Patent Application
20250136157
The present disclosure relates to selective deployment of locomotives within a consist for providing motive force to a train. More specifically, the present disclosure relates to an energy-management system of a train selectively activating or deactivating locomotives within a consist based on real-time operating conditions during a mission.
A group of locomotives, called a consist, can provide motive force for propelling a train along a route during a trip, or mission. A lead locomotive contains controls for managing behavior of the other locomotives in the consist via a communications network. Some locomotives may be configured to be individually addressable, such that the lead locomotive can provide independent commands using direct control following known protocols. Other locomotives in the consist may require group addressing via a trainline command.
Traditionally, the locomotives were powered equally. When a train engineer changed the throttle in the lead locomotive, throttles in each locomotive changed the same. To conserve fuel during a mission, however, control options now exist for setting throttles in locomotives separately to account for operational differences between these locomotives. With these options, when the throttle notch is selected in the lead locomotive, the throttle in each addressable locomotive may be set to an optimal position for providing the required power and tractive effort for the consist while enhancing fuel efficiency. These throttle options, however, do not typically account for the operating environment over the particular route being traversed.
Other control features in a consist may consider restrictions along a route to help improve fuel efficiency. Speed limits, slope changes, and noise or emission limits along the route and payload size and engine efficiencies within the train can all impact energy demands on the consist significantly. Therefore, an electronic control system within the train typically includes an energy-management system intended to automatically control the consist to manage fuel usage, speed, and other parameters. In one example, the energy-management system evaluates data relating to operating parameters for the train and characteristics of a route in real time using machine-learning algorithms or artificial intelligence. With this evaluation, the energy-management system can predict demands on the consist as the train progresses and modify throttle activity to enhance efficiency. These real-time systems do not, however, consider the activation or deactivation of selective locomotives within the consist during the mission in real time based on the predictive demands.
Other energy-management systems are a priori, or plan-based. The route of a mission is evaluated in advance to form a trip plan. Portions of the route may be identified where changes to a throttle algorithm should be made, whether for enhancing energy efficiency or for adjusting to the terrain or other operating conditions. When the train reaches those predetermined locations or situations along the route, the trip plan triggers changes to the tractive force, including possible activation or deactivation of certain locomotives within the consist.
One plan-based approach for activating and deactivating locomotives of a consist is described in U.S. Pat. No. 9,079,589 (“the '589 patent”). In the '589 patent, propulsion is needed from inactive, or isolated, locomotives during a trip because unexpected events have caused the train to slow in contravention to a trip plan. An energy-management system, in conjunction with an isolation-control system, sends instructions to activate at least one of the inactive locomotives to help the train accelerate to a plan speed, after which the locomotives are switched back to their isolated state. Implemented within a plan-based energy-management system, the '589 patent exemplifies the difficulties that can arise when actual situations on a route deviate from a trip plan. The '589 patent additionally does not contemplate a real-time energy-management system deploying locomotives in a consist based on predicted operating parameters of the train as it crosses an upcoming distance of a path.
Examples of the present disclosure are directed to overcoming deficiencies of such systems.
In an aspect of the present disclosure, a computer-implemented method includes identifying addressable locomotives within a consist of a train, the addressable locomotives being individually addressable by a control system on the train and receiving, during movement of the train along a path, current operating parameters of the train and characteristics of the path across an upcoming distance. The method further includes predicting, based at least in part on the current operating parameters and the characteristics, baseline operational metrics for the train across the upcoming distance under a baseline condition, the baseline condition being a baseline set of the addressable locomotives will be deployed, and predicting in real time, based at least in part on the current operating parameters and the characteristics, first operational metrics for the train across the upcoming distance under a first condition, the first condition being a first set of the addressable locomotives will be deployed. A baseline benefit is calculated from the baseline operational metrics, a first benefit is calculated from the first operational metrics, and the baseline benefit with the first benefit are compared. When the first benefit exceeds the baseline benefit, the method entails selecting the first set of addressable locomotives as an updated set of addressable locomotives and causing the updated set of addressable locomotives to be deployed for the train.
