), (1, 0) in the variable indicates whether it appears, 1 indicates it appears, and 0 indicates it does not appear; the quantitative parameter is a continuous variable X j , (j=1,2,3,4,5,6,7), including observation site altitude, daily average temperature, daily maximum temperature, daily minimum temperature, daily average relat...
Embodiment 2
[0142] A wire icing early warning method based on a decision tree and a regression model is a method for judging whether the wire is iced or not, and the steps are as follows:
[0143] (1) Determine the area to be judged;
[0144] (2) Collect frost, fog, dew, rain, snow, ice particles, observation field altitude, daily average temperature, daily maximum temperature, daily minimum temperature, daily average relative humidity, Meteorological parameter groups with thirteen attributes of daily average wind speed and daily maximum wind speed;
[0145] (3) Substitute the meteorological parameter groups that need to be predicted into the decision tree model established above, and classify the meteorological parameter groups that need to be predicted according to the optimal segmentation nodes and optimal segmentation attributes of each layer of the decision tree model established above, and then Obtain the result of judging the presence or absence of icing.
Embodiment 3
[0147] A wire icing warning method based on a decision tree and a regression model to predict the duration of wire icing, the steps are as follows:
[0148] (1) Using the prediction results of the decision tree model for the presence or absence of icing on conductors, extract the time series for determining the presence or absence of icing on conductors;
[0149] (2) Count the number of groups in which the conductors are iced twice or more consecutively in the time series. If there is a discontinuity in the time series, restart the statistics and record it as the second conductor icing process;
[0150] (3) Multiply the number of groups obtained in step 2 by the interval time of the forecast parameter group to obtain the duration of a wire icing;
[0151] If there is intermittent wire icing phenomenon in the forecast sequence, record the duration of wire icing respectively.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More