Real-time temperature prediction method and system for high-voltage cable core
A real-time temperature and high-voltage cable technology, which is applied to thermometers, thermometer applications, and measuring devices, can solve problems such as leading errors in prediction results, large thermal resistance and thermal capacity of cables and surrounding environments, etc., and achieve large leading errors and faster Effects of processing speed and accuracy improvement
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0087] Taking the single-circuit three-phase 110kV-YJLW02XLPE single-core power cable laid in trenches as an example, the cable skin temperature and the cable skin temperature corresponding to a historical data collection time are obtained through finite element simulation under random load conditions within 7 days. The historical skin temperature change rate of the cable skin temperature, the cable core current and the historical cable core current change rate of the cable core current corresponding to the historical data collection time, the sampling time is 168 hours, and the sampling time interval is set to In 12 minutes, the change curve of the cable core circuit is as follows Figure 4 As shown, the change curve of cable skin temperature and cable core temperature is shown in Figure 5 as shown, Figure 5 The upper part of the curve is the change curve of the cable core temperature. Figure 5 A curve in the middle and lower part is the change curve of the cable skin te...
Embodiment 2
[0089] The temperature prediction model is trained based on the neural network machine learning method, and the first input data processing method, the second input data processing method and a prediction method based on the thermal circuit model are compared and analyzed. The first input data processing method The method considers the current data collection time and one historical data collection time, the second input data processing method considers the current data collection time and three historical data collection time, and the distance between the current data collection time and the adjacent historical data collection time The collection time interval and the collection time interval between two adjacent historical data collection moments are both 1 hour.
[0090] Preferably, for the first input data processing method, the number of neurons in the input layer of the temperature prediction model is 4, the number of neurons in the output layer is 1, the number of neuron...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


