Method and equipment for predicting service life of main bearing of engine crankshaft

Abstract

The invention discloses a method and device for predicting the service life of an engine crankshaft main bearing, and relates to the technical field of engines, and the method comprises the steps: obtaining a vibration signal of the engine crankshaft main bearing, and carrying out the one-dimensional depth feature extraction of the segmented vibration signal, and obtaining a feature vector; inputting the feature vector into an SOM network for training, determining a target optimal neuron and a corresponding weight, and calculating a vibration feature life index of the main bearing; obtaining performance parameters of the main bearing, establishing a main bearing life prediction physical model, and calculating a physical model life index of the main bearing; based on the vibration characteristic life index and the physical model life index of the main bearing, determining a life comprehensive evaluation index of the main bearing; and generating a main bearing life prediction curve based on the life comprehensive evaluation index of the main bearing. The service life of the crankshaft main bearing is predicted by combining a vibration characteristic service life index obtained by deep learning and a physical model service life index obtained by material mechanics, and the prediction accuracy and reliability are improved.

Description

Technical Field

[0001] The present application relates to the field of engine technology, and in particular to a method and device for predicting the life of an engine crankshaft main bearing. Background Art

[0002] As a key rotating component in the engine, the crankshaft main bearing of the automobile engine has the dual tasks of transmitting engine power and supporting the operation of the crankshaft. Its performance stability and durability are crucial to the operating efficiency and life of the entire engine. During the operation of the engine, the crankshaft main bearing is subjected to complex loads and high-speed rotation. Over time, these loads and friction will cause wear, fatigue and even failure. Therefore, accurately predicting the life of the crankshaft main bearing is of great significance to maintaining engine performance and avoiding unexpected failures.

[0003] At present, for the life prediction of crankshaft main bearings, the traditional empirical formula method has limited accuracy, is difficult to adapt to complex and changeable actual working conditions, is overly dependent on historical data, and is not accurate enough for bearing life prediction for new engines or under special working conditions; the traditional finite element analysis method uses engineering software to perform mechanical analysis to predict the stress and strain distribution of engine crankshaft main bearings under specific working conditions, including model establishment, boundary setting, model debugging and operation calculation, which requires a large amount of simulation resources and has a long cycle and high computational cost, making it unsuitable for real-time monitoring and rapid prediction; traditional machine learning methods are highly dependent on feature engineering, and are difficult to process high-dimensional data and capture complex patterns in time series, and their accuracy needs to be improved.

[0004] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of the present application is to provide a method and device for predicting the life of an engine crankshaft main bearing, aiming to solve the technical problem of low accuracy in predicting the life of a crankshaft main bearing in the prior art.

[0006] To achieve the above objectives, the present application provides a method for predicting the life of an engine crankshaft main bearing, the method comprising:

[0007] Obtain the vibration signals of the engine crankshaft main bearing at different times during the entire life cycle, perform one-dimensional deep feature extraction on the segmented vibration signals, and obtain the feature vector corresponding to the vibration signal;

[0008] The feature vector is input into the SOM network for training, and the target best neuron and the weight of the target best neuron are determined among the neurons of the SOM network;

[0009] Based on the weight of the eigenvector and the target optimal neuron, the vibration characteristic life index of the main bearing is calculated;

[0010] Obtain the performance parameters of the main bearing and establish a physical model for predicting the life of the main bearing;

[0011] Based on the physical model for predicting the life of the main bearing, the physical model life index of the main bearing is calculated;

[0012] Based on the vibration characteristic life index and physical model life index of the main bearing, determine the comprehensive evaluation index of the main bearing life;

[0013] Based on the comprehensive evaluation index of the main bearing life, the main bearing life prediction curve is generated.

[0014] In one embodiment, the step of performing one-dimensional deep feature extraction on the segmented vibration signal to obtain a feature vector corresponding to the vibration signal includes:

[0015] The segmented vibration signal is subjected to one-dimensional deep feature extraction through a one-dimensional deep self-encoding feature unit to obtain a feature vector corresponding to the vibration signal. The one-dimensional deep self-encoding feature unit includes an encoding network and a decoding network. The encoding network includes a convolution layer and a pooling layer, and the decoding network includes an upsampling layer and a convolution layer.

[0016] In one embodiment, the feature vector is input into a SOM network for training, and the step of determining a target optimal neuron and a weight of the target optimal neuron in the neurons of the SOM network includes:

[0017] Initialize the SOM network and the weights of the neurons in the SOM network;

[0018] Input the feature vector into the SOM network and calculate the distance between the neuron and the feature vector;

[0019] The neuron with the smallest distance is regarded as the best neuron, and the neurons other than the best neuron are regarded as surrounding neurons;

[0020] Adjust the weights of the best neuron and surrounding neurons to make the surrounding neurons move closer to the best neuron;

[0021] Update the distance between the neuron and the feature vector, and re-determine the best neuron and surrounding neurons, and continue to iterate until the preset conditions are met to obtain the target best neuron and the weight of the target best neuron.

[0022] In one embodiment, the step of calculating the vibration characteristic life index of the main bearing based on the characteristic vector and the weight of the target optimal neuron includes:

[0023] Obtaining a first correspondence between a feature vector, a weight of a target optimal neuron, and a vibration feature life indicator;

[0024] Based on the eigenvector, the weight of the target optimal neuron and the first correspondence, the vibration characteristic life index of the main bearing is determined.

[0025] In one embodiment, the performance parameters at least include the rated dynamic load of the main bearing, the actual dynamic load, the main bearing life index, the crankshaft speed, the main bearing reliability correction factor, the main bearing life correction factor, the main bearing material correction factor and the lubricating oil property correction factor. The steps of obtaining the performance parameters of the main bearing and establishing the physical model for predicting the life of the main bearing include:

[0026] Based on the second corresponding relationship between the main bearing rated dynamic load, actual dynamic load, main bearing life index, crankshaft speed, main bearing reliability correction factor, main bearing life correction factor, main bearing material correction factor, lubricating oil characteristic correction factor and the physical model life index, the main bearing life prediction physical model is determined.

[0027] In one embodiment, the step of determining the comprehensive evaluation index of the life of the main bearing based on the vibration characteristic life index and the physical model life index of the main bearing includes:

[0028] The product of the physical model life index and the safety threshold weight is used as the first index threshold, and the product of the physical model life index and the health threshold weight is used as the second index threshold;

[0029] The sum of the weighted value of the vibration characteristic life index and the weighted value of the physical model life index is taken as the comprehensive weighted value of the index, the weighted value of the vibration characteristic life index is the product of the vibration characteristic life index and the corresponding weight, and the weighted value of the physical model life index is the product of the physical model life index and the corresponding weight;

[0030] Based on the first indicator threshold, the second indicator threshold and the comprehensive weighted value of the indicators, a corresponding decision strategy is determined, and based on the decision strategy, a comprehensive evaluation index of the life of the main bearing is determined.

[0031] In one embodiment, based on the first index threshold, the second index threshold and the index comprehensive weighted value, a corresponding decision strategy is determined. Based on the decision strategy, the step of determining the comprehensive evaluation index of the life of the main bearing includes:

[0032] When the vibration characteristic life index is less than the first index threshold, determining the decision strategy to be the smaller strategy;

[0033] Based on the strategy of taking the smallest value, the minimum value between the vibration characteristic life index and the physical model life index is taken as the comprehensive life evaluation index.

