Method and device for determining the degree of egr clogging

Abstract

The application provides a method and a device for determining the degree of EGR blockage. The method comprises: obtaining EGR parameters and engine parameters at multiple time points in a predetermined time period to obtain current EGR parameters and current engine parameters; analyzing the current EGR parameters and the current engine parameters using an LSTM neural network model to determine the current EGR flow, wherein the LSTM neural network model is obtained by machine learning training using multiple sets of data, each set of data in the multiple sets of data comprising: historical EGR parameters, historical engine parameters and historical EGR flow; calculating the variance of the current EGR flow, calculating the deviation index of the current EGR flow according to the variance, obtaining a preset threshold, and determining the degree of EGR blockage according to the size relationship between the deviation index and the preset threshold. Through the application, the problem that the EGR monitoring method in the prior art cannot accurately monitor the degree of EGR blockage is solved.

Description

Technical Field

[0001] This application relates to the field of determining EGR blockage, and more specifically, to a method and apparatus for determining the degree of EGR blockage. Background Technology

[0002] EGR (Exhaust Gas Recirculation) introduces a portion of exhaust gas from the exhaust pipe, cools it in the EGR cooler, and then enters the cylinders through the intake manifold after the throttle valve, thereby reducing NOx emissions. However, due to harsh environments and the engine's lifespan, residual oil in the exhaust gas or carbon deposits from the engine itself can clog the EGR valve or EGR lines, causing reduced EGR flow or even valve sticking. In severe cases, this can lead to vehicle vibration and frequent malfunctions. Furthermore, in low winter temperatures, the EGR system is prone to freezing. Currently, EGR systems have a relatively high failure rate in the market, which can easily cause customer dissatisfaction. Therefore, timely monitoring of the EGR system is necessary.

[0003] Existing EGR monitoring strategies are often limited by the computing power, data storage capacity, and cost of ECU (Electronic Control Unit) hardware, and can only utilize historical data for a limited period of time for calculations. Since the EGR system itself is a life-like system, its structure and parameters gradually change over time, leading to performance degradation as its lifespan ages. Existing methods cannot accurately reflect the changing health status of the EGR system over time, thus affecting the accuracy of estimated EGR flow.

[0004] Therefore, a method is needed to accurately monitor whether EGR is blocked. Summary of the Invention

[0005] The main objective of this application is to provide a method and apparatus for determining the degree of EGR blockage, so as to at least solve the problem that the prior art cannot accurately monitor whether EGR blockage has occurred.

[0006] To achieve the above objectives, according to one aspect of this application, a method for determining the degree of EGR blockage is provided, comprising: acquiring EGR parameters and engine parameters at multiple moments within a predetermined time period to obtain current EGR parameters and current engine parameters, wherein the EGR parameters include at least EGR upstream temperature, EGR upstream pressure, pressure difference across the EGR, and EGR valve opening, and the engine parameters include at least engine speed and engine fuel quantity; analyzing the current EGR parameters and current engine parameters using an LSTM neural network model to determine the current EGR flow rate, wherein the LSTM neural network model is obtained by machine learning training using multiple sets of data, each set of data including: historical EGR parameters, historical engine parameters, and historical EGR flow rate; calculating the variance of the current EGR flow rate, calculating the deviation index of the current EGR flow rate based on the variance, obtaining a preset threshold, and determining the degree of EGR blockage based on the relationship between the deviation index and the preset threshold.

[0007] Optionally, the variance of the current EGR flow is calculated, and the deviation index of the current EGR flow is calculated based on the variance, including: using the formula Calculate the variance of the current EGR traffic, where s represents the variance of the EGR traffic, n represents the quantity of the EGR traffic, and x... i This represents the i-th EGR flow. This represents the average of the n EGR flows; the variance is expressed by the formula... Calculate the deviation index of the current EGR flow, where A represents the average value of multiple EGR flows collected when the mileage of the vehicle containing the engine is less than a preset mileage, and k represents the deviation index.

[0008] Optionally, obtaining preset thresholds includes: obtaining EGR traffic datasets for each level of congestion, wherein the EGR traffic dataset is a dataset composed of multiple simulated EGR traffic obtained at each level of congestion; calculating the deviation index of each simulated EGR traffic in the EGR traffic dataset corresponding to each level of congestion; determining the largest deviation index in the EGR traffic dataset of the level with the lowest congestion as a first preset threshold; and determining the smallest deviation index in the EGR traffic datasets of other levels of congestion as other preset thresholds.

[0009] Optionally, the EGR flow dataset under multiple levels of congestion is obtained, including: controlling the operation of the engine equipped with EGR fault components of different congestion levels; obtaining EGR parameters and engine parameters during engine operation to obtain simulated EGR parameters and simulated engine parameters, wherein each EGR fault component corresponds to a level of congestion; and using the LSTM neural network model to analyze the simulated EGR parameters and simulated engine parameters to obtain multiple simulated EGR flows under each level of congestion, thereby obtaining the EGR flow dataset under multiple levels of congestion.

[0010] Optionally, calculating the deviation index of each simulated EGR flow in the EGR flow dataset corresponding to each level of congestion includes: calculating the average value of the EGR flow dataset for each level to obtain the average flow value at each level; calculating the variance of the EGR flow dataset at each level of congestion based on the EGR flow dataset for each level and the average flow value at each level corresponding to the EGR flow dataset for each level; and applying the variance using the formula... Calculate the deviation index of each simulated EGR flow in the EGR flow dataset at each level, where A represents the average of multiple EGR flows collected when the vehicle's mileage is less than a preset mileage, k represents the deviation index, s represents the variance of the EGR flow, and x... i This represents the i-th EGR flow.

[0011] Optionally, the degree of congestion of the EGR is determined based on the relationship between the deviation index and the preset threshold, including: if the deviation index is greater than or equal to the first preset threshold, the degree of congestion of the EGR is determined to be level one congestion; if the deviation index is less than the first preset threshold, the degree of congestion of the EGR is determined to be other levels of congestion corresponding to other preset thresholds.

[0012] Optionally, when the deviation index is less than the first preset threshold, determining the EGR's congestion level as other levels of congestion corresponding to other preset thresholds includes: when the deviation index is greater than or equal to the second preset threshold and less than the first preset threshold, determining the EGR's congestion level as level two congestion, where the second preset threshold is the preset threshold corresponding to level two congestion; when the deviation index is greater than or equal to the third preset threshold and less than the second preset threshold, determining the EGR's congestion level as level three congestion, where the third preset threshold is the preset threshold corresponding to level three congestion; when the deviation index is greater than or equal to the fourth preset threshold and less than the third preset threshold, determining the EGR's congestion level as level four congestion, where the fourth preset threshold is the preset threshold corresponding to level four congestion; and when the deviation index is greater than or equal to the fifth preset threshold and less than the fourth preset threshold, determining the EGR's congestion level as level five congestion, where the fifth preset threshold is the preset threshold corresponding to level five congestion.

[0013] Optionally, after determining the degree of EGR blockage, the determination method further includes: if the EGR blockage is determined to be Level 1 blockage, no action is taken; if the EGR blockage is determined to be Level 2 blockage, the EGR is controlled to activate an automatic cleaning function; if the EGR blockage is determined to be Level 3 blockage, a cleaning reminder message is generated and sent to the user terminal at a first preset frequency, wherein the cleaning reminder message is used to remind the user to go to a cleaning station to clean the EGR; if the EGR blockage is determined to be Level 4 blockage, the engine output torque is controlled to decrease to a first torque, and the cleaning reminder message is generated and sent to the user terminal at a second preset frequency, wherein the second preset frequency is greater than the first preset frequency; if the EGR blockage is determined to be Level 5 blockage, the engine output torque is controlled to decrease to a second torque, and the cleaning reminder message is generated and sent to the user terminal at a third preset frequency, wherein the second torque is less than the first torque, and the third preset frequency is greater than the second preset frequency.

