Prediction device for ore discharge temperature of sinter cooler equipment
A technology of mining temperature and prediction device, which is applied in the direction of prediction, instrumentation, data processing application, etc., can solve problems such as difficulties, achieve the effect of improving prediction accuracy and reducing temperature prediction error
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment approach 1
[0042] (Structure of cooler equipment)
[0043] The basic structure of the sintering line in the system according to Embodiment 1 of the present invention is the same as that described above figure 1 is the same. The cooler facility 1 (sintering cooler facility) cools the high-temperature sintered ore solidified by the sintering facility 2 and sends it to the belt conveyor 3 for conveying to downstream facilities. refer to figure 2 right figure 1 The structure of the cooler device 1 shown is illustrated. The cooler equipment 1 is equipped with the air blower 1b, the storage container 1c, and the mining machine 1d.
[0044] The air blower 1b is set at figure 2 The air supply duct inside the cylindrical storage container 1c shown in the longitudinal sectional view of , supplies cooling air to the storage container 1c. The sintered ore in the container 1c is transported by the belt conveyor 3( figure 1 ) is cooled for the target heat-resistant temperature.
[0045] Acco...
Embodiment approach 2
[0178] Next, refer to Figure 11 Embodiment 2 of the present invention will be described. Due to the characteristics of the cooler equipment that rotates as described above, periodic temperature prediction errors that are synchronized with the rotation speed of the cooler occur. Therefore, in this embodiment, a recurrent neural network (recurrent neural network) capable of predicting time-series changes in temperature prediction errors is employed. The recurrent neural network is a network structure that takes time-series information into consideration by combining the intermediate layer at the previous time with the input at the next time for learning.
[0179] In the error learning calculation unit 4e according to the second embodiment of the present invention, the recurrent neural network is made to learn the relationship between the past temperature prediction error and the current temperature prediction error. As a learning method of the recurrent neural network, common...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