[0034] In one embodiment, based on the first index threshold, the second index threshold and the index comprehensive weighted value, a corresponding decision strategy is determined. Based on the decision strategy, the step of determining the comprehensive evaluation index of the life of the main bearing includes:

[0035] When the vibration characteristic life index is greater than or equal to the first index threshold and the comprehensive weighted value of the index is greater than the second index threshold, determining the decision strategy to be a weighted average strategy;

[0036] Based on the weighted average strategy, the third corresponding relationship between the vibration characteristic life index, the physical model life index, the weight factor and the comprehensive life evaluation index is obtained;

[0037] Based on the vibration characteristic life index, the physical model life index, the weight factor of the vibration characteristic life index, the weight factor of the physical model life index and the third corresponding relationship, a comprehensive life evaluation index is determined.

[0038] In one embodiment, based on the first index threshold, the second index threshold and the index comprehensive weighted value, a corresponding decision strategy is determined. Based on the decision strategy, the step of determining the comprehensive evaluation index of the life of the main bearing includes:

[0039] When the vibration characteristic life index is greater than or equal to the first index threshold and the comprehensive weighted value of the index is less than or equal to the second index threshold, determining that the decision strategy is a dynamic adjustment strategy;

[0040] Based on the dynamic adjustment strategy, a fourth corresponding relationship among the adjustment coefficient, the vibration characteristic life index, the physical model life index and the comprehensive life evaluation index is obtained;

[0041] Based on the vibration characteristic life index, the physical model life index, the adjustment coefficient and the fourth corresponding relationship, a comprehensive life evaluation index is determined.

[0042] In addition, to achieve the above-mentioned purpose, the present application also proposes a life prediction device for a main bearing of an engine crankshaft, the life prediction device for a main bearing of an engine crankshaft comprising:

[0043] The prediction module is used to obtain the vibration signals of the engine crankshaft main bearing at different times during the whole life cycle, perform one-dimensional deep feature extraction on the segmented vibration signals, and obtain the feature vector corresponding to the vibration signal;

[0044] The prediction module is also used to input the feature vector into the SOM network for training, and determine the target best neuron and the weight of the target best neuron among the neurons of the SOM network;

[0045] The prediction module is also used to calculate the vibration characteristic life index of the main bearing based on the weight of the feature vector and the target optimal neuron;

[0046] The prediction module is also used to obtain the performance parameters of the main bearing and establish a physical model for predicting the life of the main bearing;

[0047] The prediction module is also used to calculate the physical model life index of the main bearing based on the physical model of the main bearing life prediction;

[0048] A decision module, used to determine a comprehensive evaluation index of the life of the main bearing based on the vibration characteristic life index and the physical model life index of the main bearing;

[0049] The decision module is also used to generate a main bearing life prediction curve based on the comprehensive evaluation index of the main bearing life.

[0050] In addition, to achieve the above-mentioned purpose, the present application also proposes an engine crankshaft main bearing life prediction device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the engine crankshaft main bearing life prediction method as described above.

[0051] In addition, to achieve the above-mentioned purpose, the present invention also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the engine crankshaft main bearing life prediction method as described above are implemented.

[0052] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the engine crankshaft main bearing life prediction method as described above are implemented.

[0053] The present application provides a method for predicting the life of an engine crankshaft main bearing, which includes obtaining vibration signals at different times during the entire life cycle of the engine crankshaft main bearing, performing one-dimensional deep feature extraction on the segmented vibration signal, and obtaining a feature vector corresponding to the vibration signal; inputting the feature vector into a SOM network for training, and determining a target optimal neuron and a weight of the target optimal neuron in the neurons of the SOM network; calculating a vibration characteristic life index of the main bearing based on the feature vector and the weight of the target optimal neuron; obtaining performance parameters of the main bearing, and establishing a physical model for predicting the life of the main bearing; calculating a physical model life index of the main bearing based on the physical model for predicting the life of the main bearing; determining a comprehensive evaluation index of the life of the main bearing based on the vibration characteristic life index and the physical model life index of the main bearing; and generating a main bearing life prediction curve based on the comprehensive evaluation index of the life of the main bearing. This application processes the collected vibration signal data of the crankshaft main bearing based on a deep learning model, extracts the main feature vectors reflecting the degradation state of the equipment, and then integrates the effective features to obtain the vibration characteristic life index that can reflect the health of the crankshaft main bearing, and based on the physical model of the crankshaft main bearing, combines the relevant performance parameters to calculate the physical model life index, and then combines the vibration characteristic life index obtained by deep learning with the physical model life index obtained by material mechanics to obtain a comprehensive evaluation index of the life of the crankshaft main bearing, and then constructs a life prediction curve for the crankshaft main bearing to achieve the life prediction of the crankshaft main bearing. The deep learning model can capture complex nonlinear patterns, while the physical model provides an intuitive understanding of the equipment status, effectively combining the advantages of deep learning and physical models to improve the accuracy and reliability of the prediction, and solves the technical problem of low accuracy in the prediction of the life of the engine crankshaft main bearing. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0055] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0056] Figure 1 This is a flow chart of Embodiment 1 of the method for predicting the life of the main bearing of the engine crankshaft of the present application;

[0057] Figure 2 A schematic diagram of the coding network structure of the engine crankshaft main bearing life prediction method provided in Example 1 of the present application;

[0058] Figure 3A schematic diagram of a decoding network structure of the engine crankshaft main bearing life prediction method provided in Example 1 of the present application;

[0059] Figure 4 This is a flow chart of Embodiment 2 of the method for predicting the life of the main bearing of the engine crankshaft of the present application;

[0060] Figure 5 A schematic diagram of a simplified process of a method for predicting the life of an engine crankshaft main bearing provided in Embodiment 2 of the present application;

[0061] Figure 6 This is a schematic diagram of the module structure of the engine crankshaft main bearing life prediction device according to an embodiment of the present application;

[0062] Figure 7 Schematic diagram of the equipment structure of the hardware operating environment involved in the engine crankshaft main bearing life prediction method in the embodiment of the present application.

[0063] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0064] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0065] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0066] The main solution of the embodiment of the present application is: obtaining vibration signals at different times during the entire life cycle of the engine crankshaft main bearing, performing one-dimensional deep feature extraction on the segmented vibration signal, and obtaining a feature vector corresponding to the vibration signal; inputting the feature vector into the SOM network for training, and determining the target optimal neuron and the weight of the target optimal neuron in the neurons of the SOM network; calculating the vibration characteristic life index of the main bearing based on the feature vector and the weight of the target optimal neuron; obtaining the performance parameters of the main bearing, and establishing a physical model for main bearing life prediction; calculating the physical model life index of the main bearing based on the main bearing life prediction physical model; determining the comprehensive evaluation index of the main bearing life based on the vibration characteristic life index and the physical model life index of the main bearing; and generating a main bearing life prediction curve based on the comprehensive evaluation index of the main bearing life.

