System and method for catalyst component temperature control using machine learning

Abstract

Systems, methods, and apparatus for controlling catalyst component temperature using machine learning techniques are provided. A processing circuit receives an indication of a hydrocarbon injection event in a vehicle. The processing circuitry receives a plurality of parameters related to vehicle operation within a first time range associated with a hydrocarbon injection event. The processing circuitry predicts temperature data for a second time range after the first time range using a model based on the plurality of parameters, the temperature data relating to an aftertreatment system of the vehicle. In response to the temperature data being above the threshold, the processing circuitry calculates a second injection rate based on the predicted temperature for the second time range using the model, the second injection rate being lower than the first injection rate associated with the hydrocarbon injection event. The processing circuit commands the injector to inject hydrocarbons based on the second injection rate.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of and priority to U.S. Provisional Application No. 63 / 442,696, filed February 1, 2023, entitled “Systems and Methods for Catalyst Component Temperature Control Using Machine Learning,” the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to managing the temperature of catalyst components within an aftertreatment system. More specifically, the present disclosure relates to using machine learning to control the temperature of catalyst components. Background Art

[0004] The exhaust aftertreatment system may include various catalysts (e.g., selective catalytic reduction systems, diesel oxidation catalysts, etc.), and, among other components or systems, a reductant injection system for introducing a reductant (e.g., urea, diesel exhaust fluid (DEF), ammonia solution, etc.) to reduce nitrogen oxide (NOx) emissions from the system. With emission regulations expected to become more stringent in the coming years, it is desirable to be able to effectively reduce emissions of certain exhaust components. In addition, particulate matter (e.g., soot) formed during the combustion process within the engine can be filtered by one or more components of the exhaust aftertreatment system (e.g., a diesel particulate filter (DPF)). However, over time, particulate matter may accumulate in / on one or more components of the aftertreatment system, which may adversely affect the performance of one or more components of the aftertreatment system. The accumulated particulate matter can be removed by a regeneration event that increases the temperature of one or more components, thereby burning off the accumulated particulate matter. Summary of the Invention

[0005] One embodiment relates to a computing system. The computing system includes processing circuitry comprising one or more memory devices coupled to one or more processors, the one or more memory devices configured to store instructions that, when executed by the one or more processors, cause the processing circuitry to: receive an indication of a hydrocarbon injection event in a vehicle; receive a plurality of parameters related to vehicle operation within a first time frame associated with the hydrocarbon injection event; predict temperature data for a second time frame subsequent to the first time frame using a model based on the plurality of parameters, the temperature data relating to an aftertreatment system of the vehicle; in response to the temperature data being greater than a threshold, calculate a second injection rate based on the predicted temperature for the second time frame using the model, the second injection rate being lower than a first injection rate associated with the hydrocarbon injection event; and, in response to the calculation, command an injector to inject hydrocarbons based on the second injection rate.

[0006] In some embodiments, the plurality of parameters includes at least one of the following: an inlet temperature of a catalyst component of the aftertreatment system, an outlet temperature of the catalyst component of the aftertreatment system, an exhaust flow rate from an engine coupled to the aftertreatment system, or a first injection rate of the engine. In some embodiments, to calculate the second injection rate, the instructions, when executed by one or more processors, further cause the processing circuitry to: determine a temperature difference between the temperature data and a target temperature; convert the temperature difference into a hydrocarbon amount; convert the hydrocarbon amount into a third injection rate; and calculate the second injection rate based on the difference between the first injection rate and the third injection rate.

[0007] Another embodiment relates to a method comprising: receiving, by processing circuitry including one or more memory devices coupled to one or more processors, an indication of a hydrocarbon injection event in a vehicle; receiving, by the processing circuitry, a plurality of parameters related to vehicle operation within a first time frame associated with the hydrocarbon injection event; predicting, by the processing circuitry and using a model based on the plurality of parameters, temperature data for a second time frame subsequent to the first time frame, the temperature data relating to an aftertreatment system of the vehicle; calculating, by the processing circuitry, a second injection rate based on the predicted temperature for the second time frame using the model in response to the temperature data being greater than a threshold, the second injection rate being lower than the first injection rate associated with the hydrocarbon injection event; and commanding, by the processing circuitry, in response to the calculation, an injector to inject hydrocarbons based on the second injection rate.

[0008] In some arrangements, the method includes: receiving, by the processing circuit, a target temperature associated with an outlet of a catalyst component for a hydrocarbon injection event using a first injection rate, the target temperature being determined or adjusted based on vehicle operation; determining, by the processing circuit, that the target temperature is at or above a predetermined threshold; and commanding, by the processing circuit, the injector to inject hydrocarbons using a second injection rate based on the target temperature being at or above the predetermined threshold.

[0009] Another embodiment relates to a computing system. The computing system includes one or more processors and one or more memory devices coupled to the one or more processors. The one or more memory devices are coupled to the one or more processors and store instructions that, when executed by the one or more processors, cause the one or more processors to perform the following operations: receiving at least one input to a model, including a plurality of parameters related to vehicle operation, during a first time range during which a first outlet temperature of a catalyst component of an aftertreatment system changes; using the model, correlating the plurality of parameters for the first time range with a pattern associated with a predetermined plurality of parameters, wherein the pattern indicates at least one change in the first outlet temperature of the catalyst component over time based on the plurality of parameters; receiving an indication of a second outlet temperature of the catalyst component for a second time range subsequent to the first time range based on a correlation between the plurality of parameters and the pattern; and, in response to the second outlet temperature being above a threshold, commanding at least one component of the vehicle within the first time range to reduce the second outlet temperature of the catalyst component to within the threshold within the second time range.

[0010] These and other features and their organization and mode of operation will become apparent from the following detailed description in conjunction with the accompanying drawings. Many specific details are provided to provide a thorough understanding of the embodiments of the disclosed subject matter. The features described in the disclosed subject matter can be combined in any suitable manner in one or more embodiments and / or implementations. In this regard, one or more features of one aspect of the present invention can be combined with one or more features of different aspects of the present invention. In addition, additional features may occur in certain embodiments and / or implementations, which may not be present in all embodiments or implementations. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 is a schematic diagram of an example system according to an example embodiment.

[0012] Figure 2 According to an example embodiment Figure 1 Schematic diagram of an example computing system.

[0013] Figure 3 is an example temperature profile of a catalyst component according to an example embodiment.

[0014] Figure 4A -B is a diagram of an example neural network architecture according to an example embodiment.

[0015] Figure 5A -C is a schematic diagram of an example electronic unit for controlling hydrocarbon injection using a model according to an example embodiment.

[0016] Figure 6A-B is a graph illustrating an example of prediction performance using different numbers of hidden units according to an example embodiment.

[0017] Figure 7 is a graph of example temperature adjustments according to an example embodiment.

[0018] Figure 8 is a graph of example fueling adjustments according to an example embodiment.

[0019] Figure 9A -G is a schematic diagram of an example embodiment of a model according to an example embodiment.

[0020] Figure 10 is a diagram of an example embodiment of a neural network architecture according to an example embodiment.

[0021] Figure 11 is a flow chart of a method of controlling the temperature of a component of an aftertreatment system (eg, a catalyst component) using machine learning, according to an example embodiment. DETAILED DESCRIPTION

[0022] The following is a more detailed description of various concepts and implementations of methods, apparatus, and systems for catalyst component temperature control using machine learning. The various concepts introduced above and discussed in greater detail below can be implemented in any number of ways, as the concepts described are not limited to any particular implementation. Examples of specific implementations and applications are provided primarily for illustrative purposes.

[0023] Referring generally to the accompanying drawings, various embodiments disclosed herein relate to systems, devices, and methods for using machine learning to control and / or manage the temperature of an exhaust aftertreatment system, particularly a catalyst component. The exhaust aftertreatment system is fluidly coupled to an engine and is configured to treat exhaust gas from the engine to reduce potentially environmentally harmful emissions (e.g., greenhouse gases, particulate matter, NOx, etc.). The aftertreatment system can include various catalysts (e.g., a selective catalytic reduction system, a diesel oxidation catalyst, etc.), a reductant injection system that introduces a reductant (e.g., urea, DEF, ammonia solution, etc.), and other components for reducing certain emissions from the engine (e.g., NOx emissions). In addition, the aftertreatment system can include a filter, such as a diesel particulate filter (DPF) in a diesel engine aftertreatment system, which is configured to filter particulates (e.g., soot) from the engine. As particulate matter accumulates in the DPF, a regeneration event can be triggered / initiated to remove the particulate matter (e.g., soot) from the DPF. In various arrangements, hydrocarbons (e.g., fuel) are injected into one or more cylinders of the engine (e.g., post-injection) or downstream of the engine via dedicated hydrocarbon injectors. The injected hydrocarbons can react with the diesel oxidation catalyst (DOC) to cause or provide an exothermic reaction and increase the temperature of the exhaust gas from the DOC outlet. The hydrocarbons can be injected into the system until the temperature at the DOC outlet (or the temperature of the DPF) reaches a target temperature to burn off the particulate matter. However, in some systems, controlling the injection of hydrocarbons (e.g., based on a current temperature, such as the temperature of the DPF) can result in a temperature overshoot or excessive temperature increase, which can cause damage to various components of the aftertreatment system. Factors that cause temperature overshoots may include at least one of the following: a varying exhaust flow rate carrying high temperature exhaust gas, the reactivity of the DOC with the hydrocarbons, and a delay in sensing the temperature change.

[0024] The systems and methods of the technical solutions discussed herein include a computing system in communication with a vehicle and a database. In various arrangements, the computing system is configured to train a machine learning model to predict the temperature of at least one catalyst component (e.g., a DOC, a selective catalytic reduction (SCR) catalyst, a three-way catalytic device, etc.) within an aftertreatment system. The computing system trains the model based on historical data and / or real-time data from the vehicle fleet to predict temperature data at a subsequent time point or time period (e.g., the temperature 10 seconds, 20 seconds, 30 seconds, etc. after the prediction is performed). The computing system inputs / provides vehicle parameters (e.g., sensor data, commands, etc.) that are related to changes in the catalyst component temperature to train a model (e.g., a machine learning model) to predict the temperature of one or more catalyst components or components of the aftertreatment system within a subsequent time range.

[0025] After training the model, the computing system can send the model to the vehicle (e.g., an engine control unit (ECM) or controller). The controller of the vehicle can use the model to perform predictions. In some cases, the computing system is configured to perform predictions and send commands / instructions to vehicles (e.g., multiple vehicles in a fleet). The computing system performs predictions continuously, periodically (e.g., at predetermined time intervals), or intermittently based on indications of hydrocarbon events (e.g., regeneration events of the DPF). Before a temperature overshoot is detected based on the prediction, the computing system determines the amount or rate of hydrocarbon injection to adjust or command an injector (e.g., a fuel injector, a dedicated hydrocarbon injector located downstream of the engine, etc.). For example, the computing system commands the injector based on the determined hydrocarbon injection to minimize, avoid, or potentially avoid temperature overshoot of the aftertreatment system, thereby minimizing problems such as runaway exothermic reactions caused by the presence of soot, thermal damage to aftertreatment system components, poisoning of catalyst components (e.g., SCR catalyst components), or component aging caused by excessive temperature increases. The computing system can command other components of the vehicle 108 to control the temperature below a desired temperature threshold before a temperature overshoot occurs, such as controlling a heater, engine operation (e.g., speed, torque, etc.), an exhaust gas recirculation (EGR) system, and other components of the vehicle that may affect the aftertreatment system temperature.

[0026] Now refer to Figure 1 , according to an exemplary embodiment, a system 100 is shown that includes a computing system 104 coupled to a vehicle 108 (in some embodiments, a fleet of vehicles) and a database 106. The vehicle 108 includes an engine 12, a powertrain 26 (sometimes referred to as a powertrain system), a controller 28, an operational input / output (I / O) 30 (sometimes referred to as an operational I / O device), a monitoring system 32, a telematics unit 34, and an aftertreatment system 38. The vehicle 108 can be any type of on-road or off-road vehicle, including but not limited to long-haul trucks, medium-duty trucks (e.g., pickup trucks, etc.), sedans, coupes, tanks, aircraft, ships, and any other type of vehicle. In some embodiments, the vehicle 108 can also be a stationary device (e.g., a generator or generator set, etc.). Based on these configurations, various other types of components can also be included in the system, such as a transmission, one or more gearboxes, pumps, actuators, etc.

[0027] The engine 12 can be any type of internal combustion engine. Thus, the engine 12 can be a gasoline engine, a natural gas engine, or a diesel engine, can be part of a hybrid engine system (e.g., a combination of an internal combustion engine and an electric motor), and / or any other suitable engine. Here, the engine 12 is a diesel-powered compression ignition engine. The engine 12 includes a first cylinder 14, a second cylinder 16, a third cylinder 18, a fourth cylinder 20, a fifth cylinder 22, and a sixth cylinder 24 (collectively referred to herein as "cylinders 14-24"). It should be understood that although Figure 1 Six cylinders are shown in FIG. 1 , but the number of cylinders may vary depending on system configuration and requirements. Cylinders 14 - 24 may be any type of cylinder suitable for the engine in which they are located (e.g., sized and shaped to accommodate pistons).