[0014] Optionally, the LSTM neural network model includes an input gate, a forget gate, and an output gate. The LSTM neural network model is used to analyze the EGR parameters and the engine parameters to determine the EGR flow rate. This includes using the historical EGR parameters and historical engine parameters as inputs to an initial LSTM neural network model, and the historical EGR flow rate as the output of the initial LSTM neural network model, using formula F... t =σ(wf *[H t-1 ,X t ]+b f Calculate the forgetting factor of the forgetting gate using formula I. t =σ(w i *[H t-1 ,X t ]+b i Calculate the input factor of the input gate using the formula. Calculate the memory factor using formula O t =σ(w o *[H t-1 ,X t ]+b o ) Calculate the output factor of the output gate, train the initial LSTM neural network model to obtain the LSTM neural network model, where X t w represents the parameter matrix composed of the historical EGR parameters and the historical engine parameters. f w represents the weight of the forget gate. i w represents the weight of the input gate. o b represents the weight of the output gate. f b represents the deviation of the forget gate. i b represents the deviation of the input gate. o H represents the deviation of the output gate. t-1 Let C represent the hidden state at time t-1. t-1 This represents the state of the memory cell at time t-1. The candidate memory cell state is represented by σ, which represents the sigmoid function. The current EGR parameters and the current engine parameters are input into the LSTM neural network model, and the output of the LSTM neural network model is determined as the current EGR flow.

[0015] According to another aspect of this application, an apparatus for determining the degree of EGR blockage is provided, comprising: an acquisition unit, configured to acquire EGR parameters and engine parameters at multiple moments within a predetermined time period to obtain current EGR parameters and current engine parameters, wherein the EGR parameters include at least EGR upstream temperature, EGR upstream pressure, pressure difference across the EGR, and EGR valve opening, and the engine parameters include at least engine speed and engine oil quantity; a first determination unit, configured to analyze the current EGR parameters and the current engine parameters using an LSTM neural network model to determine the current EGR flow rate, wherein the LSTM neural network model is obtained by machine learning training using multiple sets of data, each set of data including: historical EGR parameters, historical engine parameters, and historical EGR flow rate; and a second determination unit, configured to calculate the variance of the current EGR flow rate, calculate the deviation index of the current EGR flow rate based on the variance, obtain a preset threshold, and determine the degree of EGR blockage based on the relationship between the deviation index and the preset threshold.

[0016] By applying the technical solution of this application, current EGR parameters and current engine parameters are obtained. These parameters are then input into an LSTM neural network model for analysis, predicting the current EGR flow rate and calculating its variance and deviation index. A preset threshold is then obtained. The degree of EGR blockage is determined by the relationship between the preset threshold and the deviation index. This allows for accurate prediction of EGR flow rate first, followed by determination of whether EGR blockage has occurred and the degree of blockage. Compared to existing technologies where EGR monitoring strategies are limited by ECU hardware computing power, data storage capacity, and cost, often relying on historical data for only a limited period, this application can accurately predict EGR flow rate through a neural network model and further calculate the deviation index to determine the degree of EGR blockage. Therefore, it solves the problem of inaccurate monitoring of EGR blockage in existing technologies, achieving the goal of accurately monitoring the degree of EGR blockage. Attached Figure Description

[0017] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 A hardware structure block diagram of a mobile terminal for performing a method for determining the degree of EGR congestion, according to an embodiment of this application, is shown.

[0019] Figure 2 A flowchart illustrating a method for determining the degree of EGR congestion provided in an embodiment of this application is shown.

[0020] Figure 3 The illustration shows a flowchart of obtaining a preset threshold in a method for determining the degree of EGR congestion provided in an embodiment of this application;

[0021] Figure 4 A flowchart illustrating a specific method for determining the degree of EGR congestion provided in an embodiment of this application is shown;

[0022] Figure 5 The diagram illustrates the structure of an LSTM neural network model in a specific method for determining the degree of EGR congestion provided in an embodiment of this application.

[0023] Figure 6 A structural block diagram of an EGR clogging determination device provided in an embodiment of this application is shown.

[0024] The above figures include the following reference numerals:

[0025] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device. Detailed Implementation

[0026] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0027] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] For ease of description, the following explains some of the nouns or terms used in the embodiments of this application:

[0030] EGR system: Exhaust Gas Recirculation, or EGR system for short, is a device that, under certain conditions, introduces a portion of the exhaust gas discharged from the engine into the cylinder to participate in combustion again.

[0031] LSTM: Long Short-Term Memory Neural Network, is a type of time-recurrent neural network designed specifically to address the long-term dependency problem inherent in general RNNs (Recurrent Neural Networks).

[0032] As described in the background section, existing EGR monitoring methods cannot accurately monitor the degree of EGR blockage. To address the problem of inaccurate monitoring of EGR blockage, embodiments of this application provide a method and apparatus for determining the degree of EGR blockage.

[0033] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0034] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a method of determining EGR congestion level according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0035] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the EGR congestion determination method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the aforementioned networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0036] This embodiment provides a method for determining the degree of EGR congestion on a mobile terminal, computer terminal, or similar computing device. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0037] Figure 2 This is a flowchart of a method for determining the degree of EGR congestion according to an embodiment of this application. Figure 2 As shown, the method includes the following steps:

[0038] Step S201: Obtain EGR parameters and engine parameters at multiple times within a predetermined time period to obtain current EGR parameters and current engine parameters. The EGR parameters include at least EGR upstream temperature, EGR upstream pressure, EGR pressure difference across the EGR terminals, and EGR valve opening. The engine parameters include at least engine speed and engine oil quantity.

[0039] Specifically, since EGR flow rate is strongly correlated with upstream EGR temperature, upstream EGR pressure, pressure difference across the EGR, and valve opening at the EGR location, to ensure the accuracy of EGR flow rate prediction, these signals are collected, and engine parameters (engine speed and fuel level) are added to form a data matrix. The onboard ECU collects the above basic information, namely EGR parameters and engine parameters, and uploads it wirelessly to the cloud computing center. The cloud computing center stores the data and analyzes and calculates it according to the LSTM neural network model described below. The predetermined time period can be a steady-state operating condition of the engine in the medium-to-high speed load range. The current EGR parameters and current engine parameters are collected at multiple moments within the predetermined time period; that is, the current EGR parameters and current engine parameters are data corresponding to multiple moments.

[0040] Step S202: Analyze the current EGR parameters and the current engine parameters using an LSTM neural network model to determine the current EGR flow rate. The LSTM neural network model is trained using multiple sets of data through machine learning. Each set of data includes: historical EGR parameters, historical engine parameters, and historical EGR flow rate.

[0041] Specifically, the LSTM neural network model achieves dynamic changes over a fixed time scale by using self-looping within the memory unit. It utilizes the memory unit to store long-term memory information and uses hidden states to transmit short-term memory information. Therefore, the LSTM neural network model is used to predict EGR flow. Historical EGR parameters, historical engine parameters, and historical EGR flow can be collected under normal vehicle operation, selecting a steady-state condition within the mid-to-high speed load range. This primarily involves selecting a region where EGR is active, collecting one year's worth of data from a single vehicle as a reference, with missing data filled using interpolation. We will first analyze a single vehicle as an example, and later extend this to a fleet of vehicles. Under the selected operating condition, the cloud collects the EGR flow in a region with a relatively short mileage for that vehicle, calculates the average value A, and stores it. At this point, the EGR valve is still a new valve, serving as a reference threshold.

[0042] Step S203: Calculate the variance of the current EGR flow, calculate the deviation index of the current EGR flow based on the variance, obtain a preset threshold, and determine the degree of EGR congestion based on the relationship between the deviation index and the preset threshold.

[0043] Specifically, after predicting the EGR flow, the variance of the EGR flow data is calculated, and a deviation index is further calculated based on the variance. The EGR flow deviation index is a statistical indicator that measures the degree of deviation from the actual value. A higher value indicates a smaller degree of deviation and less congestion, and vice versa. Therefore, the degree of EGR congestion is determined by setting a preset threshold and the relationship between the deviation index and the preset threshold.