[0067] The present application provides a solution, which processes the collected vibration signal data of the crankshaft main bearing based on a deep learning model, extracts the main feature vector reflecting the degradation state of the equipment, then integrates the effective features to obtain the vibration characteristic life index that can reflect the health of the crankshaft main bearing, and calculates the physical model life index based on the physical model of the crankshaft main bearing in combination with relevant performance parameters, then combines the vibration characteristic life index obtained by deep learning with the physical model life index obtained by material mechanics to obtain a comprehensive evaluation index of the life of the crankshaft main bearing, and then constructs a life prediction curve for the crankshaft main bearing to achieve life prediction of the crankshaft main bearing. The deep learning model can capture complex nonlinear patterns, while the physical model provides an intuitive understanding of the equipment status, effectively combining the advantages of deep learning and physical models, improving the accuracy and reliability of the prediction, and solving the technical problem of low accuracy in life prediction of the engine crankshaft main bearing.

[0068] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, an engine crankshaft main bearing life prediction device, etc., and this embodiment does not specifically limit this. The following takes the engine crankshaft main bearing life prediction device as an example to illustrate this embodiment and the following embodiments.

[0069] The present application embodiment provides a method for predicting the life of an engine crankshaft main bearing, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the engine crankshaft main bearing life prediction method of the present application.

[0070] In this embodiment, the engine crankshaft main bearing life prediction method includes steps S10 to S70:

[0071] Step S10, obtaining vibration signals at different times during the entire life cycle of the engine crankshaft main bearing, performing one-dimensional deep feature extraction on the segmented vibration signals, and obtaining feature vectors corresponding to the vibration signals;

[0072] It should be noted that this embodiment collects vibration signals at different times during the entire life cycle of the engine crankshaft main bearing for crankshaft main bearing life prediction. The vibration signal data may be angular velocity data or angular acceleration data obtained by a sensor, which may be flexibly adjusted according to actual needs and is not specifically limited thereto.

[0073] In addition, it should be noted that the collected vibration signals need to be processed in segments, and the length of the segments can be set according to actual needs. There is no specific limitation on this. In this embodiment, the collected vibration signals are segmented according to the length of 2560.

[0074] It can be understood that the feature vector is a vector composed of the features of the vibration signal. The segmented vibration signal is subjected to one-dimensional deep feature extraction to obtain the corresponding feature vector

[0075] In a feasible implementation, the step of performing one-dimensional deep feature extraction on the segmented vibration signal to obtain a feature vector corresponding to the vibration signal includes: performing one-dimensional deep feature extraction on the segmented vibration signal through a one-dimensional deep self-encoding feature unit to obtain a feature vector corresponding to the vibration signal.

[0076] It should be noted that the present embodiment performs one-dimensional deep feature extraction through a one-dimensional deep self-encoding feature unit. The one-dimensional deep self-encoding feature unit includes an encoding network and a decoding network, the encoding network includes a convolution layer and a pooling layer, and the decoding network includes an upsampling layer and a convolution layer, wherein the number of convolution layers and pooling layers in the encoding network can be set according to actual needs, and the number of upsampling layers and convolution layers in the decoding network can be set according to actual needs, and there is no specific limitation on this.

[0077] For example, reference Figure 2 , 6 convolutional layers (Conv1~Conv6) and 4 pooling layers (Pool1~Pool4) are set in the encoding module, refer to Figure 3 , 6 convolutional layers (Conv6~Conv12) and 4 upsampling layers (Upsample1~Upsample4) are set in the decoding module.

[0078] Step S20, inputting the feature vector into the SOM network for training, and determining the target best neuron and the weight of the target best neuron among the neurons of the SOM network;

[0079] In a feasible implementation, step S20 may include steps S201 to S205:

[0080] Step S201, initializing the SOM network and initializing the weights of neurons in the SOM network;

[0081] It should be noted that the SOM network consists of an input layer and a competition layer. Multiple neurons are usually set in the competition layer. When initializing the SOM network, the number of neurons in the competition layer needs to be determined using the following calculation formula:

[0082]

[0083] In the formula, m represents the number of neurons in the competition layer, and n represents the input feature vector x h The number of features, k represents the impact factor, which is set according to actual conditions. In this embodiment, the impact factor k is 5.

[0084] Next, initialize the weights of each neuron

[0085] Step S202, inputting the feature vector into the SOM network, and calculating the distance between the neuron and the feature vector;

[0086] It is understandable that the distance between each neuron and the eigenvector needs to be calculated. The calculation relationship is as follows:

[0087] D i =‖x h -w l ‖2

[0088] Where D i represents the distance between the feature vector and the i-th neuron, x h represents the eigenvector, w l Represents the weight of the neuron. The distance between each neuron and the feature vector is calculated using the above calculation formula.

[0089] Step S203, taking the neuron with the smallest distance as the best neuron, and taking the neurons other than the best neuron as surrounding neurons;

[0090] It can be understood that all calculated distances between neurons and feature vectors are sorted, and the neuron with the smallest distance is taken as the best neuron, and the remaining neurons are peripheral neurons.

[0091] Step S204, adjusting the weights of the best neuron and the surrounding neurons so that the surrounding neurons are closer to the best neuron;

[0092] It can be understood that after the best neuron and the surrounding neurons are divided, the weights of the best neuron and the surrounding neurons are adjusted to bring the surrounding neurons closer to the best neuron.

[0093] Step S205, updating the distance between the neuron and the feature vector, and re-determining the best neuron and surrounding neurons, and continuously iterating until the preset conditions are met, to obtain the target best neuron and the weight of the target best neuron.

[0094] It should be noted that, as the surrounding neurons move closer to the best neuron, it is necessary to recalculate the distance between the neuron and the feature vector, re-determine the best neuron and the surrounding neurons according to the new distance, and adjust the weights of the best neuron and the surrounding neurons again, and then iterate continuously until the preset condition is reached. The best neuron obtained at this time is the best neuron required in the end, that is, the target best neuron. At the same time, it is necessary to determine the weight of the target best neuron so as to use the weight of the target best neuron for subsequent calculations. Among them, the preset condition can be reaching the maximum number of iterations, which is not specifically limited.

[0095] Step S30, calculating the vibration characteristic life index of the main bearing based on the characteristic vector and the weight of the target optimal neuron;

[0096] It should be noted that the vibration characteristic life index is based on vibration signal data processed by deep learning. It extracts the main feature vectors reflecting the degradation state of the equipment and focuses on extracting features from the actual monitored vibration data. It can be used to evaluate the life of the crankshaft main bearing.

[0097] In a feasible implementation, step S30 may include: obtaining a first correspondence between a feature vector, a weight and a vibration characteristic life index; and determining the vibration characteristic life index of the main bearing based on the feature vector, the weight of the target optimal neuron and the first correspondence.

[0098] It should be noted that the first corresponding relationship between the feature vector, the weight of the target optimal neuron and the vibration characteristic life index, that is, the calculation relationship of the vibration characteristic life index, is as follows:

[0099] L1=‖x h -w min ‖2

[0100] Where L1 represents the vibration characteristic life index, x h represents the eigenvector, w min represents the weight of the target optimal neuron.