[0028] Engine 12 is coupled to at least one injector, shown as fuel injector 36 and hydrocarbon injector 37. In some cases, injector 36 is directly coupled to one or more cylinders 14-24 to supply or provide fuel (e.g., hydrocarbon) (i.e., each cylinder 14-24 is equipped with one or more injectors 36). In contrast, injector 37 is located downstream of engine 12, such as at the exhaust pipe of engine 12. In this case, injector 37 is configured to supply hydrocarbon directly to aftertreatment system 38 for exothermic reaction with DOC 40. Injectors 36 and 37 can be coupled to a fuel source or a hydrocarbon source. Injector 37 can be located elsewhere within vehicle 108 depending on vehicle configuration or requirements. In addition, multiple injectors 37 may be included. Injectors 36 and 37 receive commands from controller 28 to control injection events. In some embodiments, injectors 36 and / or 37 receive commands from computing system 104 to control injection events. In response to receiving the commands, injectors 36 and / or 37 are configured to initiate or terminate the injection events accordingly. Additionally, injectors 36 and / or 37 receive commands from controller 28 to configure the injection rate for injecting hydrocarbons into DOC 40. By controlling injection events or configuring injection rates, one or more suitable machine learning techniques may be used to minimize temperature overshoot relative to a target temperature.

[0029] An aftertreatment system 38 is in fluid communication with the engine 12 to receive the exhaust gas. The aftertreatment system 38 includes a diesel oxidation catalyst (DOC) 40, a diesel particulate filter (DPF) 42, a reductant delivery system including a diesel exhaust fluid (DEF) dispenser 44, a selective catalytic reduction (SCR) 46 (e.g., an SCR system), and an ammonia slip catalyst (ASC) 48. In some cases, the aftertreatment system 38 includes a heater for heating one or more components. In some cases, the aftertreatment system 38 is coupled to an injector 37 and is configured to receive hydrocarbons from the injector 37. The spatial location of the injector 37 may vary in other embodiments, such as in other aftertreatment systems having additional or fewer components (e.g., a dual SCR system, a multi-branch aftertreatment system, etc.).

[0030] The DOC 40 is configured to receive exhaust gas from the engine 12 and oxidize hydrocarbons and carbon monoxide in the exhaust gas. The DOC 40 is configured to react with the hydrocarbons to provide an exothermic reaction, thereby increasing the temperature of the exhaust gas passing through the DOC 40. The DPF 42 is arranged or located downstream of the DOC 40 and is configured to remove particulate matter, such as soot, from the exhaust gas flow. The DPF 42 includes an inlet (for receiving the exhaust gas) and an outlet (from which the exhaust gas is discharged after substantially filtering out the particulate matter and / or converting the particulate matter into carbon dioxide). The DPF 42 can be regenerated by increasing the temperature of the DPF 42 to a desired target temperature or by increasing the temperature of the exhaust gas at the outlet of the DOC 40 to a desired target temperature to remove particulate matter or soot. For example, the temperature of the exhaust gas is increased by the exothermic reaction of the hydrocarbons with the DOC 40.

[0031] The aftertreatment system 38 may also include a reductant delivery system (e.g., DEF injector 44) that may include a decomposition chamber (e.g., a decomposition reactor, a reactor conduit, a decomposition tube, a reactor tube, etc.) to convert the reductant into ammonia. The reductant may be, for example, urea, DEF, Urea water solution (UWS), urea water solution (e.g., AUS32, etc.) and other similar fluids. DEF is added to the exhaust gas flow to assist in catalytic reduction. The reductant can be injected upstream of the SCR catalyst component through the DEF injector 44 so that the SCR catalyst component receives a mixture of reductant and exhaust gas. The reductant droplets then undergo evaporation, pyrolysis and hydrolysis processes in the decomposition chamber, SCR catalyst component and / or exhaust pipe system to form gaseous ammonia and exit the aftertreatment system 38. The aftertreatment system 38 may also include an oxidation catalyst (e.g., DOC 40) fluidically coupled to the exhaust pipe system to oxidize hydrocarbons and carbon monoxide in the exhaust gas. In order to properly assist in such reduction, the DOC 40 may need to be at a certain operating temperature. In some embodiments, the certain operating temperature is between 200-500°C. In other embodiments, the certain operating temperature is the temperature at which the conversion efficiency of the DOC 40 exceeds a predetermined threshold (e.g., the conversion of HC into less harmful compounds, i.e., HC conversion efficiency).

[0032] SCR 46 is configured to assist in reducing NOx emissions by accelerating the NOx reduction process between ammonia and NOx in the exhaust gas, reducing it to diatomic nitrogen, water and / or carbon dioxide. SCR 46 can be a system that includes at least one catalyst component. If the SCR catalyst component does not reach or exceed a certain temperature, the acceleration of the NOx reduction process will be limited and SCR 46 will not be able to operate at the necessary efficiency level to meet regulatory requirements. In some embodiments, the certain temperature is 250-300°C. The SCR catalyst component can be made of a combination of inactive material and active catalyst, so that the inactive material (e.g., ceramic metal) guides the exhaust gas to the active catalyst, which is any material suitable for catalytic reduction (e.g., base metal oxides such as vanadium, molybdenum, tungsten, or precious metals such as platinum). It should be understood that the SCR catalyst component can be formed or constructed from a variety of different materials, all of which are considered to be within the scope of the present disclosure.

[0033] The ASC 48 can be any of a variety of flow-through catalysts, such as an ammonia oxidation (AMOX) catalyst, which is configured to react with ammonia to produce primarily nitrogen. The ASC 48 is configured to remove ammonia that has passed through or escaped the SCR 46 without reacting with NOx in the exhaust. In some cases, the aftertreatment system 38 can operate with or without the ASC 48. In addition, although in Figure 1 In the figure, the ASC 48 is shown as a separate unit from the SCR 46, but in some implementations, the ASC 48 can be integrated with the SCR 46, for example, the ASC 48 and the SCR 46 can be located in the same housing. According to the present disclosure, the SCR 46 and the ASC 48 are placed in series, with the SCR 46 located before the ASC 48.

[0034] Because the aftertreatment system 38 treats the exhaust gas before it is released into the atmosphere, much of the particulate matter or chemicals that are treated or removed from the exhaust gas can accumulate in the aftertreatment system over time. For example, soot filtered from the exhaust gas by the DPF 42 can accumulate on the DPF 42 over time. Similarly, sulfur particles that may remain in the exhaust gas due to incomplete combustion of the fuel can accumulate in the SCR 46 and reduce the effectiveness of the SCR catalyst components. In addition, DEF that undergoes incomplete pyrolysis upstream of the catalyst can accumulate and form deposits on downstream components of the aftertreatment system 38. However, this accumulation on the components of the aftertreatment system 38 (and the subsequent reduction in effectiveness) can be reversible. In other words, by increasing the temperature of the exhaust gas flowing through the aftertreatment system, soot, sulfur, and DEF deposits can be substantially removed from the DPF 42 and SCR 46 to restore performance (e.g., in the case of the SCR, the efficiency of converting NOx to N2 and other compounds). These removal processes are referred to as regeneration events and may be performed on the DPF 42 , the SCR 46 , or any other deposit-forming component in the aftertreatment system.

[0035] In the illustrated example, the vehicle 108 includes a telematics unit 34. The telematics unit 34 can be configured as any type of telematics control unit. Thus, the telematics unit 34 may include, but is not limited to, one or more storage devices for storing tracking data, one or more electronic processing units for processing tracking data, and a communication interface for facilitating data exchange between the telematics unit 34 and one or more remote devices (e.g., the computing system 104 or the database 106). In this regard, the communication interface can be configured as any type of mobile communication interface or protocol, including but not limited to Wi-Fi, WiMax, Internet, radio, Bluetooth, Zigbee, satellite, radio, cellular, GSM, GPRS, LTE, and the like. The telematics unit 34 may also include a communication interface for communicating with the controller 28 of the vehicle 108. The communication interface used to communicate with the controller 28 may include any type and number of wired and wireless protocols (e.g., any standard under IEEE 802, etc.). For example, a wired connection may include a serial cable, a fiber optic cable, an SAE J1939 bus, a CAT5 cable, or any other form of wired connection. In contrast, wireless connections may include the Internet, Wi-Fi, Bluetooth, Zigbee, cellular, radio, and the like. In one embodiment, a controller area network (CAN) bus, including any number of wired and wireless connections, provides for the exchange of signals, information, and / or data between the controller 28 and the telematics unit 34. In other embodiments, a local area network (LAN), a wide area network (WAN), or an external computer (e.g., using the Internet through an Internet service provider) may provide, assist, and support communications between the telematics unit 34 and the controller 28. In yet another embodiment, communications between the telematics unit 34 and the controller 28 are conducted via the Unified Diagnostic Services (UDS) protocol. All of these variations are intended to fall within the spirit and scope of the present disclosure.

[0036] The powertrain 26 of the vehicle 108 may include the engine 12 (and potentially other components) coupled to a transmission of the vehicle 108. The transmission may be operatively coupled to a drive shaft, which may be operatively coupled to a differential that transfers the power output from the engine 12 to a final drive (e.g., wheels of the vehicle 108, or tracks in some off-highway applications) to help propel the vehicle 108. The powertrain 26 may be controlled by a controller 28 to propel the vehicle 108, for example, in response to instructions, commands, or actions of an operator. In some cases, the powertrain 26 may receive instructions from an operational I / O 30 coupled to or in electrical communication with the controller 28.

[0037] In some implementations, the powertrain system 26 may include an electric motor (not shown) and / or a motor-generator (not shown) configured to generate and provide electrical energy to one or more vehicle accessories (hence the term generator) and at least partially drive the vehicle. In some implementations, the motor-generator may be operably coupled to the engine 12 and the transmission such that, in these implementations, the vehicle 108 is configured as a hybrid vehicle (e.g., a combination of an internal combustion engine and an electric motor or electric motor / generator). In some implementations, the motor-generator may receive power from an energy source (e.g., a battery) that provides input energy to output usable work or energy, in some cases alone or in combination with the engine 12 to drive the vehicle 108. In other implementations, energy may be transferred to charge the battery or any electrically powered accessories within the vehicle. The battery may be charged through regenerative braking, a fuel cell, or a combination of both.

[0038] The powertrain 26 is configured to monitor or collect data associated with the operation of various components of the vehicle 108 (e.g., the transmission, drive shafts, differential, engine 12, etc.) and transmit / transmit the data to one or more devices within the network 102 using the telematics unit 34. The data includes one or more parameters, such as the configuration / settings of the various components, the requested energy, energy output / consumption, the requested torque, torque output, or position of the various components, as well as other information about the vehicle 108.

[0039] The operations I / O 30 may be communicatively coupled to the controller 28 so that information may be exchanged between the controller 28 and the operations I / O 30, wherein the information may relate to one or more components of the vehicle 108 or other components of the system 100, or to the judgment of the controller 28 (described below). The operations I / O 30 may enable an operator of the vehicle 108 to communicate with the controller 28 and the controller 28. Figure 1 The operational I / O 30 may communicate with one or more components of the vehicle 108. For example, the operational I / O 30 may include, but is not limited to, an interactive display, a touch screen device, one or more buttons and switches, a voice command receiver, etc. The operational I / O 30 may display a graphical user interface (GUI) to an operator (e.g., a user or customer) of the vehicle 108. The operational I / O 30 may provide one or more indications or notifications to the operator, such as a malfunction indicator light (MIL), etc.

[0040] The monitoring system 32 of the vehicle 108 is coupled to one or more components of the vehicle 108, such as the engine 12, the powertrain 26, the controller 28, the operating I / O 30, the injectors 36, the injectors 37, and / or the aftertreatment system 38. The monitoring system 32 is configured to monitor various component parameters, such as the temperature of the catalyst components (e.g., inlet temperature, bed temperature, or outlet temperature), the exhaust flow rate (e.g., sometimes referred to as mass flow rate) from the engine 12, the injection rate of the injectors 36 and 37, and the like. The monitoring system 32 monitors the temperature of the catalyst components using temperature sensors located upstream, downstream, or at the respective catalyst components. The monitoring system 32 receives temperature data from the temperature sensors. The monitoring system 32 monitors the exhaust flow rate based on at least one of pressure data sensed by a pressure sensor or flow rate data sensed by a flow rate sensor. The monitoring system 32 is configured to calculate the exhaust flow rate from the engine or the exhaust flow rate at certain catalyst components based on the pressure data, where, for example, a higher pressure corresponds to a higher flow rate. Monitoring system 32 monitors the injection rate based on commands received by injectors 36 and 37 from controller 28. In some cases, monitoring system 32 uses a gas sensor located downstream of injector 37 to monitor the injection rate of injector 37. In this case, the gas sensor is configured to sense the amount or rate of hydrocarbons injected into the exhaust gas flow.

[0041] Controller 28 is coupled to potential components such as engine 12, powertrain 26, operating I / O 30, monitoring system 32, telematics unit 34, injectors 36 and 37, and aftertreatment system 38, and is constructed or configured to at least partially control these systems / devices. Communication between components can be carried out via any number of wired or wireless connections. For example, a wired connection can include a serial cable, fiber optic cable, CAT5 cable, or any other form of wired connection. In contrast, a wireless connection can include the Internet, Wi-Fi, cellular, radio, etc. In one embodiment, a CAN bus is used to exchange signals, information, and / or data. The CAN bus includes any number of wired and wireless connections. In this regard, controller 28 can be configured to receive signals, information, data, etc. (e.g., engine operating parameter signals and / or aftertreatment system operating parameter signals) from sensors (e.g., exhaust flow rate sensors, speed sensors, pressure sensors, temperature sensors, and / or any other sensors associated with engine 12 or aftertreatment system 38).

[0042] In some arrangements, controller 28 may be configured to receive data monitored by one or more components of vehicle 108 (e.g., powertrain 26, monitoring system 32, etc.) and transmit signals, information, data, etc. to one or more devices within network 102 (e.g., computing system 104) using telematics unit 34. In some other embodiments, a telematics unit may not be included, and controller 28 may include a network interface configured to enable remote communication via network 102. In some arrangements, controller 28 is configured to use the monitored data as input to a machine learning model (sometimes collectively referred to as a model). Controller 28 receives the model from computing system 104. The model is trained by computing system 104 or another device remote from vehicle 108. In some cases, controller 28 may use the model as part of model training. Controller 28 is configured to use the model to predict the temperature of one or more components of aftertreatment system 38. For example, using one or more parameters or monitored data as input to the model, controller 28 is configured to determine the temperature at (or near) the outlet of DOC 40, DPF 42, etc., at a later timeframe.