[0044] This embodiment acquires current EGR parameters and current engine parameters, inputs them into an LSTM neural network model for analysis, predicts the current EGR flow rate, and calculates the variance and deviation index of the current EGR flow rate. A preset threshold is then obtained, and the degree of EGR blockage is determined by the relationship between the preset threshold and the deviation index. This allows for accurate prediction of EGR flow rate first, and then the determination of whether EGR blockage has occurred and the degree of blockage based on the predicted EGR flow rate. Compared to existing technologies where EGR monitoring strategies are limited by ECU hardware computing power, data storage capacity, and cost, often only utilizing historical data for a limited period and failing to accurately monitor EGR blockage and its degree, this application can accurately predict EGR flow rate through a neural network model and further calculate the deviation index to determine the degree of EGR blockage. Therefore, it solves the problem of inaccurate monitoring of EGR blockage in existing technologies, achieving the goal of accurately monitoring the degree of EGR blockage.

[0045] In the specific implementation process, the above step S203 can be achieved through the following steps: using the formula Calculate the variance of the current EGR flow, where s represents the variance of the EGR flow, n represents the quantity of the EGR flow, and x... i This represents the i-th EGR flow mentioned above. This represents the average of n EGR flow rates mentioned above; based on the variance described above, the formula is used... The deviation index of the current EGR flow is calculated, where A represents the average value of multiple EGR flows collected when the vehicle's mileage is less than a preset mileage, and k represents the deviation index. This method first calculates the variance, and then calculates the deviation index based on the variance. This allows for accurate calculation of the deviation index, thus revealing the deviation of the EGR flow and determining the degree of EGR congestion.

[0046] Specifically, after collecting multiple EGR traffic flows, the following is adopted: Calculate the variance, and then use... Calculate the deviation index k, classify the levels according to the range of the deviation index k, and determine the degree of EGR blockage.

[0047] To determine the degree of EGR congestion based on the deviation index, the above-mentioned step S203 of this application, obtaining the preset threshold, can also be achieved through the following steps, such as... Figure 3 As shown: Step S2031: Obtain EGR traffic datasets for each of the above EGR levels of congestion, wherein the above EGR traffic datasets are datasets composed of multiple simulated EGR traffic obtained at each level of congestion; Step S2032: Calculate the deviation index of each of the above simulated EGR traffic in the above EGR traffic dataset corresponding to each level of congestion, determine the largest deviation index in the above EGR traffic dataset of the lowest congestion level as the first preset threshold, and determine the smallest deviation index in the above EGR traffic datasets of other congestion levels as other preset thresholds. This method obtains EGR traffic datasets for multiple levels of congestion and calculates preset thresholds for different levels of congestion based on the EGR traffic datasets. In this way, the congestion level of EGR can be determined according to the relationship between the deviation index and the preset threshold.

[0048] In the specific implementation, multiple EGR traffic data points are acquired for each level of congestion; that is, multiple EGR traffic data points are acquired for each level of congestion, resulting in a set of EGR traffic datasets corresponding to each level of congestion. Each EGR traffic dataset may contain thousands of EGR traffic data points, and the deviation index for each EGR traffic point is calculated. Specifically, an average value is calculated for each EGR traffic dataset, and then... Calculate the variance. x represents the mean. i Let represent the i-th EGR flow, n represent the number of EGR flows mentioned above, and s represent the variance. Then, according to the formula... The deviation index corresponding to each EGR flow is calculated, where A represents the average EGR flow in the area with a smaller mileage, i.e., the average of multiple EGR flows collected when the mileage of the vehicle containing the engine is less than a preset mileage. Based on the above steps, multiple deviation indices are calculated for the EGR flow dataset corresponding to each level of congestion. Among the deviation indices corresponding to the lowest level of congestion, the one with the largest deviation index is taken as the first preset threshold. This indicates that when the deviation index is greater than the first preset threshold, the EGR congestion level is the lowest.

[0049] Step S2031 above can be achieved through the following steps: controlling the operation of the engine equipped with EGR fault components of different degrees of clogging; acquiring the EGR parameters and engine parameters during engine operation to obtain simulated EGR parameters and simulated engine parameters, wherein each EGR fault component corresponds to a level of clogging; analyzing the simulated EGR parameters and simulated engine parameters using the LSTM neural network model to obtain multiple simulated EGR flows under each level of clogging, thereby obtaining an EGR flow dataset under multiple levels of clogging. This method obtains simulated EGR parameters and simulated engine parameters by installing EGR fault components of different degrees of clogging. This allows the simulated parameters to be used as the EGR flow dataset corresponding to each level of clogging, further calculating the deviation index and determining a preset threshold.

[0050] Specifically, faulty components with EGR blockages of 20%, 40%, 60%, 80%, and 100% are fabricated. These faulty components are then sequentially installed on the engine. Simulated EGR parameters and simulated engine parameters are obtained. The aforementioned LSTM neural network model is run, and the EGR flow of the five sets of faulty components is recorded (each set may contain thousands of values). Each set of EGR flow is used as an EGR flow dataset, and the mean and deviation index of each set are calculated. In practice, the degree of blockage of the faulty components can also be classified in other ways.

[0051] To calculate the deviation index corresponding to each EGR flow dataset, in some embodiments, step S2032 can be implemented through the following steps: calculating the average value of each level of the simulated EGR flow dataset to obtain the average flow value at each level; calculating the variance of the EGR flow dataset under each level of congestion based on the EGR flow dataset at each level and the average flow value at each level corresponding to the EGR flow dataset at each level; and applying the variance using the formula... Calculate the deviation index for each simulated EGR flow in the aforementioned EGR flow datasets at each level, where A represents the average of multiple EGR flows collected when the vehicle's mileage is less than a preset mileage, k represents the deviation index, s represents the variance of the EGR flow, and x... i Let represent the i-th EGR flow. This method calculates the deviation index by calculating the variance of the EGR flow, thus accurately determining the deviation index and subsequently identifying the preset thresholds corresponding to different levels of congestion.

[0052] In the specific implementation process, the average value of each level of EGR traffic dataset is calculated, and then the variance of each EGR traffic dataset is calculated. Based on the variance, [the following steps are taken]. The deviation index for each EGR flow can be calculated.

[0053] To accurately describe and determine the degree of EGR congestion, step S203 of this application can also be implemented through the following steps: Step S2033: If the deviation index is greater than or equal to the first preset threshold, determine the degree of EGR congestion as Level 1 congestion; Step S2034: If the deviation index is less than the first preset threshold, determine the degree of EGR congestion as other levels of congestion corresponding to other preset thresholds. This method determines Level 1 congestion using the first preset threshold and determines other levels of congestion using other preset thresholds. This allows for a more accurate determination of the degree of EGR congestion.

[0054] Specifically, the degree of congestion can be divided into multiple levels, each with its own preset threshold. The lowest level of congestion is Level 1 congestion, which is slight congestion. If the congestion degree is divided into 20%, 40%, 60%, 80%, and 100%, then Level 1 congestion is defined as 20%. When the deviation index is greater than or equal to the first preset threshold, the EGR is determined to be Level 1 congestion. Correspondingly, other levels of congestion can be determined using other preset thresholds.

[0055] Step S2034 of this application can be implemented through the following steps: When the deviation index is greater than or equal to a second preset threshold and less than a first preset threshold, the EGR congestion level is determined to be level two congestion, and the second preset threshold is the preset threshold corresponding to level two congestion; when the deviation index is greater than or equal to a third preset threshold and less than the second preset threshold, the EGR congestion level is determined to be level three congestion, and the third preset threshold is the preset threshold corresponding to level three congestion; when the deviation index is greater than or equal to a fourth preset threshold and less than the third preset threshold, the EGR congestion level is determined to be level four congestion, and the fourth preset threshold is the preset threshold corresponding to level four congestion; when the deviation index is greater than or equal to a fifth preset threshold and less than the fourth preset threshold, the EGR congestion level is determined to be level five congestion, and the fifth preset threshold is the preset threshold corresponding to level five congestion. This method determines the congestion level based on each preset threshold, thus classifying the congestion level according to the preset thresholds and more accurately determining the EGR congestion level.

