Cylinder leakage detection method, device, storage medium and electronic equipment

Abstract

The present invention relates to the field of internal combustion engines, and discloses a cylinder leakage detection method, device, storage medium and electronic device. The SVR model for calculating the cylinder leakage rate can be trained based on the historical operation data of the internal combustion engine under common working conditions, and the historical operation data include: exhaust temperature before turbine, pressure after intercooler and throttle opening; when the internal combustion engine is running, the operation data of the internal combustion engine is input into the SVR model to obtain the cylinder leakage rate output by the SVR model, and the operation data at one moment corresponds to one cylinder leakage rate, and the operation data of the internal combustion engine include: exhaust temperature before turbine, pressure after intercooler and throttle opening; according to multiple cylinder leakage rates, the cylinder leakage degree of the internal combustion engine is determined online. Based on the trained SVR model, the present invention collects the operation data of the internal combustion engine when it is running, calculates the cylinder leakage rate online, and can determine the cylinder leakage degree online based on the cylinder leakage rate, without using additional equipment and offline internal combustion engines, which is relatively convenient and fast.

Description

Technical Field

[0001] The invention relates to the field of internal combustion engines, and in particular to a cylinder leakage detection method, device, storage medium and electronic equipment. Background Art

[0002] During the operation of the internal combustion engine, cylinder leakage may occur due to damage to the piston ring, wear and carbon deposits on the cylinder wall and piston, etc. Cylinder leakage will lead to reduced efficiency, increased fuel consumption, and emission problems of the internal combustion engine. If the continuous leakage problem is not solved in time, the overall life of the internal combustion engine will be significantly shortened. At present, compression testers or leakage detectors are usually used to measure the compression loss of the cylinder of the internal combustion engine. This type of equipment is complex to operate, large in size, and requires the internal combustion engine to be offline for detection, which makes it difficult to detect cylinder leakage. Summary of the invention

[0003] In view of the above problems, the present invention provides a cylinder leakage detection method, device, storage medium and electronic device that overcome the above problems or at least partially solve the above problems.

[0004] In a first aspect, a cylinder leakage detection method comprises:

[0005] Based on the historical operating data of the internal combustion engine under common working conditions, an SVR model for calculating the cylinder leakage rate is trained, wherein the common working conditions include common torque and common speed, and the historical operating data include: exhaust temperature before turbine, pressure after intercooler, and throttle opening;

[0006] When the internal combustion engine is running under the common working condition, the operating data of the internal combustion engine is input into the SVR model to obtain the cylinder leakage rate output by the SVR model, wherein the operating data at one moment corresponds to one cylinder leakage rate, and the operating data of the internal combustion engine includes: exhaust temperature before turbine, pressure after intercooler and throttle opening;

[0007] The cylinder leakage degree of the internal combustion engine is determined online based on the plurality of cylinder leakage rates.

[0008] Optionally, in some optional embodiments, the training of the SVR model for calculating the cylinder gas leakage rate based on the historical operating data of the internal combustion engine under common working conditions includes:

[0009] Filtering out historical operating data of the internal combustion engine under common working conditions from the historical data, wherein the historical operating data includes a plurality of turbine exhaust temperatures, a plurality of intercooler pressures, and a plurality of throttle openings;

[0010] Using the Z-Score method, the turbine exhaust temperature, the intercooler pressure, and the throttle opening in the historical operation data are respectively subjected to standardization preprocessing to obtain corresponding standardized data sets, wherein the standardized data sets include a plurality of standardized data corresponding to the turbine exhaust temperature, a plurality of standardized data corresponding to the intercooler pressure, and a plurality of standardized data corresponding to the throttle opening;

[0011] The SVR model for calculating cylinder leak rate is trained based on the standardized data set.

[0012] Optionally, in some optional embodiments, the training of the SVR model for calculating the cylinder leakage rate based on the standardized data set includes:

[0013] Dividing the standardized data set into a training set, a test set and a validation set, wherein the training set, the test set and the validation set all involve standardized data corresponding to the exhaust temperature before the turbine, the standardized data corresponding to the pressure after the intercooler and the standardized data corresponding to the throttle opening;

[0014] Training the SVR model using the training set;

[0015] Using the test set to evaluate the performance of the SVR model;

[0016] Using the validation set, the parameters of the SVR model were optimized by a cross-validation method.