[0043] In various arrangements, controller 28 is configured to receive a temperature prediction regarding one or more components of aftertreatment system 38 from computing system 104. In this case, controller 28 transmits data monitored by powertrain 26, monitoring system 32, and other components of vehicle 108 to computing system 104 for processing. After processing the data, controller 28 receives the prediction from computing system 104. Based on the prediction, controller 28 is configured to determine an injection rate for injector 36 and / or injector 37 to minimize or avoid a temperature overshoot of aftertreatment system 38. In some cases, based on the prediction, controller 28 is configured to determine when to initiate or terminate hydrocarbon injection from injector 37.

[0044] In various implementations, controller 28 is configured to receive commands from computing system 104. Controller 28 may delegate processing tasks to computing system 104, such as predicting the temperature of components of aftertreatment system 38 during or after a regeneration event and determining an injection rate for injector 36 and / or injector 37 based on the prediction. Thus, controller 28 may receive the determined injection rate from computing system 104 to adjust the operation of injector 36 and / or injector 37.

[0045] because Figure 1Components are shown as being embodied in vehicle 108, and controller 28 may be configured as one or more electronic control units (ECUs) or ECMs. As described herein, in some cases, controller 28 may use a model trained by computing system 104 to predict the temperature of aftertreatment system 38. Based on this prediction, controller 28 commands injector 36 and / or injector 37 to initiate or terminate an injection event of hydrocarbons or configure the injection rate of hydrocarbons. Computing system 104 may perform similar or additional features as controller 28. The functionality and structure of controller 28 or computing system 104 are at least Figure 2-10 A more detailed description is given in .

[0046] The controller 28 can be configured to directly or indirectly transmit information, such as monitoring data (e.g., the inlet temperature of the DOC 40, the outlet temperature of the DOC 40, the injection rate of the injector 36 and / or the injector 37, the exhaust flow rate from the engine 12, etc.), and receive information from the computing system 104. The controller 28 can be configured to conduct V2X (e.g., vehicle-to-everything) communications (e.g., directly communicate with the computing system 104) through the telematics unit 34. The telematics unit 34 can also include a communication interface for communicating with the controller 28 of the vehicle 108. The communication interface for communicating with the controller 28 can include any type and number of wired and wireless protocols (e.g., any standard under IEEE 802, etc.).

[0047] As described above, in some implementations, the controller 28 can be configured to conduct V2X communications without the use of a telematics unit. For example, the controller 28 can be configured to exchange information with the computing system 104 via a wide area network that communicates directly with the vehicle 108. In other embodiments, the controller 28 can communicate with the computing system 104 via the telematics unit 34.

[0048] like Figure 1As shown, computing system 104 communicates with vehicle 108 and / or at least one database 106 via network 102. Network 102 can be any type of communication protocol that facilitates information exchange between vehicle 108 and computing system 104. In this regard, network 102 can communicatively couple vehicle 108 with computing system 104. In some cases, network 102 refers to the interconnection of devices that are remote or local to one another. In this regard, devices within network 102 include computing system 104, database 106, and vehicle 108, as well as other devices connected to network 102. In one embodiment, network 102 can be configured as a wireless network. In this regard, vehicle 108 can wirelessly send and receive data to and from computing system 104. The wireless network can be any type of wireless network, such as Wi-Fi, WiMax, Geographic Information System (GIS), the Internet, radio frequency, Bluetooth, Zigbee, satellite, radio, cellular, Global System for Mobile Communications (GSM), General Packet Radio Service (GPRS), Long Term Evolution (LTE), optical signals, and the like. In an alternative embodiment, the network 102 may be configured as a wired network or a combination of wired and wireless protocols. For example, the controller 28 and / or the telematics unit 34 of the vehicle 108 may be electrically, communicatively, and / or operatively coupled to the network 102 via a fiber optic cable to selectively send and receive data wirelessly to the computing system 104.

[0049] In some embodiments, vehicle 108 may be part of a fleet. The vehicles in the fleet may have similar or different configurations and structures than vehicle 108. Each vehicle in the fleet may be coupled to computing system 104. Alternatively, only certain vehicles in the fleet may be coupled to the remote computing system. In some embodiments, operators, administrators, and the like of the fleet may be coupled to computing system 104 (e.g., via one or more computing devices, such as tablets, mobile smartphones, desktop computers, etc.).

[0050] In some implementations, computing system 104 is part of vehicle 108, such as a component of vehicle 108. In some other implementations, computing system 104 is a computing system remote from vehicle 108. Computing system 104 is configured to receive data related to the fleet (or individual vehicles 108) directly from each vehicle or from a remote database, source, or data repository (e.g., from database 106). In one embodiment, information about each vehicle in the fleet is maintained by another remote computing system (not shown). Computing system 104 may periodically receive fleet information from the remote computing system. In another embodiment, computing system 104 periodically receives information directly from one or more vehicles in the fleet via network 102. The data related to the fleet may be part of the group data. Computing system 104 is configured to store the data received from network 102 in memory (e.g., local storage) or remotely in database 106. Computing system 104 is able to access the data stored in database 106. Computing system 104 may also update the data in database 106.

[0051] The computing system 104 is configured to manage one or more machine learning models for at least one vehicle in the fleet (e.g., vehicle 108). The computing system 104 is configured to manage models for a single vehicle 108 or a fleet. Figure 2 As further described in detail in

[15] , computing system 104 is configured to train at least one model for vehicle 108. The trained model is configured to predict the temperature of components within aftertreatment system 38. Computing system 104 is configured to provide the trained model to controller 28 of vehicle 108 or to update an existing model stored on vehicle 108. In some cases, computing system 104 may use the model and provide output (e.g., predicted values) from the model to vehicle 108 (e.g., controller 28). In this case, computing system 104 receives parameters (e.g., input data) from vehicle 108, uses the parameters as input to the model, and transmits the prediction to controller 28. In some cases, computing system 104 determines an injection rate (or injection timing) to control injector 36 and / or injector 37 based on the predicted temperature and transmits the injection rate to controller 28 to control injector 36 and / or injector 37. Various data received, processed / used, or transmitted by computing system 104 may be stored in database 106.

[0052] The database 106 can be remote from the computing system 104. In some implementations, the database 106 is part of the computing system 104 or accessible by the vehicle 108. The database 106 can be referred to as a model database and is configured to store models generated, trained, or used by the computing system 104 or the vehicle 108 and other devices within the network 102. The database 106 can be accessed by the computing system 104 or the vehicle 108 with permission. The models or data stored in the database 106 can be updated, retrieved, or replaced by the computing system 104 or the vehicle 108.

[0053] Now refer to Figure 2 , showing Figure 1 1 is a schematic diagram of a computing system 104, which is a more detailed view and example implementation. The computing system 104 may be operated, owned, managed, controlled, and / or associated with a provider entity (not shown). The provider entity may be a device manufacturer (e.g., an engine manufacturer, an aftertreatment system manufacturer, a controller manufacturer, etc.), an analytics provider, a fleet operator, and / or other entities. Thus, the provider entity may own other devices coupled to the computing system 104 via the network 102, which may perform a prediction of the catalyst component temperature of the aftertreatment system 38 based on historical data stored in the database 106 or from the vehicle 108.

[0054] like Figure 2 As shown, computing system 104 includes one or more circuits and at least one communication interface 216. Computing system 104 can be communicatively coupled to vehicle 108, database 106, or other remote devices / components (e.g., a fleet or client device) via network 102. In various optional implementations, computing system 104 can include operations similarly executed on vehicle 108, such as by controller 28 configured to receive data or information from one or more components of vehicle 108. In some arrangements, database 106 can be part of computing system 104 and / or accessible by vehicle 108 (e.g., via telematics unit 34).

[0055] Still refer to Figure 2, a controller system 200 of the computing system 104 is shown, which includes processing circuitry 202, state circuitry 208, model circuitry 210, prediction circuitry 212, injection circuitry 214, and communication interface 216. In one implementation, the components of the computing system 104 are combined into a single unit. In another implementation, one or more components may be geographically dispersed. In this regard, the various components of the computing system 104 discussed below may be dispersed in separate devices or components of the computing system 104. In some implementations, the controller system 200 may correspond to or be part of a controller 28 embedded in the vehicle 108 so as to communicate directly with the injector 36 and / or injector 37 and other components of the vehicle 108. In some other implementations, the controller system 200 is remote from the vehicle 108 and is configured to communicate with the controller 28 via the network 102 using the communication interface 216.

[0056] The communication interface 216 may include any combination of wired and / or wireless interfaces (e.g., jacks, antennas, transmitters, receivers, transceivers, wired terminals) for communicating data with various systems, devices, or networks. The communication interface 216 can be configured to communicate over a local area network or a wide area network (e.g., the Internet) and can use various communication protocols (e.g., IP, LON, Bluetooth, ZigBee, radio, cellular, near field communication). In addition, the computing system 104 can use the communication interface 216 to communicate with other vehicles in one or more vehicles in the fleet. As described above, the computing system 104 can collect information, data, or parameters from various vehicles for training one or more models. The communication interface 216 is configured to receive or obtain data from vehicles within the fleet or from the database 106 that stores historical data for one or more vehicles.

[0057] In one implementation, the state circuit 208, the model circuit 210, the prediction circuit 212, or the injection circuit 214, as well as other circuits for temperature prediction or injection configuration, can be embodied as a machine or computer readable medium storing instructions executable by a processor (e.g., processor 206). The computer readable medium can include code that can be written in any programming language, including but not limited to Java or similar languages ​​and any conventional procedural programming language, such as the "C" programming language or similar programming languages. The computer readable program code can be executed on one processor or multiple processors. In the latter case, the remote processors can be connected to each other via any type of network (e.g., CAN bus, etc.).

[0058] In another implementation, state circuit 208, model circuit 210, prediction circuit 212, or injection circuit 214 may be embodied as a hardware unit, such as an electronic unit. Thus, state circuit 208, model circuit 210, prediction circuit 212, or injection circuit 214 may be embodied as one or more circuit components, including, but not limited to, processing circuitry, network interfaces, peripherals, input devices, output devices, sensors, and the like. In some implementations, one or more circuits of computing system 104 may take the form of one or more analog circuits, electronic circuits (e.g., integrated circuits (ICs), discrete circuits, system-on-a-chip (SOC) circuits, microcontrollers, etc.), telecommunications circuits, hybrid circuits, and any other type of "circuitry." In this regard, one or more circuits may include any type of component used to perform or assist in implementing the operations described herein. For example, the circuits described herein may include one or more transistors, logic gates (e.g., NAND, AND, NOR, OR, XOR, NOT, XNOR, etc.), resistors, multiplexers, registers, capacitors, inductors, diodes, wiring, and the like. One or more circuits may also include programmable hardware devices, such as field programmable gate arrays, programmable array logic, programmable logic devices, and the like. The one or more circuits may include one or more storage devices for storing instructions that can be executed by the processors of the one or more circuits. The one or more storage devices and the processor may have the same definitions provided below with respect to storage device 204 and processor 206. In some hardware unit configurations, as described above, the one or more circuits may be geographically dispersed across different locations in computing system 104. Alternatively, as shown, the one or more circuits may be embodied in or within a single unit / housing, which is shown as computing system 104.

[0059] In the example shown, the computing system 104 includes a processing circuit 202 having a processor 206 and a storage device 204. The processing circuit 202 can be configured to execute or implement the instructions, commands, and / or control processes described herein with respect to at least the state circuit 208, the model circuit 210, the prediction circuit 212, or the injection circuit 214. The depicted configuration represents one or more circuits as instructions stored in a machine or computer-readable medium. However, as described above, this illustration is not intended to be limiting, as the present disclosure contemplates other implementations in which one or more circuits can be configured as hardware units. All such combinations and variations are intended to fall within the scope of the present disclosure.

[0060] Processor 206 can be implemented as one or more processors, such as one or more application-specific integrated circuits (ASICs), one or more field-programmable gate arrays (FPGAs), a digital signal processor (DSP), a set of processing components, or other suitable electronic processing components. One or more processors can be shared by multiple circuits (e.g., state circuit 208, model circuit 210, prediction circuit 212, or injection circuit 214 can include or otherwise share the same processor, which, in some example implementations, can execute instructions stored or otherwise accessed through different memory areas). Alternatively, or in addition, one or more processors can be configured to execute or otherwise perform certain operations independently of one or more coprocessors. In other example implementations, two or more processors can be coupled via a bus to enable independent, parallel, pipelined, or multi-threaded instruction execution. All of these variations are intended to fall within the scope of the present disclosure. Storage device 204 (e.g., RAM, ROM, flash memory, hard disk storage, etc.) can store data and / or computer code to assist with the various processes described herein. Storage device 204 can be communicatively coupled to processor 206 to provide computer code or instructions to processor 206 for performing at least some of the processes described herein. In addition, the storage device 204 can be or include tangible, non-transitory volatile memory or non-volatile memory. Thus, the storage device 204 can include database components, object code components, script components, or any other type of information structure to support the various activities and information structures described herein.

[0061] One or more circuits of the computing system 104 can communicate with each other. The one or more circuits can perform the features or operations described herein to predict and / or determine the temperature of the aftertreatment system 38 and other components of the vehicle 108. In addition, the one or more circuits can perform the features or operations described herein to determine the injection rate (e.g., hydrocarbon injection rate) for operating the injector 37. The one or more circuits can be configured to operate independently of each other or simultaneously. The computing system 104 can include additional or alternative one or more circuits configured to perform these features or operations to manage the temperature of the aftertreatment system 38 and minimize or prevent the temperature from rising above a desired target level.