In another aspect of the present disclosure, a control system for a consist of locomotives includes a consist storage module configured to store status data regarding the locomotives, the status data indicating availability of the locomotives for providing motive force along a route, a prediction module, and a consist management module. The prediction module is configured to predict in real time, based at least in part on current operating parameters of a train moving along a path and characteristics of the path across an upcoming distance, baseline operational metrics expected for the train across the upcoming distance under a baseline condition, where the baseline condition is that a group of the locomotives will be deployed. The prediction module is also configured to predict in real time, based at least in part on the current operating parameters and the characteristics, alternate operational metrics expected for the train across the upcoming distance under an alternate condition, where the alternate condition is that a subset of the group of the locomotives will be deployed. The consist management module is configured to calculate a baseline benefit from the baseline operational metrics, calculate an alternate benefit from the alternate operational metrics, and compare the baseline benefit with the alternate benefit. When the alternate benefit exceeds the baseline benefit, the consist management module assigns the subset of the group of the locomotives as an updated assignment of the locomotives and causes the updated assignment of the locomotives to be deployed for the train.
In yet another aspect of the present disclosure, a consist of locomotives for a train includes a lead locomotive containing, at least in part, a control system for the train: a first locomotive individually addressable by the control system via a communication network: a second locomotive individually addressable by the control system via the communication network; and an energy-management system, at least partially contained within control system. The energy-management system is configured to receive, during movement of the train along a path, current operating parameters of the train and characteristics of the path across an upcoming distance. The energy-management system additionally predicts in real time, based at least in part on the current operating parameters and the characteristics, baseline operational metrics for the train across the upcoming distance under a baseline condition that the first locomotive and the second locomotive will be deployed, and predicts in real time, based at least in part on the current operating parameters and the characteristics, alternate operational metrics for the train across the upcoming distance under an alternate condition that the first locomotive will be deployed and the second locomotive will not be deployed. Further, the energy-management system also calculates a baseline benefit from the baseline operational metrics, calculates an alternate benefit from the alternate operational metrics, and compares the baseline benefit with the alternate benefit. When the alternate benefit exceeds the baseline benefit, the baseline set is assigned without the second locomotive as an updated set of the locomotives. The consist also includes a locomotive control system, within at least the second locomotive, configured to cause the second locomotive to operate in an isolation mode in response to the assignment of the updated set by the energy-management system.
The detailed description references the accompanying figures. In the figures, the left-most digit of a reference number identifies the figure in which the reference number first appears. The same reference numbers indicate similar or identical items.
Consistent with the principles of the present disclosure, an energy-management system and method for deploying locomotives within a consist of a train provides enhanced efficiency for running a train across an upcoming distance along a route. The energy-management system assesses in real time the operating parameters of the train and characteristics of the route using machine-learning algorithms. Portions of the energy-management system predict, based at least in part on the current operating parameters and the route characteristics, baseline operational metrics for the train across the upcoming distance under a baseline condition that a baseline set of locomotives will be deployed. The system also predicts in real time, based at least in part on the current operating parameters and the route characteristics, different operational metrics for operation of the train under a condition that a different set of locomotives will be deployed. From those predictions, a consist management module calculates a baseline benefit when the baseline set of locomotives are deployed and a value of the benefit when the different set of locomotives are deployed. The consist management module compares these operating benefits of the train derived from the predictions and identifies a number of locomotives to place in an isolation mode to provide efficient operation of the train. The number identified is provided to a driving strategy module for consideration in maneuvering the train over the upcoming distance. The following describes several examples for carrying out the principles of this disclosure.
Train 102 includes several rail cars or units, which are either powered units or non-powered units, linked together. “Powered units” refers to rail cars that are capable of self-propulsion, such as locomotives. “Nonpowered units” refers to rail cars that are incapable of self-propulsion, but which may otherwise receive electric power for other services, such as freight cars or passenger cars. In
The lead locomotive 110 serves as a center for management and control for other powered units. The term “lead” in lead locomotive 110 refers in this context to a powered unit that is designated for primary control of other locomotives within consist 118. While illustrated at the front of train 102, lead locomotive 110 may be located at any position. A propulsion system 120 within lead locomotive 110 provides power for delivering tractive effort to help cause train 102 to move along wheels 104 as well as braking effort to reduce or stop the movement. Propulsion system 120 includes electric and/or mechanical devices and components, such as diesel or electric engines, electric generators, and traction motors, used to provide tractive effort that propels train 102. Braking effort may arise from dynamic braking, rheostatic braking, frictional braking, or other means in conjunction with propulsion system 120 as known to those of ordinary skill in the field.