[0017] Optionally, in certain optional embodiments, when the internal combustion engine is operating under the common operating condition, inputting the operating data of the internal combustion engine into the SVR model to obtain the cylinder leakage rate output by the SVR model includes:

[0018] When the internal combustion engine is operating under the common operating condition, obtaining operating data of the internal combustion engine;

[0019] Performing standardization preprocessing on the operating data of the internal combustion engine based on a mean value and a standard deviation, wherein the mean value and the standard deviation are obtained in the process of performing standardization preprocessing on the pre-turbine exhaust temperature, the post-intercooler pressure, and the throttle opening in the historical operating data using a Z-Score method;

[0020] The operating data of the internal combustion engine after the standardization preprocessing is input into the SVR model to obtain the cylinder leakage rate output by the SVR model.

[0021] Optionally, in some optional embodiments, the online determination of the cylinder leakage degree of the internal combustion engine according to the plurality of cylinder leakage rates includes:

[0022] Discretization processing is performed on the cylinder leakage rates at multiple consecutive moments to obtain multiple discrete values, wherein one moment corresponds to one cylinder leakage rate, and the average value of the cylinder leakage rates at K consecutive moments is taken as a discrete value, where K is an integer greater than 1;

[0023] The cylinder blowby degree of the internal combustion engine is determined online based on the plurality of discrete values.

[0024] Optionally, in some optional embodiments, the online determination of the cylinder leakage degree of the internal combustion engine according to the plurality of discrete values ​​comprises:

[0025] If Z consecutive discrete values ​​are all within the same preset range, the cylinder leakage degree of the internal combustion engine is determined online to be a set degree corresponding to the preset range, wherein Z is an integer greater than 1, and different preset ranges correspond to different set degrees.

[0026] Optionally, in certain optional embodiments, after the online determination that the cylinder leakage degree of the internal combustion engine is a set degree corresponding to the preset range, the method further includes:

[0027] The page elements of corresponding colors are displayed through a visual interface to display the cylinder leakage degree of the internal combustion engine, wherein different cylinder leakage degrees correspond to page elements of different colors.

[0028] In a second aspect, a cylinder leakage detection device includes: a model training unit, a leakage rate obtaining unit, and a leakage degree determining unit;

[0029] The model training unit is used to train an SVR model for calculating a cylinder gas leakage rate based on historical operating data of the internal combustion engine under common operating conditions, wherein the common operating conditions include common torque and common speed, and the historical operating data include: exhaust temperature before turbine, pressure after intercooler, and throttle opening;

[0030] The leakage rate obtaining unit is used to input the operating data of the internal combustion engine into the SVR model when the internal combustion engine is operating under the common working condition, and obtain the cylinder leakage rate output by the SVR model, wherein the operating data at one moment corresponds to one cylinder leakage rate, and the operating data of the internal combustion engine includes: exhaust temperature before turbine, pressure after intercooler, and throttle opening;

[0031] The leakage degree determination unit is used to determine the cylinder leakage degree of the internal combustion engine online according to the plurality of cylinder leakage rates.

[0032] In a third aspect, a computer-readable storage medium stores a program, wherein the program, when executed by a processor, implements any of the above-mentioned cylinder leakage detection methods.

[0033] In a fourth aspect, an electronic device comprises at least one processor, and at least one memory and a bus connected to the processor; wherein the processor and the memory communicate with each other via the bus; and the processor is used to call program instructions in the memory to execute any of the above-mentioned cylinder leakage detection methods.

[0034] By means of the above technical scheme, a cylinder leakage detection method, device, storage medium and electronic device provided by the present invention can train an SVR model for calculating the cylinder leakage rate based on the historical operating data of the internal combustion engine under common operating conditions, wherein the common operating conditions include common torque and common speed, and the historical operating data include: exhaust temperature before turbine, pressure after intercooler and throttle opening; when the internal combustion engine is operating under the common operating conditions, the operating data of the internal combustion engine is input into the SVR model to obtain the cylinder leakage rate output by the SVR model, wherein the operating data at one moment corresponds to one cylinder leakage rate, and the operating data of the internal combustion engine include: exhaust temperature before turbine, pressure after intercooler and throttle opening; based on multiple cylinder leakage rates, the cylinder leakage degree of the internal combustion engine is determined online. It can be seen from this that the present invention can collect the operating data of the internal combustion engine during operation based on the trained SVR model to calculate the cylinder leakage rate online, and based on the cylinder leakage rate, the cylinder leakage degree can be determined online without using additional equipment. It is relatively convenient and quick, and there is no need to take the internal combustion engine offline.