[0062] State circuit 208 is configured or constructed to determine or identify parameters, states, characteristics, etc. of one or more components of vehicle 108 before the temperature of DPF 42 overshoots. A temperature overshoot refers to a temperature increase exceeding a desired target temperature (e.g., associated with the temperature near the outlet of DOC 40 or other catalyst component). The parameter, state, or characteristic may refer to data of one or more components of vehicle 108 that is associated with the temperature increase of DPF 42. The parameters are predefined or configured by an administrator of computing system 104. In some cases, state circuit 208 uses at least one model (e.g., a machine learning model) to identify one or more parameters associated with the temperature overshoot. The associated parameters refer to parameters that directly (or in some cases, indirectly) affect the temperature increase of DPF 42, thereby causing the temperature overshoot.

[0063] State circuit 208 is configured to collect, receive, or acquire parameters from one or more components of vehicle 108. These parameters may include, but are not limited to, injection rate, exhaust flow rate, and temperature upstream, downstream, or at certain components of aftertreatment system 38 (e.g., a catalyst component or filter). State circuit 208 is configured to receive injection rate information from controller 28 based on commands provided to injector 36 or sensor data from a gas sensor located downstream of injector 37. State circuit 208 is configured to receive exhaust flow rate information from at least one flow rate sensor located downstream of engine 12. In some cases, state circuit 208 is configured to determine exhaust flow rate based on pressure data from one or more pressure sensors in aftertreatment system 38. In various arrangements, state circuit 208 is configured to receive temperature data from one or more temperature sensors located upstream, at, or downstream of one or more components of aftertreatment system 38. For simplicity, state circuit 208 is configured to collect the temperature at the inlet of DOC 40 and the temperature at the outlet of DOC 40, although temperature data may also be collected from other components.

[0064] Although the temperature overshoot discussed herein is related to the DPF 42, other components may be similarly described and used in connection with temperature overshoot, such as the SCR 46, the ASC 48, etc. The state circuit 208 is configured to obtain data from the vehicle 108 in real time (e.g., in response to the controller 28 receiving the data) or periodically (e.g., when the controller 28 uploads the data to the database 106). In some cases, the state circuit 208 is configured to obtain data from the database 106 for training a model. The state circuit 208 collects data for the model circuit 210 to train one or more models for predicting temperature overshoot.

[0065] As described above, state circuit 208 (and other components of controller system 200) can be part of controller 28 embedded in vehicle 108. State circuit 208 is configured to continuously acquire or monitor parameters before or during operation. The acquired parameters can be fed into or input into at least one model trained by model circuit 210 to predict the temperature of at least DPF 42 at a subsequent time point or time range. In some cases, state circuit 208 receives an indication from one or more components of vehicle 108 that a regeneration event is about to occur or that hydrocarbon injection is triggered, such as based on a timer, estimated soot accumulation, or manually triggered by operating I / O 30.

[0066] Model circuitry 210 is configured or constructed to manage one or more models for temperature prediction. A model refers to a machine learning model (e.g., a file, program, logic unit, etc.) and includes various forms of code, scripts, or commands that are trained to perform various operations, including, but not limited to, tasks such as data parsing, pattern recognition, data evaluation, and / or prediction operations. Managing models includes at least one of generating, updating, storing, or training one or more models. Model circuitry 210 is configured to store the generated models in database 106 or storage device 204. Model circuitry 210 is configured to train or update the models based on parameters or information from state circuitry 208. In various implementations, model circuitry 210 receives parameters related to regeneration events or hydrocarbon injection events (e.g., DOC inlet temperature, DOC outlet temperature, exhaust flow rate, injection rate, etc.) from state circuitry 208. Model circuitry 210 is configured to use parameters from one or more vehicles to generate or train a model to identify parameters associated with DPF 42 temperature variations and predict DPF 42 temperature anomalies based on historical data or parameters. In other embodiments, temperature changes of other components in aftertreatment system 38, such as the temperature of SCR 46, ASC 48, etc., may be considered to adjust the temperature generated in aftertreatment system 38. Model circuit 210 configures the identified parameters as inputs to the model to output a predicted result.

[0067] In various implementations, each prediction is performed for a subsequent time point predefined or configured by an administrator of the computing system 104. In this case, the model circuit 210 outputs, via one or more models, a predicted temperature for a time point (e.g., 10 seconds, 20 seconds, 30 seconds, etc.) subsequent to the time at which the prediction is performed. In some cases, each prediction provides a time series of predicted temperature data as output (e.g., a prediction for a subsequent time range). In this case, based on parameter trends for a predetermined time range (e.g., parameters collected for 5 seconds, 10 seconds, 20 seconds, etc.), the model is configured to output predicted temperature data for a subsequent time range, such as predicted temperature data for 10 seconds, 20 seconds, or 30 seconds.

[0068] The model circuit 210 is configured to generate or train a model using any suitable machine learning technique or neural network technique, such as regression techniques (e.g., autoregressive moving average (ARIMA)), recurrent neural network techniques (e.g., long short-term memory (LSTM) or gated recurrent unit (GRU)), temporal convolutional network (TCN), Transformer network, etc. For example, the model circuit 210 may use LSTM to generate and train the model discussed herein for purposes of example; however, other types of machine learning techniques may also be used to train and execute the model, taking into account processing power, memory usage, efficiency, or other considerations desired by the administrator of the computing system 104.

[0069] In some implementations, the model circuit 210 can train the model online or offline. To train the model offline, the model circuit 210 obtains various types of inputs from the vehicle 108. The model circuit 210 is configured or installed with one or more machine learning techniques to perform training on the model. In response to the training, the computing system 104 or controller 28 that trained the model can use the model to perform overshoot assessment, analysis, or determination.

[0070] In various configurations, model circuitry 210 utilizes parameters from fleet vehicles to analyze the occurrence of temperature overshoots and the changes in parameters preceding the overshoots. As described in conjunction with FIGS. 4-11 , model circuitry 210 analyzes features of the various parameters preceding the overshoots to identify patterns, similarities, or correlations between these features. Thus, model circuitry 210 can provide the trained model to prediction circuitry 212 or controller 28 of vehicle 108 to perform predictions.

[0071] In some implementations, the model circuit 210 may test the model using sample data or field data to determine prediction accuracy. The model circuit 210 is configured to further train or update the model to meet a threshold prediction accuracy when used to test the field data. The field data can be data collected from vehicles operating in the field. In response to the model meeting the threshold accuracy, the model circuit 210 is configured to provide the model to the controller 28 or prediction circuit 212 to perform predictions. The model circuit 210 is configured to generate or train a corresponding model for a specific type of vehicle or aftertreatment system, for example, based on the make and model of the vehicle, the model or version of the aftertreatment system, or other similar components of the vehicle. In some cases, the model circuit 210 is configured to generate or train a single model for any make and model of vehicle 108 or aftertreatment system 38.

[0072] Prediction circuitry 212 is configured or constructed to predict the temperature of aftertreatment system 38 components to minimize or prevent temperature overshoots. Prediction circuitry 212 predicts the temperature of aftertreatment system 38 to identify or detect temperature overshoots before they occur. Injection circuitry 214 or other components of controller system 200 (or controller 28) can use the predicted temperatures to perform operations to minimize or avoid temperature overshoots before they occur, such as controlling injectors 36 and / or 37, a heater (not shown), engine speed, engine torque, the EGR system, or other components of vehicle 108 that may affect the temperature of aftertreatment system components. An aftertreatment system component may be, for example, DPF 42. Prediction circuitry 212 performs the prediction using a model trained by model circuitry 210. As described above, prediction circuitry 212 may be part of controller 28 of vehicle 108, so that the predictions can be performed locally on vehicle 108.

[0073] Prediction circuit 212 is configured to monitor parameters that serve as model inputs. In this example, the model inputs include the inlet temperature of DOC 40, the outlet temperature of DOC 40, the exhaust flow rate (e.g., the flow rate at the turbocharger outlet), and the injection rate (e.g., the regeneration fuel injection rate of injector 36 and / or injector 37). Prediction circuit 212 provides the monitored parameters as inputs to the model to perform predictions. The model processes the input parameters and provides outputs (e.g., predicted values) to prediction circuit 212. Prediction circuit 212 receives a prediction from the model, which includes at least one of a temperature at a subsequent time point or a predicted temperature for a time series. Prediction circuit 212 can provide the predicted value to injection circuit 214 for managing hydrocarbon injection.

[0074] Injection circuit 214 is configured or constructed to determine, update, adjust, or otherwise manage hydrocarbon injection from injector 36 and / or injector 37. As described above, injection circuit 214 may be part of controller 28 of vehicle 108, such that injection circuit 214 may directly command injector 36 and / or injector 37 to configure the injection rate or amount of hydrocarbon. In some cases, injection circuit 214 may be located remotely from vehicle 108. In such cases, injection circuit 214 is configured to determine the injection rate based on the predicted temperature and provide an indication of the injection rate to controller 28 for use in controlling injector 36 and / or injector 37.

[0075] In certain implementations, injector 36 and / or injector 37 initiates hydrocarbon injection at a given injection rate for a predetermined amount of time based on a command from controller 28. Controller 28 provides this command based on the current state of components within vehicle 108. However, due to certain variables, such as increased fuel demand, heater activation within aftertreatment system 38, etc., the initial hydrocarbon injection rate may cause the temperature of aftertreatment system components and other components of aftertreatment system 38 to overshoot. Therefore, injection circuitry 214 is configured to adjust the injection rate or provide another command to injector 36 and / or injector 37 to initiate hydrocarbon injection using a different injection rate based on a temperature prediction for a subsequent timeframe.

[0076] In various arrangements, when the predicted temperature is greater than the target temperature, the injection circuitry 214 is configured to determine a difference between the predicted temperature and the target temperature for the aftertreatment system component. This difference represents the amount of temperature overshoot from the target temperature. In various embodiments, the difference is determined when a temperature overshoot is detected (e.g., the predicted temperature is greater than the target temperature, etc.). Because the difference is not continuously calculated, processing power or computing resource consumption can be reduced, such that the difference calculation is performed when an overshoot is predicted to occur. Based on the difference, the injection circuitry 214 determines a corresponding hydrocarbon amount (e.g., measured in parts per million (ppm)) that caused the temperature overshoot. The injection circuitry 214 is configured to convert the hydrocarbon amount into a corresponding injection rate (e.g., in grams per second (g / s or gps)). The injection circuitry 214 can use this corresponding injection rate to reduce the initial injection rate to meet the total hydrocarbon amount and minimize the temperature overshoot.

[0077] In some arrangements, injection circuitry 214 is configured to command injector 36 and / or injector 37 to terminate an injection event when the total hydrocarbon amount is met. For example, injection circuitry 214 identifies the injection rate and injection duration commanded to injector 36 and / or injector 37 for a regeneration event. Injection circuitry 214 determines an initial hydrocarbon amount based on the injection rate and injection duration (e.g., the product of these two values). Injection circuitry 214 determines the total hydrocarbon amount to be injected by subtracting the excess hydrocarbon amount predicted to cause a temperature overshoot from the initial hydrocarbon amount. Based on the total hydrocarbon amount, injection circuitry 214 determines an adjusted injection hydrocarbon duration for a given initial injection rate. Thus, in response to the expiration of the adjusted duration, injection circuitry 214 is configured to command injector 36 and / or injector 37 to terminate hydrocarbon injection, thereby minimizing or preventing a temperature overshoot.

[0078] In some arrangements, the computing system 104 uses the communication interface 216 to communicate with other remote devices (e.g., the vehicle 108 or the database 106) over the network 102 to obtain data, train models, provide predictions, etc. The computing system 104 (or one or more circuits of the computing system 104) can use the communication interface 216 to provide one or more models and other information to at least one of the vehicle 108 or one or more remote devices. The computing system 104 is configured to receive data (e.g., parameters), including raw data or processed data, from various components, devices, or systems within the network 102 and store the data in a local memory (e.g., the storage device 204) for access by at least one other device within the network 102. In some other cases, the computing system 104 can, for example, use the communication interface 216 to relay information received from other devices within the network 102 to the database 106 for storage.

[0079] Reference Figure 3 , according to an example implementation, a graph 300 of an example temperature of a catalyst component is shown. The catalyst component is shown as DOC 40. In some embodiments, the temperature may be the temperature of another component of the aftertreatment system 38, such as DPF 42 or SCR 46. Graph 300 includes data points for the inlet temperature of DOC 40 (line 302), the outlet temperature of DOC 40 (line 304), and a target temperature (line 306). As shown in at least portions 308 and 310 of graph 300, the difference between the temperature at the outlet of DOC 40 (line 304) and the configured target temperature is measured / monitored. Based on the difference being greater than an upper threshold / limit (e.g., 35 degrees Celsius, 50 degrees Celsius, etc.), the temperature may be considered an overshoot. The temperature at the outlet of DOC 40 may correspond to, for example, the temperature at the inlet of an aftertreatment system component or the temperature at the aftertreatment system component. Thus, a high DOC outlet temperature reflects a high DPF temperature. Features discussed herein may predict at least one of the DOC outlet temperature, the DPF inlet temperature, or the temperature of an aftertreatment system component to identify an overshoot. Furthermore, the features discussed herein may determine the amount or rate of hydrocarbons to reduce in order to minimize or prevent such overshoot.

[0080] Reference Figure 4A -B, depicts a schematic diagram of an example neural network architecture according to an example implementation. For simplicity, an LSTM machine learning model is trained for this feature, but other machine learning techniques may also be used in addition or alternatively.