The powered units within consist 118 other than lead locomotive 110 may be referred to as trailing units or trailing locomotives. Consist 118 of
Additionally, the powered locomotives within consist 118 include at least some portion or form of control system 122 for electronically monitoring, managing, controlling, or executing various operations within consist 118. Typically, lead locomotive 110 contains a control system 122, whether implemented as a discrete control system within that lead locomotive or a portion of a collective control system distributed across each locomotive in consist 118, that acts as a central controller for operations within itself and the other locomotives within consist 118. In some examples, control system 122 resides at least within lead locomotive 110 within an onboard controller and embodies one or more computer processors that include a means for operating and/or controlling train 102 based on information obtained from sensors monitoring various train components, from data stored in memory, from communications received external to the locomotive, and other sources. Control system 122 may include computer-readable media in the form of a memory, a secondary storage device, a processor, and any other components for executing instructions stored on the computer-readable media. The memory may include a non-transitory computer-readable medium, such as RAM, ROM, FLASH memory, CD ROM, magnetic devices (e.g., disks, tape, etc.), and/or other types of memory. Various other circuits may be associated with a controller such as power supply circuitry, signal conditioning circuitry, solenoid driver circuitry, and other types of circuitry. Different modules and functions of control system 122 are discussed in more detail below.
Train 102 also includes communication system 150 at least for wirelessly transmitting and receiving electrical signals with offboard remote electronics external to the train. communication system 150 is communicatively coupled with control system 122 in at least one or more of the powered locomotives and includes one or more antennas for facilitating wireless communication. Communication system 150 may include electronic components for facilitating communication within a railroad network under protocols and standards known to those of ordinary skill in the field. Examples include a cellular connection following TDMA, 3G, 4G, 5G, or other protocol, a wireless local area network (WLAN), WiMax (Worldwide Interoperability for Microwave Access), satellite communications, and the like. In one example, communication system 150 connects wirelessly with a back-office system 152, which may include mainframe computers and other storage devices for housing information relevant to the operation, routing, and scheduling of train 102 along route 106.
As depicted in
In some examples, communication system 150 in conjunction with control system 122 coordinates operations of various components and subsystems, such as propulsion system 120 within each of the powered units of consist 118. The communication system 150 within lead locomotive 110 may be referred to as a lead communication unit that initiates a message to back-office system 152 containing information on an operational state of lead locomotive 110, such as a throttle setting, a brake setting, readiness for dynamic braking, the tripping of a circuit breaker, or other operational characteristics. Communication system 150 within lead locomotive 110 may communicate with the communication system 150 within other powered units of consist 118, which are interconnected via a network connection 124. In one example, network connection 124 includes a net port and jumper cable that extends along train 102 and between each control system 122 of the powered units. The network connection 124 may be a multiple unit cable (MU cable), an Electrically Controlled Pneumatic brake line (ECPB), a fiber optic cable, other cable configuration, or a wireless connection. Similarly, network connection 124 may connect and communicatively couple propulsion system 120 of the locomotives in consist 118. High speed communication networks may be employed in consist 118 using a common ethernet communication bus. The ethernet communication signals may be superimposed over other digital communication signals carried by some of the wires in MU cables. MU cables provide trainline communications between the locomotives in a consist, including operational commands such as throttle control settings. The networking standard employed in the consist may include an inter-consist communication (ICC) router and an ethernet bridge device on each car.
In some examples, network connection 124 includes several channels over which network data is communicated with each channel representing a different pathway for network data. In some examples, one or more channels within network connection 124 may communicate a trainline command accessible by all rail cars within train 102. In this situation, the information conveyed within the trainline command is general to all rail cars and is generally not specific to a particular locomotive. In other examples, control system 122 within one or more of the locomotives, such as addressable locomotives 114 within consist 118, have the capacity to be directly and uniquely addressed by lead locomotive 110. For instance, control system 122 within addressable locomotives 114 may abide by LCCM protocols and enable direct communication from control system 122 in lead locomotive 110. On the other hand, some powered units in train 102, such as nonaddressable locomotive 112, may not have the capacity to be directly addressed by lead locomotive 110 and must rely on trainline communications. These communication options are provided as examples and are not limiting to the present disclosure. Various options for communicating between railcars are available and within the knowledge of those of ordinary skill in the field.
