Track quality prediction method and system based on improved grey combination model
A track quality and combined model technology, applied in the field of rail transit data analysis and prediction, can solve the problems of not taking into account the different degrees of model influence, not considering the impact, and the accuracy of background value calculation needs to be improved.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0079] Aiming at the problems and shortcomings of existing track quality prediction methods, the present invention proposes an improved gray combination model track quality prediction method, which can well reflect the development trend of TQI sequence. Improvements are made on the basis of traditional gray forecasting modeling. The weight distribution coefficient is introduced into the cumulative generation process to optimize the weight distribution of the time interval. At the same time, the weight matrix is introduced to solve the model parameters. The closer the prediction time is, the greater the weight is given. The prediction accuracy has been greatly improved; the use of the integral area to optimize the solution of the background value of the difference equation has solved the problem of large errors caused by the calculation of the trapezoidal area in the past; the PSO-Elman residual correction model can reduce data fluctuations and reduce Improved unequal time int...
Embodiment 2
[0099] The invention provides a track quality prediction method based on the improved gray combination model, which has the following obvious advantages and beneficial effects compared with the existing technology:
[0100] (1) Improve on the basis of traditional gray prediction modeling, introduce the weight distribution coefficient into the cumulative generation process, optimize the weight distribution of time intervals, and introduce the weight matrix to solve the model parameters, and the closer the prediction time is, the greater the value given to the data weight, the prediction accuracy has been greatly improved.
[0101] (2) Using the integral area to optimize the solution of the background value of the difference equation solves the problem of large errors caused by the calculation of the trapezoidal area in the past, reduces the prediction error of the model, and enhances the reliability of the prediction results.
[0102] (3) The TQI sequence belongs to the time se...
Embodiment 3
[0165] like Figure 5 As shown, although the single gray model has a good linear change trend, the error between it and the actual measured value is relatively large. If the track maintenance department arranges maintenance tasks with reference to its predicted value, it will not be able to carry out targeted track maintenance operations. Using the PSO-Elman model for residual correction, it can be seen from the figure that the error of the final prediction result is lower than that of the preliminary prediction model.
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com