[0035] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0037] Figure 1 A flow chart showing a first cylinder gas leakage detection method provided by the present invention is shown;

[0038] Figure 2 A flow chart showing a second cylinder gas leakage detection method provided by the present invention;

[0039] Figure 3 A flow chart showing a third cylinder gas leakage detection method provided by the present invention is shown;

[0040] Figure 4 A flow chart showing a fourth cylinder gas leakage detection method provided by the present invention is shown;

[0041] Figure 5 A flow chart showing a fifth cylinder gas leakage detection method provided by the present invention is shown;

[0042] Figure 6 A schematic structural diagram of a cylinder leakage detection device provided by the present invention is shown;

[0043] Figure 7 A structural schematic diagram of an electronic device provided by the present invention is shown. DETAILED DESCRIPTION

[0044] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0045] like Figure 1 As shown, the present invention provides a cylinder leakage detection method, comprising: S100, S200 and S300;

[0046] S100, training an SVR model for calculating a cylinder gas leakage rate based on historical operating data of an internal combustion engine under common operating conditions, wherein the common operating conditions include common torque and common speed, and the historical operating data include: exhaust temperature before turbine, pressure after intercooler, and throttle opening;

[0047] Optionally, when various types of internal combustion engines are in operation, the operating conditions and corresponding operating data during operation can be stored in a database, so as to facilitate query and screening from the database when necessary, and the present invention does not limit this. It should be noted that the internal combustion engines referred to in the present invention may include multiple types, such as internal combustion engines installed in different vehicle types, including different vehicle types such as trains, trucks, tractors and family cars.

[0048] Optionally, in addition to storing the historical operation data of the internal combustion engine, the present invention can also obtain and store the historical operation data of different internal combustion engines under common working conditions through a large number of experiments. That is, the historical operation data mentioned in the present invention can also be understood as experimental data, and the present invention does not limit this.

[0049] Optionally, the historical operating data of the common working conditions of the internal combustion engines corresponding to different vehicle types are different. Therefore, the present invention can query the historical operating data of the common working conditions corresponding to the internal combustion engines of different vehicle types from the database, and then train to obtain the corresponding SVR model, and the present invention is not limited to this.

[0050] Optionally, the common operating conditions of internal combustion engines of different vehicle types are different, and the common operating conditions of the same internal combustion engine when installed on vehicles of different vehicle types are also different. Therefore, the present invention can predetermine the common operating conditions of different internal combustion engines of different vehicle types through a large number of experiments or statistics of a large amount of data, and the present invention does not limit this.

[0051] Optionally, the common operating conditions are characterized by the common torque and common speed of the internal combustion engine, that is, the present invention can obtain corresponding historical operating data based on the common torque and common speed query, and the present invention is not limited to this.

[0052] Optionally, parameters such as exhaust temperature before turbine, pressure after intercooler, and throttle opening are well-known concepts in the art, and the present invention does not describe them in detail. For details, please refer to the relevant explanations in the art. It should be noted that the historical operation data referred to in the present invention may include data at multiple historical moments, one historical moment corresponds to one piece of data, and one piece of data includes three items, namely: exhaust temperature before turbine, pressure after intercooler, and throttle opening.

[0053] Optionally, the SVR model is a supervised learning algorithm, mainly used for regression analysis problems. It maps the original feature space to a high-dimensional space by introducing a kernel function, and searches for an optimal hyperplane in the high-dimensional space for regression analysis. The SVR model is an extended method based on the support vector machine (SVM) and is used to solve nonlinear problems.