As embodied within lead locomotive 110, control system 122 also includes an automatic train operation (ATO) system that has an energy-management system 130 embedded within or in electronic communication at least with propulsion system 120. Energy-management system 130 is an automated module or software engine within control system 122 configured to generate command signals to optimize control of train 102 under currently detected circumstances. In this context, a module or engine refers to hardware, software, or combinations of hardware and software configured to store and execute computer-readable instructions for a particular task. Control system 122 may be configured to receive input signals from a variety of sensors and other inputs indicative of operating parameters of the train and to generate output signals for achieving optimum control of the train. The automated energy management system may be configured to make these output signals or requests based on a variety of measured operational parameters, track conditions, freight loads, trip plans, and predetermined maps or other stored data with a goal of improving availability, safety, timeliness, overall fuel economy and emissions output for the entire train. For example, energy-management system 130 may generate command signals for automatically controlling throttle, braking, and or other aspects of train 102 associated with propulsion system 120 based on the current operating parameters and health condition of the locomotive, along with the current mission, location, and topography, in order to achieve optimum performance while accomplishing mission goals and objectives. Mission goals and objectives may include achieving performance goals (e.g., performance levels, efficiency levels, etc.), adhering to schedules, and obeying laws (e.g., speed limits). The energy-management system may also control or modify operating parameters for train 102 to maximize fuel efficiency and/or to minimize emissions. As discussed further below, decisions and command signals generated within energy-management system 130 may involve and be shared with other locomotives within consist 118.
As indicated in
In some examples, energy-management system 130 also includes a prediction module 134, which is a machine-learning modeling engine configured to create one or more models of train operation in a variety of conditions. The machine-learning modeling engine may be based on training data focused on historical operational data acquired by various sensors associated with one or more locomotives during actual train runs. The sensors may include brake temperature sensors, exhaust sensors, fuel level sensors, pressure sensors, knock sensors, reductant level or temperature sensors, generator power output sensors, voltage or current sensors, speed sensors, motion detection sensors, location sensors, wheel temperature or bearing temperature sensors, or any other sensor known in the art for monitoring various train operational parameters.
During historical runs of train 102 or similar trains over a variety of conditions, i.e., different grades, speeds, climates, payloads, fuels, locomotives, etc., the sensors (not shown) provide operational data that is fed as training data to prediction module 134. The operational data can include a state of a particular locomotive in a consist, a representation or state of an environment surrounding the consist, including behavior of other trains or locomotives on the same or interconnected tracks in the same geographical area, and commands, instructions, or other communications received from other entities. The training data may also include data indicative of actions taken by a train operator, or directly or indirectly resulting from actions taken by the train operator, under a large variety of operating conditions, and on trains with the same or different equipment, different operational characteristics, and different parameters. For example, a learning system according to prediction module 134 can be trained to learn how a train such as train 102 and experienced human engineers respond to different inputs under various operating conditions, such as during the automatic implementation of train control commands by trip optimizer programs, positive train control (PTC) algorithms, and automatic train operations (ATO), during extreme weather conditions, during emergency conditions caused by other train traffic or equipment failure on the train, while approaching and maneuvering in train yards, and under other train operating conditions. The methods and algorithms for training a learning system such as prediction module 134 with historical operational data fed from sensors during actual runs of a train such as train 102 are within the knowledge of those of ordinary skill in the art.
Using its trained learning system, prediction module 134 within energy-management system 130 can be configured to generate predicted data for operation of train 102.
