A control method for eliminating the accumulation of hc emissions in the doc unit of an aftertreatment

Abstract

The present application relates to the technical field of vehicle aftertreatment, and discloses a control method for eliminating HC emission accumulation of a DOC unit of an aftertreatment device, comprising the following steps: step S101, obtaining an accumulated time length during which the exhaust temperature of an aftertreatment system is less than or equal to a preset exhaust temperature threshold; step S102, when the accumulated time length is greater than or equal to a preset time threshold, entering step S103; step S103, when an engine operating state parameter meets a regeneration condition, entering step S104; step S104, inputting an engine operating state sequence into a first neural network model, and outputting a value representing an opening degree adjustment value of a throttle valve; and step S105, recording an operating time length of the engine at the opening degree adjustment value of the throttle valve, and marking the operating time length as a regeneration time length; according to the present application, the accumulated low-temperature exhaust time length of the aftertreatment system and the engine operating state under the condition that the engine operating state meets the regeneration condition are monitored, the exhaust temperature is actively increased to oxidize the accumulated hydrocarbon in the DOC, and pollution of the environment caused by direct discharge is avoided.

Description

Technical Field

[0001] This invention relates to the field of vehicle aftertreatment technology, and more specifically, to a control method for eliminating HC emission buildup in the DOC unit of an aftertreatment system. Background Technology

[0002] To meet the emission requirements of China VI diesel engines, the mainstream aftertreatment technology used by major engine manufacturers for China VI diesel engines is DOC (oxidation catalyst) + DPF (wall-flow particulate filter) + SCR (selective catalytic reduction) + ASC (ammonia oxidation catalyst). Among them, DOC uses a catalyst containing precious metals such as platinum, palladium, and rhodium to further oxidize HC (hydrocarbons) and CO (carbon monoxide) in engine exhaust into non-toxic and harmless CO2 (carbon dioxide) and H2O (water) at temperatures above 250°C. At the same time, it can absorb soluble organic components and some carbon particles, reducing some PM (particulate matter) emissions. In addition, DOC can partially convert NO (nitric oxide) into NO2 (nitric oxide), and use NO2 as an oxidant to remove carbon particles in DPF to generate CO2. That is, DOC can be used for passive regeneration of DPF.

[0003] However, when the engine runs at idle speed or under low load for a long time, the engine exhaust temperature is low, the DOC conversion efficiency for HC is low, and the HC components in the exhaust cannot be further oxidized in the DOC. Some HC will be adsorbed on the surface of the DOC, and over time, HC will continuously deposit and accumulate on the surface of the DOC. Summary of the Invention

[0004] This invention provides a control method for eliminating HC emission buildup in the DOC unit of the after-processor, solving the technical problem in related technologies where HC continuously deposits and accumulates on the surface of the DOC when the engine operates at idle speed or low load for a long time.

[0005] This invention provides a control method for eliminating HC emission buildup in the DOC unit of a post-processor, comprising the following steps:

[0006] Step S101: Obtain the cumulative duration during which the exhaust temperature of the aftertreatment system is less than or equal to a preset exhaust temperature threshold.

[0007] Step S102: Determine whether the cumulative duration is greater than or equal to the preset duration threshold. If the cumulative duration is greater than or equal to the preset duration threshold, proceed to step S103; otherwise, return to step S101.

[0008] Step S103: Within the preset acquisition time period M, obtain the engine operating status parameters at n time points according to the preset time interval N, and determine whether the engine operating status parameters at the nth time point meet the regeneration conditions. If the regeneration conditions are met, proceed to step S104; otherwise, return to step S101.

[0009] Engine operating parameters include: engine speed, engine coolant temperature, after-treatment system temperature, fuel injection volume, vehicle speed, and gear.

