Diesel engine fault identification method and device

Abstract

The application discloses a diesel engine fault identification method and device. In the method, first, a first vibration signal of the diesel engine is collected; then, information entropy of the first vibration signal is calculated to obtain an entropy value corresponding to the first vibration signal; finally, the entropy value is identified through a fault identification model to obtain a fault identification result of the diesel engine, wherein the fault identification model is obtained by training a preset identification model according to entropy values corresponding to vibration signal samples of the diesel engine in multiple states and state labels of the entropy values. It can be seen that the method takes the entropy value corresponding to the vibration signal of the diesel engine as a characteristic parameter, effectively represents different states of the diesel engine, and can more accurately identify the fault of the diesel engine based on the more obvious entropy value, thereby improving the accuracy of diesel engine fault identification.

Description

Technical Field

[0001] This application relates to the field of fault identification technology, and in particular to a diesel engine fault identification method and apparatus. Background Technology

[0002] Diesel engines are widely used in transportation, shipbuilding, construction, generator sets, and many other fields. However, due to their complex structure and multi-faceted internal interactions, diesel engines frequently experience malfunctions during operation, severely impacting their usability. Therefore, accurately identifying diesel engine faults is crucial for ensuring the stability and safety of their operation.

[0003] In the existing technology, the method for diesel engine fault identification usually involves extracting statistical features such as mean and variance from the vibration signal of the diesel engine, and then using the statistical features to identify the faults of the diesel engine through a diesel engine fault identification model to obtain the fault identification results.

[0004] However, in the above-mentioned diesel engine fault identification methods, the vibration signal is affected by the strong background noise during the operation of the diesel engine, and the extracted statistical features are not obvious. They cannot effectively characterize the normal or fault state of the diesel engine, which easily leads to inaccurate diesel engine fault identification results and low accuracy of diesel engine fault identification. Summary of the Invention

[0005] In view of this, embodiments of this application provide a diesel engine fault identification method and apparatus, aiming to improve the accuracy of diesel engine fault identification.

[0006] In a first aspect, embodiments of this application provide a diesel engine fault identification method, the method comprising:

[0007] Collect the first vibration signal of the diesel engine;

[0008] The information entropy of the first vibration signal is calculated to obtain the entropy value corresponding to the first vibration signal;

[0009] The entropy value is identified by a fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is obtained by training a preset identification model based on the entropy value corresponding to the vibration signal sample of the diesel engine in various states and the state label of the entropy value.

[0010] Optionally, the step of calculating the information entropy of the first vibration signal to obtain the entropy value corresponding to the first vibration signal specifically involves:

[0011] The phase entropy of the first vibration signal is calculated to obtain the phase entropy value corresponding to the first vibration signal.

[0012] Optionally, the first vibration signal includes multiple vibration signal points, and the step of calculating the phase entropy of the first vibration signal to obtain the phase entropy value corresponding to the first vibration signal includes:

[0013] Calculate each vibration signal point according to a preset sequence formula to obtain two sequence values ​​corresponding to each vibration signal point;

[0014] Based on the arctangent function, the ratio of the two sequence values ​​is converted into an angle;

[0015] Divide the angle into equal parts to obtain multiple equally divided angles;

[0016] Calculate the cumulative slope of each evenly divided angle, and calculate the ratio of the cumulative slope to the slope corresponding to the angle, to obtain the distribution value of each evenly divided angle;

[0017] The phase entropy value is obtained according to the preset calculation formula and the distribution value.

[0018] Optionally, before calculating the information entropy of the first vibration signal to obtain the entropy value corresponding to the first vibration signal, the method further includes:

[0019] According to a preset time scale, the first vibration signal is divided into multiple second vibration signals;

[0020] The step of calculating the information entropy of the first vibration signal to obtain the entropy value corresponding to the first vibration signal specifically involves:

[0021] The information entropy of the plurality of second vibration signals is calculated to obtain a plurality of entropy values ​​corresponding to the plurality of second vibration signals;

[0022] The step of identifying the entropy value using a fault identification model to obtain the fault identification result of the diesel engine specifically involves:

[0023] Each entropy value corresponding to each second vibration signal is identified by the fault identification model to obtain the fault identification result of the diesel engine.

