Data-driven valve core deformation detection method and system

By calculating the extreme deviation coefficient and periodic deviation ratio in the stroke volume data of the needle valve, combined with the heat engine circulation cylinder deformation adjustment coefficient, a cyclic correlation correction sequence is generated, which solves the problem of insufficient time correlation in the traditional detection method and realizes the needle valve Accurate detection of deformation.

CN118462903BActive Publication Date: 2024-11-15DONGGUAN SHOUCI HARDWARE PROD CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410571861.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-10
Publication Date
2024-11-15
Estimated Expiration
2044-05-10

AI Technical Summary

Technical Problem

When detecting needle valve deformation, traditional abnormal data detection algorithm only considers the data size, and does not consider the time correlation, resulting in the detected abnormal values ​​that may not be true deformation values, increasing the detection difficulty.

Method used

By collecting the stroke volume data of the needle valve, the extreme deviation coefficient, period deviation ratio and cylinder stroke cycle deviation coefficient of each stroke cycle period are calculated, combined with the heat engine cycle cylinder deformation adjustment coefficient, a cycle correlation correction sequence is generated, and an abnormal data monitoring algorithm is used for Valve spool deformation detection.

Benefits of technology

The accuracy of needle valve deformation detection is improved, and the stroke variation of the needle valve during different stroke cycles can be accurately evaluated, and abnormal data are corrected to achieve accurate measurement of needle valve deformation in cylinder injector.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118462903B_ABST
    Figure CN118462903B_ABST
Patent Text Reader

Abstract

The present application relates to the field of valve core deformation measurement technology, and specifically to a valve core deformation detection method and system based on data drive, the method comprising: collecting a periodic stroke sequence; obtaining a stroke extreme value sequence; determining the extreme deviation coefficient of each stroke cycle; obtaining the periodic deviation ratio of each stroke cycle in the local comparison time window of each cylinder; obtaining the cylinder stroke period deviation coefficient of each stroke cycle in each cylinder; obtaining the maximum extreme deviation ratio between each cylinder and other cylinders in each thermal engine cycle; obtaining the thermal engine cycle cylinder deformation adjustment coefficient of each cylinder in each thermal engine cycle; determining the cycle correlation correction sequence of each stroke cycle of each cylinder; and detecting the valve core deformation based on the cycle correlation correction sequence. The present application improves the detection accuracy of abnormal strokes, thereby realizing accurate measurement of needle valve deformation in cylinder injectors.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of valve core deformation measurement, and in particular to a data-driven valve core deformation detection method and system. Background Art

[0002] Valves are an indispensable component in fluid control systems. They control the flow of fluids by adjusting the degree of opening and closing. There are many types of valves, which are widely used in different application scenarios and industry fields. The needle valve in the automobile fuel injector is a type of valve, which plays a vital role in the internal combustion engine. The needle valve is located inside the fuel injector and is responsible for controlling the injection amount and injection time of the fuel. By accurately controlling the opening and closing of the needle valve, precise control of the fuel supply can be achieved. However, the needle valve works in a high temperature and high pressure environment for a long time, and it is easy to deform due to factors such as wear and tear. Therefore, it is necessary to detect the deformation of the needle valve.

[0003] Since the overall volume of the needle valve is small and it is located inside the engine, it cannot be disassembled and measured anytime and anywhere, so it is necessary to adopt a measurement method based on a data-driven sensor. However, during the operation of the engine, the engine has different speeds according to different road conditions, so the opening and closing state of the needle valve is constantly changing, that is, the needle valve stroke data is fluctuating, and is affected by fuel quality, temperature changes and pressure fluctuations, which increases the difficulty of detecting the needle valve deformation. An abnormal data detection algorithm is usually used to realize the abnormal value detection of the needle valve stroke fluctuation data, but the traditional abnormal data detection algorithm only considers the size of the data, and does not consider the time correlation of the needle valve stroke change, so it is easy to cause the detected abnormal value to be non-real deformation value. In response to the above problems, the present invention proposes a valve core deformation detection method and system based on data-driven. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a valve core deformation detection method and system based on data drive, and the technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a valve core deformation detection method based on data drive, the method comprising the following steps:

[0006] Collecting each periodic stroke sequence composed of the stroke amount of each needle valve at each sampling moment; obtaining the stroke extreme value sequence of each stroke cycle based on the periodic stroke sequence;

[0007] Based on the stroke extreme value sequence and periodic stroke sequence in each stroke cycle of each cylinder in the automobile engine, the extreme value deviation coefficient of each stroke cycle is determined; according to the number and duration of the periodic stroke sequence and the stroke cycle, the periodic deviation ratio of each stroke cycle in the local comparison time window of each cylinder is obtained; according to the extreme value deviation coefficient, the periodic deviation ratio, and the elements in the periodic stroke sequence, the cylinder stroke periodic deviation coefficient of each stroke cycle in each cylinder is obtained;