In coordination with path characteristics 160 such as for upcoming distance 108 for train 102, prediction module 134 may also be configured to receive certain “real-time” operating parameters of the train, namely, operating parameters 162. “Real time” in this context, generally means concurrently with train 102 traveling along route 106 and approaching upcoming distance 108. Real-time path characteristics 160 of the train include any sensed or measured parameter relating to the functioning of train 102 along upcoming distance 108, which may include data from at least the same sensors considered in building the learning system within prediction module 134. That is, the sensors and their associated measured parameters may include brake temperature sensors, exhaust sensors, fuel level sensors, pressure sensors, knock sensors, reductant level or temperature sensors, generator power output sensors, voltage or current sensors, speed sensors, motion detection sensors, location sensors, wheel temperature or bearing temperature sensors, or any other sensor known in the art for monitoring various train operational parameters. Using real-time operating parameters 162 of train 102, prediction module 134 may consider the current situation for the trip, which will tend to correlate to the future parameters when traversing upcoming distance 108 of a few miles. Thus, sensed operating parameters 162 of fuel level, speed, emissions, power output, etc. for a current or real-time situation for the particular locomotives and payload in train 102 may provide a starting point for prediction module 134 for predictive analysis of the behavior of consist 118 and train 102 in the future along upcoming distance 108.
Among the computations performable within energy-management system 130, prediction module 134 may evaluate future operational metrics for train 102 based on a varying number and identity of addressable locomotives 114 being active. In this context, being active or activated refers to the locomotive providing at least some tractive effort for train 102. In contrast, being inactive or deactivated refers to the locomotive operating in an isolation mode or in an idle condition where tractive effort is not provided for train 102. Thus, while consist 118 may have a plurality of addressable locomotives 114, such as the three depicted in the example of
Referring to
Receiving different potential sets of locomotives, prediction module 134 in some examples evaluates those different sets using its learning machine with historical data, path characteristics 160, and operating parameters 162 to predict future operational metrics. In particular, prediction module 134 may calculate baseline metrics 170 for the baseline set of locomotives 164, first metrics 172 for the first set of locomotives 166, and second metrics 174 for the second set of locomotives 168. As discussed above, these operational metrics may indicate a wide variety of features for consist 118 and train 102, such as fuel usage, fuel amount, energy consumption, exhaust emissions, engine temperature, engine diagnostics, train speed, noise emissions, among others.
In accordance with the principles of the present disclosure, energy-management system 130 additionally includes a consist management module 136 configured to manage deployment of selected ones of the locomotives within consist 118. Typically, energy-management system 130 is hardware, software, or combinations of hardware and software configured to store and execute computer-readable instructions for determining a number of locomotives within consist 118 to activate or deactivate based on the processing of prediction module 134. In general, consist management module 136 receives results of prediction analysis from prediction module 134, such as baseline metrics 170, first metrics 172, and second metrics 174 predicted for train 102 along upcoming distance 108, and analyzes and compares the results. The analysis and comparison can entail any basis or mathematical function appropriate for the situation. In some examples, consist management module 136 may compare results of baseline metrics 170 for baseline set of locomotives 164 derived by prediction module 134 with first metrics 172 for first set of locomotives 166, or baseline metrics 170 may be compared with second metrics 174 for second set of locomotives 168 derived by prediction module 134. Consist management module 136 may be configured to identify from the comparison whether baseline set of locomotives 164, first set of locomotives 166, or second set of locomotives 168 provides more benefit to operation of train 102, represented respectively by baseline benefit 176, first benefit 178, and second benefit 180. The benefit may be any operational parameter deemed significant to operation of train 102, typically to provide energy efficiency, such as fuel savings, decreased emissions, or lower noise, as discussed further below with respect to
To provide further detail with respect to calculating a benefit of deploying certain ones of addressable locomotives 114,
As indicated in
Similarly, as shown in
In another example, benefit calculator 202 may consider a force available/required ratio 218, which is a ratio of the amount of motive force available from a particular configuration of addressable locomotives 114 to an amount of motive force required to maintain speed approximate to the permitted speed limit through upcoming distance 108. The amount of motive force available from a configuration may be viewed as a force fraction, i.e., a fraction of the total motive force possible from addressable locomotives 114 that the particular configuration may provide. Therefore, if train 102 has a consist 118 with lead locomotive 110, nonaddressable locomotive 112, first addressable locomotive 114A, second addressable locomotive 114B, and third addressable locomotive 114C providing train system with 100% of the collective motive force, and each of the locomotives is capable of providing 20% of the collective motive force, then a configuration of addressable locomotive 114A and addressable locomotive 114C would have a force fraction of (20+20) for the addressable locomotives 114 and (20+20) for lead locomotive 110 and nonaddressable locomotive 112, i.e., 0.8 or 80%. Consequently, if consist 118 only needs to provide 70% of its capacity to maintain the speed limit through upcoming distance 108, then force available/required ratio 218 might be 0.8/0.7, and consist management module 136 might conclude that at least two of addressable locomotives 114 maybe pulled into an isolation state to operate through upcoming distance 108 most efficiently. Other variations of force available/required ratio 218 for evaluating an excess or lack of motive force will be apparent from this discussion to those of ordinary skill in the art.