[0010] Step S104: Construct an engine operating state sequence based on engine operating state parameters at n time points, and input the engine operating state sequence into the first neural network model. The output value represents the throttle valve opening adjustment value. The engine operating state sequence is represented as: A={a1…a n}, where a1…a n These represent the engine operating state parameters from the 1st time point to the nth time point; the first neural network model includes n time steps, with the i-th sequence unit of the engine operating state sequence being input at the i-th time step, where 1≤i≤n, and the value output at the n-th time step representing the throttle valve opening adjustment value;

[0011] Step S105: Adjust the throttle valve opening to the opening adjustment value, record the engine running time at the throttle valve opening, and mark it as the regeneration time.

[0012] Step S106: Determine whether the regeneration time is greater than or equal to the preset regeneration time threshold. If the regeneration time is greater than or equal to the preset regeneration time threshold, restore the throttle valve opening to the throttle valve opening before adjustment in step S105. Otherwise, continue to execute step S106.

[0013] Furthermore, the preset exhaust temperature threshold is a custom parameter.

[0014] Furthermore, the preset duration threshold is a custom parameter.

[0015] Furthermore, M and N are both user-defined parameters, and n = M / N.

[0016] Furthermore, the regeneration conditions include: engine speed greater than or equal to a preset engine speed threshold; engine coolant temperature greater than or equal to a preset engine coolant temperature threshold; after-treatment system temperature greater than or equal to a preset after-treatment system temperature threshold; cyclic fuel injection quantity greater than or equal to a preset cyclic fuel injection quantity threshold; vehicle speed greater than or equal to a preset vehicle speed threshold; and gear position greater than or equal to a preset gear position threshold. The preset engine speed threshold, preset engine coolant temperature threshold, preset after-treatment system temperature threshold, preset cyclic fuel injection quantity threshold, preset vehicle speed threshold, and preset gear position threshold are all user-defined parameters. All of these regeneration conditions must be met before proceeding to step S104.

[0017] Furthermore, the formula for calculating the t-th time step of the first neural network model includes:

[0018] Forgotten Gate f t The calculation formula is as follows: f t =σ(W f x t +U f h t-1 +b f ), where W f U represents the first weight parameter. f b represents the second weighting parameter. f Indicates the first bias parameter;

[0019] Input gate i t The calculation formula is as follows: i t =σ(W i x t +U i h t-1 +b i ), where W i U represents the third weighting parameter. i This represents the fourth weighting parameter, b. i Indicates the second bias parameter;

[0020] intermediate state The calculation formula is as follows: Among them W C U represents the fifth weighting parameter. C This represents the sixth weighting parameter, b. C Indicates the third bias parameter;

[0021] Cell state C t The calculation formula is as follows: Where C t-1 This represents the cell state at time step t;

[0022] Output gate o t The calculation formula is as follows: o t =σ(W o x t +U o h t-1 +b o ), where W o U represents the seventh weighting parameter. o This represents the eighth weighting parameter, b. o Indicates the fourth bias parameter;

[0023] Output status h t The calculation formula is as follows: h t =o t ⊙tanh(Ct );

[0024] Definition: 1≤t≤n, x t h represents the t-th sequence unit of the engine operating state sequence input at the t-th time step. t-1 Let represent the output state at time step t-1, σ represent the sigmoid activation function, tanh represent the hyperbolic tangent function, ⊙ represent pointwise multiplication, h0 = 0, and C0 = 0.

[0025] Furthermore, the sample labels corresponding to the training samples in the training dataset used to train the first neural network model are obtained through simulation experiments, including the following steps:

[0026] Step S201: Randomly generate engine operating state parameters that meet the regeneration conditions;

[0027] Step S202: Input the engine operating status parameters into the simulation platform and randomly output a set of throttle valve opening adjustment values ​​of quantity K.

[0028] Step S203: According to the opening adjustment value of the number K throttle valves, adjust the opening of the throttle valves to the opening adjustment value respectively, and record the HC content on the DOC surface within the same regeneration time corresponding to each opening adjustment value.