[0008] Obtain the maximum deviation ratio between each cylinder and other cylinders in each thermal engine cycle according to the periodic stroke sequence and the stroke extreme value sequence; obtain the thermal engine cycle cylinder deformation adjustment coefficient of each cylinder in each thermal engine cycle according to the maximum deviation ratio and the cylinder stroke period deviation coefficient; determine and obtain the cycle association correction sequence of each stroke cycle of each cylinder by combining the thermal engine cycle cylinder deformation adjustment coefficient and the periodic stroke sequence;

[0009] The valve core deformation is detected based on a cyclically associated correction sequence.

[0010] Further, the step of obtaining a stroke extreme value sequence of each stroke cycle based on the periodic stroke sequence includes:

[0011] The periodic stroke sequence in each stroke cycle is used as the input of the fitting algorithm, and the periodic stroke curve is outputted. The maximum value points on the periodic stroke curve are sorted in chronological order to construct the stroke extreme value sequence of the stroke cycle;

[0012] The stroke cycle includes an intake stroke, a compression stroke, a power stroke and an exhaust stroke of the cylinder.

[0013] Further, the determination of the extreme deviation coefficient of each stroke cycle period includes:

[0014] For each stroke cycle of each cylinder, the difference between the maximum value in the periodic stroke sequence and each element in the stroke extreme value sequence is calculated as the first difference, and the sum of all the first differences in each stroke cycle of each cylinder is taken as the periodic stroke sum value, and the periodic stroke sum value is negatively correlated with the extreme value deviation coefficient.

[0015] Further, the step of obtaining the period deviation ratio of each stroke cycle in the local comparison time window of each cylinder includes:

[0016] Taking each periodic travel sequence as the end point, a preset number of periodic travel sequences are selected forward to form a local comparison time window;

[0017] In the local comparison time window of each cylinder, the difference between the sampling time of the maximum value in the periodic stroke sequence of each stroke cycle and the sampling time of the maximum value in the periodic stroke sequence of other stroke cycles is calculated as the second difference, the number of stroke cycles between each stroke cycle and other stroke cycles is calculated as the first category number, and the average duration of all stroke cycles is calculated;

[0018] The period deviation ratio is positively correlated with the second difference, negatively correlated with the first category number, and negatively correlated with the average duration.

[0019] Further, the obtaining of the cylinder stroke cycle deviation coefficient of each stroke cycle in each cylinder includes:

[0020] For the local comparison time window where each stroke cycle in each cylinder is located, the absolute value of the difference between the period deviation ratio of other stroke cycles in the local comparison time window and the number 1 is calculated as the first absolute value of the difference, the absolute value of the difference between the maximum value in the periodic stroke sequence of each stroke cycle and the maximum value in the periodic stroke sequence of other stroke cycles in the local comparison time window is calculated as the second absolute value of the difference, and the mean of all the first absolute values ​​of the difference in the local comparison time window is taken as the first mean; the cylinder stroke period deviation coefficient is positively correlated with the first mean, positively correlated with the second absolute value of the difference, and positively correlated with the extreme value deviation coefficient.

[0021] Further, the step of obtaining the maximum deviation ratio between each cylinder and other cylinders in each heat engine cycle includes:

[0022] The cycle consisting of the stroke cycles of all cylinders in the engine is taken as each heat engine cycle;

[0023] For each thermal engine cycle, the ratio of the maximum value of the elements in the periodic stroke sequence of the stroke cycle of each cylinder in each thermal engine cycle to the maximum value of the elements in the periodic stroke sequence of the stroke cycle of other cylinders is calculated as the first ratio, the absolute value of the difference between the first ratio and the number 1 is calculated as the third absolute value of the difference, the correlation between the stroke extreme value sequence in the stroke cycle of each cylinder and the stroke extreme value sequence in the stroke cycle of other cylinders is calculated as the stroke correlation, the extreme value deviation ratio is positively correlated with the third absolute value of the difference, and is negatively correlated with the stroke correlation.

[0024] Further, the step of obtaining the heat engine cycle cylinder deformation adjustment coefficient of each cylinder in each heat engine cycle period includes:

[0025] In each thermal engine cycle, the absolute value of the difference between the cylinder stroke cycle deviation coefficient of each stroke cycle of each cylinder and the cylinder stroke cycle deviation coefficient of each stroke cycle of other cylinders is calculated as the fourth absolute value of the difference. The thermal engine cycle cylinder deformation adjustment coefficient is positively correlated with the fourth absolute value of the difference and is positively correlated with the extreme deviation ratio.

