Engine oil injection fault diagnosis method, device and equipment

Abstract

The invention provides an engine oil injection fault diagnosis method, device and equipment, belongs to the technical field of computer information processing, and solves the problems that the existing engine oil injection fault diagnosis depends on manual inspection and is low in diagnosis efficiency, low in diagnosis accuracy and insufficient in timeliness. The method comprises the steps that at least one kind of sound wave signal data generated in the engine oil injection process is obtained; preprocessing the at least one kind of sound wave signal data to obtain preprocessed sound wave data; performing feature extraction processing on the preprocessed sound wave data to obtain feature data; and carrying out identification and diagnosis processing on the feature data to obtain engine oil injection fault diagnosis data. According to the scheme, the sound wave signals generated in the oil injection process are collected and analyzed in a non-intrusive mode, and comprehensive, real-time and accurate diagnosis of the working states of the electronic injection control unit, the wire harness and the oil injector is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer information processing, and in particular to a method, device and equipment for diagnosing engine fuel injection faults. Background Art

[0002] As the main power source for general small aircraft, piston aircraft engines have the advantages of high reliability, low fuel consumption and high cost performance. With the advancement of technology, more and more piston aircraft engine fuel supply systems are using electronic fuel injection technology. However, there are subtle differences in the hardware, wiring harnesses and injectors of electronic fuel injection controllers of the same batch and specification. In order to optimize the performance of piston aircraft engines, it is necessary to quantify the injection-related changes caused by the differences between the controller hardware, wiring harnesses and injectors. At present, most of the diagnostic processes are separate diagnostic tests for controller hardware, wiring harnesses, and injectors, which are not organically combined, or can only be tested once, and cannot simulate the continuous operation during engine operation.

[0003] The injection parameters that affect the performance of piston aircraft engines mainly include injection timing and injection amount. The injection timing determines the distribution and uniformity of the fuel in the mixture in the cylinder during the combustion process, as well as the pressure and temperature in the cylinder during combustion. The injection amount determines the heat release during the combustion process, which has a direct impact on the power of the piston aircraft engine. During the operation of the engine, the injection timing and injection amount jointly determine the heat release rate curve. The appropriate heat release rate can ensure the maximum effective power of the engine and the lowest fuel consumption. However, too early combustion start may cause knocking in the combustion process in the engine cylinder, causing mechanical damage to the cylinder and cylinder liner, causing the engine to work abnormally, and ultimately shortening the service life of the engine. Too late combustion start will cause serious afterburning in the combustion process in the cylinder, causing the engine temperature to rise, increasing the wear and failure rate of the engine, and also producing a large amount of nitrogen oxides, causing environmental pollution.

[0004] Existing methods for piston aircraft engine fuel injection fault diagnosis mostly rely on separate tests of controller hardware, wiring harness and injector, lacking comprehensive diagnostic capabilities under actual engine operating conditions. In addition, traditional methods mostly rely on manual inspection, disassembly inspection and static testing, which are not fast enough and timely. Summary of the invention

[0005] The present invention provides an engine fuel injection fault diagnosis method, device and equipment, which solves the problem that the current engine fuel injection fault diagnosis relies on manual inspection, has low diagnostic efficiency, low diagnostic accuracy and insufficient timeliness.

[0006] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0007] An embodiment of the present invention provides an engine fuel injection fault diagnosis method, comprising:

[0008] Acquiring at least one type of acoustic signal data generated during the engine fuel injection process;

[0009] Preprocessing the at least one acoustic wave signal data to obtain preprocessed acoustic wave data;

[0010] Performing feature extraction processing on the preprocessed sound wave data to obtain feature data;

[0011] The characteristic data is identified and diagnosed to obtain engine fuel injection fault diagnosis data.

[0012] Optionally, the step of acquiring at least one type of acoustic signal data generated during the engine fuel injection process includes:

[0013] At least one kind of sound wave signal data generated during the engine fuel injection process is acquired through a microphone array and a data acquisition card installed at the fuel injection component position of the engine.

[0014] Optionally, preprocessing the at least one acoustic wave signal data to obtain preprocessed acoustic wave data includes:

[0015] Performing filtering processing on the at least one sound wave signal data to obtain sound wave data after filtering processing;

[0016] The filtered sound wave data is subjected to denoising to obtain pre-processed sound wave data.

[0017] Optionally, performing feature extraction processing on the preprocessed sound wave data to obtain feature data includes:

[0018] Extracting characteristic parameters of the fuel injection state from the pre-processed acoustic wave data to obtain key characteristic parameters of the fuel injection state;

[0019] Performing statistical analysis on key characteristic parameters of the fuel injection state to obtain a characteristic vector;

[0020] The feature vector is normalized to obtain feature data.