[0029] Step S204: Select the aperture adjustment value corresponding to the minimum HC content on the DOC surface as a sample label for a training sample;

[0030] Step S205: Repeat steps S201 to S204 until sample labels for J training samples are generated.

[0031] Furthermore, both K and J are user-defined parameters.

[0032] Furthermore, the preset regeneration time threshold is obtained through a second neural network model. The input of the second neural network model is the same as that of the first neural network model, and the output value represents the preset regeneration time threshold and the regeneration time limit value under the throttle valve opening adjustment value output by the first neural network model.

[0033] Furthermore, the sample labels corresponding to the training samples in the training dataset used to train the second neural network model are obtained through simulation experiments, including the following steps:

[0034] Step S301: Randomly generate engine operating status parameters that meet the regeneration conditions;

[0035] Step S302: Input the engine operating status parameters into the simulation platform and randomly output a set of L throttle valve opening adjustment values, where L is a user-defined parameter.

[0036] Step S303: According to the opening adjustment values ​​of the L throttle valves, adjust the opening of the throttle valves to the opening adjustment values ​​respectively, and record the time for each opening adjustment value to reach the passive regeneration condition.

[0037] Step S304: Select the average duration of time required to achieve passive regeneration conditions as a training label for a training sample;

[0038] Step S305: Repeat steps S301 to S304 until sample labels for P training samples are generated, where P is a user-defined parameter.

[0039] The beneficial effects of this invention are as follows: This invention monitors the cumulative low-temperature exhaust time of the aftertreatment system and determines that the engine operating status meets the regeneration conditions to actively trigger the driving regeneration function, thereby increasing the exhaust temperature to oxidize the hydrocarbons accumulated in DOC and generate H2O and CO2 for discharge. This avoids the direct discharge of accumulated hydrocarbons and polluting the environment, and also avoids the emission of large amounts of HC to produce white smoke, which could mislead users into thinking that the engine is malfunctioning and affect the engine's market reputation. Attached Figure Description

[0040] Figure 1 This is a flowchart of a control method for eliminating HC emission buildup in the DOC unit of a post-processor according to the present invention;

[0041] Figure 2 This is a flowchart of the present invention, which obtains the sample labels corresponding to the training samples of the training dataset used to train the first neural network model through simulation experiments.

[0042] Figure 3 This is a flowchart illustrating the process of obtaining sample labels corresponding to training samples in a training dataset used to train a second neural network model through simulation experiments, as described in this invention. Detailed Implementation

[0043] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0044] It should be noted that, unless otherwise defined, the technical or scientific terms used in one or more embodiments of the present invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in one or more embodiments of the present invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0045] like Figures 1-3 As shown, a control method for eliminating HC emission buildup in the DOC unit of a post-processor includes the following steps:

[0046] Step S101: Obtain the cumulative duration during which the exhaust temperature of the aftertreatment system is less than or equal to a preset exhaust temperature threshold.

[0047] Step S102: Determine whether the cumulative duration is greater than or equal to the preset duration threshold. If the cumulative duration is greater than or equal to the preset duration threshold, proceed to step S103; otherwise, return to step S101.

[0048] Step S103: Within the preset acquisition time period M, obtain the engine operating status parameters at n time points according to the preset time interval N, and determine whether the engine operating status parameters at the nth time point meet the regeneration conditions. If the regeneration conditions are met, proceed to step S104; otherwise, return to step S101.

[0049] Engine operating parameters include: engine speed, engine coolant temperature, after-treatment system temperature, fuel injection volume, vehicle speed, and gear.

[0050] Step S104: Construct an engine operating state sequence based on engine operating state parameters at n time points, and input the engine operating state sequence into the first neural network model. The output value represents the throttle valve opening adjustment value.