[0081] Figure 4AA schematic diagram of an LSTM architecture with a single cell 400 is shown. A single cell can refer to a single layer of the model. Cell 400 includes a forget gate, an input gate, an output gate, and a cell state. Cell 400 includes two input ports and two output ports. The input includes input data and a cell state. The input data includes at least one temperature data point (e.g., the inlet temperature and / or outlet temperature of the DOC 40), exhaust flow rate, or injection rate. The cell state includes weights for different types of input data, for example, based on the iterations of processing the input data. For example, the output from the output port includes output data from cell 400 and the cell state. The output data includes processed or filtered input data. The cell state includes a copy of the input data filtered by at least one forget gate or input gate. Unfiltered input data is assigned a higher weight than filtered data. The cell state is sent back to the cell as input for subsequent data processing iterations. Cell 400 includes a sigmoid function and a tanh function. Each sigmoid function can be used, for example, as a gating function and is configured to process or manipulate raw input values ​​(e.g., recurrent neural network (RNN) values). Each tanh function may be used to adjust the output of unit 400 , for example, to maintain the value between various values ​​(eg, −1 and 1).

[0082] The forget gate is configured to filter out information that is not relevant to a desired output (e.g., a temperature prediction of a DOC outlet temperature or identifying a temperature overshoot). The input gate filters data that has relatively low relevance to the desired output, such as filtering out parameters that may not be relevant to the temperature prediction or retaining parameters that are directly relevant to the temperature prediction. The input gate includes data from the engine 12, and the output gate includes filtered or processed data from the input gate. Thus, after a certain number of iterations (e.g., a predefined number or based on the time series of the input), the input data can be filtered to only data that has a relatively high relevance to the desired output (e.g., a temperature overshoot or a temperature prediction). Highly relevant input data is assigned a higher weight and / or bias than other data that has a lower relevance to the desired output.

[0083] In various arrangements, to train the model, controller system 200 (e.g., model circuitry 210) provides various parameters as input signals to the model (e.g., unit 400 in this example). These parameters are historical data from vehicles in the fleet, capturing data for one or more components within the timeframe of an overshoot. For training purposes, the parameters may include an indication of a temperature overshoot, such as the time or duration of the overshoot detected. In response to providing the parameters to unit 400, unit 400 retrieves the state (e.g., parameters) of the vehicle 108 component prior to the overshoot. Unit 400 is configured to generate a correlation matrix to compare / correlate the overshoot with the component's previous state (e.g., parameter behavior or characteristics). Unit 400, at a forget gate, discards one or more states that remain consistent before and after the temperature increase or are unrelated to the temperature overshoot, such as NOx sensor data, efficiency data for aftertreatment system 38 components, commands to the DEF dispenser 44, etc. Unit 400 further filters the states / parameters at an input gate to a subset of states with a relatively high correlation with the temperature overshoot (e.g., states directly related to the temperature increase, thereby causing the overshoot). Unit 400 is configured to implement and assign relatively higher weights or biases to certain states that have a relatively higher correlation with temperature overshoot than other states. Thus, controller system 200 can train a model using historical data from the vehicle (e.g., field data) or simulation data generated for testing or training the model to determine the pattern / behavior / characteristics of certain parameters over time and the resulting DOC outlet temperature within the next time window (e.g., the next 10 seconds, 20 seconds, etc.).

[0084] Figure 4B At least two interconnected units 402, 404 are shown. Each unit 402, 404 is similar to unit 400. For example, rather than iterating over a single unit 400, the output of unit 402 is used as the input to the next unit 404. Although two units 402, 404 are shown, more units can be connected at the output of unit 404 and any subsequent units to provide additional iterations for processing the input data.

[0085] Reference Figure 5A -C, which shows a schematic diagram of example logic 500 - 504 for controlling hydrocarbon injection using a model. Figure 5A The logic of -C may be part of the controller system 200 of the computing system 104 and / or the controller 28 of the vehicle 108. In this case, the features, functions, and operations of the electronic unit are configured to be similarly performed by the controller 28 or the controller system 200.

[0086] Now refer to the Figure 5AAs shown in a schematic diagram of logic 500, controller system 200 (e.g., model circuit 210) or controller 28 is configured to use a model to predict the temperature at the outlet of DOC 40. The DOC outlet temperature may correspond to or represent the inlet temperature of an aftertreatment system component or the temperature of the aftertreatment system component itself (e.g., bed temperature). Controller system 200 provides various parameters (e.g., input data) to the model. In some cases, controller system 200 provides the model with certain types of parameters relevant to determining or predicting changes in DOC outlet temperature. Controller system 200 obtains one or more parameters from various sensors embedded or installed in vehicle 108 (e.g., temperature sensors, flow rate sensors, engine speed sensors, etc.). These sensors may be real or virtual (i.e., non-physical sensors that are configured into program logic within controller system 200 or controller 28 to make various estimates or determinations). For example, the flow rate sensor may be a real or virtual sensor configured to measure or otherwise obtain data, values, or information indicative of the exhaust flow rate of engine 12 (typically expressed in revolutions per minute). The sensor is strategically positioned to couple with at least one portion of the engine 12 or aftertreatment system 38 (when configured as a real sensor) and is configured to send a signal indicative of the exhaust flow rate of the engine 12 to the controller 28 or controller system 200. When configured as a virtual sensor, the controller 28 may use at least one input in an algorithm, model, lookup table, etc. to determine or estimate a parameter associated with the engine 12 (e.g., exhaust flow rate, generated exhaust byproducts, engine speed, power output, etc.). Other sensors may also be real or virtual.

[0087] The parameter includes at least one of a DOC inlet temperature, a DOC outlet temperature, exhaust flow rate data, and a hydrocarbon command (e.g., a hydrocarbon injection rate or amount). The parameter may be data captured at a certain point in time by one or more components of the vehicle 108. In some cases, the parameter may be time series data captured within a certain (e.g., predefined or predetermined) duration / time window, such as before or during a regeneration event (e.g., a hydrocarbon injection event).

[0088] In various implementations, the parameters used as model inputs are similar to the parameters used to train the model. The controller system 200 can use the current parameters of the vehicle 108 to train the model to perform subsequent predictions. These parameters can be referred to as input signals to the model. Although Figure 5A The input parameters shown are DOC inlet temperature, DOC outlet temperature, exhaust gas data, and hydrocarbon command, but other parameters related to changes in DOC outlet temperature may also be used as inputs.

[0089] The controller system 200 (e.g., prediction circuit 212) uses a model to process the parameters. The model (e.g., trained by model circuit 210) is configured to correlate the parameters with patterns or historical data to determine an expected or predicted DOC outlet temperature within the next predetermined time range (e.g., 10 seconds, 20 seconds, etc.). For simplicity and to provide an example, the model is configured to determine the increase in DOC outlet temperature caused by the exothermic reaction with the DOC 40 based on the DOC inlet temperature, the DOC outlet temperature, the exhaust flow rate, and the commanded hydrocarbon injection rate.

[0090] In some implementations, the DOC inlet temperature, exhaust flow rate, and hydrocarbon injection rate can determine the temperature increase at the outlet of the DOC 40. For example, the hydrocarbon injection rate defines the amount of hydrocarbons (e.g., grams) provided to the exhaust flow over a period of time (e.g., per second). A higher injection rate indicates that more hydrocarbons are injected into the exhaust flow, which can result in a relatively greater exothermic reaction compared to a lower injection rate. A higher DOC inlet temperature can result in a higher DOC outlet temperature, for example, as the temperature increases due to the reaction of the DOC 40 with the hydrocarbons. The exhaust flow rate can indicate the rate at which the injected hydrocarbons pass through the DOC 40. Therefore, having a relatively higher exhaust flow rate can accelerate the change in temperature (e.g., a faster rate of temperature increase during an injection event).

[0091] As described above, based on the DOC inlet temperature, exhaust flow data, and hydrocarbon command, the model is configured to determine the change in the current DOC outlet temperature. The temperature change is determined as a time series over a time range, e.g., during a hydrocarbon injection event until the point in time at which the event completes (e.g., as indicated by the hydrocarbon injection command). Given the historical data used to train the model, the model can correlate current parameters with historical parameters (e.g., patterns) to determine the expected change in the current DOC outlet temperature. In various embodiments, the controller system 200 using the model correlates the current DOC inlet temperature, DOC outlet temperature, and exhaust flow rate with the corresponding historical parameters. These parameters can be used to predict the initial data point of the time series. Furthermore, the controller system 200 correlates the current injection rate (e.g., hydrocarbon command) with the historical injection rates. Based on this correlation, the controller system 200 determines the change in DOC outlet temperature over time caused by the exothermic reaction between the hydrocarbon and the DOC 40, which is used to map the remaining time series for at least the duration of the injection event. Thus, using the model, the controller system 200 is configured to at least predict the DOC outlet temperature (e.g., 10 seconds, 20 seconds ahead, etc.) to identify any overshoot that may occur. For simplicity, the controller system 200 is configured to obtain a time series of temperatures as the output of the model. Examples of the time series are at least as follows: Figure 7 shown.

[0092] In various implementations, the controller system 200 (e.g., the prediction circuit 212) utilizes the model to perform predictions continuously, periodically, or aperiodically. For example, the controller system 200 performs predictions based on a predetermined time period (e.g., every 5 minutes, 10 minutes, 20 minutes, etc.). In another example, the controller system 200 performs predictions in response to receiving a hydrocarbon injection command or when a regeneration event is triggered or is about to be triggered. A regeneration event is triggered in response to soot levels exceeding a threshold, a timer, or other configuration controlled by the controller 28.

[0093] After receiving the output from the model, the controller system 200 receives or identifies a target temperature command. The target temperature command comprises a target temperature for a regeneration event, a hydrocarbon injection event, or other event intended to increase the temperature of an aftertreatment system component. For simplicity, the target temperature may be associated with a regeneration event or a hydrocarbon injection event. The target temperature indicates the desired temperature to be achieved by injecting hydrocarbons. The controller system 200 feeds the target temperature indication and prediction (e.g., output from the model) to the hysteresis block.

[0094] The controller system 200 is configured to use a hysteresis block to adjust (e.g., reduce) the injection rate or hydrocarbon fuel supply until the temperature overshoot is minimized to below a lower limit. In this case, a temperature overshoot refers to the DOC outlet temperature being at least greater than the target temperature. A temperature overshoot can also refer to the DOC outlet temperature being greater than an upper limit set above the target temperature. The controller system 200 monitors the time series of predicted temperatures to identify any overshoots that occur within a subsequent time range. Minimizing temperature overshoots refers to maintaining or reducing the difference between the predicted temperature and the target temperature at different time points in the time series below a predetermined threshold.

[0095] The hysteresis block is configured to receive an input signal comprising a predicted temperature and a target temperature at different points in time in a time series. In some cases, the input signal comprises a time series of differences between the predicted temperature data and the target temperature. The hysteresis block can be configured with an upper threshold, an upper bound, or a lower threshold, and receive the previous value output by the hysteresis block. The upper threshold can be greater than the lower threshold, for example, 35 and 15, 40 and 20, or 50 and 15, respectively, as well as other combinations. The upper and lower thresholds can be configured by an administrator, operator, or other entity operating the computing system 104 or the controller 28.

[0096] When the difference between the predicted temperature and the target temperature at a certain time point in the time series is greater than or equal to an upper threshold, the hysteresis block can activate its logic or operation. By activating the logic, the hysteresis block is configured to output a signal to trigger the operation of other electronic units for fuel adjustment (e.g., hydrocarbon injection adjustment). The output signal from the hysteresis block can be a binary signal (1 or 0) indicating whether the hydrocarbon injection should be adjusted. Signal "1" can indicate the activation of the hydrocarbon adjustment logic, and signal "0" can indicate the deactivation of the hydrocarbon adjustment logic, and vice versa. In some embodiments, each iteration of activating the hysteresis block can be delayed according to a unit delay of 1 / Z (i.e., the Z transform operator). For example, the hysteresis block can delay receiving the input according to a configurable unit delay, thereby delaying the processing of the input to generate the output signal.

[0097] Activating logic or issuing a fuel adjustment operation signal from the hysteresis block enables the controller system 200 to continue executing the operations described in the logic diagram 502. In various implementations, the controller system 200 may repeat the operations of the logic diagram 500 after making the initial adjustment to the fuel supply (e.g., hydrocarbon supply), for example by reusing the model or hysteresis block to determine whether additional adjustments are needed based on an updated temperature prediction (e.g., using another current DOC inlet temperature, DOC outlet temperature, exhaust flow rate, and / or adjusted hydrocarbon command). In some implementations, the controller system 200 maintains the activation of the hysteresis block logic until the predicted temperature is below a lower threshold after at least one adjustment to the hydrocarbon injection rate.

[0098] Reference Figure 5B As shown in the logic diagram 502 of FIG. 2 , the controller system 200 converts the difference between the predicted temperature (e.g., the highest predicted temperature in the time series) and the target temperature into an amount of hydrocarbons in response to triggering / activating the hysteresis block logic. The temperature difference can be converted to a corresponding amount of hydrocarbons to be injected into the exhaust stream based on a predefined conversion factor. For example, a certain amount of hydrocarbons (e.g., 1000 ppm, etc.) can correspond to a predefined temperature change of a certain degree (e.g., 14 degrees Celsius, etc.). In this example, the controller system 200 converts the temperature difference into an amount of hydrocarbons to be reduced using the following formula (1):

[0099] (1)

[0100] Although the above conversion factor uses 1000 ppm hydrocarbons and 14 degrees Celsius, other hydrocarbon and temperature combinations can be configured in equation (1). Using equation (1), controller system 200 is configured to determine the amount of hydrocarbons to be reduced. For example, if the temperature difference is 56 degrees Celsius, controller system 200 determines that the total amount of hydrocarbons to be reduced is 4000 ppm.