[0056] In practical implementation, the congestion level is divided into 20%, 40%, 60%, 80%, and 100%. Each level of congestion has a corresponding preset threshold. As mentioned above, if the deviation index is greater than or equal to the first preset threshold, it is a 20% level 1 congestion; if the deviation index is greater than or equal to the second preset threshold but less than the first preset threshold, the EGR congestion level is determined to be a level 2 congestion (40%); if the deviation index is greater than or equal to the third preset threshold but less than the second preset threshold, the EGR congestion level is determined to be a level 3 congestion (60%); if the deviation index is greater than or equal to the fourth preset threshold but less than the third preset threshold, the EGR congestion level is determined to be a level 4 congestion (80%); and if the deviation index is greater than or equal to the fifth preset threshold but less than the fourth preset threshold, the EGR congestion level is determined to be a level 5 congestion (100%). In practical applications, the first preset threshold is usually 2, the second preset threshold is usually 1.67, the third preset threshold is usually 1.33, the fourth preset threshold is usually 1, and the fifth preset threshold is usually 0.67.

[0057] To address different levels of blockage, the method of this application further includes the following steps: after determining the blockage level of the EGR, the determination method further includes: if the blockage level of the EGR is determined to be Level 1 blockage, no action is taken; if the blockage level of the EGR is determined to be Level 2 blockage, the EGR is controlled to activate an automatic cleaning function; if the blockage level of the EGR is determined to be Level 3 blockage, a cleaning reminder message is generated and sent to the user terminal at a first preset frequency, wherein the cleaning reminder message is used to prompt the user to go to a cleaning station to clean the EGR. The method involves cleaning the EGR system. If the EGR blockage is determined to be at level four, the engine output torque is reduced to a first torque, and a cleaning prompt is generated. This prompt is then sent to the user at a second preset frequency, where the second preset frequency is greater than the first preset frequency. If the EGR blockage is determined to be level five, the engine output torque is reduced to a second torque, and a cleaning prompt is generated. This prompt is then sent to the user at a third preset frequency, where the second torque is less than the first torque, and the third preset frequency is greater than the second preset frequency. This method employs different processing methods for different blockage levels, allowing for tailored EGR management to ensure stable engine and vehicle operation.

[0058] In the specific implementation, the frequency of prompts and torque limitations are set according to the severity of the blockage. At the initial stage of blockage, the blockage is less severe, resulting in a lower frequency of prompts and less torque limitation. The more severe the blockage, the higher the frequency of prompts and the greater the torque limitation. Specifically, it can be set as follows: For a level 1 blockage (20% blockage), no action is taken; for a 40% blockage, the EGR system activates automatic cleaning; for a 60% blockage, a weekly cleaning reminder is sent to the driver via a first preset frequency (once a week); for an 80% blockage, indicating severe blockage, initial torque limiting is implemented (controlling the engine torque below a certain threshold), and a daily cleaning reminder is sent to the driver via a second preset frequency (once a day); for a 100% blockage, severe torque limiting is implemented (controlling the engine torque further below a certain threshold), and a cleaning reminder is sent to the driver hourly via a third preset frequency (once an hour). This application does not limit the specific values ​​of the preset frequencies and torques mentioned above. In the case of 100% blockage, the torque is limited to the greatest extent, that is, the second torque is less than the first torque mentioned above, and the first torque may include the second torque.

[0059] Step S202 above can be implemented as follows: The LSTM neural network model includes an input gate, a forget gate, and an output gate. The LSTM neural network model is used to analyze the EGR parameters and engine parameters to determine the EGR flow rate. This includes using the historical EGR parameters and historical engine parameters as inputs to the initial LSTM neural network model, and the historical EGR flow rate as the output of the initial LSTM neural network model, through formula F... t =σ(w f *[H t-1 ,X t ]+b f Calculate the forgetting factor of the forgetting gate mentioned above using formula I. t =σ(w i *[H t-1 ,X t ]+b i Calculate the input factor of the above input gate using the formula. Calculate the memory factor using formula O t =σ(w o *[H t-1 ,X t ]+b oCalculate the output factor of the above output gate, train the above initial LSTM neural network model to obtain the above LSTM neural network model, where X t w represents the parameter matrix composed of the aforementioned historical EGR parameters and the aforementioned historical engine parameters. f w represents the weight of the forget gate mentioned above. i w represents the weight of the input gate mentioned above. o b represents the weight of the output gate mentioned above. f b represents the deviation of the forgetting gate mentioned above. i b represents the deviation of the above input gate. o H represents the deviation of the above output gate. t-1 Let C represent the hidden state at time t-1. t-1 This represents the state of the memory cell at time t-1. The state of the candidate memory cell is represented by σ, which represents the sigmoid function. The current EGR parameters and the current engine parameters are input into the LSTM neural network model, and the output of the LSTM neural network model is determined as the current EGR flow.

[0060] Specifically, the LSTM neural network introduces three gates (input gate, output gate, and forget gate) and two memory states (hidden state H). t Memory unit state C t The main function of the forget gate is to filter and discard transmitted information. In the forget gate, this is achieved through formula F. t =σ(w f *[H t-1 ,X t ]+b f The forgetting factor F was calculated. t And use this factor to process the memory unit C passed from the previous time. t-1 Forgetting is performed. The input gate is used to calculate the information of the current time, through formula I. t =σ(w i *[H t-1 ,X t ]+b i )and The input factor I was obtained respectively. t and candidate memory factors Both are through formula The memory factor C of the current step size is calculated. t The output gate is based on the output factor O. t and memory factor C t Output the hidden factor H of the current step size t Through formula O t =σ(w o *[Ht-1 ,X t ]+b o ) and H t =O t *tanh(C t ) to obtain, where X t This indicates EGR parameters and engine parameters collected from the cloud, w f w i w o w c These are the forget gate weights, input gate weights, output gate weights, and candidate memory cell state weights, respectively. f b i b o b c These are the forget gate bias, input gate bias, output gate bias, and candidate memory cell state bias, respectively. t-1 Let C represent the hidden state at time t-1. t Indicates the state of the memory cell. Represents the candidate memory cell state, σ represents the sigmoid function, and is expressed by the formula... The tanh function is obtained through the formula. To improve prediction accuracy, the number of network layers can be increased. Increasing the number of network layers enhances the model's learning ability and reduces bias. The EGR traffic predicted by the LSTM model is compared with the actual collected traffic, and the number of network layers and weight coefficients w are continuously optimized. f w i w o w c This ensures that the deviation between the model value and the actual value is less than a certain acceptable range.

[0061] To enable those skilled in the art to better understand the technical solution of this application, the implementation process of the method for determining the degree of EGR blockage in this application will be described in detail below with reference to specific embodiments.

[0062] This embodiment relates to a specific method for determining the degree of EGR blockage, such as... Figure 4 As shown, it includes the following steps:

[0063] Step S1: Begin;

[0064] Step S2: Select a steady-state operating condition within the medium-to-high speed composite region;

[0065] Step S3: Collect signals (EGR parameters and engine parameters) from the cloud for a vehicle under selected operating conditions over a period of time, including upstream EGR temperature, upstream EGR pressure, pressure difference across the EGR, EGR location, engine speed, and fuel quantity.

[0066] Step S4: Calculate the average EGR flow A during the new vehicle stage;

[0067] Step S5: Calculate the EGR flow prediction model using the LSTM algorithm. Figure 5 This is a schematic diagram of the LSTM neural network model. LSTM introduces three gates (input gate I...). t Output gate O t Forgotten Gate F t The two memory states are the hidden state H. t and memory cell state C t C t-1 H is a memory unit transferred from the previous time. t-1 Let X represent the hidden state at time t-1. t The table contains the EGR parameters and engine parameters obtained. Represents the candidate memory cell state, σ represents the sigmoid function, and tanh represents... function;

[0068] Step S6: Calculate the EGR flow deviation index k using the flow values ​​predicted by the model;

[0069] Step S7: Fabricate the EGR blockage fault component and install it on the engine;

[0070] Step S8: Run the same road spectrum (LSTM neural network model) as the LSTM model;

[0071] Step S9: Record the EGR flow rate of the 5 groups of faulty components and calculate the EGR flow rate deviation index;

[0072] Step S10: Divide the deviation index into 5 zones, representing the degree of EGR blockage, as shown in Table 1. When the EGR flow deviation index k≥2, there is 20% blockage (Level 1 blockage), no action is taken; when 1.67≤k<2, there is 40% blockage (Level 2 blockage), and the self-cleaning function is activated; when 1.33≤k<1.67, there is 60% blockage (Level 3 blockage), torque is not limited temporarily, and the service station cleans the valve; when 1≤k<1.33, there is 80% blockage (Level 4 blockage), primary torque is limited, and the service station cleans the valve; when 0.67≤k<1, there is 100% blockage (Level 5 blockage), severe torque is limited, and a new valve is replaced.

