The paper raises the problem of insufficient quality control of the operating modes of electric locomotives, carried out by on-board systems. The problem of determining the recommended values of the traction forces is presented as an iterative determination of the setting of the traction force for several subsequent time steps on the basis of a dataset known in advance and describing the past states of the locomotive. To solve this problem, artificial recurrent neural networks (ARNN) of the long short-term memory (LSTM) architecture are used, which are insensitive to time gaps between events and the effect of forgetting. As a result of comparing the data accumulated in the course of experimental studies and comparing them with the results of simulation modeling, it was found that when the locomotive implements the recommended values of traction and braking forces, a decrease in the values of the specific energy consumption of traction of trains per trip is expected on average by 1%.
Control of Mainline Freight Electric Locomotives with Adjustment of Operating Modes
Lect. Notes in Networks, Syst.
International School on Neural Networks, Initiated by IIASS and EMFCSC ; 2022 ; St.Petersburg, Russia February 08, 2022 - February 10, 2022
Networked Control Systems for Connected and Automated Vehicles ; Chapter : 24 ; 251-259
2022-11-16
9 pages
Article/Chapter (Book)
Electronic Resource
English
America's first diesel mainline freight locomotives completed
Engineering Index Backfile | 1941
|America's first diesel mainline freight locomotives completed
Engineering Index Backfile | 1941
|America's first diesel mainline freight locomotives completed
Engineering Index Backfile | 1941
|Diesel-electric locomotives for French mainline services
Engineering Index Backfile | 1945