[0026] Further, the determining and obtaining of the cycle-related correction sequence of each stroke cycle of each cylinder includes:

[0027] Taking the absolute value of the obtained data after first-order difference processing on the periodic stroke sequence of each cylinder in each stroke cycle, obtaining the stroke residual sequence of each cylinder in each stroke cycle, and inserting the number 0 at the beginning of the stroke residual sequence as the first data of the stroke residual sequence;

[0028] For each stroke cycle of each cylinder, data in the periodic stroke sequence that is in the same position as the stroke amount at each sampling moment is obtained as the first data, the product of the first data and the heat engine cycle cylinder deformation adjustment coefficient of each cylinder in each heat engine cycle is taken as the isotopic stroke product, the calculation result of an exponential function with a natural constant as the base and the isotopic stroke product as the exponent is calculated as the isotopic stroke index of each stroke amount in the periodic stroke sequence, the product of each stroke amount in the periodic stroke sequence and the isotopic stroke index of each stroke amount is calculated as the cyclic correlation correction value of each stroke amount, and the cyclic correlation correction values ​​of each stroke amount are sorted according to the order of each stroke amount in its periodic stroke sequence to obtain the cyclic correlation correction sequence of each stroke cycle of each cylinder.

[0029] Further, the detection of valve core deformation based on the cyclic associated correction sequence includes:

[0030] For the cyclic correlation correction sequence of each cylinder in each stroke cycle, the cyclic correlation correction sequence is used as the input of the abnormal data monitoring algorithm, and the abnormal score of each element in the cyclic correlation correction sequence is output. The abnormal score of each element in the cyclic correlation correction sequence is used as the input of the threshold segmentation method. The algorithm outputs the abnormal threshold, and the elements in the cyclic correlation correction sequence whose abnormal scores are less than or equal to the abnormal threshold are marked as normal elements, and the elements whose abnormal scores are greater than the abnormal threshold are marked as abnormal elements;

[0031] The ratio of the number of abnormal elements in each stroke cycle of each cylinder to the total number at the sampling time is calculated as the second ratio. If the second ratio is greater than the preset value, it indicates that the needle valve of the cylinder is deformed in the current stroke cycle; otherwise, it indicates that the needle valve of the cylinder is not deformed in the current stroke cycle.

[0032] In a second aspect, an embodiment of the present application also provides a data-driven valve core deformation detection system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods when executing the computer program.

[0033] This application has at least the following beneficial effects:

[0034] This application measures the stroke data of the needle valve in each cylinder injector in the engine, obtains the cylinder stroke cycle deviation coefficient based on the stroke data of a single cylinder in a single stroke cycle and the difference with other cycles in the local comparison time window, reflects the difference of a single cylinder in different stroke cycles through horizontal comparison, and evaluates the stroke variation in the current cycle, and combines the difference between different cylinders in the thermal engine cycle to obtain the thermal engine cycle cylinder deformation adjustment coefficient, which reflects the abnormal degree of the needle valve stroke data of the cylinder in the vertical comparison, and finally obtains the cycle correlation correction sequence, and realizes abnormal data monitoring. It solves the problem that the traditional abnormal data detection algorithm only considers the numerical value but not the time correlation characteristics between the stroke data. This application amplifies the abnormal value according to the time correlation characteristics of the needle valve stroke data, improves the detection accuracy of the abnormal stroke, and realizes the accurate measurement of the needle valve deformation in the cylinder injector. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 A flowchart of the steps of a valve core deformation detection method based on data drive provided in one embodiment of the present application;

[0037] Figure 2 It is a schematic diagram of the stroke cycle;

[0038] Figure 3 Schematic diagram of the heat engine cycle. DETAILED DESCRIPTION

[0039] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the data-driven valve core deformation detection method and system proposed in the present application, its specific implementation method, structure, features and effects are described in detail below in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0041] The specific scheme of the valve core deformation detection method and system based on data-driven provided by the present application is described in detail below with reference to the accompanying drawings.

[0042] See also Figure 1 , which shows a flow chart of the steps of a valve core deformation detection method based on data drive provided by an embodiment of the present application, the method comprising the following steps:

[0043] Step S001, obtaining the stroke data of the needle valve in the fuel injection nozzle through a high-precision displacement sensor.

[0044] The completion of the intake stroke, compression stroke, power stroke and exhaust stroke for a single cylinder in the engine is called a stroke cycle, and the needle valve of the fuel injector corresponding to the cylinder will open in the compression stroke, and the injection amount will be controlled by the engine control unit ECU, and the needle valve will be closed in other stroke times. In addition, a single engine contains multiple cylinders. The most common one on the market is the four-cylinder engine. This embodiment takes the four-cylinder engine as an example. Different cylinders in the engine are often arranged horizontally opposite to each other, and the cylinders are connected by a crankshaft, that is, when the first cylinder is in the intake stroke, the second cylinder is in the compression stroke, and so on.