[0024] Optionally, the multiple states include a normal state and at least one fault state.

[0025] Optionally, the training steps of the fault identification model include:

[0026] Obtain the vibration signal samples and state labels of the diesel engine under the various states;

[0027] The vibration signal samples are identified by a preset identification model to obtain the fault identification results of the diesel engine;

[0028] Based on the fault identification results, the status labels, and the loss function of the preset identification model, the parameters of the preset identification model are trained.

[0029] The preset recognition model that has been trained is determined as the fault recognition model.

[0030] Optionally, the preset recognition model is specifically a smoothed support vector machine (SSVM).

[0031] Secondly, embodiments of this application provide a diesel engine fault identification device, the device comprising:

[0032] The acquisition module is used to acquire the first vibration signal of the diesel engine;

[0033] The calculation module is used to calculate the information entropy of the first vibration signal to obtain the entropy value corresponding to the first vibration signal;

[0034] The identification module is used to identify the entropy value through a fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is obtained by training a preset identification model based on the entropy value corresponding to the vibration signal sample of the diesel engine in multiple states and the state label of the entropy value.

[0035] Thirdly, embodiments of this application provide a diesel engine fault identification device, the device comprising:

[0036] Memory, used to store computer programs;

[0037] A processor is configured to execute the computer program to cause the device to perform the diesel engine fault identification method described in the first aspect above.

[0038] Fourthly, embodiments of this application provide a computer storage medium storing a computer program. When the computer program is run, the device running the computer program implements the diesel engine fault identification method described in the first aspect above.

[0039] Compared with the prior art, the embodiments of this application have the following beneficial effects:

[0040] This application provides a diesel engine fault identification method and apparatus. In this method, firstly, a first vibration signal of the diesel engine is acquired; then, the information entropy of the first vibration signal is calculated to obtain an entropy value corresponding to the first vibration signal; finally, the entropy value is used to identify the fault through a fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is obtained by training a preset identification model based on the entropy values ​​corresponding to vibration signal samples of the diesel engine under various states, and the state labels of the entropy values. It can be seen that this method uses the entropy value corresponding to the vibration signal of the diesel engine as a feature parameter, effectively representing different states of the diesel engine. Based on the more obvious representation of the entropy value, the fault of the diesel engine can be identified more accurately, thereby improving the accuracy of diesel engine fault identification. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings used in the description of the embodiment or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This application scenario illustrates a diesel engine fault identification method provided in this embodiment.

[0043] Figure 2 A flowchart of a diesel engine fault identification method provided in this application embodiment;

[0044] Figure 3 This is a schematic diagram of the structure of a diesel engine fault identification device provided in an embodiment of this application. Detailed Implementation

[0045] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0046] Currently, existing diesel engine fault identification methods typically extract statistical features such as mean and variance from the diesel engine's vibration signals, and then use these statistical features to identify faults using a diesel engine fault identification model. However, in this method, the vibration signals are affected by strong background noise during diesel engine operation, resulting in indistinct extracted statistical features that cannot effectively characterize the normal or faulty state of the diesel engine. This leads to inaccurate fault identification results and low accuracy in diesel engine fault identification.

[0047] To address the aforementioned problems, this application provides a diesel engine fault identification method and apparatus. In this method, firstly, a first vibration signal from the diesel engine is acquired; then, information entropy is calculated on the first vibration signal to obtain an entropy value corresponding to the first vibration signal; finally, the entropy value is used to identify the fault through a fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is obtained by training a preset identification model based on the entropy values ​​corresponding to vibration signal samples of the diesel engine under various states, and the state labels of the entropy values. It is evident that this method uses the entropy value corresponding to the vibration signal of the diesel engine as a feature parameter, effectively characterizing different states of the diesel engine. Based on the more clearly represented entropy value, the faults of the diesel engine can be identified more accurately, thereby improving the accuracy of diesel engine fault identification.