[0101] It can be understood that the characteristic vector and the weight of the target optimal neuron are substituted into the above-mentioned first corresponding relationship to calculate the vibration characteristic life index of the main bearing.

[0102] Step S40, obtaining the performance parameters of the main bearing and establishing a physical model for predicting the life of the main bearing;

[0103] It should be noted that the main bearing life prediction physical model is a physical model for predicting the main bearing life, and this embodiment uses performance parameters to construct the main bearing life prediction physical model. The selected main bearing performance parameters at least include the main bearing rated dynamic load, actual dynamic load, main bearing life index, crankshaft speed, main bearing reliability correction factor, main bearing life correction factor, main bearing material correction factor and lubricant property correction factor, and these parameters can provide comprehensive input data for the life prediction and performance evaluation of the main bearing.

[0104] In a feasible implementation, step S40 may include: determining the main bearing life prediction physical model based on the second correspondence between the main bearing rated dynamic load, actual dynamic load, main bearing life index, crankshaft speed, main bearing reliability correction factor, main bearing life correction factor, main bearing material correction factor, lubricating oil characteristic correction factor and the physical model life index.

[0105] It should be noted that the second corresponding relationship between the main bearing rated dynamic load, actual dynamic load, main bearing life index, crankshaft speed, main bearing reliability correction factor, main bearing life correction factor, main bearing material correction factor, lubricant property correction factor and the physical model life index, that is, the calculation relationship satisfied by the physical model life index, is as follows:

[0106]

[0107] Where L2 represents the life index of the physical model in hours, C d Indicates the rated dynamic load of the main bearing, f equ represents the actual dynamic load of the main bearing, n represents the crankshaft speed, unit: revolutions per minute (r / min), ε represents the main bearing life index, β1 represents the main bearing reliability correction factor, β2 represents the main bearing life correction factor, β2 represents the main bearing material correction factor, β4 represents the lubricating oil characteristic correction factor, E represents the elastic modulus of the material, H represents the Brinell hardness of the material, s ex represents the material property index, ρ represents the density of the lubricating oil, μ represents the dynamic viscosity of the lubricating oil, and v ex Indicates the lubricating oil performance index. This embodiment uses the second corresponding relationship to construct a physical model for predicting the life of the main bearing.

[0108] In addition, it should be noted that the main bearing dynamic load rating C d It is a theoretical value determined by the bearing manufacturer based on the bearing type, bearing structure size, material characteristics and process design, indicating the maximum load that the bearing can withstand under specific conditions without permanent deformation of the bearing. It is provided in the bearing product catalog or technical documentation. equIt refers to the load that the bearing bears under actual working conditions. It can characterize the stress condition of the main bearing and affect the fatigue life of the main bearing material. The main bearing life index ε is a constant related to the characteristics of the bearing material. It is usually provided by the bearing manufacturer. For different types of rolling bearings, the value of the life index may be different. For example, the life index of ball bearings is usually 3, and the life index of roller bearings is usually 10 / 3. It depends on the type and size of the roller and is not specifically limited. The life of the bearing is also related to the crankshaft speed n, because the speed affects the friction and heat generation inside the bearing, which will affect the operating state of the main bearing, the average pressure of the main bearing and the generation of the lubricating oil film. The main bearing life correction factor β2 is obtained based on the relationship curve between the main bearing's lubricating oil dynamic viscosity ratio, lubricating oil contamination coefficient, and fatigue load limit value, and is provided by the bearing manufacturer.

[0109] Generally speaking, the main bearing reliability correction factor β1 is related to reliability and is affected by multiple factors such as bearing type, bearing geometry, material properties, etc. It is provided by the bearing manufacturer. The specific value can be referred to Table 1 (only as an example, other appropriate values ​​can also be set). Reliability 90% means that a group of sufficiently large and identical bearings are running under the same conditions, and 90% of the bearings do not experience fatigue failure. The fatigue life is expected to be reached or exceeded, also known as the rated life L 10 Correspondingly, reliability 95% means that a group of sufficiently large and identical bearings are running under the same conditions, and 95% of the bearings do not fail due to fatigue. The fatigue life is expected to be reached or exceeded, also known as the rated life L5. Usually, the rated life L is used. 10 As a measure of bearing life, define It can be used to determine the bearing size sufficient to avoid fatigue failure. For a given load, the bearing size calculated using a high reliability (e.g. 99%) and a reliability correction factor will result in the selection of an oversized bearing. The higher the reliability, the larger the bearing size is. Therefore, the reliability and reliability correction factor can be flexibly adjusted according to the actual application.

[0110] Table 1

[0111] reliability(%) Rated life <![CDATA[Reliability correction factor β1]]> 90 <![CDATA[L 10 ]]> 1 95 <![CDATA[L5]]> 0.64 96 <![CDATA[L4]]> 0.55 97 <![CDATA[L3]]> 0.47 98 <![CDATA[L2]]> 0.37 99 <![CDATA[L1]]> 0.25

[0112] It is understandable that the main bearing of the engine is usually composed of an upper and lower bearing shell. The bearing shell is divided into a two-layer structure bearing shell and a three-layer structure bearing shell. The two-layer structure bearing shell is composed of a steel back and a friction-reducing alloy layer, and the three-layer structure bearing shell is composed of a steel back, a friction-reducing alloy layer and a soft coating. Therefore, it is necessary to set the main bearing material correction coefficient β3 to characterize the influence of material properties on the life of the main bearing. The elastic modulus E of the material can reflect the elastic properties of the main bearing material, the Brinell hardness H of the material can reflect the surface hardness of the material and affect the wear characteristics of the main bearing, and the material property index s ex , which represents the influence parameter of main bearing material characteristics on main bearing life, needs to be set according to actual conditions.

[0113] In addition, it can be understood that the lubricant characteristic parameters include lubricant density, lubricant dynamic viscosity and lubricant performance index. The lubricant characteristic parameters affect its flow state and oil film distribution state. Common engine main bearings use pressure lubrication. During operation, a sufficient thickness of oil film is formed on the main bearing surface to reduce friction and ensure that the main bearing has good lubrication performance and service life. Therefore, it is necessary to set the lubricant characteristic correction factor to characterize the effect of the lubricant characteristic parameters on the main bearing life. Lubricant density ρ affects the stability of the main bearing fluid dynamic lubricant film, lubricant dynamic viscosity μ affects the thickness and load-bearing capacity of the lubricant film, and lubricant performance index v ex , which indicates the influence of the physical properties of lubricating oil on the life of the main bearing, needs to be set according to actual conditions.

[0114] It should be understood that the actual bearing life is affected by a combination of factors, such as bearing type, lubrication conditions, bearing installation accuracy, fatigue characteristics of bearing materials, dynamic characteristics of the crankshaft, ignition sequence of multi-cylinder engines, bearing geometry, lubrication conditions, bearing clearance, material properties, environmental factors, etc. Therefore, in practical applications, when constructing a physical model for predicting the life of the main bearing, factors such as the fatigue life, wear life and lubrication life of the bearing can also be considered. The calculation relationship used in the model can be flexibly adjusted according to actual conditions to ensure the reliability of the main bearing design and performance optimization.