[0051] The engine operating state sequence is represented as: A = {a1…a2} n}, where a1…a n These represent the engine operating status parameters from the 1st time point to the nth time point;

[0052] The first neural network model includes n time steps. The i-th time step is input to the i-th sequence unit of the engine operating state sequence, where 1≤i≤n. The value output at the n-th time step represents the throttle valve opening adjustment value.

[0053] Step S105: Adjust the throttle valve opening to the opening adjustment value, record the engine running time at the throttle valve opening, and mark it as the regeneration time.

[0054] Step S106: Determine whether the regeneration time is greater than or equal to the preset regeneration time threshold. If the regeneration time is greater than or equal to the preset regeneration time threshold, restore the throttle valve opening to the throttle valve opening before adjustment in step S105. Otherwise, continue to execute step S106.

[0055] In one embodiment of the present invention, the preset exhaust temperature threshold is a custom parameter, preferably set to 100°C.

[0056] Specifically, an exhaust gas analyzer can be connected to the engine exhaust pipe and the exhaust temperature can be adjusted to measure the HC (hydrocarbon) content in the exhaust under different temperature conditions. When the HC content is greater than or equal to the HC content threshold of the China VI standard, the corresponding temperature is selected as the preset exhaust temperature threshold. For example, the HC content threshold of the China VI standard is 0.1g / km.

[0057] In one embodiment of the present invention, the preset duration threshold is a custom parameter, preferably set to 30 minutes.

[0058] Specifically, the engine can be run at the exhaust temperature corresponding to the preset exhaust temperature threshold, and different exhaust durations can be adjusted. The HC content on the surface of DOC (oxidation catalyst) can be detected by a hydrocarbon analyzer at different exhaust durations. When the HC content is greater than or equal to the set threshold, the corresponding exhaust duration is selected as the preset duration threshold, for example, the set threshold is set to 50mg / m².

[0059] In one embodiment of the present invention, M and N are both custom parameters, n = M / N. Preferably, M is set to 1 minute and N is set to 5 seconds, then n is 12.

[0060] In one embodiment of the present invention, the engine speed is obtained by a crankshaft sensor, the engine coolant temperature and aftertreatment system temperature are obtained by a temperature sensor, the cyclic fuel injection quantity is obtained by a fuel injection quantity sensor, the vehicle speed is obtained by a sensor, and the gear position is obtained by a sensor inside the gearbox or transmission system.

[0061] In one embodiment of the present invention, the regeneration conditions include: engine speed greater than or equal to a preset engine speed threshold; engine coolant temperature greater than or equal to a preset engine coolant temperature threshold; after-treatment system temperature greater than or equal to a preset after-treatment system temperature threshold; cyclic fuel injection quantity greater than or equal to a preset cyclic fuel injection quantity threshold; vehicle speed greater than or equal to a preset vehicle speed threshold; and gear position greater than or equal to a preset gear threshold. The preset engine speed threshold, preset engine coolant temperature threshold, preset after-treatment system temperature threshold, preset cyclic fuel injection quantity threshold, preset vehicle speed threshold, and preset gear threshold are all user-defined parameters. All regeneration conditions must be met before proceeding to step S104.

[0062] Preferably, the preset engine speed threshold is set to the engine's rated speed; the preset engine coolant temperature threshold is set to the lower limit of the engine's normal operating coolant temperature, for example, if the normal operating coolant temperature range of the engine is between 85℃ and 100℃, then the preset engine coolant temperature threshold is set to 85℃; the preset aftertreatment system temperature threshold is set to the lower limit of the aftertreatment system's normal operating temperature, for example, if the normal operating temperature range of the aftertreatment system is between 250℃ and 350℃, then the preset aftertreatment system temperature threshold is set to 250℃; and the preset cyclic fuel injection quantity threshold is set to the lower limit of the cyclic fuel injection quantity of the engine during normal operation, for example, if the cyclic fuel injection quantity range of the engine during normal operation is 5mm... 3 / st (stroke) to 20mm 3 Between / st, the preset cyclic injection quantity threshold is set to 5mm. 3 / st; The preset vehicle speed threshold is set to the lower limit of the vehicle speed corresponding to normal engine operation. For example, if the engine speed is 3000 rpm, the tire diameter is 0.6 meters, and the gearbox ratio is 3.5 when the engine is operating normally, then the corresponding preset vehicle speed threshold can be obtained according to the vehicle speed calculation formula. The calculation formula for vehicle speed v is as follows: Where R represents the tire diameter, r represents the engine speed, and Per represents the gearbox ratio, which is provided by the engine manufacturer. Therefore, the vehicle speed v is 31.8 km / h, meaning the preset vehicle speed threshold is set to 31.8 km / h; the preset gear threshold is set to second gear.