[0073] Step S11: Compare k with the partition range;

[0074] Step S12: Determine if EGR blockage is ≥60. If yes, proceed to step S2; otherwise, proceed to step S13.

[0075] Step S13: Remind the driver to perform maintenance via mobile APP (cleaning reminder information);

[0076] Step S14: End.

[0077] Table 1

[0078]

[0079] This application also provides an apparatus for determining the degree of EGR congestion. It should be noted that this apparatus can be used to execute the method for determining the degree of EGR congestion provided in this application. This apparatus is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0080] The following describes the apparatus for determining the degree of EGR blockage provided in the embodiments of this application.

[0081] Figure 6 This is a schematic diagram of an apparatus for determining the degree of EGR blockage according to an embodiment of this application. Figure 6 As shown, the device includes:

[0082] The acquisition unit 10 is used to acquire EGR parameters and engine parameters at multiple moments within a predetermined time period to obtain the current EGR parameters and current engine parameters. The EGR parameters include at least the upstream temperature of the EGR, the upstream pressure of the EGR, the pressure difference between the two ends of the EGR, and the opening of the EGR valve. The engine parameters include at least the engine speed and the engine oil quantity.

[0083] Specifically, since EGR flow rate is strongly correlated with upstream EGR temperature, upstream EGR pressure, pressure difference across the EGR, and valve opening at the EGR location, to ensure the accuracy of EGR flow rate prediction, these signals are collected, and engine parameters (engine speed and fuel level) are added to form a data matrix. The onboard ECU collects the above basic information, namely EGR parameters and engine parameters, and uploads it wirelessly to the cloud computing center. The cloud computing center stores the data and analyzes and calculates it according to the LSTM neural network model described below. The predetermined time period can be a steady-state operating condition of the engine in the medium-to-high speed load range. The current EGR parameters and current engine parameters are collected at multiple moments within the predetermined time period; that is, the current EGR parameters and current engine parameters are data corresponding to multiple moments.

[0084] The first determining unit 20 is used to analyze the current EGR parameters and the current engine parameters using an LSTM neural network model to determine the current EGR flow rate. The LSTM neural network model is trained using multiple sets of data through machine learning. Each set of data includes: historical EGR parameters, historical engine parameters, and historical EGR flow rate.

[0085] Specifically, the LSTM neural network model achieves dynamic changes over a fixed time scale by using self-looping within the memory unit. It utilizes the memory unit to store long-term memory information and uses hidden states to transmit short-term memory information. Therefore, the LSTM neural network model is used to predict EGR flow. Historical EGR parameters, historical engine parameters, and historical EGR flow can be collected under normal vehicle operation, selecting a steady-state condition within the mid-to-high speed load range. This primarily involves selecting a region where EGR is active, collecting one year's worth of data from a single vehicle as a reference, with missing data filled using interpolation. We will first analyze a single vehicle as an example, and later extend this to a fleet of vehicles. Under the selected operating condition, the cloud collects the EGR flow in a region with a relatively short mileage for that vehicle, calculates the average value A, and stores it. At this point, the EGR valve is still a new valve, serving as a reference threshold.

[0086] The second determining unit 30 is used to calculate the variance of the current EGR flow, calculate the deviation index of the current EGR flow based on the variance, obtain a preset threshold, and determine the degree of EGR congestion based on the relationship between the deviation index and the preset threshold.

[0087] Specifically, after predicting the EGR flow, the variance of the EGR flow data is calculated, and a deviation index is further calculated based on the variance. The EGR flow deviation index is a statistical indicator that measures the degree of deviation from the actual value. A higher value indicates a smaller degree of deviation and less congestion, and vice versa. Therefore, the degree of EGR congestion is determined by setting a preset threshold and the relationship between the deviation index and the preset threshold.

[0088] This embodiment acquires current EGR parameters and current engine parameters, inputs them into an LSTM neural network model for analysis, predicts the current EGR flow rate, and calculates the variance and deviation index of the current EGR flow rate. A preset threshold is then obtained, and the degree of EGR blockage is determined by the relationship between the preset threshold and the deviation index. This allows for accurate prediction of EGR flow rate first, and then the determination of whether EGR blockage has occurred and the degree of blockage based on the predicted EGR flow rate. Compared to existing technologies where EGR monitoring strategies are limited by ECU hardware computing power, data storage capacity, and cost, often only utilizing historical data for a limited period and failing to accurately monitor EGR blockage and its degree, this application can accurately predict EGR flow rate through a neural network model and further calculate the deviation index to determine the degree of EGR blockage. Therefore, it solves the problem of inaccurate monitoring of EGR blockage in existing technologies, achieving the goal of accurately monitoring the degree of EGR blockage.

[0089] In the specific implementation process, the second determining unit includes a first calculation module and a second calculation module, wherein the first calculation module is used to determine the formula. Calculate the variance of the current EGR flow, where s represents the variance of the EGR flow, n represents the quantity of the EGR flow, and x... i This represents the i-th EGR flow mentioned above. This represents the average of n EGR flows mentioned above; the second calculation module is used to calculate the variance using the formula... The deviation index of the current EGR flow is calculated, where A represents the average value of multiple EGR flows collected when the vehicle's mileage is less than a preset mileage, and k represents the deviation index. This method first calculates the variance, and then calculates the deviation index based on the variance. This allows for accurate calculation of the deviation index, thus revealing the deviation of the EGR flow and determining the degree of EGR congestion.

[0090] Specifically, after collecting multiple EGR traffic flows, the following is adopted: Calculate the variance, and then use... Calculate the deviation index k, classify the levels according to the range of the deviation index k, and determine the degree of EGR blockage.

[0091] To determine the EGR congestion level based on the deviation index, the second determining unit further includes a first acquisition module and a first determining module. The first acquisition module acquires EGR traffic datasets for each of the multiple congestion levels, where each EGR traffic dataset consists of multiple simulated EGR traffic data acquired at each congestion level. The first determining module calculates the deviation index of each simulated EGR traffic data in the EGR traffic dataset corresponding to each congestion level, determines the largest deviation index in the EGR traffic dataset for the lowest congestion level as a first preset threshold, and determines the smallest deviation index in the EGR traffic datasets for other congestion levels as other preset thresholds. This method acquires EGR traffic datasets for multiple congestion levels and calculates preset thresholds for different levels of congestion based on the EGR traffic datasets, thus determining the EGR congestion level based on the relationship between the deviation index and the preset thresholds.

[0092] In the specific implementation, multiple EGR traffic data points are acquired for each level of congestion; that is, multiple EGR traffic data points are acquired for each level of congestion, resulting in a set of EGR traffic datasets corresponding to each level of congestion. Each EGR traffic dataset may contain thousands of EGR traffic data points, and the deviation index for each EGR traffic point is calculated. Specifically, an average value is calculated for each EGR traffic dataset, and then... Calculate the variance. x represents the mean. i Let represent the i-th EGR flow, n represent the number of EGR flows mentioned above, and s represent the variance. Then, according to the formula... The deviation index corresponding to each EGR flow is calculated, where A represents the average EGR flow in the area with a smaller mileage, i.e., the average of multiple EGR flows collected when the mileage of the vehicle containing the engine is less than a preset mileage. Based on the above steps, multiple deviation indices are calculated for the EGR flow dataset corresponding to each level of congestion. Among the deviation indices corresponding to the lowest level of congestion, the one with the largest deviation index is taken as the first preset threshold. This indicates that when the deviation index is greater than the first preset threshold, the EGR congestion level is the lowest.

[0093] The first acquisition module includes a control submodule, an acquisition submodule, and a first determination submodule. The control submodule controls the operation of the engine equipped with EGR fault components of varying degrees of congestion. The acquisition submodule acquires the EGR parameters and engine parameters during engine operation, obtaining simulated EGR parameters and simulated engine parameters, where each EGR fault component corresponds to a level of congestion. The first determination submodule analyzes the simulated EGR parameters and simulated engine parameters using the LSTM neural network model to obtain multiple simulated EGR flows at each level of congestion, thus obtaining an EGR flow dataset for multiple levels of congestion. This method acquires simulated EGR parameters and simulated engine parameters by installing EGR fault components with different degrees of congestion. This allows the simulated parameters to be used as the EGR flow dataset corresponding to each level of congestion for further calculation of the deviation index and determination of a preset threshold.