[0048] For example, one scenario in the embodiments of this application can be applied to, such as Figure 1 The scenario shown includes a vibration sensor 101 and a processor 102. The vibration sensor 101 collects a first vibration signal from the diesel engine and sends the first vibration signal from the diesel engine to the processor 102. The processor 102 uses the implementation method provided in this application to identify diesel engine faults.

[0049] First, in the above application scenarios, although the action descriptions of the implementation methods provided in this application are executed by the processor 102, the implementation methods of this application are not limited in terms of the execution subject, as long as the actions disclosed in the implementation methods provided in this application are executed.

[0050] Secondly, the above scenario is only one example provided by the embodiments of this application, and the embodiments of this application are not limited to this scenario.

[0051] The following detailed description, in conjunction with the accompanying drawings and embodiments, illustrates the specific implementation of the diesel engine fault identification method and device in this application.

[0052] See Figure 2 The figure is a flowchart of a diesel engine fault identification method provided in an embodiment of this application, combined with... Figure 2 As shown, it can specifically include:

[0053] S201: Collect the first vibration signal of the diesel engine.

[0054] The system collects the first vibration signal generated by the diesel engine during operation. For example, it can collect the first vibration signal generated by the cylinder head or the cylinder block of the diesel engine. Of course, it can also collect the first vibration signal from other locations of the diesel engine without affecting the implementation of the embodiments of this application.

[0055] In one possible implementation, the first vibration signal generated by the cylinder head vibration during diesel engine operation can be collected. This is because, in the event of conventional faults such as misfires, gear or valve clearance issues, the vibration generated by the cylinder head is more representative than vibrations generated in other parts of the diesel engine, allowing for the acquisition of a more clearly defined first vibration signal. For example, taking a 6-cylinder diesel engine as the research object, a unidirectional vibration sensor can be placed at the cylinder head 1 to collect the first vibration signal of the diesel engine.

[0056] S202: Calculate the information entropy of the first vibration signal to obtain the entropy value corresponding to the first vibration signal.

[0057] Information entropy is calculated on the first vibration signal obtained, yielding an entropy value corresponding to the first vibration signal. Information entropy, as a complexity indicator that directly measures the amount of information in a time series, can deeply mine the rich state information within the time series and describe the uncertainty of various possible events occurring from the information source. In this embodiment, the first vibration signal is used as the information source for information entropy calculation, and the resulting entropy value corresponding to the first vibration signal can effectively characterize the state of the diesel engine during operation.

[0058] Information entropy includes various algorithms such as approximate entropy, sample entropy, fuzzy entropy, or phase entropy. The embodiments of this application do not specifically limit the information entropy algorithm used. For ease of understanding, the following description is based on a possible implementation method.

[0059] In one possible implementation, phase entropy can be calculated on the first vibration signal to obtain the phase entropy value corresponding to the first vibration signal. Phase entropy refers to a complexity index of data information quantity measured in a two-dimensional phase space, i.e., a second-order difference diagram. Compared to algorithms such as approximate entropy, sample entropy, and fuzzy entropy, the phase entropy algorithm has fewer parameters, making it easier to adjust and avoiding application problems caused by multiple parameters in the algorithm and the lack of clear parameter adjustment rules in practical applications.

[0060] The first vibration signal may specifically include multiple vibration signal points. Phase entropy calculation is performed on the first vibration signal to obtain the corresponding phase entropy value. Specifically, this may involve: first, calculating for each vibration signal point according to a preset sequence formula to obtain two sequence values ​​corresponding to each vibration signal point; then, converting the ratio of the two sequence values ​​into an angle using the arctangent function; next, dividing the angle equally to obtain multiple equally divided angles; then, calculating the cumulative slope of each equally divided angle and the ratio of the cumulative slope to the slope corresponding to the angle to obtain the distribution value of each equally divided angle; finally, obtaining the phase entropy value corresponding to the first vibration signal according to the preset calculation formula and the distribution value.