[0101] In some cases, the conversion factor may vary based on one or more parameters, such as a varying DOC outlet temperature. In such cases, the controller system 200 performs the conversion using a table or matrix. For example, at 350 degrees Celsius, 1000 ppm of hydrocarbon injection may correspond to a 14-degree Celsius temperature increase; at 400 degrees Celsius, 1000 ppm of hydrocarbon injection may correspond to a 13-degree Celsius temperature increase; at 450 degrees Celsius, 1000 ppm of hydrocarbon injection may correspond to a 12-degree Celsius temperature increase, and so on.

[0102] In various arrangements, the controller system 200 considers the rate of temperature change (e.g., rate of increase) to determine a multiplier for reducing the amount of hydrocarbons. The controller system 200 can use this multiplier to further adjust the amount of hydrocarbon reduction to reduce the intensity of the temperature increase, thereby avoiding temperature overshoot. The controller system 200 determines the multiplier based on the slope of the average predicted DOC outlet temperature change, such as in combination with Figure 5C As stated.

[0103] Reference Figure 5C In logic 504, the controller system 200 is configured to calculate the slope of the DOC outlet temperature using a counter block. The controller system 200 calculates the slope of the average predicted DOC outlet temperature based on the values ​​predicted or estimated within a predetermined (duration) window size. The window size can be configurable, for example, a window size of 4, 5, 6, etc. The value during the window represents the average DOC outlet temperature predicted for the duration of the window. For example, in response to identifying or determining a temperature overshoot, the controller system 200 begins recording values ​​associated with the time range of the temperature overshoot. The controller system 200 continues to record values ​​for a maximum window size, for example, incrementing a counter by 1 for each value recorded and terminating the recording operation once the counter reaches the maximum window size. In this case, the controller system 200 performs a moving average (e.g., shown as a moving average block) to determine the average DOC outlet temperature change over the entire window.

[0104] In some implementations, the controller system 200 can perform operations such as a moving maximum, moving median, or moving minimum to identify one or more values ​​corresponding to a time window from the temperature data (e.g., the predicted DOC outlet temperature). In some cases, the values ​​corresponding to the time window from the temperature data can be used, at least in part, as input to the model. In some embodiments, in response to the counter reaching the maximum window size, the counter block can signal the moving average block to initiate its operation / logic. Simultaneously, the signal from the counter block can be used to reset the counter value (e.g., after a unit delay of 1 / Z).

[0105] The controller system 200 is configured to record consecutive DOC outlet temperatures to determine an average DOC outlet temperature change (e.g., slope). In some cases, consecutive temperatures are averaged with each other. For example, the controller system 200 calculates an average between the temperatures at a first time point during a predicted temperature overshoot (e.g., after two counter increments) and a second time point. Based on the first and second time points, the controller system 200 calculates another average between a third time point and the previous average (e.g., at the third counter increment), and so on, until the counter reaches a predetermined maximum window size.

[0106] In some implementations, the controller system 200 may compare average values ​​of the DOC outlet temperatures. The controller system 200 is configured to apply a multiplier to the adjustment of the hydrocarbon injection amount. For example, if at least one average DOC outlet temperature (e.g., slope) calculated over the entire time series is equal to or higher than a predetermined maximum temperature change rate (e.g., 5 degrees Celsius per second, etc.), the controller system 200 may apply the highest corresponding multiplier (e.g., 2 times, 2.5 times, etc.) to the combined value. Figure 5B If the slope is less than a minimum temperature change rate (e.g., 1 degree Celsius per second, etc.), the controller system 200 applies a multiplier of 1, which does not change the determined amount of hydrocarbons to be reduced. Otherwise, if the slope is between a predetermined minimum and maximum temperature change rate, the controller system 200 may apply a multiplier proportional to the calculated temperature change rate.

[0107] In another example, the determined hydrocarbon amount may be 4000 ppm. If the temperature change rate exceeds the maximum change rate, the controller system 200 may multiply the hydrocarbon amount by 2. In this example, the controller system 200 adjusts the determined hydrocarbon amount to 8000 ppm to reduce the injection of hydrocarbons. In some other implementations, the controller system 200 may not be configured with a counter block, so that the controller system 200 may continue to use the previously determined hydrocarbon amount for fuel adjustments.

[0108] Return to reference Figure 5B After determining the hydrocarbon amount (whether or not a multiplier is applied), the controller system 200 converts the hydrocarbon amount to a hydrocarbon injection rate, for example, in units of amount / time (e.g., milliliters / second). The controller system 200 uses this converted hydrocarbon injection rate to reduce the commanded hydrocarbon injection rate. To perform the conversion, the controller system 200 determines an injection duration based on the hydrocarbon command. The controller system 200 is configured to divide the hydrocarbon amount by the duration to determine the amount of hydrocarbons to be injected per a predetermined duration (e.g., per second). For example, if the duration is 20 seconds and the hydrocarbon amount is 4000 ppm, the injection rate for adjustment is 200 ppm per second. This injection rate can be referred to as the reduction rate for reducing the current hydrocarbon injection rate.

[0109] In certain implementations, the controller system 200 determines whether to adjust the injection rate based on at least one of the current DOC outlet temperature or a target temperature. The target temperature may be predefined based on at least a minimum temperature that results in the removal of particulate matter from the aftertreatment system components. In some cases, the target temperature is determined or adjusted relative to or based on the current temperature of the exhaust gas, certain components of the aftertreatment system 38, the operating conditions of the engine 12 (e.g., relatively hot conditions compared to relatively cold conditions may change the target temperature), etc. If at least one of these temperatures is not equal to or above the respective threshold value, the controller system 200 may not adjust the injection rate because the temperature is within an acceptable temperature range (e.g., normal operating temperatures for the aftertreatment system 38 components). As an example, in Figure 5B In the example, the threshold value of the target temperature can be predefined as 530, 540 or 550 degrees Celsius, the threshold value of the DOC outlet temperature (eg, for the current DOC outlet temperature) can be predefined as 500, 510 or 520 degrees Celsius, and other values ​​configurable by the administrator.

[0110] If at least one of these temperatures is equal to or greater than the respective threshold, the controller system 200 is configured to adjust the injection rate. Using the current hydrocarbon command, the controller system 200 reduces the current hydrocarbon injection rate by the determined reduction rate to obtain an adjusted hydrocarbon injection rate (or second injection rate). Thus, the controller system 200 may disable the model and provide the adjusted or final hydrocarbon command along with the adjusted injection rate to the injector 36 and / or injector 37. Providing the adjusted hydrocarbon command may mean updating an existing command (e.g., a first injection rate) provided to the injector 36 and / or injector 37, or terminating the existing command and providing another command to the injector 36 and / or injector 37. By adjusting the injection rate, temperature overshoots may be minimized or avoided before they occur.

[0111] Figure 6A -B is a graph 600-602 showing examples of prediction performance using different numbers of hidden units (e.g., from an LSTM) according to an example implementation. Specifically, the y-axis represents a histogram (pdf) of the proportion (e.g., number) of overshoot data points relative to the x-axis value, and the x-axis value represents the temperature difference between the predicted temperature at the DOC outlet and the actual temperature. The predicted temperature refers to the predicted DOC outlet temperature output by the above-mentioned machine learning model (e.g., a neural network model). The actual temperature refers to the actual DOC outlet temperature monitored / observed / measured, for example, by a temperature sensor on the vehicle (e.g., captured in a test dataset). Depending on the configuration of the model or the data used for training or prediction, more or fewer hidden units can be used. In some cases, too many hidden units can lead to overfitting, while too few hidden units can lead to underfitting.

[0112] In the case shown in graphs 600-602, the predicted temperature is estimated 10 seconds ahead of the actual temperature, so there is a 10-second offset. The temperature at other predetermined time points can be predicted by computing system 104 or controller 28 using the above-mentioned techniques, such as 15 seconds, 20 seconds, etc.

[0113] The data points of graphs 600-602 represent situations where the target temperature is above a predetermined threshold (e.g., 520 degrees Celsius), and at least one of the difference between the predicted temperature and the target temperature or the difference between the actual temperature (at the time corresponding to the predicted temperature) and the target temperature is above a predetermined upper threshold. In graph 600, the upper threshold is 35 degrees Celsius. In graph 602, the upper threshold is 25 degrees Celsius. The data point closest to zero on the x-axis produces the best results because the predicted temperature is closest to the actual measured temperature.

[0114] like Figure 6A As shown in Figure 1-B, under these conditions, a machine learning model using six hidden units produced the most accurate results in predicting the DOC outlet temperature, with the median of the data points closer to zero variance on the x-axis. However, other numbers of units may be used depending on the configuration of vehicle 108, different upper thresholds, and other variables. Underestimation of the model's predictions or outputs may result in a weaker response to mitigate temperature overshoots. Overestimation of the model's predictions may result in a stronger response to temperature overshoots, leading to undershoots of the DOC outlet temperature in some cases.

[0115] Reference Figure 7 , depicts a graph 700 illustrating an example temperature adjustment according to an example implementation. Graph 700 displays an initial target temperature (labeled "Target" in the legend), an adjusted target temperature (labeled "Target-50" in the legend), a predicted DOC outlet temperature (labeled "DOC Outlet" in the legend), and a simulated DOC outlet temperature after adjusting the injection rate or target temperature. In some implementations, the computing system 104 (or controller 28) predicts the DOC outlet temperature based on a hydrocarbon command based on the initial target temperature and other parameters. At portion 702, the computing system 104 predicts the occurrence of a temperature overshoot. The overshoot is predicted to be due to injecting hydrocarbons to meet the initial target temperature (e.g., injecting 40% more regeneration fuel than desired to meet a predetermined temperature range for the initial target temperature). In response to identifying the overshoot, the computing system 104 is configured to reduce the high heat release to below a desired level so that the DOC outlet temperature for the regeneration event is within a threshold (e.g., an upper temperature limit or a lower temperature limit).

[0116] As shown, the computing system 104 can use simulation results from the model to adjust the hydrocarbon injection rate. For the simulation, the computing system 104 determines the difference between the predicted temperature data (e.g., the highest predicted temperature) and the initial target temperature in section 702. In some cases, the computing system 104 determines the difference between the predicted temperature data and at least one upper limit of the initial target temperature. The computing system 104 can lower the initial target temperature based on the difference to obtain an adjusted target temperature. The computing system 104 determines an adjusted injection rate (or adjusts the injection rate of the hydrocarbon command) based on the adjusted target temperature. Using the adjusted injection rate as input to the model, the computing system 104 is configured to determine a simulated DOC outlet temperature associated with the adjusted target temperature. Therefore, the computing system 104 (or the controller 28) can use the adjusted injection rate to reduce the peak DOC outlet temperature from approximately 645 degrees Celsius to approximately 600 degrees Celsius, as shown in section 702.

[0117] In some implementations, the controller 28 uses the adjusted injection rate for the hydrocarbon command to achieve the simulated DOC outlet temperature. In some other implementations, the computing system 104 performs a second prediction using the adjusted target temperature as an input to the model to achieve the simulated DOC outlet temperature. In this case, the computing system 104 may further adjust the target temperature to achieve the desired simulated DOC outlet temperature.

[0118] Reference Figure 8 , depicts a graph 800 showing an example fueling adjustment according to one example implementation. Graph 800 includes a hydrocarbon injection rate (fueling rate) on the y-axis and associated time on the x-axis. Graph 800 includes a commanded injection rate, labeled "Field Fueling" in the legend. Graph 800 includes an adjusted (or simulated) injection rate, labeled "Simulated Fueling" in the legend.

[0119] In some implementations, portion 802 of graph 800 can be associated with portion 702 of graph 700, such as in conjunction with Figure 7 For example, when using the commanded injection rate of portion 802, in response to identifying or predicting an overshoot at portion 702, the computing system 104 (or controller 28) can adjust the injection rate by lowering the target temperature. For example, the computing system 104 can use the adjusted injection rate of portion 802 to obtain the simulated DOC outlet temperature of portion 702.

[0120] In some implementations, the "DOC outlet" data points and the "measured delivery" data points are data collected from at least one vehicle in the fleet. The computing system 104 or controller 28 can input the collected data into a model to obtain simulated results by reducing the injection rate based on the reduced target temperature. In this case, the collected data is used to train or test the model. The computing system 104 or controller 28 can perform similar simulations on real-time data to predict the DOC outlet temperature, as described herein.

[0121] Reference Figure 9A -G, a schematic diagram depicting an example implementation of the model, such as in at least Figure 1-8 The model described in . Figure 9A An overview of a process 900 for implementing a model to predict the temperature of a component or system (e.g., DOC outlet temperature) is shown. The process 900 includes blocks 902-914 (e.g., logic units) for predicting temperature, although more or fewer logic units may be implemented. The process 900 may be performed by Figure 1-2 Components of the process 900 are executed, so they may be referenced to help explain the process 900. As discussed below as an overview, the process 900 includes receiving input data (902). The process 900 includes converting the input data (904). The process 900 includes processing the input data (906). The process 900 includes using a model (e.g., an LSTM model) (908). The process 900 includes processing output from the model (910). The process 900 includes converting the output data (912). The process 900 includes obtaining a predicted temperature (914). The process 900 may include combining at least Figure 5A -C describes the operation.

[0122] At block 902, the computing system 104 (e.g., the state circuit 208) or the controller 28 is configured to receive input data from components (e.g., sensors, etc.) of the vehicle 108. The input data includes one or more parameters configured as inputs to a model, such as inlet and outlet temperatures of the DOC 40, exhaust flow rate, or a hydrocarbon command (e.g., injection rate), etc.