[0054] That is, Figure 2 As shown, in some optional embodiments, the S100 includes: S110, S120 and S130;

[0055] S110, filtering out historical operating data of the internal combustion engine under common working conditions from the historical data, wherein the historical operating data includes a plurality of exhaust temperatures before turbine, a plurality of pressures after intercooler, and a plurality of throttle openings;

[0056] S120, using a Z-Score method to perform standardization preprocessing on the turbine exhaust temperature, the intercooler pressure, and the throttle opening in the historical operation data, respectively, to obtain a corresponding standardized data set, wherein the standardized data set includes a plurality of standardized data corresponding to the turbine exhaust temperature, a plurality of standardized data corresponding to the intercooler pressure, and a plurality of standardized data corresponding to the throttle opening;

[0057] Optionally, the Z-Score method is a mathematical tool for measuring the standardized distance of a value relative to the average value. It is obtained by calculating the difference between a value and the average and then dividing it by the standard deviation. Taking the turbine pre-exhaust temperature as an example, the historical operation data mentioned in the present invention may include 1000 turbine pre-exhaust temperatures. The present invention can calculate the average value of these 1000 turbine pre-exhaust temperatures, and then for any turbine pre-exhaust temperature, calculate the difference between the turbine pre-exhaust temperature and the average value of these 1000 turbine pre-exhaust temperatures as the standardized data corresponding to the turbine pre-exhaust temperature, thereby obtaining 1000 standardized data corresponding to the 1000 turbine pre-exhaust temperatures, and the present invention does not limit this.

[0058] Optionally, in the process of using the Z-Score method to calculate the standardized data set, the present invention can also obtain the corresponding mean and standard deviation for direct application in the subsequent step S200. It should be noted that the mean and standard deviation are well-known concepts involved in the Z-Score method, and the present invention will not make further descriptions on this. For details, please refer to the relevant explanations in the art.

[0059] S130 . Based on the standardized data set, train the SVR model for calculating the cylinder leakage rate.

[0060] For example, in some optional embodiments, the S130 includes: step 1.1, step 1.2, step 1.3 and step 1.4;

[0061] Step 1.1, dividing the standardized data set into a training set, a test set and a validation set, wherein the training set, the test set and the validation set all involve standardized data corresponding to the exhaust temperature before the turbine, the standardized data corresponding to the pressure after the intercooler and the standardized data corresponding to the throttle opening;

[0062] Optionally, the present invention can divide the training set, the test set and the validation set in a ratio of 7:2:1. Take 1000 pieces of standardized data corresponding to the exhaust temperature before the turbine, 1000 pieces of standardized data corresponding to the pressure after the intercooler and 1000 pieces of standardized data corresponding to the throttle opening as an example. 700 pieces of standardized data corresponding to the exhaust temperature before the turbine, 700 pieces of standardized data corresponding to the pressure after the intercooler and 700 pieces of standardized data corresponding to the throttle opening are divided into a training set. 200 pieces of standardized data corresponding to the exhaust temperature before the turbine, 200 pieces of standardized data corresponding to the pressure after the intercooler and 200 pieces of standardized data corresponding to the throttle opening are divided into a training set. 100 pieces of standardized data corresponding to the exhaust temperature before the turbine, 100 pieces of standardized data corresponding to the pressure after the intercooler and 100 pieces of standardized data corresponding to the throttle opening are divided into a training set.

[0063] It should be noted that: in order to ensure the effect of model training, the data at the same time are divided into the same data set. For example, the standardized data corresponding to the exhaust temperature before the turbine, the standardized data corresponding to the pressure after the intercooler, and the standardized data corresponding to the throttle opening at the same time are divided into the training set, and the data at the same time cannot be divided into different data sets (for example, the standardized data corresponding to the exhaust temperature before the turbine is divided into the training set, the standardized data corresponding to the pressure after the intercooler is divided into the test set, and the standardized data corresponding to the throttle opening is divided into the verification set). The present invention does not limit this.

[0064] Step 1.2, training the SVR model using the training set;

[0065] Optionally, the present invention can use the data of the training set to train the SVR model. In this step, the model will try to find a hyperplane to minimize the error between the predicted value and the actual value while keeping the complexity of the model under control, which is not limited by the present invention.

[0066] Step 1.3, using the test set to evaluate the performance of the SVR model;

[0067] Optionally, the present invention may use a test set to evaluate the performance of the model, and commonly used evaluation indicators include mean square error (MSE) and root mean square error (RMSE) and the like.

[0068] Step 1.4: Use the validation set to optimize the parameters of the SVR model through a cross-validation method.

[0069] Optionally, the present invention can optimize model parameters by cross-validation and other methods, such as adjusting the regularization C parameter, tolerance bandwidth epsilon and kernel function, so as to improve the prediction accuracy of the model.