[0061] For example, if the first vibration signal includes N vibration signal points, that is, the length of the first vibration signal is N, it can be represented as {G=g(1), g(2), …, g(N)}, where G is the first vibration signal and g(N) are the vibration signal points. The specific calculation process for calculating the phase entropy of the first vibration signal to obtain the phase entropy value corresponding to the first vibration signal is as follows:

[0062] 1) Calculate the values ​​of N vibration signal points according to a preset sequence formula to obtain two sequence values ​​corresponding to each vibration signal point. The preset sequence formula can be as follows:

[0063] Y(N)=g(N+2)-g(N+1), X(N)=g(N+1)-g(N)

[0064] Where Y(N) and X(N) are the two sequence values ​​corresponding to the vibration signal point g(N).

[0065] 2) Based on the arctangent function, the ratio of the two sequence values ​​Y(N) and X(N) is converted into an angle, which can be calculated using the following formula:

[0066]

[0067] Where θ[N] refers to the angle corresponding to the vibration signal point g(N).

[0068] 3) Divide the angle θ[N] into k equal parts to obtain k equal angles, and calculate the cumulative slope of each equal angle. The formula for calculating the cumulative slope is as follows:

[0069]

[0070] Among them, S θ (i) refers to the cumulative slope, M i This refers to the number of slope values ​​within each equally divided angle.

[0071] 4) Calculate the ratio of the cumulative slope to the slope corresponding to the angle, and obtain the distribution value p(i) for each evenly divided angle.

[0072] 5) Based on the preset calculation formula and the distribution value p(i), obtain the phase entropy value corresponding to the first vibration signal. The preset calculation formula can be as follows:

[0073]

[0074] Here, PhaseEn refers to the phase entropy value.

[0075] S203: The entropy value is identified through the fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is obtained by training a preset identification model based on the entropy value corresponding to the vibration signal sample of the diesel engine in various states and the state label of the entropy value.

[0076] The entropy value corresponding to the first vibration signal is identified by a pre-trained fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is trained on the preset identification model based on the entropy value corresponding to the vibration signal sample of the diesel engine in various different states and the state label corresponding to the entropy value.

[0077] In this application, the embodiments may not specifically limit the various states. For ease of understanding, the following description is based on a possible implementation method.

[0078] In one possible implementation, the multiple states may specifically include a normal state and at least one fault state. For example, at least one fault state may specifically include: a clogged air filter, an excessively large intake valve clearance, or an excessively small intake valve clearance. Correspondingly, the normal state refers to a state where there is no fault and the system can operate normally, that is, the intake valve clearance is at a preset standard value and the air filter is not clogged.

[0079] The specific implementation method of the fault identification model training process is not limited in this application embodiment. For ease of understanding, the following description is based on a possible implementation method.

[0080] In one possible implementation, the training process of the fault identification model may specifically include: first, acquiring vibration signal samples of the diesel engine under various states, and different state labels corresponding to the vibration signal samples; then, identifying the vibration signal samples using a preset identification model to obtain the fault identification result of the diesel engine, i.e., the state of the diesel engine; then, adjusting the parameters of the preset identification model based on the fault identification results obtained from multiple training sessions, the corresponding state labels, and the loss function of the preset identification model, so that the fault identification result corresponds to the state label; finally, determining the trained preset identification model as the fault identification model.

[0081] In one possible implementation, the preset identification model can specifically be a smoothed support vector machine (SSVM). SVM is a commonly used nonlinear classifier in fault identification methods, while SSVM is a smoothed version of SVM, transforming a constrained convex optimization problem into an unconstrained strongly convex optimization problem. Training with SSVM as the preset identification model can improve the efficiency of the training process and the overall performance of the fault identification model.

[0082] For example, the preset recognition model can also be SVM, Bayesian network, artificial neural network, or deep learning model. Of course, other preset recognition models can also be used without affecting the implementation of the embodiments of this application.