[0123] At block 904, the computing system 104 is configured to convert the input data for processing. The computing system 104 converts the input data into a uniform / corresponding data type for use between different models to perform predictions. In some cases, the input data may already have been converted into a form that can be used for further processing or for use in a model.

[0124] At block 906, the computing system 104 (e.g., state circuit 208) is configured to process / preprocess the input data (e.g., raw data or transformed data) for input into the model. The computing system 104 preprocesses the raw data by normalizing the raw data in the input processing logic. Figure 9B , depicting the logic involved in the input processing logic unit of block 906. As shown, the computing system 104 receives raw data converted into DOC inlet temperature, DOC outlet temperature, turbine / turbocharger outlet flow rate (in grams per second (gps)), and regeneration fuel supply rate (injection rate) (in gps), although other parameters can also be used as input. The computing system 104 pre-processes the injection rate into a certain amount of hydrocarbons, measured in ppm (labeled "Regeneration Fuel Concentration ppm"). The amount of hydrocarbons is calculated based on the flow rate of the turbocharger, the molecular weight of the fuel used (e.g., which can be configured based on the type of fuel used by the vehicle 108), the injection rate of the hydrocarbons, and / or the exhaust flow rate.

[0125] The computing system 104 (eg, the state circuit 208) is configured to normalize the parameters (eg, DOC inlet temperature, DOC outlet temperature, hydrocarbon amount (in ppm), and flow rate (in gps)). The normalization of the parameters may be based on Figure 9B After normalization, the computing system 104 provides the pre-processed data as input to the model. Figure 9B In the reference, "muln" refers to the mean of a value (e.g., a calibrable one), and "signIn" refers to the standard deviation of a value (e.g., a fixed parameter).

[0126] At block 908, the computing system 104 (e.g., the model circuit 210 or the prediction circuit 212) uses the model to process the input data. The use of the model (or model logic unit) can be Figure 9D Described in. Figure 9D , depicts the model logic of a model for performing temperature prediction. The model includes weights and biases for making temperature predictions in a normalized form (e.g., a normalized prediction of the predicted temperature). The model includes a model layer (e.g., an LSTM layer) 916 and a fully connected layer 918. In some arrangements, the fully connected layer 918 includes two or more calibrable items. The model layer 916 includes various tables that can be divided into different calibration items (e.g., vectors).

[0127] Reference Figure 9E, a fully connected layer 918 is shown that includes the output states from the model layer 916, various weights, various biases, and a matrix whose entries correspond to the number of hidden units (e.g., six hidden units, etc.). Various weights can be applied to the matrix (e.g., a 6x6 matrix for a six hidden unit model). The computing system 104 can bias the matrix of the model so that the model produces an output that includes a prediction of the DOC outlet temperature. The fully connected layer 918 is configured to convert the various outputs from the model layer 916 into a single output (e.g., a predicted DOC outlet temperature), which can be, for example, part of an LSTM. Referring to Figure 9F -G, depicts Figure 9D The logic 920A-B of the model layer 916 is described in more detail below.

[0128] At block 910, computing system 104 (e.g., prediction circuitry 212) is configured to process output from the model, as in Figure 9C The output processing logic unit is shown in the following figure. Figure 9C , depicting the Figure 9E The output processing logic converts the normalized DOC outlet temperature prediction into physical units. As shown, the output processing logic includes two calibrable items. The computing system 104 provides the model output to the output processing logic to convert the output signal from the model into physical units (e.g., physical quantities). In this case, the computing system 104 converts the output signal into a predicted DOC outlet temperature expressed in corresponding units, such as Celsius, Fahrenheit, or Kelvin, depending on the configuration of the computing system 104 or the controller 28.

[0129] At block 912, the computing system 104 (e.g., prediction circuitry 212) is configured to convert the output data (e.g., the predicted temperature in physical units). The computing system 104 converts the output from block 910 to convert the predicted temperature into the desired data type. Thus, at block 914, the computing system 104 obtains the predicted temperature from block 912. In various implementations, the operations of process 900 are performed by the controller 28 such that the controller 28 uses the model trained by the computing system 104 to predict the DOC outlet temperature. In some arrangements, the controller 28 is configured to receive the temperature prediction from the computing system 104.

[0130] Reference Figure 10 , depicts a schematic diagram 1000 of an example implementation of a neural network architecture according to an example implementation. Figure 10 The neural network architecture includes Figure 4A-B has a structure or component similar to at least one unit 400-404. The neural network architecture includes various components, such as an input gate, a forget gate, a cell state (e.g., a cell candidate), and a forget gate, similar to the units 400-404. The components of the neural network architecture (e.g., the input gate, the forget gate, the cell candidate, and the output gate) can be described using the following formula shown in Table 1.

[0131] part formula Input Gate <![CDATA[(2)i t =σ g (W i x t +R i h t-1 +b i )]]> Forget Gate <![CDATA[(3)f t =σ g (W f x t +R f h t-1 +b f )]]> Unit Candidates <![CDATA[(4)g t =σ g (W g x t +R g h t-1 +b g )]]> Output Gate <![CDATA[(5)o t =σ g (W o x t +R o h t-1 +b i )]]>

[0132] Table 1

[0133] exist Figure 10 In , x represents the input at the corresponding time step t, h represents the hidden state at the corresponding time step t, and c represents the cell state at the corresponding time step t. The following formula can be associated with x, h, and c.

[0134] (6)

[0135] (7)

[0136] (8)

[0137] As described in the paper, n corresponds to the number of inputs accepted by the model, and m corresponds to the number of hidden units (e.g., 3, 4, 6, or any number based on the model configuration). Based on formula (2), the following formula (9) for the input gate can be derived.

[0138] (9)

[0139]

[0140] W of formula (9) represents the degree of influence of the input on the hidden state or unit state (e.g., weight). W can be a multiplier of the corresponding value, for example, 1 times for a relatively normal influence, 0.5 times for a relatively low influence, or 2 times for a relatively high influence. The total matrix size of W described in this article can be calculated as follows: W matrix total size = 4×m×n. R of formula (9) represents the degree of influence of the previous hidden state and unit state on the current hidden state and unit state (e.g., hidden matrix). The total matrix size of R can be calculated as follows: R matrix total size = 4×m×m. b of formula (9) represents the bias used in at least formula (12). The total matrix size of b can be calculated as follows: b matrix total size = 4×m. The matrices describing W, R, and b correspond to formulas (10)-(12), respectively.

[0141] (10)

[0142]

[0143] (11)

[0144]

[0145] (12)

[0146] Furthermore, in the regression layer, the total matrix size corresponds to m. The m hidden states converge into a single output in the regression layer, consisting of a column vector of size m×1 and a single bias. Using the above formula, the computing system 104 or the controller 28 can use the model to perform temperature prediction. In various implementations, based on the above formula, the final states of all hidden units can be obtained, and the matrix can indicate the weights applied to each final state and the corresponding bias for moving the state.

[0147] Figure 11 is implemented according to an example for using at least Figure 1-10 Flowchart of method 1100 for controlling catalyst component temperature using a machine learning model as described in Figure 1-2 1100. The method 1100 discussed below includes steps 1102-1110, as well as other processes (or other operations) for predicting temperature overshoot and adjusting hydrocarbon injection rate. In various implementations, certain processes may be performed before or after another process.

[0148] At step 1102, computing system 104 (e.g., state circuit 208) or controller 28 receives an indication of a hydrocarbon injection event (e.g., a regeneration event) directed to a catalyst component in the vehicle. The catalyst component is at least one of DOC 40 or other components of aftertreatment system 38. In this context, a hydrocarbon injection event is an event in which hydrocarbons are injected to exothermically react with DOC 40. In some cases, the hydrocarbon injection event is directed to an aftertreatment system component, such as to remove or burn particulate matter (e.g., soot) from the aftertreatment system component. In some embodiments, the aftertreatment system component may be DPF 42. In other embodiments, the aftertreatment system component may be SCR 46 or other components of aftertreatment system 38.

[0149] Computing system 104 receives an indication from at least one component of vehicle 108 (e.g., operating I / O 30 or monitoring system 32). For example, computing system 104 receives a signal from monitoring system 32 regarding an amount of soot accumulation in an aftertreatment system component that triggers a regeneration event. In another example, computing system 104 receives an indication from operating I / O 30 to initiate a regeneration event. In some cases, regeneration events are periodic. In such cases, computing system 104 determines that a regeneration event is about to be initiated based on a timer.

[0150] In step 1104, the computing system 104 (e.g., the state circuit 208) or the controller 28 receives various parameters regarding the catalyst component (e.g., the DOC 40) of the engine 12 and the vehicle 108. The computing system 104 receives these parameters within a first time range (e.g., parameters monitored or observed over a period of time). The parameters include at least one of an inlet temperature of the catalyst component, an outlet temperature of the catalyst component, an exhaust flow rate of the engine 12 or a turbocharger, and / or a first / initial injection rate of the engine 12. The computing system 104 obtains the injection rate from an initial command (e.g., an instruction) to initiate a hydrocarbon injection event. Other parameters may also be used in the operation of the method 1100.

[0151] At step 1106, the computing system 104 (e.g., model circuitry 210 or prediction circuitry 212) or controller 28 predicts temperature data using a parameter-based model. The computing system 104 inputs these parameters into the model as input data. Based on the initial command to initiate the hydrocarbon injection event (e.g., an operation to inject hydrocarbons or fuel) and the current conditions of the vehicle 108 monitored within the first timeframe, the computing system 104 predicts catalyst component temperature data (e.g., DOC outlet temperature) within a second timeframe subsequent to the first timeframe. The predicted temperature data can be presented in the form of a time series for at least the duration of the commanded hydrocarbon injection event. Thus, the computing system 104 can predict the catalyst component outlet temperature for at least the duration after the regeneration event is initiated (e.g., at least 10 seconds, 20 seconds, etc., from the start of the regeneration event).

[0152] At step 1108, the computing system 104 (e.g., injection circuit 214) or controller 28 calculates a second injection rate for injector 36 and / or injector 37 in response to or based on the predicted temperature data being above a threshold (e.g., above a predetermined target temperature set for the regeneration event or above a predetermined upper threshold of the target temperature). The computing system 104 uses the model to calculate the second injection rate. The second injection rate may be different from (e.g., lower than) the first injection rate to be commanded to injector 36 and / or injector 37. The computing system 104 determines the second injection rate based on the predicted temperature for at least a portion of the second time frame.

[0153] In various implementations, to calculate the second injection rate, computing system 104 determines the temperature difference between the temperature data and at least the target temperature. If the temperature data is equal to or below the target temperature, computing system 104 determines that there is no overshoot and no adjustment is required. If the temperature data is above the target temperature, computing system 104 determines that the injection rate needs to be adjusted or updated. In some cases, computing system 104 determines to adjust the first injection rate or provide the second injection rate when the temperature data is above an upper threshold of the target temperature (e.g., 25 degrees, 30 degrees, etc. above the target temperature).

[0154] After determining the adjusted injection rate, the computing system 104 converts the temperature difference into a hydrocarbon amount (e.g., a number of hydrocarbons or physical units). The computing system 104 converts the hydrocarbon amount into a third injection rate, which represents a reduced injection rate for reducing the predicted temperature to at least below a lower threshold of the target temperature. The lower threshold is less than an upper threshold of the target temperature. The computing system 104 is configured to perform a second prediction of the temperature data (e.g., a second predicted temperature data) to predict whether the adjusted injection rate (e.g., the second injection rate) will reduce the peak catalyst member outlet temperature to at least below the upper threshold of the target temperature. After determining the third injection rate, the computing system 104 calculates the second injection rate based on the difference between the first injection rate and the third injection rate (e.g., subtracting the third injection rate from the first injection rate).

[0155] In step 1110, the computing system 104 (e.g., the injection circuit 214) or the controller 28, in response to determining the second injection rate, commands the injector 36 and / or the injector 37 of the vehicle 108 to inject hydrocarbons based on the second injection rate. For example, the commanded second injection rate may implement an injection duration similar to that previously commanded. In some cases, the computing system 104 may not use the second injection rate, but instead reduce the injection duration, intermittently pause the injection operation, or prematurely terminate the regeneration event based on the total amount of hydrocarbons injected into the aftertreatment system 38.

[0156] For example, the computing system 104 may calculate the reduced or adjusted amount of hydrocarbons based on the difference between the first amount of hydrocarbons according to the initial command to the injector 36 and / or the injector 37 and the amount of hydrocarbons converted from the temperature difference (e.g., the amount of hydrocarbons to be reduced). The computing system 104 monitors the total amount of hydrocarbons injected into the engine 12 or the aftertreatment system 38. Therefore, when the total amount of hydrocarbons injected reaches the reduced amount of hydrocarbons, the computing system 104 may terminate the hydrocarbon injection event.

[0157] In some implementations, computing system 104 determines whether to use the second injection rate or update command based on whether at least one of the catalyst component temperature (e.g., outlet temperature), predicted temperature data, or a target temperature is equal to or greater than a corresponding threshold. In this case, computing system 104 receives a target temperature for a hydrocarbon event. Computing system 104 determines whether at least one of the target temperature is equal to or greater than a first predetermined threshold, or whether the catalyst component outlet temperature (e.g., predicted or actual) is equal to or greater than a second predetermined threshold, to command the injector. If at least one of the conditions is met, computing system 104 commands injector 36 and / or injector 37 to initiate the hydrocarbon injection event using the second injection rate. Otherwise, computing system 104 may not command injector 36 and / or injector 37, causing injector 36 and / or injector 37 to initiate the hydrocarbon injection event using the first injection rate (in which case, the temperature overshoot is still within a tolerable range predefined by an administrator or the like).