[0021] Optionally, identification and diagnosis processing is performed on the characteristic data to obtain engine fuel injection fault diagnosis data, including:

[0022] The characteristic data is input into a fault diagnosis model for processing to obtain fault diagnosis data; wherein the fault diagnosis model processes the characteristic data through an activation function to obtain an intermediate result, and the intermediate result is classified to obtain engine injection fault diagnosis data.

[0023] Optionally, the fault diagnosis model processes the characteristic data through an activation function to obtain an intermediate result, and classifies the intermediate result to obtain engine injection fault diagnosis data, including:

[0024] The fault diagnosis model uses the activation function:

[0025]

[0026] Process the feature data to obtain intermediate results;

[0027] pass:

[0028]

[0029] Normalizing the intermediate results to obtain engine fuel injection fault diagnosis data;

[0030] Among them, y is the intermediate result vector, x is the feature data, P(y i ) is the predicted probability of the i-th category, z i is the i-th element of the intermediate result vector, i and j are positive integers and the range of i and j is [0, K], K=C-1, and C is the dimension of y.

[0031] Optionally, the engine fuel injection fault diagnosis method further includes:

[0032] Outputting an engine fuel injection fault diagnosis report, wherein the engine fuel injection fault diagnosis report includes at least one of a sound wave waveform diagram, a spectrum analysis diagram, and a fault diagnosis result.

[0033] The embodiment of the present invention further provides an engine fuel injection fault diagnosis device, comprising:

[0034] An acquisition module, used to acquire at least one type of acoustic wave signal data generated during the engine fuel injection process;

[0035] A processing module, used to preprocess the at least one acoustic wave signal data to obtain preprocessed acoustic wave data; perform feature extraction processing on the preprocessed acoustic wave data to obtain feature data;

[0036] The generating module is used to perform identification and diagnosis processing on the characteristic data to obtain engine fuel injection fault diagnosis data.

[0037] An embodiment of the present invention further provides a computing device, comprising: a processor and a memory storing a computer program, wherein the computer program executes the above method when executed by the processor.

[0038] An embodiment of the present invention further provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the above method.

[0039] The technical solution of the present invention includes at least the following effects:

[0040] The above-mentioned scheme of the present invention obtains at least one kind of sound wave signal data generated during the engine injection process; pre-processes the at least one kind of sound wave signal data to obtain pre-processed sound wave data; performs feature extraction processing on the pre-processed sound wave data to obtain feature data; and performs identification and diagnosis processing on the feature data to obtain engine injection fault diagnosis data. The technical scheme provided by the present invention non-invasively collects and analyzes the sound wave signals generated during the injection process, and utilizes the different sound wave frequencies and amplitudes generated at various stages during the injection process of the injector to perform injection-related fault diagnosis, thereby solving the problem that the current engine injection fault diagnosis relies on manual inspection, has low diagnostic efficiency, low diagnostic accuracy, and insufficient timeliness, and realizes a comprehensive, real-time, and accurate diagnosis of the working status of the electronic injection control unit, wiring harness, and injector. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a flow chart of an engine fuel injection fault diagnosis method provided by an embodiment of the present invention;

[0042] Figure 2 is a structural diagram of an engine fuel injection fault diagnosis device provided by an embodiment of the present invention;

[0043] Figure 3 It is a schematic diagram of the structure of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

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

[0045] like Figure 1 As shown, an embodiment of the present invention provides an engine fuel injection fault diagnosis method, comprising:

[0046] Step 11, obtaining at least one acoustic wave signal data generated during the engine fuel injection process;

[0047] Step 12, preprocessing the at least one acoustic wave signal data to obtain preprocessed acoustic wave data;

[0048] Step 13, performing feature extraction processing on the pre-processed sound wave data to obtain feature data;

[0049] Step 14: performing identification and diagnosis processing on the characteristic data to obtain engine injection fault diagnosis data.

[0050] In this embodiment, in order to obtain fault diagnosis data, the acoustic wave signal data is first collected; the acoustic wave signal data includes multiple types, and at least one of them needs to be collected; a special acoustic wave sensor, such as a high-sensitivity microphone array, is used in the collection process to capture the acoustic wave signal generated during the engine injection process; the microphone or sensor needs to be installed near the engine injector to ensure that the acoustic wave generated at the moment of injection can be accurately captured; the acoustic wave signal collection system has a high signal-to-noise ratio and a wide bandwidth to capture various acoustic wave components from low frequency to high frequency; at the same time, the collection system has real-time transmission capability to transmit the collected acoustic wave signal to a subsequent processing unit in real time; the frequency range of the high-sensitivity microphone array should match the expected acoustic wave signal;