[0070] For example, the present invention obtains the following parameter values ​​through experiments:

[0071] 1. It is recommended to use logarithmic scale search to adjust the regularization C parameter, such as {0.001, 0.01, 0.1, 1, 10,100, 1000}, etc.;

[0072] 2. The recommended tolerance bandwidth range is 0.01~0.05;

[0073] 3. It is recommended to use the radial basis function (RBF) kernel as the kernel function.

[0074] S200, when the internal combustion engine is operating under the common working condition, inputting the operating data of the internal combustion engine into the SVR model to obtain the cylinder leakage rate output by the SVR model, wherein the operating data at one moment corresponds to one cylinder leakage rate, and the operating data of the internal combustion engine includes: exhaust temperature before turbine, pressure after intercooler, and throttle opening;

[0075] Optionally, as mentioned above, after training, the SVR model can be inferred to calculate the corresponding cylinder leakage rate. Therefore, after the SVR model is trained, the present invention can directly apply the SVR model to calculate the cylinder leakage rate, and the present invention does not limit this.

[0076] For example, Figure 3 As shown, in some optional embodiments, the S200 includes: S210, S220 and S230;

[0077] S210, when the internal combustion engine is operating under the common operating condition, obtaining operating data of the internal combustion engine;

[0078] S220, performing standardization preprocessing on the operating data of the internal combustion engine based on a mean value and a standard deviation, wherein the mean value and the standard deviation are obtained in the process of performing standardization preprocessing on the pre-turbine exhaust temperature, the post-intercooler pressure, and the throttle opening in the historical operating data using a Z-Score method;

[0079] S230: Inputting the normalized pre-processed operating data of the internal combustion engine into the SVR model to obtain the cylinder leakage rate output by the SVR model.

[0080] Optionally, the present invention can calculate the cylinder leakage rate of the internal combustion engine online in real time based on real-time operating data (which can also be understood as networked data), so as to obtain the cylinder leakage rate at multiple consecutive moments, so as to facilitate the subsequent determination of the cylinder leakage degree based on the cylinder leakage rates at multiple consecutive moments. The present invention does not impose any restrictions on this.

[0081] S300: Determine the cylinder leakage degree of the internal combustion engine online according to the plurality of cylinder leakage rates.

[0082] For example, Figure 4 As shown, in some optional embodiments, the S300 includes: S310 and S320;

[0083] S310, discretizing the cylinder leakage rates at multiple consecutive moments to obtain multiple discrete values, wherein one moment corresponds to one cylinder leakage rate, and the average value of the cylinder leakage rates at K consecutive moments is taken as a discrete value, where K is an integer greater than 1;

[0084] S320: Determine the cylinder leakage degree of the internal combustion engine online according to the plurality of discrete values.

[0085] Optionally, in some optional implementations, the S320 includes: step 2.1;

[0086] Step 2.1: If Z consecutive discrete values ​​are all within the same preset range, the cylinder leakage degree of the internal combustion engine is determined online to be a set degree corresponding to the preset range, wherein Z is an integer greater than 1, and different preset ranges correspond to different set degrees.

[0087] Optionally, the present invention can count the average value of the leakage rate of 100 consecutive cylinders (i.e., K is equal to 100), and discretize it. Each discrete value of the discretization process is called a judgment unit (initial judgment). Discrete values ​​in different preset ranges correspond to different set degrees, as shown below:

[0088] (1) 0-5% is considered “normal”;

[0089] (2) 5% to 10% is classified as “mild”;

[0090] (3) 10% to 20% was recorded as “moderate”;

[0091] (4) 20% or more is considered “severe”.

[0092] Optionally, if 5 consecutive (i.e., Z is equal to 5) judgment units are the same, then the judgment result is determined to be established (secondary judgment). For example, if the judgment results of 5 consecutive judgment units are all "severe", then the judgment result is determined to be established, that is, the cylinder leakage degree of the internal combustion engine is "severe", and the present invention does not limit this.

[0093] Optionally, in certain optional embodiments, after the online determination in step 2.1 that the cylinder leakage degree of the internal combustion engine is a set degree corresponding to the preset range, the method further comprises: step 3.1;

[0094] Step 3.1: Display page elements of corresponding colors through a visual interface to display the cylinder leakage degree of the internal combustion engine, wherein different cylinder leakage degrees correspond to page elements of different colors.