[0063] In one embodiment of the present invention, the calculation formula for the t-th time step of the first neural network model includes:

[0064] Forgotten Gate f t The calculation formula is as follows: f t =σ(W f x t +U f h t-1 +b f ), where W f U represents the first weight parameter. f b represents the second weighting parameter.f Indicates the first bias parameter;

[0065] Input gate i t The calculation formula is as follows: i t =σ(W i x t +U i h t-1 +b i ), where W i U represents the third weighting parameter. i This represents the fourth weighting parameter, b. i Indicates the second bias parameter;

[0066] intermediate state The calculation formula is as follows: Among them W C U represents the fifth weighting parameter. C This represents the sixth weighting parameter, b. C Indicates the third bias parameter;

[0067] Cell state C t The calculation formula is as follows: Where C t-1 This represents the cell state at time step t;

[0068] Output gate o t The calculation formula is as follows: o t =σ(W o x t +U o h t-1 +b o ), where W o U represents the seventh weighting parameter. o This represents the eighth weighting parameter, b. o Indicates the fourth bias parameter;

[0069] Output status h t The calculation formula is as follows: h t =o t ⊙tanh(C t );

[0070] Definition: 1≤t≤n, x t h represents the t-th sequence unit of the engine operating state sequence input at the t-th time step. t-1 Let represent the output state at time step t-1, σ represent the sigmoid activation function, tanh represent the hyperbolic tangent function, ⊙ represent pointwise multiplication, h0 = 0, and C0 = 0.

[0071] In one embodiment of the present invention, the sample labels corresponding to the training samples in the training dataset used to train the first neural network model are obtained through simulation experiments, such as... Figure 2 As shown, it includes the following steps:

[0072] Step S201: Randomly generate engine operating state parameters that meet the regeneration conditions;

[0073] Step S202: Input the engine operating status parameters into the simulation platform and randomly output a set of throttle valve opening adjustment values ​​of quantity K.

[0074] Step S203: According to the opening adjustment value of the number K throttle valves, adjust the opening of the throttle valves to the opening adjustment value respectively, and record the HC content on the DOC surface within the same regeneration time corresponding to each opening adjustment value.

[0075] Step S204: Select the aperture adjustment value corresponding to the minimum HC content on the DOC surface as a sample label for a training sample;

[0076] Step S205: Repeat steps S201 to S204 until sample labels for J training samples are generated.

[0077] In one embodiment of the present invention, K and J are both custom parameters. Preferably, K is set to 5 and J is set to 200.

[0078] It should be noted that step S203 can also record the duration of reaching the passive regeneration condition for each opening adjustment value, and select the opening adjustment value corresponding to the shortest duration of reaching the passive regeneration condition as a sample label for a training sample. The passive regeneration condition is that the DOC temperature reaches 250°C.

[0079] In one embodiment of the present invention, the first neural network model may also be constructed based on RNN (Recurrent Neural Network) or GRU (Gated Recurrent Unit).

[0080] In one embodiment of the present invention, the preset regeneration time threshold is obtained through a second neural network model. The input of the second neural network model is the same as the input of the first neural network model, and the output value represents the preset regeneration time threshold and the regeneration time limit value under the throttle valve opening adjustment value output by the first neural network model.