[0158] It should be understood that no claim element herein shall be interpreted under the provisions of 35 U.S.C. § 112(f) unless the element expressly uses the phrase "means for..." The above-mentioned schematic flow charts and method diagrams are generally presented in the form of logical flow charts. Therefore, the depicted order and labeled steps are merely indicative of representative embodiments. Other steps, sequences and methods can be conceived that are equivalent in function, logic or effect to one or more steps or portions thereof of the method shown in the diagram. In addition, references in this specification to "one embodiment," "embodiment," "example embodiment" or similar language mean that the particular features, structures or characteristics associated with the embodiment are included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment," "in an embodiment," "in an example embodiment" and similar language appearing in this specification may, but do not necessarily, refer to the same embodiment.

[0159] Furthermore, the format and symbols used are intended to explain the logical steps of the schematic diagrams and should not be construed as limiting the scope of the methods illustrated in these diagrams. Although various arrow types and line types may be employed in the schematic diagrams, they should not be construed as limiting the scope of the corresponding methods. Indeed, some arrows or other connectors may be used solely to indicate the logical flow of the method. For example, an arrow may indicate a waiting or monitoring period of unspecified duration between enumerated steps of the depicted method. Furthermore, the order in which a particular method occurs may not strictly adhere to the order of the corresponding steps shown. It should also be noted that each block of the block diagrams and / or flow charts, as well as combinations of blocks in the block diagrams and / or flow charts, may be implemented by a dedicated hardware-based system that performs the specified functions or behaviors, or a combination of dedicated hardware and program code.

[0160] Many of the functional units described in this specification are labeled as circuits to more specifically emphasize their implementation independence. For example, a circuit can be implemented as a hardware circuit containing custom very large scale integrated circuits (VLSI) or gate arrays, off-the-shelf semiconductors (such as logic chips, transistors), or other discrete components. A circuit can also be implemented in programmable hardware devices such as field programmable gate arrays, programmable array logic, or programmable logic devices.

[0161] As mentioned above, the circuits may also be implemented in a machine-readable medium for use by various types of processors (e.g., Figure 2 The identified circuitry of the executable code may, for example, comprise one or more physical or logical blocks of computer instructions, which may, for example, be organized into objects, procedures, or functions. However, the executable files of the identified circuitry need not be physically located together, but may comprise different instructions stored in different locations, which, when logically connected together, constitute the circuitry and achieve the intended purpose of the circuitry. In practice, the circuitry of the computer-readable program code may be a single instruction or many instructions, and may even be distributed across several different code segments, different programs, and across several storage devices. Similarly, operational data may be identified and illustrated within the circuitry and may be embodied in any suitable form and organized in any suitable type of data structure. The operational data may be collected as a single data set or distributed across different locations, including different storage devices, and may exist, at least in part, simply as electronic signals on a system or network.

[0162] A computer-readable medium (also referred to herein as a machine-readable medium or machine-readable content) can be a tangible computer-readable storage medium that stores computer-readable program code. A computer-readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, holographic, micro-electromechanical, or semiconductor system, device, or apparatus, or any suitable combination of the foregoing. As described above, examples of computer-readable storage media can include, but are not limited to, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), an optical storage device, a magnetic storage device, a holographic storage medium, a micro-electromechanical storage device, or any suitable combination of the foregoing. In the context of this article, a computer-readable storage medium can be any tangible medium that can contain and / or store computer-readable program code for use by and / or in connection with an instruction execution system, device, or apparatus.

[0163] Computer-readable media may also be computer-readable signal media. A computer-readable signal medium may include a propagated data signal in which computer-readable program code is embedded, for example, in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including but not limited to electrical, electromagnetic, magnetic, optical, or any suitable combination thereof. A computer-readable signal medium may be any computer-readable medium that is not a computer-readable storage medium and is capable of communicating, propagating, or transmitting computer-readable program code for use by or in connection with an instruction execution system, device, or apparatus. As described above, the computer-readable program code embedded in a computer-readable signal medium may be transmitted using any suitable medium, including but not limited to wireless, wired, fiber optic cable, radio frequency (RF), or any suitable combination of the foregoing. In one embodiment, the computer-readable medium may include a combination of one or more computer-readable storage media and one or more computer-readable signal media. For example, the computer-readable program code may be propagated as an electromagnetic signal via a fiber optic cable for execution by a processor, or may be stored on a RAM storage device for execution by a processor.

[0164] The computer readable program code for performing the operations of various aspects of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, Smalltalk, C++ or similar languages, and conventional procedural programming languages ​​such as the "C" programming language or similar programming languages. The computer readable program code may be executed entirely on a local computer (e.g., via Figure 1 and 2 The software may be executed partially on the local computer, as a stand-alone computer-readable software package, partially on the local computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using the Internet through an Internet service provider).

[0165] The program code may also be stored in a computer-readable medium that can direct a computer, other programmable data processing device or other device to operate in a specific manner so that the instructions stored in the computer-readable medium produce an article of manufacture that includes instructions for implementing the functions / behaviors specified in one or more blocks of the schematic flowchart and / or schematic block diagram.

[0166] Therefore, the present disclosure may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects as illustrative only and not restrictive. The scope of the present disclosure is therefore indicated by the appended claims rather than by the foregoing description. All variations that come within the meaning and range of equivalence of the claims are intended to be included within their scope.

Claims

1. A computing system, characterized in that: include: A processing circuit comprising one or more memory devices coupled to one or more processors, the one or more memory devices being configured to store instructions that, when executed by the one or more processors, cause the processing circuit to: receiving an indication of a hydrocarbon injection event in a vehicle; receiving a plurality of parameters associated with operation of the vehicle during a first time range associated with the hydrocarbon injection event; predicting temperature data for a second time range subsequent to the first time range using a model based on the plurality of parameters, the temperature data being related to an aftertreatment system of the vehicle; in response to the temperature data being greater than a threshold, calculating a second injection rate based on the predicted temperature for the second time frame using the model, the second injection rate being lower than a first injection rate associated with the hydrocarbon injection event; as well as In response to the calculating, the injector is commanded to inject hydrocarbons based on the second injection rate.

2. The computing system according to claim 1, wherein: The plurality of parameters includes at least one of an inlet temperature of a catalyst component of the aftertreatment system, an outlet temperature of the catalyst component of the aftertreatment system, an exhaust flow rate from an engine coupled to the aftertreatment system, or the first injection rate of the engine.

3. The computing system according to claim 1, wherein: To calculate the second injection rate, the instructions, when executed by the one or more processors, further cause the processing circuit to perform the following operations: determining a temperature difference between the temperature data and a target temperature; converting the temperature difference into a hydrocarbon amount; converting the hydrocarbon amount into a third injection rate; as well as The second injection rate is calculated based on a difference between the first injection rate and the third injection rate.

4. The computing system according to claim 1, wherein: When the instructions are executed by the one or more processors, the processing circuitry is further caused to perform the following operations: receiving a target temperature associated with a temperature near an outlet of a catalyst component of the aftertreatment system, the target temperature for the hydrocarbon injection event using the first injection rate; determining that the target temperature is equal to or greater than a predetermined threshold; as well as The injector is commanded to inject hydrocarbons using the second injection rate based on the target temperature being equal to or greater than the predetermined threshold. The computing system according to claim 1 , wherein: When the instructions are executed by the one or more processors, the processing circuitry is further caused to perform the following operations: receiving an outlet temperature of a catalyst component of the aftertreatment system associated with the hydrocarbon injection event using the first injection rate; and determining that the outlet temperature of the catalyst component is equal to or greater than a predetermined threshold; as well as The injector is commanded to inject hydrocarbons using the second injection rate based on the outlet temperature of the catalyst component being equal to or greater than the predetermined threshold. The computing system according to claim 1 , wherein: The operation of the vehicle is related to at least one of a speed of an engine of the vehicle, a torque of the engine of the vehicle, or an operating state of a heater of the vehicle.

7. The computing system according to claim 1, wherein: The temperature data regarding the aftertreatment system corresponds to a temperature near an outlet of a catalyst component of the aftertreatment system.

8. The computing system according to claim 7, wherein: To predict the temperature data, when the instructions are executed by the one or more processors, the processing circuitry is further caused to perform the following operations: using the model, correlating the plurality of parameters associated with the operation of the vehicle for the first time frame with a predetermined plurality of parameters associated with a second operation of a second vehicle within a third time frame prior to the first time frame, wherein the predetermined plurality of parameters is associated with a pattern indicative of at least one change to the temperature near the outlet of the catalyst member over a predetermined duration; as well as The at least one change to the outlet temperature of the catalyst member within the second time frame is determined based on correlations between the plurality of parameters and the predetermined plurality of parameters and the pattern associated with the predetermined plurality of parameters.

9. The computing system according to claim 1, wherein: When the instructions are executed by the one or more processors, the processing circuitry is further caused to perform the following operations: identifying a plurality of values ​​corresponding to a plurality of time windows from the temperature data for the second time range, the plurality of time windows having a predetermined duration window size; inputting the plurality of values ​​into the model; receiving a rate of change of the plurality of values ​​within the second time range; as well as Using the model, the second injection rate for the injector is calculated based on the rate of change of the plurality of values.

10. The computing system according to claim 1, wherein: When the instructions are executed by the one or more processors, the processing circuitry further causes: receiving a second plurality of parameters relating to a second operation of at least one second vehicle different from the vehicle; providing the second plurality of parameters of the at least one second vehicle as input for training the model; as well as After training the model, the trained model is deployed to predict the temperature data for the second time range, wherein the trained model applies at least one of weights or biases to the plurality of parameters to predict the temperature data.

11. A method, characterized in that include: receiving, by processing circuitry comprising one or more memory devices coupled to one or more processors, an indication of a hydrocarbon injection event in a vehicle; receiving, by the processing circuit, a plurality of parameters associated with operation of the vehicle within a first time range associated with the hydrocarbon injection event; predicting, by the processing circuit and using a model based on the plurality of parameters, temperature data for a second time range subsequent to the first time range, the temperature data being related to an aftertreatment system of the vehicle; calculating, by the processing circuitry, a second injection rate based on a predicted temperature for the second timeframe using the model in response to the temperature data being above a threshold, the second injection rate being lower than a first injection rate associated with the hydrocarbon injection event; as well as In response to the calculation, an injector is commanded, by the processing circuit, to inject hydrocarbons based on the second injection rate.

12. The method according to claim 11, characterized in that The plurality of parameters includes at least one of an inlet temperature of a catalyst component of the aftertreatment system, an outlet temperature of the catalyst component of the aftertreatment system, an exhaust flow rate from an engine coupled to the aftertreatment system, or the first injection rate of the engine.

13. The method according to claim 11, characterized in that Also includes: determining, by the processing circuit, a temperature difference between the temperature data and a target temperature; converting the temperature difference into a hydrocarbon amount by the processing circuit; converting the hydrocarbon amount into a third injection rate via the processing circuit; as well as The second injection rate is calculated, by the processing circuit, based on a difference between the first injection rate and the third injection rate.

14. The method according to claim 11, characterized in that Also includes: receiving, by the processing circuit, a target temperature associated with a temperature near an outlet of a catalyst component of the aftertreatment system, the target temperature for the hydrocarbon injection event using the first injection rate determining, by the processing circuit, that the target temperature is equal to or greater than a predetermined threshold; as well as The injector is commanded, by the processing circuit, to inject hydrocarbons using the second injection rate based on the target temperature being equal to or greater than the predetermined threshold.

15. The method according to claim 11, characterized in that The temperature data regarding the aftertreatment system corresponds to a temperature near an outlet of a catalyst component of the aftertreatment system.

16. The method according to claim 15, comprising: correlating, by the processing circuitry, using the model, the plurality of parameters associated with the operation of the vehicle for the first time frame with a predetermined plurality of parameters associated with a second operation of a second vehicle for a third time frame prior to the first time frame, wherein the predetermined plurality of parameters is associated with a pattern indicative of at least one change to the temperature near the outlet of the catalyst member over a predetermined duration; as well as The at least one change to the outlet temperature of the catalyst member within the second time frame is determined, by the processing circuitry, based on correlations between the plurality of parameters and the predetermined plurality of parameters and the pattern associated with the predetermined plurality of parameters.

17. A computing system, characterized in that include: one or more processors; as well as One or more storage devices coupled to the one or more processors, the one or more storage devices storing instructions that, when executed by the one or more processors, cause the one or more processors to: receiving at least one input to a model, the input comprising a plurality of parameters related to vehicle operation, during a first time range during which a first outlet temperature of a catalyst component of the aftertreatment system changes; using the model, correlating the plurality of parameters for the first time frame with a pattern associated with a predetermined plurality of parameters, wherein the pattern indicates at least one change in the first outlet temperature of the catalyst member; receiving a predicted second outlet temperature of the catalyst component associated with a second time frame subsequent to the first time frame based on a correlation between the plurality of parameters and the pattern; as well as In response to the predicted second outlet temperature being greater than a threshold, at least one component of the vehicle is commanded during the first time frame to reduce the predicted second outlet temperature of the catalyst component to equal to or below the threshold during the second time frame.

18. The computing system according to claim 17, wherein: The at least one component of the vehicle includes at least one of an injector configured to inject hydrocarbons, a heater, or an engine of the vehicle.

19. The computing system according to claim 17, wherein: The plurality of parameters includes at least one of an inlet temperature of the catalyst component, the first outlet temperature of the catalyst component, an exhaust flow rate from an engine of the vehicle, or an injection rate of the engine.

20. The computing system of claim 17, wherein: The predetermined plurality of parameters is associated with second operation of a second vehicle within a third time frame prior to the first time frame, and wherein the predetermined plurality of parameters is used to train the model to predict temperature data associated with the second outlet temperature of the catalyst component within the second time frame.

Images