[0115] Step S50, calculating the physical model life index of the main bearing based on the main bearing life prediction physical model;

[0116] It should be noted that the physical model life index focuses on the main bearing performance parameters, and uses the main bearing life prediction physical model for calculation to evaluate the life of the crankshaft main bearing.

[0117] It is understandable that, in specific implementation, the actual values ​​of the main bearing performance parameters are substituted into the main bearing life prediction physical model to calculate the physical model life index of the main bearing.

[0118] Step S60, determining a comprehensive evaluation index of the life of the main bearing based on the vibration characteristic life index and the physical model life index of the main bearing;

[0119] It should be noted that, in this embodiment, the vibration characteristic life index and the physical model life index are combined according to actual conditions to obtain a comprehensive life evaluation index.

[0120] It is understandable that the deep learning model can capture complex nonlinear patterns, while the physical model provides an intuitive understanding of the equipment status. The vibration signal analysis model based on deep learning and the safety factor based on the physical model in this embodiment can significantly improve the accuracy of the prediction. In addition, since this embodiment is not completely dependent on a single data set or model assumption, this embodiment can be generalized to new or unknown situations and has better generalization ability.

[0121] Step S70, generating a main bearing life prediction curve based on the comprehensive evaluation index of the main bearing life.

[0122] It should be noted that the final life prediction result is obtained based on the comprehensive evaluation index of the main bearing life.

[0123] The present embodiment provides a method for predicting the life of an engine crankshaft main bearing, which obtains vibration signals at different times during the entire life cycle of the engine crankshaft main bearing, performs one-dimensional deep feature extraction on the segmented vibration signal, and obtains a feature vector corresponding to the vibration signal; inputs the feature vector into a SOM network for training, and determines a target optimal neuron and a weight of the target optimal neuron in the neurons of the SOM network; calculates a vibration characteristic life index of the main bearing based on the feature vector and the weight of the target optimal neuron; obtains performance parameters of the main bearing, and establishes a main bearing life prediction physical model; calculates a physical model life index of the main bearing based on the main bearing life prediction physical model; determines a comprehensive life evaluation index of the main bearing based on the vibration characteristic life index and the physical model life index of the main bearing; and generates a main bearing life prediction curve based on the comprehensive life evaluation index of the main bearing. The collected vibration signal data of the crankshaft main bearing is processed based on the deep learning model, and the main feature vectors reflecting the degradation state of the equipment are extracted. Then the effective features are integrated to obtain the vibration characteristic life index that can reflect the health of the crankshaft main bearing. Based on the physical model of the crankshaft main bearing and combined with relevant performance parameters, the physical model life index is calculated. Then, the vibration characteristic life index obtained by deep learning and the physical model life index obtained by material mechanics are combined to obtain a comprehensive evaluation index of the crankshaft main bearing life, and then the crankshaft main bearing life prediction curve is constructed to realize the crankshaft main bearing life prediction. The deep learning model can capture complex nonlinear patterns, while the physical model provides an intuitive understanding of the equipment status, effectively combining the advantages of deep learning and physical models.

[0124] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 4 , step S60 may include steps S601 to S603:

[0125] Step S601, taking the product of the physical model life index and the safety threshold weight as the first index threshold, and taking the product of the physical model life index and the health threshold weight as the second index threshold;

[0126] It should be noted that the first indicator threshold and the second indicator threshold are two thresholds set in this embodiment. The first indicator threshold is the product of the physical model life indicator L2 and the safety threshold weight α, that is, L2*α, and the second indicator threshold is the product of the physical model life indicator L2 and the health threshold weight β, that is, L2*β. Among them, the specific values ​​of the safety threshold weight and the health threshold weight need to be determined according to the actual situation, and there is no specific limitation on this.

[0127] Step S602, taking the sum of the vibration characteristic life index weighted value and the physical model life index weighted value as the comprehensive index weighted value;

[0128] It should be noted that the weighted value of the vibration characteristic life index is the value obtained after the vibration characteristic life index is weighted. In specific calculations, the weighted value of the vibration characteristic life index is the product of the vibration characteristic life index L1 and the corresponding weight γ1. The weighted value of the physical model life index is the value obtained after the physical model life index is weighted. In specific calculations, the weighted value of the physical model life index is the product of the physical model life index L2 and the corresponding weight γ2.

[0129] It can be understood that by adding the weighted value of the vibration characteristic life index and the weighted value of the physical model life index, the comprehensive weighted value of the index can be calculated, that is, the comprehensive weighted value of the index is equal to γ1*L1+γ2*L2.

[0130] Step S603, based on the first index threshold, the second index threshold and the index comprehensive weighted value, determine the corresponding decision strategy, and based on the decision strategy, determine the comprehensive evaluation index of the life of the main bearing.

[0131] It should be noted that this embodiment sets three key values, namely: the first indicator threshold, the second indicator threshold and the comprehensive weighted value of the indicator. Three value ranges are set using these key values. Different value ranges use different methods to calculate the comprehensive evaluation index of life. The decision strategy is the method used to calculate the comprehensive evaluation index of life.

[0132] In a feasible implementation, step S603 may include steps A11 to A12:

[0133] Step A11, when the vibration characteristic life index is less than the first index threshold, determining the decision strategy to be the smaller strategy;

[0134] It should be noted that the smaller strategy refers to selecting the smaller indicator between the vibration characteristic life indicator and the physical model life indicator as the final comprehensive life evaluation indicator. Under this strategy, the most conservative approach is taken to ensure the safety of the equipment under extreme conditions. By comparing the life evaluation indicators predicted by the vibration signal analysis model and the physical analysis model, the smaller value of the two is selected as the final comprehensive life evaluation indicator. This strategy is suitable for application scenarios with extremely high safety requirements to avoid overly optimistic life predictions.

[0135] It can be understood that if the vibration characteristic life index is less than the first index threshold, that is, L1<L2*α, the smaller strategy is adopted to determine the comprehensive evaluation index of life.

[0136] Step A12: Based on the minimum strategy, the minimum value of the vibration characteristic life index and the physical model life index is used as the comprehensive life evaluation index.

[0137] It can be understood that according to the strategy of taking the smaller value, find the indicator with a smaller value between the vibration characteristic life index and the physical model life index, that is, find the minimum value, and use this minimum value as the comprehensive life evaluation index, as shown below:

[0138] LPI=min(L1,L2)

[0139] Where LPI represents the comprehensive evaluation index of life, L1 represents the vibration characteristic life index, and L2 represents the physical model life index.

[0140] In another feasible implementation, step S603 may include steps B11 to B12:

[0141] Step B11, when the vibration characteristic life index is greater than or equal to the first index threshold and the comprehensive weighted value of the index is greater than the second index threshold, determining the decision strategy to be a weighted average strategy;

[0142] It should be noted that the weighted average strategy refers to the calculation of the weighted average life by assigning different weight factors to the indicators calculated by the two different prediction methods. The weight factor allocation is based on the importance and reliability evaluation of each prediction method. Therefore, the combined impact of the two prediction methods can be considered to ensure that the prediction results of the two methods can be weighted according to their reliability and importance, so as to make a more balanced and comprehensive prediction of the life of the main bearing.