[0095] Optionally, the determination result of the degree of cylinder leakage can be updated with a visualization result on a BI (Business Intelligence) tool, wherein different degrees of leakage such as normal, mild leakage, moderate leakage, severe leakage, etc. are respectively displayed as page elements of different colors such as green page elements, yellow page elements, orange page elements, and red page elements. The page elements can be text, graphics, or animations, etc., and the present invention is not limited to this.

[0096] In order to further clearly illustrate the detailed scheme of the present invention, the present invention provides as follows Figure 5 The flowchart shown in the figure, please refer to the above explanation for understanding the content of the figure, and the present invention will not elaborate on it any more.

[0097] In summary, the present invention calculates the cylinder leakage rate online based on networked data, ensuring the efficiency and convenience of feedback of the results, which is significantly better than traditional measurement methods such as compression testers or leakage detectors; the present invention adopts working condition screening, data standardization, secondary judgment and other scheme designs to improve the accuracy of the early warning results.

[0098] like Figure 6 As shown, the present invention provides a cylinder leakage detection device, comprising: a model training unit 100, a leakage rate obtaining unit 200 and a leakage degree determining unit 300;

[0099] The model training unit 100 is used to train an SVR model for calculating a cylinder gas leakage rate based on historical operating data of an internal combustion engine under common operating conditions, wherein the common operating conditions include common torque and common speed, and the historical operating data include: exhaust temperature before turbine, pressure after intercooler, and throttle opening;

[0100] The leakage rate obtaining unit 200 is used to input the operating data of the internal combustion engine into the SVR model when the internal combustion engine is operating under the common working condition, and obtain the cylinder leakage rate output by the SVR model, wherein the operating data at one moment corresponds to one cylinder leakage rate, and the operating data of the internal combustion engine includes: exhaust temperature before turbine, pressure after intercooler, and throttle opening;

[0101] The leakage degree determination unit 300 is used to determine the cylinder leakage degree of the internal combustion engine online according to the plurality of cylinder leakage rates.

[0102] Optionally, in some optional implementations, the model training unit 100 includes: a data screening subunit, a preprocessing subunit and a model training subunit;

[0103] The data screening subunit is used to screen out historical operating data of the internal combustion engine under common working conditions from historical data, wherein the historical operating data includes a plurality of turbine exhaust temperatures, a plurality of intercooler pressures and a plurality of throttle openings;

[0104] The preprocessing subunit is used to perform standardization preprocessing on the turbine exhaust temperature, the intercooler pressure and the throttle opening in the historical operation data using a Z-Score method to obtain a corresponding standardized data set, wherein the standardized data set includes a plurality of standardized data corresponding to the turbine exhaust temperature, a plurality of standardized data corresponding to the intercooler pressure and a plurality of standardized data corresponding to the throttle opening;

[0105] The model training subunit is used to train the SVR model for calculating the cylinder leakage rate based on the standardized data set.

[0106] Optionally, in certain optional embodiments, the model training subunit includes: a data set division subunit, a training subunit, a testing subunit and a verification subunit;

[0107] The data set division subunit is used to divide the standardized data set into a training set, a test set and a validation set, wherein the training set, the test set and the validation set all involve standardized data corresponding to the exhaust temperature before the turbine, the standardized data corresponding to the pressure after the intercooler and the standardized data corresponding to the throttle opening;

[0108] The training subunit is used to train the SVR model using the training set;

[0109] The testing subunit is used to evaluate the performance of the SVR model using the test set;

[0110] The validation subunit is used to optimize the parameters of the SVR model by a cross-validation method using the validation set.

[0111] Optionally, in some optional embodiments, the air leakage rate obtaining unit 200 includes: an operation data obtaining unit, a standardization processing subunit and an air leakage rate obtaining subunit;

[0112] The operating data obtaining unit is used to obtain the operating data of the internal combustion engine when the internal combustion engine is operating under the common operating condition;

[0113] The standardization processing subunit is used to perform standardization preprocessing on the operating data of the internal combustion engine based on a mean value and a standard deviation, wherein the mean value and the standard deviation are obtained in the process of performing standardization preprocessing on the pre-turbine exhaust temperature, the post-intercooler pressure and the throttle opening in the historical operating data using a Z-Score method;

[0114] The leakage rate obtaining subunit is used to input the operating data of the internal combustion engine after the standardization preprocessing into the SVR model to obtain the cylinder leakage rate output by the SVR model.