[0081] In one embodiment of the present invention, the sample labels corresponding to the training samples in the training dataset used to train the second neural network model are obtained through simulation experiments, such as... Figure 3 As shown, it includes the following steps:

[0082] Step S301: Randomly generate engine operating status parameters that meet the regeneration conditions;

[0083] Step S302: Input the engine operating status parameters into the simulation platform and randomly output a set of L throttle valve opening adjustment values.

[0084] Step S303: According to the opening adjustment values ​​of the L throttle valves, adjust the opening of the throttle valves to the opening adjustment values ​​respectively, and record the time for each opening adjustment value to reach the passive regeneration condition.

[0085] Step S304: Select the average duration of time required to achieve passive regeneration conditions as a training label for a training sample;

[0086] Step S305: Repeat steps S301 to S304 until sample labels for P training samples are generated.

[0087] In one embodiment of the present invention, L and P are both custom parameters. Preferably, L is set to 5 and P is set to 100.

[0088] In one embodiment of the present invention, the preset regeneration time threshold can also be set to a fixed value, for example, the preset regeneration time threshold can be set to 5 minutes.

[0089] In one embodiment of the present invention, in addition to increasing the DOC temperature by adjusting the opening of the throttle valve, the DOC temperature can also be increased by installing an electronic heating device. Accordingly, the value output by the first neural network model represents the temperature adjustment value of the electronic heating device.

[0090] The embodiments of this example have been described above. However, this example is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this example, and all of them are within the protection scope of this example.

Claims

1. A control method for eliminating HC emission buildup in a DOC (Discharge Control Organism) unit of a post-processor, characterized in that, Includes the following steps: Step S101: Obtain the cumulative duration during which the exhaust temperature of the aftertreatment system is less than or equal to a preset exhaust temperature threshold. Step S102: Determine whether the cumulative duration is greater than or equal to the preset duration threshold. If the cumulative duration is greater than or equal to the preset duration threshold, proceed to step S103; otherwise, return to step S101. Step S103: Within the preset acquisition time period M, obtain the engine operating status parameters at n time points according to the preset time interval N, and determine whether the engine operating status parameters at the nth time point meet the regeneration conditions. If the regeneration conditions are met, proceed to step S104; otherwise, return to step S101. Engine operating parameters include: engine speed, engine coolant temperature, after-treatment system temperature, fuel injection volume, vehicle speed, and gear. Step S104: Construct an engine operating state sequence based on engine operating state parameters at n time points, and input the engine operating state sequence into the first neural network model. The output value represents the throttle valve opening adjustment value; the engine operating state sequence is represented as: ,in These represent the engine operating state parameters from the 1st time point to the nth time point; the first neural network model includes n time steps, with the i-th sequence unit of the engine operating state sequence being input at the i-th time step, where 1≤i≤n, and the value output at the n-th time step representing the throttle valve opening adjustment value; The sample labels corresponding to the training samples in the training dataset used to train the first neural network model were obtained through simulation experiments, including the following steps: Step S201: Randomly generate engine operating state parameters that meet the regeneration conditions; Step S202: Input the engine operating status parameters into the simulation platform and randomly output a set of throttle valve opening adjustment values ​​of quantity K. Step S203: According to the opening adjustment value of the number K throttle valves, adjust the opening of the throttle valves to the opening adjustment value respectively, and record the HC content on the DOC surface within the same regeneration time corresponding to each opening adjustment value. Step S204: Select the aperture adjustment value corresponding to the minimum HC content on the DOC surface as a sample label for a training sample; Step S205: Repeat steps S201 to S204 until sample labels for J training samples are generated; K and J are user-defined parameters. Step S105: Adjust the throttle valve opening to the opening adjustment value, record the engine running time at the throttle valve opening, and mark it as the regeneration time. Step S106: Determine whether the regeneration time is greater than or equal to the preset regeneration time threshold. If the regeneration time is greater than or equal to the preset regeneration time threshold, restore the throttle valve opening to the throttle valve opening before adjustment in step S105. Otherwise, continue to execute step S106.