[0094] Specifically, faulty components with EGR blockages of 20%, 40%, 60%, 80%, and 100% are fabricated. These faulty components are then sequentially installed on the engine. Simulated EGR parameters and simulated engine parameters are obtained. The aforementioned LSTM neural network model is run, and the EGR flow of the five sets of faulty components is recorded (each set may contain thousands of values). Each set of EGR flow is used as an EGR flow dataset, and the mean and deviation index of each set are calculated. In practice, the degree of blockage of the faulty components can also be classified in other ways.

[0095] To calculate the deviation index corresponding to each EGR flow dataset, in some embodiments, the first determining module includes a first calculation submodule, a second calculation submodule, and a third calculation submodule. The first calculation submodule is used to calculate the average value of each level of the simulated EGR flow dataset to obtain the average flow value at each level. The second calculation submodule is used to calculate the variance of the EGR flow dataset under each level of congestion based on the EGR flow dataset at each level and the average flow value at each level corresponding to the EGR flow dataset at each level. The third calculation submodule is used to calculate the variance using the formula... Calculate the deviation index for each simulated EGR flow in the aforementioned EGR flow datasets at each level, where A represents the average of multiple EGR flows collected when the vehicle's mileage is less than a preset mileage, k represents the deviation index, s represents the variance of the EGR flow, and x... i Let represent the i-th EGR flow. This method calculates the deviation index by calculating the variance of the EGR flow, thus accurately determining the deviation index and subsequently identifying the preset thresholds corresponding to different levels of congestion.

[0096] In the specific implementation process, the average value of each level of EGR traffic dataset is calculated, and then the variance of each EGR traffic dataset is calculated. Based on the variance, [the following steps are taken]. The deviation index for each EGR flow can be calculated.

[0097] To accurately describe and determine the degree of EGR congestion, the second determining unit further includes a second determining module and a third determining module. The second determining module is used to determine the EGR congestion level as Level 1 congestion when the deviation index is greater than or equal to the first preset threshold. The third determining module is used to determine the EGR congestion level as other levels of congestion corresponding to other preset thresholds when the deviation index is less than the first preset threshold. This method determines Level 1 congestion using the first preset threshold and other levels of congestion using other preset thresholds. This allows for a more accurate determination of the EGR congestion level.

[0098] Specifically, the degree of congestion can be divided into multiple levels, each with its own preset threshold. The lowest level of congestion is Level 1 congestion, which is slight congestion. If the congestion degree is divided into 20%, 40%, 60%, 80%, and 100%, then Level 1 congestion is defined as 20%. When the deviation index is greater than or equal to the first preset threshold, the EGR is determined to be Level 1 congestion. Correspondingly, other levels of congestion can be determined using other preset thresholds.

[0099] The third determining module includes a second determining submodule, a third determining submodule, a fourth determining submodule, and a fifth determining submodule. The second determining submodule is used to determine the EGR congestion level as level two congestion when the deviation index is greater than or equal to a second preset threshold and less than a first preset threshold. The second preset threshold is a preset threshold corresponding to level two congestion. The third determining submodule is used to determine the EGR congestion level as level three congestion when the deviation index is greater than or equal to a third preset threshold and less than a second preset threshold. The third preset threshold is a preset threshold corresponding to level three congestion. The fourth determining submodule is used to determine the EGR congestion level as level four congestion when the deviation index is greater than or equal to a fourth preset threshold and less than a third preset threshold. The fourth preset threshold is a preset threshold corresponding to level four congestion. The fifth determining submodule is used to determine the EGR congestion level as level five congestion when the deviation index is greater than or equal to a fifth preset threshold and less than a fourth preset threshold. The fifth preset threshold is a preset threshold corresponding to level five congestion. This method determines the congestion level based on each preset threshold, allowing for a more accurate determination of the EGR congestion level by classifying the congestion level according to the preset thresholds.

[0100] In practical implementation, the congestion level is divided into 20%, 40%, 60%, 80%, and 100%. Each level of congestion has a corresponding preset threshold. As mentioned above, if the deviation index is greater than or equal to the first preset threshold, it is a 20% level 1 congestion; if the deviation index is greater than or equal to the second preset threshold but less than the first preset threshold, the EGR congestion level is determined to be a level 2 congestion (40%); if the deviation index is greater than or equal to the third preset threshold but less than the second preset threshold, the EGR congestion level is determined to be a level 3 congestion (60%); if the deviation index is greater than or equal to the fourth preset threshold but less than the third preset threshold, the EGR congestion level is determined to be a level 4 congestion (80%); and if the deviation index is greater than or equal to the fifth preset threshold but less than the fourth preset threshold, the EGR congestion level is determined to be a level 5 congestion (100%). In practical applications, the first preset threshold is usually 2, the second preset threshold is usually 1.67, the third preset threshold is usually 1.33, the fourth preset threshold is usually 1, and the fifth preset threshold is usually 0.67.

[0101] To address different levels of blockage, the device described in this application further includes a processing unit, a control unit, a first sending unit, a second sending unit, and a third sending unit. The processing unit is configured to perform no processing if the EGR is determined to be at level one blockage. The control unit is configured to activate the automatic cleaning function of the EGR if the EGR is determined to be at level two blockage. The first sending unit is configured to generate a cleaning reminder message and send it to the user at a first preset frequency if the EGR is determined to be at level three blockage. The cleaning reminder message prompts the user to go to a cleaning station to clean the EGR. The first unit performs cleaning; the second sending unit, when determining that the EGR blockage level is level four, controls the engine output torque to decrease to a first torque, generates the cleaning prompt information, and sends the cleaning prompt information to the user terminal at a second preset frequency, wherein the second preset frequency is greater than the first preset frequency; the third sending unit, when determining that the EGR blockage level is level five, controls the engine output torque to decrease to a second torque, generates the cleaning prompt information, and sends the cleaning prompt information to the user terminal at a third preset frequency, wherein the second torque is less than the first torque, and the third preset frequency is greater than the second preset frequency. This method employs different processing methods for different blockage levels, allowing for different measures to be taken to treat the EGR according to the actual situation, ensuring stable operation of the engine and vehicle.

[0102] In the specific implementation, the frequency of prompts and torque limitations are set according to the severity of the blockage. At the initial stage of blockage, the blockage is less severe, resulting in a lower frequency of prompts and less torque limitation. The more severe the blockage, the higher the frequency of prompts and the greater the torque limitation. Specifically, it can be set as follows: For a level 1 blockage (20% blockage), no action is taken; for a 40% blockage, the EGR system activates automatic cleaning; for a 60% blockage, a weekly cleaning reminder is sent to the driver via a first preset frequency (once a week); for an 80% blockage, indicating severe blockage, initial torque limiting is implemented (controlling the engine torque below a certain threshold), and a daily cleaning reminder is sent to the driver via a second preset frequency (once a day); for a 100% blockage, severe torque limiting is implemented (controlling the engine torque further below a certain threshold), and a cleaning reminder is sent to the driver hourly via a third preset frequency (once an hour). This application does not limit the specific values ​​of the preset frequencies and torques mentioned above. In the case of 100% blockage, the torque is limited to the greatest extent, that is, the second torque is less than the first torque mentioned above, and the first torque may include the second torque.