[0045] In order to obtain the deformation of the needle valve of the fuel injector in the cylinder, a miniature high-precision displacement sensor is arranged at the needle valve of the fuel injector in each cylinder, and the stroke data of the needle valve is measured at equal intervals, and the sampling frequency is set to 10KHz. The stroke sequences of the four needle valves are obtained, which are represented by A1, A2, A3, and A4 respectively, where A1 represents the stroke sequence of the first cylinder. In a single stroke sequence, taking the first cylinder as an example, A1 = {a1(k)|k∈[1,N]}, a1(k) represents the stroke amount of the first stroke sequence at the kth sampling moment, and N represents the total number of samples. It should be noted that this application only samples data during the operation of the car, and the size of N depends on the specific situation. In order to reduce the impact of environmental noise and sensor noise on data analysis, this embodiment will obtain the stroke sequence and use Gaussian filtering for noise reduction. Gaussian filtering is a well-known technology and will not be repeated in this application.

[0046] Step S002, according to the distribution characteristics of the stroke sequence of a single cylinder in a single stroke cycle, and the difference with other stroke cycles in the local comparison time window, the cylinder stroke cycle deviation coefficient is obtained, and the maximum stroke distance difference between different cylinders in a single thermal engine cycle is calculated to obtain the thermal engine cycle cylinder deformation adjustment coefficient, and the cycle correlation correction sequence is obtained in combination with the stroke residual sequence.

[0047] Ideally, the travel distribution of the needle valve in the injection nozzle of a single cylinder in one stroke cycle is as follows: Figure 2 As shown in the figure, the needle valve opens only in the compression stroke to allow the injector to spray oil, and the needle valve gradually closes when the injection is completed. The stroke cycle of a single cylinder is continuous. Since a stroke cycle is short, and considering that the vehicle operating conditions do not change significantly in a short period of time. Therefore, in the adjacent cycles of the stroke cycle, the needle valve stroke sequence often has a certain periodicity. However, if the needle valve in the cylinder is affected by the injection pressure, it will produce a certain deformation, resulting in a large difference in the needle valve stroke sequence in different stroke cycles, destroying the original periodic characteristics.

[0048] During the operation of the vehicle, the displacement sensor continuously obtains the stroke data of the injection nozzle needle valve in the cylinder. When the acquisition time is long enough, it is not conducive to data analysis. In this embodiment, the engine control unit ECU obtains the stroke cycle of each cylinder, thereby dividing the infinitely long stroke sequence into a sequence based on the stroke cycle. Represents the periodic stroke sequence of the first cylinder in the engine during the mth stroke cycle.

[0049] In order to measure the fluctuation degree and periodic characteristics of the needle valve periodic stroke sequence, taking the first cylinder as an example, the periodic stroke sequence in the mth stroke cycle is As the input of the Smooth smooth fitting algorithm, the algorithm outputs the cycle stroke curve, calculates the maximum value points for the cycle stroke curve, arranges the maximum values ​​in the time sequence of the maximum value points to obtain the stroke extreme value sequence, and uses The Smooth smoothing fitting algorithm is a well-known technology and will not be described in detail in this application.

[0050] Furthermore, the extreme value deviation coefficient of each stroke cycle in each cylinder is determined according to the stroke extreme value sequence and the periodic stroke sequence in each stroke cycle of each cylinder in the automobile engine, and the acquisition method is:

[0051] For each stroke cycle of each cylinder, the difference between the maximum value in the periodic stroke sequence and each element in the stroke extreme value sequence is calculated as the first difference, and the sum of all the first differences in each stroke cycle of each cylinder is taken as the periodic stroke sum value, and the periodic stroke sum value is negatively correlated with the extreme value deviation coefficient.

[0052] Specifically, the present application proposes an embodiment in which the reciprocal of the periodic stroke and value is used as the extreme value deviation coefficient.

[0053] It should be noted that a negative correlation means that the dependent variable decreases as the independent variable increases, and increases as the independent variable decreases. The negative correlation can be a subtraction relationship, a division relationship, etc., which can be determined according to actual conditions and will not be elaborated in this application.

[0054] Furthermore, taking each stroke cycle as the last stroke cycle, W stroke cycles are selected forward to construct a local comparison time window for each stroke cycle. The first W stroke cycles when the automobile engine just starts running are not processed. In this embodiment, W=10.