[0143] It can be understood that if the vibration characteristic life index is greater than or equal to the first index threshold and the comprehensive weighted value of the index is greater than the second index threshold, that is, L1≥L2*α and γ1*L1+γ2*L2>L2*β, the weighted average strategy is used to calculate the comprehensive evaluation index of life.

[0144] Step B12, based on the weighted average strategy, obtain the third corresponding relationship between the vibration characteristic life index, the physical model life index, the weight factor and the comprehensive life evaluation index, and determine the comprehensive life evaluation index based on the vibration characteristic life index, the physical model life index, the weight factor of the vibration characteristic life index, the weight factor of the physical model life index and the third corresponding relationship.

[0145] It should be noted that the weight factor is the weight of the vibration characteristic life index / physical model life index in the process of calculating the comprehensive evaluation index of life. The third corresponding relationship between the vibration characteristic life index, the physical model life index, the weight factor and the comprehensive evaluation index of life is the calculation relationship of the comprehensive evaluation index of life when the weighted average strategy is adopted, as shown below:

[0146] LPI=w1·L1+w2·L2

[0147] Wherein, LPI represents the comprehensive evaluation index of life, L1 represents the vibration characteristic life index, L2 represents the physical model life index, w1 represents the weight factor of the vibration characteristic life index, w2 represents the weight factor of the physical model life index, and w1+w2=1.

[0148] It can be understood that the vibration characteristic life index, the physical model life index, the weight factor of the vibration characteristic life index, and the weight factor of the physical model life index are substituted into the above third corresponding relationship to calculate the comprehensive life evaluation index.

[0149] In another feasible implementation, step S603 may include steps C11 to C12:

[0150] Step C11, when the vibration characteristic life index is greater than or equal to the first index threshold and the index comprehensive weighted value is less than or equal to the second index threshold, determining that the decision strategy is a dynamic adjustment strategy;

[0151] It should be noted that the dynamic adjustment strategy refers to the calculation of the comprehensive evaluation index of life by introducing an adjustment coefficient. This strategy comprehensively considers the life prediction of the two prediction methods and introduces an adjustment coefficient to balance the influence of the two methods, thus realizing dynamic and adaptive prediction of the main bearing life.

[0152] It can be understood that if the vibration characteristic life index is greater than or equal to the first index threshold and the comprehensive weighted value of the index is less than or equal to the second index threshold, that is, L1≥L2*α and γ1*L1+γ2*L2≤L2*β, the dynamic adjustment strategy is adopted.

[0153] Step C12, based on the dynamic adjustment strategy, obtain the fourth corresponding relationship between the adjustment coefficient, the vibration characteristic life index, the physical model life index and the comprehensive life evaluation index, and determine the comprehensive life evaluation index based on the vibration characteristic life index, the physical model life index, the adjustment coefficient and the fourth corresponding relationship.

[0154] It should be noted that the fourth corresponding relationship between the adjustment coefficient, the vibration characteristic life index, the physical model life index and the comprehensive life evaluation index is the calculation relationship of the comprehensive life evaluation index when the dynamic adjustment strategy is adopted, as shown below:

[0155]

[0156] Wherein, LPI represents the comprehensive evaluation index of life, L1 represents the vibration characteristic life index, L2 represents the physical model life index, and γ represents the adjustment coefficient. Substitute the vibration characteristic life index, the physical model life index, and the adjustment coefficient into the fourth corresponding relationship to calculate the comprehensive evaluation index of life.

[0157] It can be understood that the adjustment coefficient γ reflects the real-time feedback of the current main bearing operating status or the impact of external environmental factors on the life, and can be dynamically adjusted according to the main bearing operating status and external conditions. When γ>1, it means that the current conditions are conducive to extending the life, and when γ<1, it means that the current conditions may shorten the life.

[0158] In summary, this embodiment sets three decision strategies: the minimum strategy, the weighted average strategy, and the dynamic adjustment strategy. The strategy selection method is as follows:

[0159]

[0160] Wherein, L1 represents the vibration characteristic life index, L2 represents the physical model life index, α represents the safety threshold weight, β represents the health threshold weight, γ1 represents the weight of the vibration characteristic life index, and γ2 represents the weight of the physical model life index.

[0161] The present embodiment provides a method for predicting the life of an engine crankshaft main bearing. The collected vibration signal data of the crankshaft main bearing is processed based on a deep learning model to extract the main feature vector reflecting the degradation state of the equipment. Then, the effective features are integrated to obtain a vibration characteristic life index that can reflect the health of the crankshaft main bearing. Based on the physical model of the crankshaft main bearing and in combination with relevant performance parameters, the physical model life index is calculated. Then, the vibration characteristic life index obtained by deep learning and the physical model life index obtained by material mechanics are combined to obtain a comprehensive evaluation index of the life of the crankshaft main bearing. Then, a life prediction curve of the crankshaft main bearing is constructed to realize the life prediction of the crankshaft main bearing. The deep learning model can capture complex nonlinear patterns, while the physical model provides an intuitive understanding of the equipment status, effectively combining the advantages of deep learning and physical models.

[0162] For example, in order to help understand the implementation process of the engine crankshaft main bearing life prediction method obtained by combining this embodiment with the above-mentioned embodiment 2, please refer to Figure 5 , Figure 5 A brief flow chart of a method for predicting the life of an engine crankshaft main bearing is provided, specifically:

[0163] Step 1: Data preparation, collecting vibration signals at different times during the entire life cycle of the crankshaft main bearing;

[0164] Step 2: One-dimensional deep feature extraction: input the segmented vibration signal into a one-dimensional deep self-encoding feature unit for training to obtain a feature vector. The one-dimensional deep self-encoding feature unit consists of two parts: an encoding network and a decoding network.

[0165] Step 3: The feature vector is input into the SOM network for training;

[0166] Step 4: Calculate the life evaluation index 1 of the crankshaft main bearing, which is recorded as the vibration characteristic life index of the main bearing;

[0167] Step 5: Obtain the performance parameters of the crankshaft main bearing and establish a physical model for evaluating the main bearing life prediction;

[0168] Step 6: Calculate the life evaluation index 2 of the crankshaft main bearing, which is recorded as the physical model life index of the main bearing;

[0169] Step 7: The crankshaft main bearing life decision unit combines the vibration characteristic life index and the physical model life index of the main bearing according to the actual situation to calculate the comprehensive evaluation index of life;

[0170] Step 8: Obtain the crankshaft main bearing life degradation curve.

[0171] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the engine crankshaft main bearing life prediction method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0172] This application also provides a device for predicting the life of the main bearing of the engine crankshaft. Figure 6 , the engine crankshaft main bearing life prediction device includes:

[0173] The prediction module 10 is used to obtain the vibration signals of the engine crankshaft main bearing at different times during the whole life cycle, perform one-dimensional deep feature extraction on the segmented vibration signals, and obtain the feature vector corresponding to the vibration signal;

[0174] The prediction module 10 is further used to input the feature vector into the SOM network for training, and determine the target best neuron and the weight of the target best neuron among the neurons of the SOM network;

[0175] The prediction module 10 is further used to calculate the vibration characteristic life index of the main bearing based on the characteristic vector and the weight of the target optimal neuron;

[0176] The prediction module 10 is also used to obtain the performance parameters of the main bearing and establish a physical model for predicting the life of the main bearing;

[0177] The prediction module 10 is further used to calculate the physical model life index of the main bearing based on the physical model of the main bearing life prediction;

[0178] A decision module 20 is used to determine a comprehensive evaluation index of the life of the main bearing based on the vibration characteristic life index and the physical model life index of the main bearing;

[0179] The decision module 20 is also used to generate a main bearing life prediction curve based on the comprehensive evaluation index of the main bearing life.

