Real-time control of off-lining of locomotives for energy management

A train control system including a machine learning engine and a virtual system modeling engine, wherein the train control system is configured to use machine learning for controlling one or more locomotives in a train by modeling and classifying sections of track along which the train is traveling, the train control system comprising:

• an effective weight module programmed to acquire a first set of real-time and historical data relating to grade vs. location for the section of track and a first set of corresponding structural information on each of the one or more locomotives and on non-powered rail cars of the train traveling over the section of track, and calculate an effective weight profile of the train in the section of track using the first set of real-time and historical data and the first set of corresponding structural information;
• an effective friction module programmed to acquire a second set of real-time and historical data relating to speed of the train vs. location for the section of track and a second set of corresponding structural information on each of the one or more locomotives and on non-powered rail cars of the train traveling over the section of track, and calculate an effective friction profile of the train in the section of track using the second set of real-time and historical data and the second set of corresponding structural information;
• a power availability module programmed to acquire a third set of real-time and historical data relating to a tractive effort force vs. speed and a dynamic braking force for each of the one or more locomotives traveling over the section of track, and calculate an effective power availability profile of the train in the section of track using the third set of real-time and historical data;
• the virtual system modeling engine being programmed to use the acquired data and information and the machine learning engine to train one or more virtual system models to recognize designated areas of the track where the total tractive effort force or dynamic braking force applied by all of the locomotives in the train is less than the tractive effort force or dynamic braking force, respectively, that can be provided by a subset of the available locomotives in the train for a time duration greater than the time needed to power down and power back up one or more unneeded locomotives or consists of locomotives in the train;
• memory for storing the one or more virtual system models simulated by the virtual system modeling engine, wherein each of the one or more virtual system models includes a mapping between different combinations of the first, second, and third sets of real-time and historical data and corresponding designated areas of the track;
• and
• an energy management system associated with one or more locomotives of the train and configured to off-line the one or more unneeded locomotives or consists of locomotives for the time duration based at least in part on one of the virtual system models.

A locomotive control system, comprising:

• a machine learning engine configured to receive real-time and historical data for use as training data for a virtual system modeling engine configured to model and classify sections of track along which one or more locomotives of a train are traveling, wherein
• the machine learning engine comprising:
• an effective weight module programmed to acquire a first set of real-time and historical data relating to grade vs. location for each of the sections of track and a first set of corresponding structural information on each of the one or more locomotives and on non-powered rail cars of the train traveling over each section of track, and calculate the effective weight profile of the train in each section of track using the first set of real-time and historical data and the first set of corresponding structural information;
• an effective friction module programmed to acquire a second set of real-time and historical data relating to speed of the train vs. location for each of the sections of track and a second set of corresponding structural information on each of the one or more locomotives and on non-powered rail cars of the train traveling over each section of track, and calculate the effective friction profile of the train in each section of track using the second set of real-time and historical data and the second set of corresponding structural information;
• and
• a power availability module programmed to acquire a third set of real-time and historical data relating to a tractive effort force vs. speed and a dynamic braking force for each of the one or more locomotives traveling over each section of track, and calculate the effective power availability profile of the train in each section of track using the third set of real-time and historical data;
• and
• the virtual system modeling engine being programmed to use the acquired data and information to generate one or more virtual system models identifying designated areas of the track where the total tractive effort force or dynamic braking force applied by all of the locomotives in the train is less than a tractive effort force or a dynamic braking force, respectively, that can be provided by a subset of the available locomotives in the train for a time duration greater than the time needed to power down and power back up one or more unneeded locomotives or consists of locomotives in the train;
• memory for storing the one or more virtual system models simulated by the virtual system modeling engine, wherein each of the one or more virtual system models includes a mapping between different combinations of the first, second, and third sets of real-time and historical data and corresponding designated areas of the track;
• and
• an energy management system associated with one or more locomotives of the train and configured to off-line the one or more unneeded locomotives or consists of locomotives for the time duration based at least in part on one of the virtual system models.