[0083] In addition, in an optional embodiment of this application, to further characterize the temporal complexity of the vibration signal and improve its characterization capability, the first vibration signal can be divided into multiple second vibration signals according to a preset time scale. Correspondingly, information entropy can be calculated for the multiple second vibration signals to obtain multiple entropy values ​​corresponding to the multiple second vibration signals, i.e., calculating the multi-scale information entropy corresponding to the first vibration signal. Correspondingly, each entropy value corresponding to each second vibration signal can be identified through a fault identification model to obtain the fault identification result of the diesel engine. Here, multi-scale information entropy refers to a feature index that, combined with a coarse-grained process, can characterize the complexity of a time series at multiple time scales. Calculating the multi-scale information entropy corresponding to the first vibration signal based on the calculated information entropy improves the ability to mine information entropy at a single time scale, making the characterization of the first vibration signal more obvious, and is beneficial to further improving the accuracy of diesel engine fault identification.

[0084] For example, if the first vibration signal includes N vibration signal points, that is, the length of the first vibration signal is N, it can be represented as {X = x(1), x(2), ..., x(N)}, where X is the first vibration signal and x(N) are the vibration signal points. Then the first vibration signal can be divided into multiple second vibration signals. The specific calculation formula is as follows:

[0085]

[0086] Where s is a preset time scale, taking the value of an integer greater than 0, and X is divided into a second vibration signal E with a quantity of N / s. j (s).

[0087] For example, if the length of the first vibration signal X is 10, that is, the first vibration signal is {x(1), x}

[0088] (2), ..., x(10)}, s is preset to 5, then E1=(x(1)+x(2)+x(3)+x(4)+x(5)) / 5, E2=(x(6)+x(7)+x(8)+x(9)+x(10)) / 5, that is, the first vibration signal X is divided into two second vibration signals.

[0089] In one possible implementation, a multi-scale phase entropy value corresponding to the first vibration signal can be calculated. This involves dividing the first vibration signal into multiple second vibration signals, calculating the phase entropy of each second vibration signal to obtain multiple phase entropy values. Based on each phase entropy value corresponding to each second vibration signal, a fault identification model is used to identify the fault in the diesel engine. The fault identification model is obtained by training a preset identification model using the multi-scale phase entropy values ​​corresponding to vibration signal samples under various states, along with the state labels of these multi-scale phase entropy values. By combining the multi-scale process and the phase entropy algorithm, the calculated multi-scale phase entropy value corresponding to the first vibration signal is used as a fault feature parameter of the diesel engine. This approach inherits the advantage of the complexity of phase entropy evaluation time series while overcoming the limited information mining capability of phase entropy under a single scale. It forms a feature index for deeply mining the fault state information rich in diesel engine vibration signals, effectively characterizing different states of the diesel engine and improving the accuracy of diesel engine fault identification.

[0090] Based on the aforementioned S201-S203, in this embodiment, firstly, a first vibration signal of the diesel engine is acquired; then, the information entropy of the first vibration signal is calculated to obtain the entropy value corresponding to the first vibration signal; finally, the entropy value is identified through a fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is obtained by training a preset identification model based on the entropy values ​​corresponding to vibration signal samples of the diesel engine under various states, and the state labels of the entropy values. It is evident that this method uses the entropy value corresponding to the vibration signal of the diesel engine as a feature parameter, effectively characterizing different states of the diesel engine. Based on the more obvious characterization of the entropy value, the faults of the diesel engine can be identified more accurately, thereby improving the accuracy of diesel engine fault identification.

[0091] The above are some specific implementations of the diesel engine fault identification method provided in the embodiments of this application. Based on this, this application also provides a corresponding device. The device provided in the embodiments of this application will be described below from the perspective of functional modularity.

[0092] See Figure 3 The figure is a schematic diagram of the structure of a diesel engine fault identification device 300 provided in an embodiment of this application. The device 300 may include:

[0093] Acquisition module 301 is used to acquire the first vibration signal of the diesel engine;

[0094] The calculation module 302 is used to calculate the information entropy of the first vibration signal to obtain the entropy value corresponding to the first vibration signal;

[0095] The identification module 303 is used to identify the entropy value through the fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is obtained by training a preset identification model based on the entropy value corresponding to the vibration signal sample of the diesel engine in various states and the state label of the entropy value.