[0115] Optionally, in some optional embodiments, the air leakage degree determination unit 300 includes: a discretization processing subunit and an air leakage degree determination subunit;

[0116] The discretization processing subunit is used to discretize the cylinder leakage rates at multiple consecutive moments to obtain multiple discrete values, wherein one moment corresponds to one cylinder leakage rate, and the average value of the cylinder leakage rates at K consecutive moments is taken as a discrete value, where K is an integer greater than 1;

[0117] The leakage degree determination subunit is used to determine the cylinder leakage degree of the internal combustion engine online according to the multiple discrete values.

[0118] Optionally, in certain optional embodiments, the air leakage degree determination subunit includes: a preset degree determination subunit;

[0119] The preset degree determination subunit is used to online determine the cylinder leakage degree of the internal combustion engine to be a set degree corresponding to the preset range if Z consecutive discrete values ​​are all within the same preset range, wherein Z is an integer greater than 1, and different preset ranges correspond to different set degrees.

[0120] Optionally, in certain optional embodiments, the device further comprises: a result display unit;

[0121] The result display unit is used to display page elements of corresponding colors through a visual interface to display the cylinder leakage degree of the internal combustion engine after the online determination that the cylinder leakage degree of the internal combustion engine is a set degree corresponding to the preset range, wherein different cylinder leakage degrees correspond to page elements of different colors.

[0122] The present invention provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the cylinder leakage detection method described in any one of the above items is implemented.

[0123] like Figure 7 As shown, the present invention provides an electronic device 70, which includes at least one processor 701, and at least one memory 702 and a bus 703 connected to the processor 701; wherein the processor 701 and the memory 702 communicate with each other through the bus 703; the processor 701 is used to call the program instructions in the memory 702 to execute any of the above-mentioned cylinder leakage detection methods.

[0124] In the present invention, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0125] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0126] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined in the present invention may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown in the present invention, but will conform to the widest scope consistent with the principles and novel features disclosed in the present invention.

[0127] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A cylinder leakage detection method, characterized in that: include: Based on the historical operation data of the internal combustion engine under common working conditions, the SVR model for calculating the cylinder leakage rate is trained, including: selecting the historical operation data of the internal combustion engine under common working conditions from the historical data, wherein the historical operation data includes a plurality of turbine exhaust temperatures, a plurality of intercooler pressures, and a plurality of throttle openings; using the Z-Score method to perform standardization preprocessing on the turbine exhaust temperatures, the intercooler pressures, and the throttle openings in the historical operation data, respectively, to obtain corresponding standardized data sets, wherein the standardized data sets include standardized data corresponding to a plurality of turbine exhaust temperatures, a plurality of intercooler pressures, and a plurality of throttle openings; based on the standardized data sets, the SVR model for calculating the cylinder leakage rate is trained; wherein the common working conditions include common torques and common speeds, and the historical operation data include: turbine exhaust temperatures, intercooler pressures, and throttle openings; When the internal combustion engine is running under the common working condition, the operating data of the internal combustion engine is input into the SVR model to obtain the cylinder leakage rate output by the SVR model, wherein the operating data at one moment corresponds to one cylinder leakage rate, and the operating data of the internal combustion engine includes: exhaust temperature before turbine, pressure after intercooler and throttle opening; Based on the multiple cylinder leakage rates, the cylinder leakage degree of the internal combustion engine is determined online, including: discretizing the cylinder leakage rates at multiple consecutive moments to obtain multiple discrete values, wherein one moment corresponds to one cylinder leakage rate, and the average value of the cylinder leakage rates at K consecutive moments is taken as a discrete value, and K is an integer greater than 1; based on the multiple discrete values, the cylinder leakage degree of the internal combustion engine is determined online.

2. The method according to claim 1, characterized in that The step of training the SVR model for calculating the cylinder leakage rate based on the standardized data set comprises: Dividing the standardized data set into a training set, a test set and a validation set, wherein the training set, the test set and the validation set all involve standardized data corresponding to the exhaust temperature before the turbine, the standardized data corresponding to the pressure after the intercooler and the standardized data corresponding to the throttle opening; Training the SVR model using the training set; Using the test set to evaluate the performance of the SVR model; Using the validation set, the parameters of the SVR model were optimized by a cross-validation method.