[0103] The aforementioned LSTM neural network model includes an input gate, a forget gate, and an output gate. The first determining unit includes a training module and a fourth determining module. The training module uses the historical EGR parameters and historical engine parameters as inputs to the initial LSTM neural network model, and the historical EGR flow rate as the output of the initial LSTM neural network model, through formula F. t =σ(w f *[H t-1 ,X t ]+b f Calculate the forgetting factor of the forgetting gate mentioned above using formula I. t =σ(w i *[H t-1 ,X t ]+b i Calculate the input factor of the above input gate using the formula. Calculate the memory factor using formula O t =σ(w o *[H t-1 ,X t ]+b o Calculate the output factor of the above output gate, train the above initial LSTM neural network model to obtain the above LSTM neural network model, where Xt w represents the parameter matrix composed of the aforementioned historical EGR parameters and the aforementioned historical engine parameters. f w represents the weight of the forget gate mentioned above. i w represents the weight of the input gate mentioned above. o b represents the weight of the output gate mentioned above. f b represents the deviation of the forgetting gate mentioned above. i b represents the deviation of the above input gate. o H represents the deviation of the above output gate. t-1 Let C represent the hidden state at time t-1. t-1 This represents the state of the memory cell at time t-1. The state of the candidate memory unit is represented by σ, which represents the sigmoid function. The fourth determining module is used to input the current EGR parameters and the current engine parameters into the LSTM neural network model, and determine the current EGR flow rate by the output of the LSTM neural network model. This method predicts the EGR flow rate through the LSTM neural network model, thus determining the EGR flow rate more accurately.

[0104] Specifically, the LSTM neural network introduces three gates (input gate, output gate, and forget gate) and two memory states (hidden state H). t Memory unit state C t The main function of the forget gate is to filter and discard transmitted information. In the forget gate, this is achieved through formula F. t =σ(w f *[H t-1 ,X t ]+b f The forgetting factor F was calculated. t And use this factor to process the memory unit C passed from the previous time. t-1 Forgetting is performed. The input gate is used to calculate the information of the current time, through formula I. t =σ(w i *[H t-1 ,X t ]+b i )and The input factor I was obtained respectively. t and candidate memory factors Both are through formula The memory factor C of the current step size is calculated. t The output gate is based on the output factor O. t and memory factor C t Output the hidden factor H of the current step size t Through formula O t =σ(w o *[H t-1 ,Xt ]+b o ) and H t =O t *tanh(C t ) to obtain, where X t This indicates EGR parameters and engine parameters collected from the cloud, w f w i w o w c These are the forget gate weights, input gate weights, output gate weights, and candidate memory cell state weights, respectively. f b i b o b c These are the forget gate bias, input gate bias, output gate bias, and candidate memory cell state bias, respectively. t-1 Let C represent the hidden state at time t-1. t Indicates the state of the memory cell. Represents the candidate memory cell state, σ represents the sigmoid function, and is expressed by the formula... The tanh function is obtained through the formula. To improve prediction accuracy, the number of network layers can be increased. Increasing the number of network layers enhances the model's learning ability and reduces bias. The EGR traffic predicted by the LSTM model is compared with the actual collected traffic, and the number of network layers and weight coefficients w are continuously optimized. f w i w o w c This ensures that the deviation between the model value and the actual value is less than a certain acceptable range.

[0105] The aforementioned EGR congestion determination device includes a processor and a memory. The acquisition unit, the first determination unit, and the second determination unit are all stored as program units in the memory. The processor executes these program units stored in the memory to achieve the corresponding functions. All of the above modules are located in the same processor; alternatively, the modules may be located in different processors in any combination.

[0106] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and adjusting kernel parameters can accurately monitor the level of EGR congestion.

[0107] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0108] This invention provides a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the method for determining the degree of EGR congestion.

[0109] Specifically, the methods for determining the degree of EGR blockage include:

[0110] Step S201: Obtain EGR parameters and engine parameters at multiple times within a predetermined time period to obtain current EGR parameters and current engine parameters. The EGR parameters include at least EGR upstream temperature, EGR upstream pressure, EGR pressure difference across the EGR terminals, and EGR valve opening. The engine parameters include at least engine speed and engine oil quantity.

[0111] Specifically, since EGR flow rate is strongly correlated with upstream EGR temperature, upstream EGR pressure, pressure difference across the EGR, and valve opening at the EGR location, to ensure the accuracy of EGR flow rate prediction, these signals are collected, and engine parameters (engine speed and fuel level) are added to form a data matrix. The onboard ECU collects the above basic information, namely EGR parameters and engine parameters, and uploads it wirelessly to the cloud computing center. The cloud computing center stores the data and analyzes and calculates it according to the LSTM neural network model described below. The predetermined time period can be a steady-state operating condition of the engine in the medium-to-high speed load range. The current EGR parameters and current engine parameters are collected at multiple moments within the predetermined time period; that is, the current EGR parameters and current engine parameters are data corresponding to multiple moments.

[0112] Step S202: Analyze the current EGR parameters and the current engine parameters using an LSTM neural network model to determine the current EGR flow rate. The LSTM neural network model is trained using multiple sets of data through machine learning. Each set of data includes: historical EGR parameters, historical engine parameters, and historical EGR flow rate.

[0113] Specifically, the LSTM neural network model achieves dynamic changes over a fixed time scale by using self-looping within the memory unit. It utilizes the memory unit to store long-term memory information and uses hidden states to transmit short-term memory information. Therefore, the LSTM neural network model is used to predict EGR flow. Historical EGR parameters, historical engine parameters, and historical EGR flow can be collected under normal vehicle operation, selecting a steady-state condition within the mid-to-high speed load range. This primarily involves selecting a region where EGR is active, collecting one year's worth of data from a single vehicle as a reference, with missing data filled using interpolation. We will first analyze a single vehicle as an example, and later extend this to a fleet of vehicles. Under the selected operating condition, the cloud collects the EGR flow in a region with a relatively short mileage for that vehicle, calculates the average value A, and stores it. At this point, the EGR valve is still a new valve, serving as a reference threshold.

[0114] Step S203: Calculate the variance of the current EGR flow, calculate the deviation index of the current EGR flow based on the variance, obtain a preset threshold, and determine the degree of EGR congestion based on the relationship between the deviation index and the preset threshold.

[0115] Specifically, after predicting the EGR flow, the variance of the EGR flow data is calculated, and a deviation index is further calculated based on the variance. The EGR flow deviation index is a statistical indicator that measures the degree of deviation from the actual value. A higher value indicates a smaller degree of deviation and less congestion, and vice versa. Therefore, the degree of EGR congestion is determined by setting a preset threshold and the relationship between the deviation index and the preset threshold.

[0116] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0117] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0118] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0119] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0120] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0121] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0122] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0123] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0124] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0125] As can be seen from the above description, the embodiments of this application achieve the following technical effects:

[0126] 1) In the method for determining the degree of EGR blockage in this application, the current EGR parameters and engine parameters are obtained, and then input into an LSTM neural network model for analysis. The EGR flow rate is predicted, and the variance and deviation index of the EGR flow rate are calculated. A preset threshold is then obtained, and the degree of EGR blockage is determined by the relationship between the preset threshold and the deviation index. This allows for accurate prediction of the EGR flow rate first, and then the determination of whether EGR blockage has occurred and the degree of blockage is based on the predicted EGR flow rate. Compared with existing technologies where EGR monitoring strategies are limited by ECU hardware computing power, data storage capacity, and cost, often only able to use historical data for calculations within a limited timeframe and thus unable to accurately monitor whether EGR is blocked and the degree of blockage, this application can accurately predict EGR flow rate through a neural network model and further calculate the deviation index to determine the degree of EGR blockage. Therefore, it can solve the problem of inaccurate monitoring of EGR blockage in existing technologies, achieving the goal of accurately monitoring the degree of EGR blockage.

[0127] 2) In the EGR clogging degree determination device of this application, the current EGR parameters and engine parameters are acquired, and then input into an LSTM neural network model for analysis. The EGR flow rate is predicted, and the variance and deviation index of the EGR flow rate are calculated. A preset threshold is then obtained. The degree of EGR clogging is determined by the relationship between the preset threshold and the deviation index. This allows for accurate prediction of the EGR flow rate first, and then the determination of whether EGR clogging has occurred and the degree of clogging based on the predicted EGR flow rate. Compared with existing technologies where EGR monitoring strategies are limited by ECU hardware computing power, data storage capacity, and cost, often only able to use historical data for calculations within a limited timeframe and thus unable to accurately monitor whether EGR is clogged and the degree of clogging, this application can accurately predict EGR flow rate through a neural network model and further calculate the deviation index to determine the degree of EGR clogging. Therefore, it can solve the problem of inaccurate monitoring of EGR clogging in existing technologies, achieving the goal of accurately monitoring the degree of EGR clogging.