As shown in
Finally, as shown in
Turning from the structure and organization of train 102 to a method involving those structures,
As reflected in
In a second step 302 within method 300, consist storage module 132 sets, or stores in a memory, a number N and/or an identification of addressable locomotives 114 to be inactive. In some examples, N corresponds to the number of addressable locomotives currently inactive in consist 118 when method 300 begins. In other examples, consist storage module 132 may select one or more of addressable locomotives 114 to be inactive to begin the method, such as selecting in memory first addressable locomotive 114A to be in an isolation state. Coincident with the action, consist storage module 132 may also set, or store in a memory, a number N−1 of addressable locomotives to be inactive. Thus, if N for step 302 had been a value of one with first addressable locomotive 114A being deactivated, then step 304 would indicate that zero (N−1) addressable locomotives would be inactive. Likewise, for step 306, consist storage module 132 also sets, or stores in a memory, a number N+1 (i.e., two) of the addressable locomotives to be inactive. For example, consist storage module 132 could identify that first addressable locomotive 114A and second addressable locomotive 114B would be selected to enter an isolation state.
Based on the setting of addressable locomotives to be placed into an isolation state in steps 302, 304, and 306, prediction module 134 of energy-management system 130 then assesses the future performance of train 102 in each of those scenarios. Thus, as shown in
Following the activity of prediction module 134, benefit calculator 202 within consist management module 136 calculates and evaluates benefits from each of the predictions from steps 308, 310, and 312. As indicated in
As generally discussed above with respect to
In comparing the benefits flowing from the different result sets, consist management module 136 determines the most advantageous deployment of locomotives for propelling the train X miles ahead, i.e., across upcoming distance 108. In step 320, consist management module 136 evaluates whether the benefits from result set A are better than the benefits from result set C. In addition, as shown in
In step 324, consist management module 136 evaluates whether the benefits from result set B (N+1 locomotives inactive) are better than the benefits from result set C by a predetermined threshold T2. Threshold T2 may be the same or different from threshold T1 and also provides a form of hysteresis in adding more locomotives to the set of inactive ones. If the benefits from result set B exceed the benefits from result set C by the threshold T2, consist management module 136 will conclude that the N locomotives currently set as being inactive should be changed to N+1 locomotives (result set B) at step 326. If at step 324, consist management module 136 determines that the benefits from result set B do not exceed the benefits from result set C, or do not exceed above the threshold T2, then consist management module 136 will not make a change to the N locomotives currently set as being inactive (step 328).
In step 330, consist management module 136 will select or otherwise pass the setting for inactive locomotives determined by one of steps 322, 326, and 328. This setting will become the value of N for recurring method 300. If result set C indicates the highest benefit for train 102, then the value of N will remain unchanged from what it had been within step 308. If consist management module 136 determines that result set A (steps 320 and 322) or result set B (steps 324 and 326) provides superior benefits X miles ahead, then a new value (either N−1 or N+1) will become the new value for N. As shown in
In addition to returning to consist storage module 132, the new value of N and any characteristic data of those N inactive locomotives may also be passed to driving strategy module 138. Driving strategy module 138 can then process its throttle algorithm and other controls for consist 118 based on the number of locomotives available for providing tractive effort as determined by
As shown in activated locomotives graph 420, consist management module 136 selectively determines that different numbers of locomotives may be employed over the course of the journey to provide most efficient operation. Specifically, the plot of activated engines 422 depicts a quantity of addressable locomotives activated during the run, between 1.0 and 3.0. Slightly after the beginning of the simulation, consist management module 136 determines at first location 424 that 1.0 locomotives are most efficient for passing over the downward elevation shown in elevation plot 410 while most closely meeting the speed limit 432 correspondingly shown in simulated results 430. Perhaps due to an increase in grade shown in elevation plot 410, consist management module 136 next concludes that an increased number of active locomotives to 3.0 would be required to achieve the greatest benefit at second location 426 in activated locomotives graph 420. As the elevation in elevation plot 410 then again slopes downwardly, the number of active locomotives in the run determined by consist management module 136 is decreased to 1.0 at third location 428. Similar changes to the number of locomotives during the run can be traced by comparing elevation plot 410, activated locomotives graph 420, and simulated results 430.