[0096] In this embodiment, through the cooperation of the acquisition module 301, the calculation module 302, and the identification module 303, the following steps are taken: First, a first vibration signal of the diesel engine is acquired; then, the information entropy of the first vibration signal is calculated to obtain the entropy value corresponding to the first vibration signal; finally, the entropy value is identified through a fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is trained using the entropy values ​​corresponding to vibration signal samples of the diesel engine under various states, and the state labels of the entropy values. It is evident that this method uses the entropy value corresponding to the vibration signal of the diesel engine as a feature parameter, effectively characterizing different states of the diesel engine. Based on the more obvious characterization of the entropy value, the faults of the diesel engine can be identified more accurately, thereby improving the accuracy of diesel engine fault identification.

[0097] As one implementation method, the computing module 302 can specifically be used for:

[0098] The phase entropy of the first vibration signal is calculated to obtain the phase entropy value corresponding to the first vibration signal.

[0099] As one implementation method, the computing module 302 may specifically include:

[0100] The first calculation unit is used to calculate each vibration signal point according to a preset sequence formula to obtain two sequence values ​​corresponding to each vibration signal point.

[0101] The conversion unit is used to convert the ratio of two sequence values ​​into an angle based on the arctangent function;

[0102] The equal division unit is used to divide the angle equally to obtain multiple equal division angles;

[0103] The second calculation unit is used to calculate the cumulative slope of each evenly divided angle and the ratio of the cumulative slope to the slope corresponding to the angle, so as to obtain the distribution value of each evenly divided angle.

[0104] The acquisition unit is used to obtain the phase entropy value according to the preset calculation formula and distribution value.

[0105] In one embodiment, the diesel engine fault identification device 300 may further include:

[0106] The segmentation module is used to segment the first vibration signal into multiple second vibration signals according to a preset time scale;

[0107] Accordingly, the calculation module 302 can be specifically used for:

[0108] Information entropy is calculated for multiple second vibration signals to obtain multiple entropy values ​​corresponding to the multiple second vibration signals;

[0109] Accordingly, the identification module 303 can be specifically used for:

[0110] Each entropy value corresponding to each second vibration signal is identified through the fault identification model to obtain the fault identification result of the diesel engine.

[0111] As one implementation method, the multiple states in the identification module 303 may specifically include:

[0112] Normal state and at least one fault state.

[0113] As one implementation method, the training steps of the fault identification model in the identification module 303 may specifically include:

[0114] The acquisition unit is used to acquire vibration signal samples and status labels of the diesel engine under various conditions;

[0115] The identification unit is used to identify vibration signal samples through a preset identification model to obtain the fault identification results of the diesel engine;

[0116] The training unit is used to train the parameters of the preset identification model based on the fault identification results, status labels, and the loss function of the preset identification model.

[0117] The determination unit is used to determine the pre-set recognition model after training as the fault recognition model.

[0118] As one implementation method, the preset recognition model in the recognition unit can specifically be a smoothed support vector machine (SSVM).

[0119] This application also provides corresponding devices and computer storage media for implementing the solutions provided in this application.

[0120] The device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to enable the device to perform the diesel engine fault identification method according to any embodiment of this application.

[0121] The computer storage medium stores a computer program. When the code is run, the device running the computer program implements the diesel engine fault identification method described in any embodiment of this application.

[0122] In the embodiments of this application, the terms "first" and "second" (if they exist) are used only as name identifiers and do not represent the order of first and second.