[0128] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for determining the degree of EGR blockage, characterized in that, Including: Obtain EGR parameters and engine parameters at multiple moments within a predetermined time period to obtain current EGR parameters and current engine parameters. Among them, the EGR parameters at least include EGR upstream temperature, EGR upstream pressure, differential pressure across EGR, and EGR valve opening, and the engine parameters at least include engine speed and engine fuel quantity; Use the LSTM neural network model to analyze the current EGR parameters and the current engine parameters to determine the current EGR flow rate. Among them, the LSTM neural network model is obtained through machine learning training using multiple sets of data, and each set of data in the multiple sets of data includes: historical EGR parameters, historical engine parameters, and historical EGR flow rate; Calculate the variance of the current EGR flow rate, calculate the deviation index of the current EGR flow rate according to the variance, obtain a preset threshold, and determine the degree of blockage of the EGR according to the size relationship between the deviation index and the preset threshold; Calculating the variance of the current EGR flow rate and calculating the deviation index of the current EGR flow rate according to the variance includes: Calculate the variance of the current EGR flow rate through the formula where s represents the variance of the EGR flow rate, n represents the number of the EGR flow rates, and x i represents the i-th EGR flow rate, and represents the average value of the n EGR flow rates; According to the variance, through the formula calculate the deviation index of the current EGR flow rate, where A represents the average value of multiple EGR flow rates collected when the driving mileage of the vehicle where the engine is located is less than the preset mileage, and k represents the deviation index; Obtaining a preset threshold includes: Respectively obtain EGR flow rate data sets of the EGR under multiple degrees of blockage. Among them, the EGR flow rate data set is a data set composed of multiple simulated EGR flow rates obtained under each degree of blockage; Calculate the deviation index of each simulated EGR flow rate in the EGR flow rate data set corresponding to each degree of blockage, and determine the largest deviation index in the EGR flow rate data set with the smallest degree of blockage as the first preset threshold, and determine the smallest deviation index in the EGR flow rate data sets with other degrees of blockage as other preset thresholds.

2. The determination method according to claim 1, wherein Respectively obtaining EGR flow rate data sets of the EGR under multiple degrees of blockage includes: Control the engine equipped with EGR fault components with different degrees of blockage to run; Obtain the EGR parameters and engine parameters during the operation of the engine to obtain simulated EGR parameters and simulated engine parameters. Among them, each EGR fault component corresponds to one degree of blockage; Use the LSTM neural network model to analyze the simulated EGR parameters and the simulated engine parameters to obtain multiple simulated EGR flow rates under each degree of blockage, so as to obtain the EGR flow rate data set under multiple degrees of blockage.

3. The determination method according to claim 1, wherein Calculating the deviation index of each simulated EGR flow rate in the EGR flow rate data set corresponding to each degree of blockage includes: Respectively calculate the average value of each EGR flow rate data set to obtain the average flow rate of each level; Calculate the variance of the EGR flow rate data set under each degree of blockage according to each EGR flow rate data set and the average flow rate of each level corresponding to each EGR flow rate data set; According to the variance, through the formula calculate the deviation index of each of the simulated EGR flow rates in the EGR flow rate datasets at each level, where A represents the average value of multiple EGR flow rates collected when the driving mileage of the vehicle where the engine is located is less than the preset mileage, k represents the deviation index, s represents the variance of the EGR flow rate, and x i represents the i-th EGR flow rate.

4. The determination method according to claim 1, wherein Determining the degree of blockage of the EGR according to the size relationship between the deviation index and the preset threshold includes: In the case where the deviation index is greater than or equal to the first preset threshold, determine that the degree of blockage of the EGR is a first-level blockage; When the deviation index is less than the first preset threshold, determine that the degree of blockage of the EGR is other levels of blockage corresponding to the other preset thresholds.

5. The determination method according to claim 4, wherein When the deviation index is less than the first preset threshold, determining that the degree of blockage of the EGR is other levels of blockage corresponding to the other preset thresholds includes: When the deviation index is greater than or equal to the second preset threshold and less than the first preset threshold, determine that the degree of blockage of the EGR is secondary blockage, and the second preset threshold is the preset threshold corresponding to the secondary blockage; When the deviation index is greater than or equal to the third preset threshold and less than the second preset threshold, determine that the degree of blockage of the EGR is tertiary blockage, and the third preset threshold is the preset threshold corresponding to the tertiary blockage; When the deviation index is greater than or equal to the fourth preset threshold and less than the third preset threshold, determine that the degree of blockage of the EGR is quaternary blockage, and the fourth preset threshold is the preset threshold corresponding to the quaternary blockage; When the deviation index is greater than or equal to the fifth preset threshold and less than the fourth preset threshold, determine that the degree of blockage of the EGR is quinary blockage, and the fifth preset threshold is the preset threshold corresponding to the quinary blockage.

6. The determination method according to claim 5, wherein After determining the degree of blockage of the EGR, the determination method further includes: When it is determined that the degree of blockage of the EGR is the primary blockage, no treatment is performed; When it is determined that the degree of blockage of the EGR is the secondary blockage, control the EGR to enable the automatic cleaning function; When it is determined that the degree of blockage of the EGR is the tertiary blockage, generate a cleaning prompt message and send the cleaning prompt message to the user terminal at a first preset frequency, where the cleaning prompt message is used to prompt the user to clean the EGR at a cleaning station; When it is determined that the degree of blockage of the EGR is the quaternary blockage, control the output torque of the engine to be reduced to a first torque, generate the cleaning prompt message, and send the cleaning prompt message to the user terminal at a second preset frequency, where the second preset frequency is greater than the first preset frequency; When it is determined that the degree of blockage of the EGR is the quinary blockage, control the output torque of the engine to be reduced to a second torque, generate the cleaning prompt message, and send the cleaning prompt message to the user terminal at a third preset frequency, where the second torque is less than the first torque and the third preset frequency is greater than the second preset frequency.

7. The determination method according to any one of claims 1 to 6, characterized in that, The LSTM neural network model includes an input gate, a forget gate, and an output gate. Using the LSTM neural network model to analyze the current EGR parameters and the current engine parameters to determine the current EGR flow rate includes: Taking the historical EGR parameters and historical engine parameters as the inputs of the initial LSTM neural network model, and the historical EGR flow rate as the output of the initial LSTM neural network model, through the formula calculate the forgetting factor of the forgetting gate, through the formula calculate the input factor of the input gate, through the formula calculate the memory factor, through the formula calculate the output factor of the output gate, and train the initial LSTM neural network model to obtain the LSTM neural network model, where represents the parameter matrix composed of the historical EGR parameters and the historical engine parameters, represents the weight of the forgetting gate, represents the weight of the input gate, represents the weight of the output gate, represents the bias of the forgetting gate, represents the bias of the input gate, represents the bias of the output gate, represents the hidden state at time t - 1, represents the memory cell state at time t - 1, represents the candidate memory cell state, and σ represents the sigmoid function; Input the current EGR parameters and the current engine parameters into the LSTM neural network model, and determine the output of the LSTM neural network model as the current EGR flow rate.

8. An apparatus for determining the degree of EGR clogging, characterized in that, The determination device is applied to execute the method according to any one of claims 1 to 7, and the device includes: An acquisition unit is configured to acquire EGR parameters and engine parameters at multiple moments within a predetermined time period, and obtain current EGR parameters and current engine parameters. Among them, the EGR parameters at least include the EGR upstream temperature, the EGR upstream pressure, the pressure difference across the EGR, and the EGR valve opening degree, and the engine parameters at least include the engine speed and the engine fuel quantity; A first determination unit is configured to analyze the current EGR parameters and the current engine parameters by using an LSTM neural network model to determine the current EGR flow rate. Among them, the LSTM neural network model is obtained by machine learning training using multiple sets of data, and each set of data in the multiple sets of data includes: historical EGR parameters, historical engine parameters, and historical EGR flow rate; A second determination unit is configured to calculate the variance of the current EGR flow rate, calculate the deviation index of the current EGR flow rate according to the variance, and obtain a preset threshold. According to the magnitude relationship between the deviation index and the preset threshold, determine the degree of EGR blockage.

Images