[0055] Furthermore, the period deviation ratio of each stroke cycle in the local comparison time window of each cylinder is obtained according to the number and duration of the periodic stroke sequence and the stroke cycle, and the acquisition method is:

[0056] In the local comparison time window of each stroke cycle of each cylinder, the difference between the sampling time of the maximum value in the periodic stroke sequence of each stroke cycle and the sampling time of the maximum value in the periodic stroke sequence of other stroke cycles in the local comparison time window is calculated as the second difference, the number of stroke cycles between each stroke cycle and other stroke cycles is calculated as the first category number, and the average duration of all stroke cycles is calculated; the period deviation ratio is positively correlated with the second difference, negatively correlated with the first category number, and negatively correlated with the average duration.

[0057] Specifically, the present application proposes an embodiment, calculating the product of the first category number and the average duration as the first product, and taking the ratio of the second difference to the first product as the period deviation ratio.

[0058] It should be noted that the positive correlation relationship can be a multiplication, addition, or idempotent relationship of an exponential function, which can be determined based on actual conditions and will not be elaborated in this application; the negative correlation relationship can be a subtraction relationship, a division relationship, etc., which can be determined based on actual conditions and will not be elaborated in this application.

[0059] Furthermore, the cylinder stroke period deviation coefficient of each stroke cycle in each cylinder is obtained according to the extreme value deviation coefficient, the period deviation ratio, and the elements in the period stroke sequence. The obtaining method is:

[0060] For the local comparison time window where each stroke cycle in each cylinder is located, the absolute value of the difference between the period deviation ratio of other stroke cycles in the local comparison time window and the number 1 is calculated as the first absolute value of the difference, the absolute value of the difference between the maximum value in the periodic stroke sequence of each stroke cycle and the maximum value in the periodic stroke sequence of other stroke cycles in the local comparison time window is calculated as the second absolute value of the difference, and the mean of all the first absolute values ​​of the difference in the local comparison time window is taken as the first mean; the cylinder stroke period deviation coefficient is positively correlated with the first mean, positively correlated with the second absolute value of the difference, and positively correlated with the extreme value deviation coefficient.

[0061] Specifically, the present application proposes an embodiment in which the product of the first mean value, the absolute value of the second difference and the extreme value deviation coefficient is used as the cylinder stroke cycle deviation coefficient.

[0062] It should be noted that the positive correlation relationship can be an idempotent relationship of multiplication, addition, or exponential function, which can be determined based on actual conditions and will not be elaborated in this application.

[0063] If the needle valve is deformed during the opening and closing process in the current stroke cycle due to factors such as injection pressure fluctuations, it will cause premature full opening or incomplete full opening, and cause large fluctuations in the periodic stroke sequence. Therefore, there are more maximum values ​​in the fitted periodic stroke curve, resulting in a decrease in the deviation between the maximum and the extreme value, and a larger value of the extreme value deviation coefficient. Due to the deformation, there is a deviation in the position of the maximum value in the current stroke cycle, so the ratio of the interval between the maximum values ​​to the interval between the stroke cycles is significantly different from 1. In addition, deformation will also cause changes in the size of the opening and closing of the needle valve, resulting in a large deviation from the maximum stroke of other stroke cycles in the local comparison time window, which ultimately makes the cylinder stroke cycle deviation coefficient On the contrary, if the current cylinder does not deform and maintains stable operation, the periodic stroke sequence fluctuation is small in the current stroke cycle, and the periodic stroke curve is smoother, with fewer maximum values ​​or only one maximum value, and the difference with other stroke cycles in the comparison time window is small, and the result is The value of is smaller.

[0064] The cylinder stroke cycle deviation coefficient is obtained by analyzing the differences in the cycle stroke sequences of a single cylinder in different stroke cycles. If the driver drives the car more violently, there will be slight differences in the cycle stroke sequences in different stroke cycles within the comparison time window. In order to reduce the interference of slight differences on the needle valve deformation, the differences between different cylinders in the stroke cycle are analyzed based on the above comparison time window analysis.

[0065] In this application, a four-cylinder engine is used as an example. Therefore, there is a stroke time difference between different cylinders. In order to obtain the same engine working condition, Figure 3 As shown, this embodiment divides the heat engine cycle period, and obtains the complete stroke cycle period of different cylinders in the same heat engine cycle period to analyze the deformation of the needle valve under the same working condition of the engine.

[0066] Combining the differences in the periodic stroke sequences of different cylinder stroke cycles in the thermal engine cycle, the thermal engine cycle cylinder deformation adjustment coefficient is obtained:

[0067] Furthermore, the maximum deviation ratio between each cylinder and other cylinders in each heat engine cycle is obtained according to the periodic stroke sequence and the stroke extreme value sequence. The obtaining method is:

[0068] For each thermal engine cycle, the ratio of the maximum value of the elements in the periodic stroke sequence of the stroke cycle of each cylinder in each thermal engine cycle to the maximum value of the elements in the periodic stroke sequence of the stroke cycle of other cylinders is calculated as the first ratio, the absolute value of the difference between the first ratio and the number 1 is calculated as the third absolute value of the difference, the correlation between the stroke extreme value sequence in the stroke cycle of each cylinder and the stroke extreme value sequence in the stroke cycle of other cylinders is calculated as the stroke correlation, the extreme value deviation ratio is positively correlated with the third absolute value of the difference, and is negatively correlated with the stroke correlation.