The graph of simulated results 430 depicts differences in speed and ultimate performance attained under the simulation. The plot of all-activated speed 434 shows speed for simulated train 102 with all available locomotives being activated, while the plot of partial-deactivated speed 436 shows the speed across the run with some of the locomotives being deactivated as indicated by activated locomotives graph 420. The top of the simulated results 430 summarizes the performance advantages that the simulation concluded when employing method 300. Specifically, the run saved 49 gallons of fuel while 7.2 minutes of travel time were lost due to the intermittent decrease in active locomotives.
Those of ordinary skill in the field will appreciate that the principles of this disclosure are not limited to the specific examples discussed or illustrated in the figures. For example, while the system and method has been discussed in the context of changing a consist by activating or deactivating one locomotive, other sequences of analyzing change to the consist are feasible and could be implemented. Moreover, the process of comparing benefits between groups of locomotives may also take into account specific characteristics of each locomotive and consider the value of activating or deactivating each locomotive individually.
The present disclosure provides systems and methods to assess in real time the operating parameters of a train and characteristics of a route for the train using machine-learning algorithms. A consist management module within the energy-management system evaluates predictions of future performance by the train when different numbers of locomotives within a consist are inactive. The consist management module calculates and compares operating benefits of the train derived from the predictions with respect to travel across an upcoming distance and identifies a number of locomotives to place in an isolation mode to provide efficient operation of the train. The number identified is provided to a driving strategy module for consideration in maneuvering the train over the upcoming distance.
As noted above with respect to
In the examples of the present disclosure, the energy-management system using real-time analysis with a machine-learning engine can effectively determine the proper amount of tractive effort to apply for a given situation as the train moves continuously along a route 106. As such, the train need not rely on a predetermined plan for adjusting activation of locomotives during a mission in case situations arise during the mission that deviate from the plan. Instead, the reaction of the energy-management system a posteriori to situations along upcoming distance 108 considering current operating parameters of the train, characteristics of the path, and historical data of previous runs from which the machine-learning engine has been trained can yield more nimble reactions to changes along the route. As a result, train 102 can run more efficiently, minimizing fuel consumption, exhaust emissions, and noise, as desired, without losing significant travel time for the mission.
Unless explicitly excluded, the use of the singular to describe a component, structure, or operation does not exclude the use of plural such components, structures, or operations or their equivalents. As used herein, the word “or” refers to any possible permutation of a set of items. For example, the phrase “A, B, or C” refers to at least one of A, B, C, or any combination thereof, such as any of: A: B; C: A and B: A and C: B and C: A, B, and C; or multiple of any item such as A and A: B, B, and C: A, A, B, C, and C: etc.
Terms of approximation are meant to include ranges of values that do not change the function or result of the disclosed structure or process. For instance, the term “about” generally refers to a range of numeric values that one of skill in the art would consider equivalent to the recited numeric value or having the same function or result. Similarly, the antecedent “substantially” means largely, but not wholly, the same form, manner or degree, and the particular element will have a range of configurations as a person of ordinary skill in the art would consider as having the same function or result.
While aspects of the present disclosure have been particularly shown and described with reference to the embodiments above, it will be understood by those skilled in the art that various additional embodiments may be contemplated by the modification of the disclosed systems and methods without departing from the spirit and scope of what is disclosed. Such embodiments should be understood to fall within the scope of the present disclosure as determined based upon the claims and any equivalents thereof.