[0180] In a feasible implementation, the prediction module 10 is also used to perform one-dimensional deep feature extraction on the segmented vibration signal through a one-dimensional deep self-encoding feature unit to obtain a feature vector corresponding to the vibration signal. The one-dimensional deep self-encoding feature unit includes an encoding network and a decoding network. The encoding network includes a convolution layer and a pooling layer, and the decoding network includes an upsampling layer and a convolution layer.

[0181] In a feasible implementation, the prediction module 10 is also used to initialize the SOM network and initialize the weights of neurons in the SOM network;

[0182] Input the feature vector into the SOM network and calculate the distance between the neuron and the feature vector;

[0183] The neuron with the smallest distance is regarded as the best neuron, and the neurons other than the best neuron are regarded as surrounding neurons;

[0184] Adjust the weights of the best neuron and surrounding neurons to make the surrounding neurons move closer to the best neuron;

[0185] Update the distance between the neuron and the feature vector, and re-determine the best neuron and surrounding neurons, and continue to iterate until the preset conditions are met to obtain the target best neuron and the weight of the target best neuron.

[0186] In a feasible implementation, the prediction module 10 is further used to obtain a first corresponding relationship between a feature vector, a weight of a target optimal neuron, and a vibration feature life index;

[0187] Based on the eigenvector, the weight of the target optimal neuron and the first correspondence, the vibration characteristic life index of the main bearing is determined.

[0188] In a feasible implementation manner, the performance parameters include at least the rated dynamic load of the main bearing, the actual dynamic load, the main bearing life index, the crankshaft speed, the main bearing reliability correction factor, the main bearing life correction factor, the main bearing material correction factor and the lubricating oil characteristic correction factor. The prediction module 10 is also used to determine the main bearing life prediction physical model based on the second corresponding relationship between the main bearing rated dynamic load, the actual dynamic load, the main bearing life index, the crankshaft speed, the main bearing reliability correction factor, the main bearing life correction factor, the main bearing material correction factor, the lubricating oil characteristic correction factor and the physical model life index.

[0189] In a feasible implementation manner, the decision module 20 is further configured to use the product of the physical model life index and the safety threshold weight as the first index threshold, and use the product of the physical model life index and the health threshold weight as the second index threshold;

[0190] The sum of the weighted value of the vibration characteristic life index and the weighted value of the physical model life index is taken as the comprehensive weighted value of the index, the weighted value of the vibration characteristic life index is the product of the vibration characteristic life index and the corresponding weight, and the weighted value of the physical model life index is the product of the physical model life index and the corresponding weight;

[0191] Based on the first indicator threshold, the second indicator threshold and the comprehensive weighted value of the indicators, a corresponding decision strategy is determined, and based on the decision strategy, a comprehensive evaluation index of the life of the main bearing is determined.

[0192] In a feasible implementation manner, the decision module 20 is further configured to determine that the decision strategy is a smaller strategy when the vibration characteristic life index is less than the first index threshold;

[0193] Based on the strategy of taking the smallest value, the minimum value between the vibration characteristic life index and the physical model life index is taken as the comprehensive life evaluation index.

[0194] In a feasible implementation manner, the decision module 20 is further configured to determine that the decision strategy is a weighted average strategy when the vibration characteristic life index is greater than or equal to a first index threshold and the index comprehensive weighted value is greater than a second index threshold;

[0195] Based on the weighted average strategy, the third corresponding relationship between the vibration characteristic life index, the physical model life index, the weight factor and the comprehensive life evaluation index is obtained;

[0196] Based on the vibration characteristic life index, the physical model life index, the weight factor of the vibration characteristic life index, the weight factor of the physical model life index and the third corresponding relationship, a comprehensive life evaluation index is determined.

[0197] In a feasible implementation manner, the decision module 20 is further configured to determine that the decision strategy is a dynamic adjustment strategy when the vibration characteristic life index is greater than or equal to the first index threshold and the index comprehensive weighted value is less than or equal to the second index threshold;

[0198] Based on the dynamic adjustment strategy, a fourth corresponding relationship among the adjustment coefficient, the vibration characteristic life index, the physical model life index and the comprehensive life evaluation index is obtained;

[0199] Based on the vibration characteristic life index, the physical model life index, the adjustment coefficient and the fourth corresponding relationship, a comprehensive life evaluation index is determined.

[0200] The engine crankshaft main bearing life prediction device provided by the present application adopts the engine crankshaft main bearing life prediction method in the above embodiment, which can solve the technical problem of low accuracy in engine crankshaft main bearing life prediction. Compared with the prior art, the engine crankshaft main bearing life prediction device provided by the present application has the same beneficial effects as the engine crankshaft main bearing life prediction method provided by the above embodiment, and other technical features in the engine crankshaft main bearing life prediction device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0201] The present application provides an engine crankshaft main bearing life prediction device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the engine crankshaft main bearing life prediction method in the above-mentioned embodiment one.

[0202] Reference below Figure 7, which shows a schematic diagram of the structure of an engine crankshaft main bearing life prediction device suitable for implementing the embodiment of the present application. The engine crankshaft main bearing life prediction device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 7 The engine crankshaft main bearing life prediction device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0203] like Figure 7 As shown, the engine crankshaft main bearing life prediction device may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. Various programs and data required for the operation of the engine crankshaft main bearing life prediction device are also stored in the RAM 1004. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 may allow the engine crankshaft main bearing life prediction device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows an engine crankshaft main bearing life prediction device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0204] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0205] The engine crankshaft main bearing life prediction device provided by the present application adopts the engine crankshaft main bearing life prediction method in the above embodiment, which can solve the technical problem of low accuracy in engine crankshaft main bearing life prediction. Compared with the prior art, the engine crankshaft main bearing life prediction device provided by the present application has the same beneficial effects as the engine crankshaft main bearing life prediction method provided by the above embodiment, and other technical features in the engine crankshaft main bearing life prediction device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0206] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0207] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0208] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the engine crankshaft main bearing life prediction method in the above-mentioned embodiment.

[0209] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0210] The computer-readable storage medium may be included in the engine crankshaft main bearing life prediction device; or may exist independently without being assembled into the engine crankshaft main bearing life prediction device.

[0211] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the engine crankshaft main bearing life prediction device, the engine crankshaft main bearing life prediction device is enabled to: obtain vibration signals at different times in the entire life cycle of the engine crankshaft main bearing, perform one-dimensional deep feature extraction on the segmented vibration signal, and obtain a feature vector corresponding to the vibration signal; input the feature vector into the SOM network for training, and determine the target optimal neuron and the weight of the target optimal neuron in the neurons of the SOM network; calculate the vibration characteristic life index of the main bearing based on the feature vector and the weight of the target optimal neuron; obtain the performance parameters of the main bearing and establish a main bearing life prediction physical model; calculate the physical model life index of the main bearing based on the main bearing life prediction physical model; determine the comprehensive evaluation index of the main bearing life based on the vibration characteristic life index and the physical model life index of the main bearing; and generate a main bearing life prediction curve based on the comprehensive evaluation index of the main bearing life.