[0069] Specifically, the present application proposes an embodiment in which the ratio between the absolute value of the third difference and the stroke correlation is used as the maximum extreme deviation ratio.

[0070] It should be noted that the positive correlation relationship can be an idempotent relationship of multiplication, addition, or exponential function, which can be determined based on actual conditions and will not be elaborated in this application.

[0071] Furthermore, the heat engine cycle cylinder deformation adjustment coefficient of each cylinder in each heat engine cycle is obtained according to the maximum value extreme deviation ratio and the cylinder stroke cycle deviation coefficient. The acquisition method is:

[0072] In each thermal engine cycle, the absolute value of the difference between the cylinder stroke cycle deviation coefficient of each stroke cycle of each cylinder and the cylinder stroke cycle deviation coefficient of each stroke cycle of other cylinders is calculated as the fourth absolute value of the difference. The thermal engine cycle cylinder deformation adjustment coefficient is positively correlated with the fourth absolute value of the difference, and is positively correlated with the maximum extreme deviation ratio.

[0073] Specifically, the present application proposes an embodiment in which the product of the maximum extreme deviation ratio and the absolute value of the fourth difference is used as the heat engine cycle cylinder deformation adjustment coefficient.

[0074] It should be noted that the positive correlation relationship can be an idempotent relationship of multiplication, addition, or exponential function, which can be determined based on actual conditions and will not be elaborated in this application.

[0075] Since the four cylinders are responding to the same engine working condition in the same heat engine cycle, and the possibility of needle valve deformation in the four cylinders at the same time is extremely low. Therefore, when the needle valve deformation occurs in the i-th cylinder in the heat engine cycle, the periodic stroke sequence of this cylinder is significantly different from that of other cylinders. Specifically, the maximum value difference between the periodic stroke sequence of this cylinder and the periodic stroke sequence of other cylinders is large, and the correlation between the stroke extreme value sequences is small due to the influence of needle valve deformation, and the value of the maximum value extreme deviation ratio is large. At the same time, the difference between the cylinder stroke cycle deviation coefficient of this cylinder and other cylinders is large, and finally the value of the heat engine cycle cylinder deformation adjustment coefficient is large.

[0076] After performing first-order difference processing on the periodic stroke sequence of each cylinder in each stroke cycle, the absolute value of the obtained data is taken to obtain the stroke residual sequence of each cylinder in each stroke cycle, and the number 0 is inserted at the beginning of the stroke residual sequence as the first data of the stroke residual sequence, so that the length of the periodic stroke sequence and the stroke residual sequence of each cylinder in each stroke cycle is the same. Further, the periodic stroke sequence is corrected by combining the heat engine cycle cylinder deformation adjustment coefficient and the stroke residual sequence, and the cyclic correlation correction sequence of each stroke cycle of each cylinder is obtained. The acquisition method is:

[0077] For each stroke cycle of each cylinder, data in the periodic stroke sequence that is in the same position as the stroke amount at each sampling moment is obtained as the first data, the product of the first data and the heat engine cycle cylinder deformation adjustment coefficient of each cylinder in each heat engine cycle is taken as the isotopic stroke product, the calculation result of an exponential function with a natural constant as the base and the isotopic stroke product as the exponent is calculated as the isotopic stroke index of each stroke amount in the periodic stroke sequence, the product of each stroke amount in the periodic stroke sequence and the isotopic stroke index of each stroke amount is calculated as the cyclic correlation correction value of each stroke amount, and the cyclic correlation correction values ​​of each stroke amount are sorted according to the order of each stroke amount in its periodic stroke sequence to obtain the cyclic correlation correction sequence of each stroke cycle of each cylinder.

[0078] Traverse all sampling moments in the mth stroke cycle of the current cylinder to obtain the corresponding cycle correlation correction sequence D i m .

[0079] If the needle valve is deformed or has been deformed at the current sampling time, the obtained thermal engine cycle cylinder deformation adjustment coefficient The value of is large, and affected by the deformation, the element value band of the periodic stroke sequence is large, resulting in a large residual value between adjacent sampling moments, thereby amplifying the abnormal stroke amount at the sampling moment, thereby realizing the detection of the abnormal stroke amount of the needle valve.

[0080] Step S003, according to the obtained cyclic correlation correction sequence, an abnormal value is obtained through an algorithm, and an abnormal threshold is set to determine the detection of the deformation of the engine cylinder needle valve.