2. The control method for eliminating HC emission buildup in the DOC unit of the post-processor according to claim 1, characterized in that, The preset exhaust temperature threshold is a custom parameter.

3. The control method for eliminating HC emission buildup in the DOC unit of the post-processor according to claim 1, characterized in that, The preset duration threshold is a custom parameter.

4. The control method for eliminating HC emission buildup in the DOC unit of the post-processor according to claim 1, characterized in that, Both M and N are user-defined parameters, and n = M / N.

5. The control method for eliminating HC emission buildup in the DOC unit of the post-processor according to claim 1, characterized in that, The regeneration conditions include: engine speed greater than or equal to a preset engine speed threshold; engine coolant temperature greater than or equal to a preset engine coolant temperature threshold; after-treatment system temperature greater than or equal to a preset after-treatment system temperature threshold; cyclic fuel injection quantity greater than or equal to a preset cyclic fuel injection quantity threshold; vehicle speed greater than or equal to a preset vehicle speed threshold; and gear position greater than or equal to a preset gear threshold. The preset engine speed threshold, preset engine coolant temperature threshold, preset after-treatment system temperature threshold, preset cyclic fuel injection quantity threshold, preset vehicle speed threshold, and preset gear threshold are all user-defined parameters. All of these regeneration conditions must be met before proceeding to step S104.

6. The control method for eliminating HC emission buildup in the DOC unit of the post-processor according to claim 1, characterized in that, The formula for calculating the t-th time step of the first neural network model includes: Forgotten Gate The calculation formula is as follows: ,in This represents the first weight parameter. This represents the second weighting parameter. Indicates the first bias parameter; Input gate The calculation formula is as follows: ,in This represents the third weighting parameter. This represents the fourth weighting parameter. Indicates the second bias parameter; intermediate state The calculation formula is as follows: ,in This represents the fifth weighting parameter. This represents the sixth weight parameter. Indicates the third bias parameter; Cell state The calculation formula is as follows: ,in This represents the cell state at time step t-1; Output gate The calculation formula is as follows: ,in This represents the seventh weighting parameter. This represents the eighth weighting parameter. Indicates the fourth bias parameter; Output status The calculation formula is as follows: ; Definition: 1≤t≤n, This represents the t-th sequence unit of the engine operating state sequence input at time step t. This indicates the output state at time step (t-1). denoted by sigmoid, and tanh by hyperbolic tangent. This indicates point-by-point multiplication. , .

7. The control method for eliminating HC emission buildup in the DOC unit of the post-processor according to claim 1, characterized in that, The preset regeneration time threshold is obtained through a second neural network model. The input of the second neural network model is the same as that of the first neural network model. The output value represents the preset regeneration time threshold and the regeneration time limit value under the throttle valve opening adjustment value output by the first neural network model.

8. The control method for eliminating HC emission buildup in the DOC unit of the post-processor according to claim 7, characterized in that, The sample labels corresponding to the training samples in the training dataset used to train the second neural network model were obtained through simulation experiments, including the following steps: Step S301: Randomly generate engine operating status parameters that meet the regeneration conditions; Step S302: Input the engine operating status parameters into the simulation platform and randomly output a set of L throttle valve opening adjustment values, where L is a user-defined parameter. Step S303: According to the opening adjustment value of the L throttle valves, adjust the opening of the throttle valves to the opening adjustment value respectively, and record the time to reach the passive regeneration condition corresponding to each opening adjustment value, wherein the passive regeneration condition is that the DOC temperature reaches 250℃. Step S304: Select the average duration of time required to achieve passive regeneration conditions as a training label for a training sample; Step S305: Repeat steps S301 to S304 until sample labels for P training samples are generated, where P is a user-defined parameter.

Images