[0212] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0213] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0214] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0215] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned engine crankshaft main bearing life prediction method, and can solve the technical problem of low accuracy in engine crankshaft main bearing life prediction. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the engine crankshaft main bearing life prediction method provided by the above-mentioned embodiment, and will not be repeated here.

[0216] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned engine crankshaft main bearing life prediction method when executed by a processor.

[0217] The computer program product provided by the present application can solve the technical problem of low accuracy in predicting the life of the main bearing of the engine crankshaft. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the method for predicting the life of the main bearing of the engine crankshaft provided by the above embodiment, and will not be described in detail here.

[0218] The above are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for predicting the life of an engine crankshaft main bearing, characterized in that: The method comprises: Acquire vibration signals at different times during the entire life cycle of the engine crankshaft main bearing, perform one-dimensional deep feature extraction on the segmented vibration signals, and obtain feature vectors corresponding to the vibration signals; Inputting the feature vector into a SOM network for training, and determining a target optimal neuron and a weight of the target optimal neuron among neurons of the SOM network; Calculating a vibration characteristic life index of the main bearing based on the characteristic vector and the weight of the target optimal neuron; Acquiring the performance parameters of the main bearing and establishing a physical model for predicting the life of the main bearing; Calculating a physical model life index of the main bearing based on the main bearing life prediction physical model; Determining a comprehensive evaluation index of the life of the main bearing based on the vibration characteristic life index and the physical model life index of the main bearing; Based on the comprehensive evaluation index of the life of the main bearing, a main bearing life prediction curve is generated.

2. The method according to claim 1, characterized in that The step of performing one-dimensional deep feature extraction on the segmented vibration signal to obtain a feature vector corresponding to the vibration signal comprises: A one-dimensional deep feature extraction is performed on the segmented vibration signal through a one-dimensional deep self-encoding feature unit to obtain a feature vector corresponding to the vibration signal. The one-dimensional deep self-encoding feature unit includes an encoding network and a decoding network. The encoding network includes a convolution layer and a pooling layer, and the decoding network includes an upsampling layer and a convolution layer.

3. The method according to claim 1, characterized in that The step of inputting the feature vector into the SOM network for training and determining the target optimal neuron and the weight of the target optimal neuron in the neurons of the SOM network comprises: Initializing a SOM network and initializing the weights of neurons in the SOM network; Inputting the feature vector into the SOM network, and calculating the distance between the neuron and the feature vector; The neuron with the smallest distance is regarded as the best neuron, and the neurons other than the best neuron are regarded as surrounding neurons; Adjust the weights of the best neuron and surrounding neurons to make the surrounding neurons move closer to the best neuron; The distance between the neuron and the feature vector is updated, and the best neuron and the surrounding neurons are re-determined, and the iteration is continued until the preset conditions are met to obtain the target best neuron and the weight of the target best neuron.

4. The method according to claim 1, characterized in that The step of calculating the vibration characteristic life index of the main bearing based on the characteristic vector and the weight of the target optimal neuron includes: Obtaining a first correspondence between a feature vector, a weight of a target optimal neuron, and a vibration feature life indicator; Based on the feature vector, the weight of the target optimal neuron and the first corresponding relationship, a vibration characteristic life index of the main bearing is determined.

5. The method according to claim 1, characterized in that The performance parameters at least include the rated dynamic load of the main bearing, the actual dynamic load, the main bearing life index, the crankshaft speed, the main bearing reliability correction factor, the main bearing life correction factor, the main bearing material correction factor and the lubricating oil property correction factor. The steps of obtaining the performance parameters of the main bearing and establishing a physical model for predicting the life of the main bearing include: Based on the second corresponding relationship between the main bearing rated dynamic load, actual dynamic load, main bearing life index, crankshaft speed, main bearing reliability correction factor, main bearing life correction factor, main bearing material correction factor, lubricating oil property correction factor and the physical model life index, the main bearing life prediction physical model is determined.

6. The method according to claim 1, characterized in that The step of determining the comprehensive evaluation index of the life of the main bearing based on the vibration characteristic life index and the physical model life index of the main bearing comprises: The product of the physical model life index and the safety threshold weight is used as the first index threshold, and the product of the physical model life index and the health threshold weight is used as the second index threshold; The sum of the weighted value of the vibration characteristic life index and the weighted value of the physical model life index is taken as the comprehensive weighted value of the index, wherein the weighted value of the vibration characteristic life index is the product of the vibration characteristic life index and the corresponding weight, and the weighted value of the physical model life index is the product of the physical model life index and the corresponding weight; Based on the first index threshold, the second index threshold and the comprehensive weighted value of the index, a corresponding decision strategy is determined, and based on the decision strategy, a comprehensive evaluation index of the life of the main bearing is determined.

7. The method according to claim 6, characterized in that The step of determining a corresponding decision strategy based on the first indicator threshold, the second indicator threshold and the comprehensive indicator weighted value, and determining the comprehensive evaluation index of the life of the main bearing based on the decision strategy includes: When the vibration characteristic life index is less than the first index threshold, determining the decision strategy to be a smaller strategy; Based on the minimization strategy, the minimum value between the vibration characteristic life index and the physical model life index is used as the comprehensive life evaluation index.

8. The method according to claim 6, characterized in that The step of determining a corresponding decision strategy based on the first indicator threshold, the second indicator threshold and the comprehensive indicator weighted value, and determining the comprehensive evaluation index of the life of the main bearing based on the decision strategy includes: When the vibration characteristic life index is greater than or equal to the first index threshold and the comprehensive weighted value of the index is greater than the second index threshold, determining that the decision strategy is a weighted average strategy; Based on the weighted average strategy, a third corresponding relationship among the vibration characteristic life index, the physical model life index, the weight factor and the comprehensive life evaluation index is obtained; The comprehensive life evaluation index is determined based on the vibration characteristic life index, the physical model life index, the weight factor of the vibration characteristic life index, the weight factor of the physical model life index and the third corresponding relationship.

9. The method according to claim 6, characterized in that The step of determining a corresponding decision strategy based on the first indicator threshold, the second indicator threshold and the comprehensive indicator weighted value, and determining the comprehensive evaluation index of the life of the main bearing based on the decision strategy includes: When the vibration characteristic life index is greater than or equal to the first index threshold and the comprehensive weighted value of the index is less than or equal to the second index threshold, determining that the decision strategy is a dynamic adjustment strategy; Based on the dynamic adjustment strategy, a fourth corresponding relationship among the adjustment coefficient, the vibration characteristic life index, the physical model life index and the comprehensive life evaluation index is obtained; The comprehensive life evaluation index is determined based on the vibration characteristic life index, the physical model life index, the adjustment coefficient and the fourth corresponding relationship.

10. An engine crankshaft main bearing life prediction device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the method for predicting the life of an engine crankshaft main bearing according to any one of claims 1 to 9.

Images