[0123] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that all or part of the steps in the methods of the above embodiments can be implemented by means of software plus a general-purpose hardware platform. Based on this understanding, the technical solution of this application can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as a read-only memory (ROM) / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, a server, or a network communication device such as a router) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0124] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments. The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separate. The components indicated as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment solution according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0125] The above description is merely one specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A diesel engine fault identification method, characterized in that, The method includes: Collect the first vibration signal of the diesel engine; The information entropy of the first vibration signal is calculated to obtain the entropy value corresponding to the first vibration signal; The entropy value is identified through a fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is obtained by training a preset identification model based on the entropy value corresponding to the vibration signal sample of the diesel engine in multiple states and the state label of the entropy value. Specifically, the step of calculating the information entropy of the first vibration signal to obtain the entropy value corresponding to the first vibration signal involves: The phase entropy of the first vibration signal is calculated to obtain the phase entropy value corresponding to the first vibration signal; The first vibration signal includes multiple vibration signal points. The step of calculating the phase entropy of the first vibration signal to obtain a phase entropy value corresponding to the first vibration signal includes: Calculate each vibration signal point according to a preset sequence formula to obtain two sequence values ​​corresponding to each vibration signal point; Based on the arctangent function, the ratio of the two sequence values ​​is converted into an angle; Divide the angle into equal parts to obtain multiple equally divided angles; Calculate the cumulative slope of each evenly divided angle, and calculate the ratio of the cumulative slope to the slope corresponding to the angle, to obtain the distribution value of each evenly divided angle; The phase entropy value is obtained according to the preset calculation formula and the distribution value.

2. The method according to claim 1, characterized in that, Before calculating the information entropy of the first vibration signal to obtain the entropy value corresponding to the first vibration signal, the method further includes: According to a preset time scale, the first vibration signal is divided into multiple second vibration signals; The step of calculating the information entropy of the first vibration signal to obtain the entropy value corresponding to the first vibration signal specifically involves: The information entropy of the plurality of second vibration signals is calculated to obtain a plurality of entropy values ​​corresponding to the plurality of second vibration signals; The step of identifying the entropy value using a fault identification model to obtain the fault identification result of the diesel engine specifically involves: Each entropy value corresponding to each second vibration signal is identified by the fault identification model to obtain the fault identification result of the diesel engine.

3. The method according to claim 1, characterized in that, The multiple states include a normal state and at least one fault state.

4. The method according to claim 1, characterized in that, The training steps of the fault identification model include: Obtain the vibration signal samples and state labels of the diesel engine under the various states; The vibration signal samples are identified by a preset identification model to obtain the fault identification results of the diesel engine; Based on the fault identification results, the status labels, and the loss function of the preset identification model, the parameters of the preset identification model are trained. The preset recognition model that has been trained is determined as the fault recognition model.

5. The method according to claim 4, characterized in that, The preset recognition model is specifically a smoothed support vector machine (SSVM).

6. A diesel engine fault identification device, characterized in that, The device includes: The acquisition module is used to acquire the first vibration signal of the diesel engine; The calculation module is used to calculate the information entropy of the first vibration signal to obtain the entropy value corresponding to the first vibration signal; The identification module is used to identify the entropy value through a fault identification model to obtain the fault identification result of the diesel engine. The fault identification model is obtained by training a preset identification model based on the entropy value corresponding to the vibration signal sample of the diesel engine in multiple states and the state label of the entropy value. The calculation module is specifically used for: The phase entropy of the first vibration signal is calculated to obtain the phase entropy value corresponding to the first vibration signal; The calculation module specifically includes: The first calculation unit is used to calculate each vibration signal point according to a preset sequence formula to obtain two sequence values ​​corresponding to each vibration signal point. The conversion unit is used to convert the ratio of two sequence values ​​into an angle based on the arctangent function; The equal division unit is used to divide the angle equally to obtain multiple equal division angles; The second calculation unit is used to calculate the cumulative slope of each evenly divided angle and the ratio of the cumulative slope to the slope corresponding to the angle, so as to obtain the distribution value of each evenly divided angle. The acquisition unit is used to obtain the phase entropy value according to the preset calculation formula and distribution value.

7. A diesel engine fault identification device, characterized in that, The device includes: Memory, used to store computer programs; A processor is configured to execute the computer program to cause the device to perform the steps of the diesel engine fault identification method as described in any one of claims 1-5.

8. A computer storage medium, characterized in that, The computer storage medium stores a computer program, and when the computer program is run, the device running the computer program implements the steps of the diesel engine fault identification method as described in any one of claims 1-5.

Images