[0081] According to step S002, the cyclic correlation correction sequence of a single cylinder in the stroke cycle period is obtained. In this embodiment, the cyclic correlation correction sequence is used as the input of the isolation forest IF algorithm, the number of isolated trees is set to 100, and the algorithm outputs the abnormal score of each element in the cyclic correlation correction sequence; wherein, the abnormal data detection algorithm selected in this embodiment is the isolation forest IF algorithm, and the implementer can select other abnormal data detection algorithms according to actual conditions.

[0082] Further, the abnormality score of each element in the cyclic association correction sequence is used as the input of the Otsu threshold method, and the algorithm outputs the abnormality threshold, and the elements with abnormality scores less than or equal to the abnormality threshold in the cyclic association correction sequence are marked as normal elements, and the elements with abnormality scores greater than the abnormality threshold are marked as abnormal elements. The isolation forest IF algorithm and the Otsu threshold method are well-known technologies and will not be described in detail in this embodiment.

[0083] Thus, the ratio of the number of abnormal elements in the current stroke cycle of the cylinder to the total number at the sampling time is counted. If this ratio is greater than 0.1%, it indicates that the needle valve of the cylinder is deformed in the current stroke cycle, and the abnormal situation is prompted to the driver through the display platform of the on-board computer. Vehicle maintenance is required to avoid greater deformation of the needle valve, so as to reduce the hidden danger to the driver's personal and property safety; otherwise, it indicates that the needle valve of the cylinder is not deformed in the current stroke cycle.

[0084] Based on the same inventive concept as the above method, an embodiment of the present application also provides a data-driven valve core deformation detection system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, it implements the steps of any one of the above-mentioned data-driven valve core deformation detection methods.

[0085] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0086] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0087] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. Data-driven valve core deformation detection method, It is characterized in that The method comprises the following steps: Collecting each periodic stroke sequence composed of the stroke amount of each needle valve at each sampling moment; obtaining the stroke extreme value sequence of each stroke cycle based on the periodic stroke sequence; Based on the stroke extreme value sequence and periodic stroke sequence in each stroke cycle of each cylinder in the automobile engine, the extreme value deviation coefficient of each stroke cycle is determined; according to the number and duration of the periodic stroke sequence and the stroke cycle, the periodic deviation ratio of each stroke cycle in the local comparison time window of each cylinder is obtained; according to the extreme value deviation coefficient, the periodic deviation ratio, and the elements in the periodic stroke sequence, the cylinder stroke periodic deviation coefficient of each stroke cycle in each cylinder is obtained; Obtain the maximum deviation ratio between each cylinder and other cylinders in each thermal engine cycle according to the periodic stroke sequence and the stroke extreme value sequence; obtain the thermal engine cycle cylinder deformation adjustment coefficient of each cylinder in each thermal engine cycle according to the maximum deviation ratio and the cylinder stroke period deviation coefficient; determine and obtain the cycle association correction sequence of each stroke cycle of each cylinder by combining the thermal engine cycle cylinder deformation adjustment coefficient and the periodic stroke sequence; Detect valve core deformation based on cyclic correlation correction sequence; The detecting of valve core deformation based on the cyclic associated correction sequence comprises: For the cyclic correlation correction sequence of each cylinder in each stroke cycle, the cyclic correlation correction sequence is used as the input of the abnormal data monitoring algorithm, and the abnormal score of each element in the cyclic correlation correction sequence is output. The abnormal score of each element in the cyclic correlation correction sequence is used as the input of the threshold segmentation method. The algorithm outputs the abnormal threshold, and the elements in the cyclic correlation correction sequence whose abnormal scores are less than or equal to the abnormal threshold are marked as normal elements, and the elements whose abnormal scores are greater than the abnormal threshold are marked as abnormal elements; The ratio of the number of abnormal elements in each stroke cycle of each cylinder to the total number at the sampling time is calculated as the second ratio. If the second ratio is greater than the preset value, it indicates that the needle valve of the cylinder is deformed in the current stroke cycle; otherwise, it indicates that the needle valve of the cylinder is not deformed in the current stroke cycle.

2. The valve core deformation detection method based on data drive according to claim 1, It is characterized in that The step of obtaining a stroke extreme value sequence of each stroke cycle based on a periodic stroke sequence includes: The periodic stroke sequence in each stroke cycle is used as the input of the fitting algorithm, and the periodic stroke curve is outputted. The maximum value points on the periodic stroke curve are sorted in chronological order to construct the stroke extreme value sequence of the stroke cycle; The stroke cycle includes an intake stroke, a compression stroke, a power stroke and an exhaust stroke of the cylinder.