3. The method according to claim 1, characterized in that When the internal combustion engine is operating under the common working condition, inputting the operating data of the internal combustion engine into the SVR model to obtain the cylinder leakage rate output by the SVR model includes: When the internal combustion engine is operating under the common operating condition, obtaining operating data of the internal combustion engine; Performing standardization preprocessing on the operating data of the internal combustion engine based on a mean value and a standard deviation, wherein the mean value and the standard deviation are obtained in the process of performing standardization preprocessing on the pre-turbine exhaust temperature, the post-intercooler pressure, and the throttle opening in the historical operating data using a Z-Score method; The operating data of the internal combustion engine after the standardization preprocessing is input into the SVR model to obtain the cylinder leakage rate output by the SVR model.

4. The method according to claim 1, characterized in that: The step of determining the cylinder leakage degree of the internal combustion engine online based on the plurality of discrete values ​​comprises: If Z consecutive discrete values ​​are all within the same preset range, the cylinder leakage degree of the internal combustion engine is determined online to be a set degree corresponding to the preset range, wherein Z is an integer greater than 1, and different preset ranges correspond to different set degrees.

5. The method according to claim 4, characterized in that After the online determination that the cylinder leakage degree of the internal combustion engine is a set degree corresponding to the preset range, the method further includes: The page elements of corresponding colors are displayed through a visual interface to display the cylinder leakage degree of the internal combustion engine, wherein different cylinder leakage degrees correspond to page elements of different colors.

6. A cylinder leakage detection device, characterized in that: include: Model training unit, air leakage rate obtaining unit and air leakage degree determining unit; The model training unit is used to train an SVR model for calculating a cylinder gas leakage rate based on historical operating data of the internal combustion engine under common operating conditions, wherein the common operating conditions include common torque and common speed, and the historical operating data include: exhaust temperature before turbine, pressure after intercooler, and throttle opening; The model training unit includes: a data screening subunit, a preprocessing subunit and a model training subunit; The data screening subunit is used to screen out historical operating data of the internal combustion engine under common working conditions from historical data, wherein the historical operating data includes a plurality of turbine exhaust temperatures, a plurality of intercooler pressures and a plurality of throttle openings; The preprocessing subunit is used to perform standardization preprocessing on the turbine exhaust temperature, the intercooler pressure and the throttle opening in the historical operation data using a Z-Score method to obtain a corresponding standardized data set, wherein the standardized data set includes a plurality of standardized data corresponding to the turbine exhaust temperature, a plurality of standardized data corresponding to the intercooler pressure and a plurality of standardized data corresponding to the throttle opening; The model training subunit is used to train the SVR model for calculating the cylinder leakage rate based on the standardized data set; The leakage rate obtaining unit is used to input the operating data of the internal combustion engine into the SVR model when the internal combustion engine is operating under the common working condition, and obtain the cylinder leakage rate output by the SVR model, wherein the operating data at one moment corresponds to one cylinder leakage rate, and the operating data of the internal combustion engine includes: exhaust temperature before turbine, pressure after intercooler, and throttle opening; The air leakage degree determination unit is used to determine the air leakage degree of the cylinder of the internal combustion engine online according to the air leakage rates of the plurality of cylinders; The air leakage degree determination unit comprises: a discretization processing subunit and an air leakage degree determination subunit; The discretization processing subunit is used to discretize the cylinder leakage rates at multiple consecutive moments to obtain multiple discrete values, wherein one moment corresponds to one cylinder leakage rate, and the average value of the cylinder leakage rates at K consecutive moments is taken as a discrete value, where K is an integer greater than 1; The leakage degree determination subunit is used to determine the cylinder leakage degree of the internal combustion engine online according to the multiple discrete values.

7. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the cylinder gas leakage detection method according to any one of claims 1 to 5 is implemented.

8. An electronic device, characterized in that: The electronic device includes at least one processor, and at least one memory and a bus connected to the processor; wherein the processor and the memory communicate with each other via the bus; the processor is used to call program instructions in the memory to execute the cylinder leakage detection method as described in any one of claims 1 to 5.

Images