3. The valve core deformation detection method based on data drive according to claim 1, It is characterized in that Determining the extreme value deviation coefficient of each stroke cycle period includes: For each stroke cycle of each cylinder, the difference between the maximum value in the periodic stroke sequence and each element in the stroke extreme value sequence is calculated as the first difference, and the sum of all the first differences in each stroke cycle of each cylinder is taken as the periodic stroke sum value, and the periodic stroke sum value is negatively correlated with the extreme value deviation coefficient.

4. The valve core deformation detection method based on data drive according to claim 1, It is characterized in that The step of obtaining the period deviation ratio of each stroke cycle in the local comparison time window of each cylinder includes: Taking each periodic travel sequence as the end point, a preset number of periodic travel sequences are selected forward to form a local comparison time window; In the local comparison time window of each cylinder, the difference between the sampling time of the maximum value in the periodic stroke sequence of each stroke cycle and the sampling time of the maximum value in the periodic stroke sequence of other stroke cycles is calculated as the second difference, the number of stroke cycles between each stroke cycle and other stroke cycles is calculated as the first category number, and the average duration of all stroke cycles is calculated; The period deviation ratio is positively correlated with the second difference, negatively correlated with the first category number, and negatively correlated with the average duration.

5. The valve core deformation detection method based on data drive according to claim 1, It is characterized in that The step of obtaining the cylinder stroke cycle deviation coefficient of each stroke cycle in each cylinder includes: For the local comparison time window where each stroke cycle in each cylinder is located, the absolute value of the difference between the period deviation ratio of other stroke cycles in the local comparison time window and the number 1 is calculated as the first absolute value of the difference, the absolute value of the difference between the maximum value in the periodic stroke sequence of each stroke cycle and the maximum value in the periodic stroke sequence of other stroke cycles in the local comparison time window is calculated as the second absolute value of the difference, and the mean of all the first absolute values ​​of the difference in the local comparison time window is taken as the first mean; the cylinder stroke period deviation coefficient is positively correlated with the first mean, positively correlated with the second absolute value of the difference, and positively correlated with the extreme value deviation coefficient.

6. The data-driven valve core deformation detection method according to claim 1, It is characterized in that The step of obtaining the maximum deviation ratio between each cylinder and other cylinders in each heat engine cycle includes: The cycle consisting of the stroke cycles of all cylinders in the engine is taken as each heat engine cycle; For each thermal engine cycle, the ratio of the maximum value of the elements in the periodic stroke sequence of the stroke cycle of each cylinder in each thermal engine cycle to the maximum value of the elements in the periodic stroke sequence of the stroke cycle of other cylinders is calculated as the first ratio, the absolute value of the difference between the first ratio and the number 1 is calculated as the third absolute value of the difference, the correlation between the stroke extreme value sequence in the stroke cycle of each cylinder and the stroke extreme value sequence in the stroke cycle of other cylinders is calculated as the stroke correlation, the extreme value deviation ratio is positively correlated with the third absolute value of the difference, and is negatively correlated with the stroke correlation.

7. The valve core deformation detection method based on data drive according to claim 1, It is characterized in that The step of obtaining the heat engine cycle cylinder deformation adjustment coefficient of each cylinder in each heat engine cycle period includes: In each thermal engine cycle, the absolute value of the difference between the cylinder stroke cycle deviation coefficient of each stroke cycle of each cylinder and the cylinder stroke cycle deviation coefficient of each stroke cycle of other cylinders is calculated as the fourth absolute value of the difference. The thermal engine cycle cylinder deformation adjustment coefficient is positively correlated with the fourth absolute value of the difference and is positively correlated with the extreme deviation ratio.

8. The valve core deformation detection method based on data drive according to claim 1, It is characterized in that The step of determining and obtaining a cyclic correlation correction sequence of each stroke cycle period of each cylinder includes: Taking the absolute value of the obtained data after first-order difference processing on the periodic stroke sequence of each cylinder in each stroke cycle, obtaining the stroke residual sequence of each cylinder in each stroke cycle, and inserting the number 0 at the beginning of the stroke residual sequence as the first data of the stroke residual sequence; For each stroke cycle of each cylinder, data in the periodic stroke sequence that is in the same position as the stroke amount at each sampling moment is obtained as the first data, the product of the first data and the heat engine cycle cylinder deformation adjustment coefficient of each cylinder in each heat engine cycle is taken as the isotopic stroke product, the calculation result of an exponential function with a natural constant as the base and the isotopic stroke product as the exponent is calculated as the isotopic stroke index of each stroke amount in the periodic stroke sequence, the product of each stroke amount in the periodic stroke sequence and the isotopic stroke index of each stroke amount is calculated as the cyclic correlation correction value of each stroke amount, and the cyclic correlation correction values ​​of each stroke amount are sorted according to the order of each stroke amount in its periodic stroke sequence to obtain the cyclic correlation correction sequence of each stroke cycle of each cylinder.

9. A valve core deformation detection system based on data drive, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.