[0051] After obtaining the sound wave signal data, it is necessary to preprocess the sound wave signal data; the preprocessing operation includes filtering and denoising the sound wave signal to eliminate interference factors such as environmental noise and mechanical vibration. The filtering process can use digital filters, such as low-pass filters, band-pass filters, etc., and select the appropriate filter type and parameters according to the characteristics of the sound wave signal. The denoising process can use advanced denoising techniques such as wavelet transform and empirical mode decomposition to further improve the signal quality;

[0052] After the acoustic signal data is preprocessed, information that can reflect the essential characteristics of the signal is extracted from it; specifically, based on the physical characteristics of the injection process, key characteristic parameters that can characterize the injection state are extracted, such as the acoustic peak value at the injection moment, the specific frequency components of the acoustic spectrum, the distribution of acoustic energy, etc. These characteristic parameters will serve as input for subsequent pattern recognition;

[0053] The extracted feature parameters are input into the trained pattern recognition model, which is built based on machine learning algorithms such as support vector machines, neural networks, random forests, etc. The model contains a preset fault mode library, which stores the acoustic feature parameter ranges and pattern features corresponding to different fault types; the model compares and analyzes the input feature parameters, and determines whether the current injection state is normal or has a certain fault by calculating indicators such as similarity or distance measurement. If a fault exists, the model will further determine the fault type, location and severity, and generate an engine injection fault diagnosis report;

[0054] In addition, the method can also include a result display and feedback process, that is, the fault diagnosis results are intuitively displayed to the user through a display system; the display system can be a computer screen, a touch screen or a mobile device interface, etc., and the display content includes sound wave waveforms, spectrum analysis diagrams, fault diagnosis reports, etc.; users can quickly locate the fault location based on the report content and take corresponding maintenance or replacement measures; when the model predicts a possible fault trend in the future, it will send an early warning signal to the user in advance so that the user can take preventive maintenance measures in time. The early warning signal can be notified by status code, sound or light.

[0055] Compared with the existing technology, this technical solution has the following advantages:

[0056] (1) Non-intrusive diagnosis: no need to disassemble the engine or touch the internal components of the injection system, reducing the risk and cost of the diagnostic process.

[0057] (2) Real-time and accuracy: It can collect and analyze acoustic signals in real time while the engine is running, thus improving the timeliness and accuracy of fault diagnosis.

[0058] (3) Comprehensiveness and comprehensiveness: It can comprehensively evaluate the performance differences and fault conditions of the electronic fuel injection controller, wiring harness and injector, providing comprehensive data support for optimizing engine performance.

[0059] (4) Intelligence and automation: Use machine learning algorithms to perform pattern recognition and fault diagnosis, realize the intelligence and automation of the diagnostic process, and improve diagnostic efficiency and accuracy.

[0060] In an optional embodiment of the present invention, step 11 may include:

[0061] Step 111, obtaining at least one kind of sound wave signal data generated during the engine fuel injection process by means of a microphone array and a data acquisition card installed at the fuel injection component of the engine.

[0062] In this embodiment, the microphone array is a device composed of multiple microphones arranged and combined in a certain pattern, which is used to capture and record sound wave signals; these sound wave signals contain rich information about the working state of the injector, such as injection time, injection amount, injection pressure, etc.; each microphone can independently receive sound waves and convert them into electrical signals; the sound wave sensor is a device for detecting sound wave signals and converting them into measurable signals; different from the microphone array, the sound wave sensor may be smaller and lighter, and suitable for different application scenarios;

[0063] Whether using a microphone array or an acoustic wave sensor, the specific steps for obtaining acoustic wave signal data generally include: selecting a suitable microphone array or acoustic wave sensor according to the application scenario and requirements, and arranging it in a suitable position. In this embodiment, a high-sensitivity microphone array needs to be installed near the fuel injection component of the engine; connecting the device to a data acquisition system or computer, and performing necessary configuration and calibration. Start the data acquisition software or program to start recording the acoustic wave signal data received by the microphone array or acoustic wave sensor.

[0064] In an optional embodiment of the present invention, step 12 may include:

[0065] Step 121, filtering the sound wave signal data to obtain sound wave data after filtering;

[0066] Step 122, performing denoising on the sound wave data after the filtering process to obtain pre-processed sound wave data.

[0067] In this embodiment, the main purpose of filtering is to remove unnecessary frequency components from the original sound wave signal data, thereby retaining information useful for subsequent analysis; in the sound wave signal processing, filtering can reduce noise interference and improve the quality of the signal; first, according to the characteristics of the sound wave signal and the frequency range to be processed, a suitable filter type is selected, such as a low-pass filter or a band-pass filter; after filtering, the noise and interference components in the sound wave signal data will be effectively removed, retaining the frequency components useful for subsequent analysis, and improving the accuracy and reliability of the sound wave data;

[0068] Although filtering has removed some of the noise, in some cases, there may still be residual noise or interference; denoising is intended to further reduce these noise components to improve the clarity and readability of the acoustic data; denoising can use denoising techniques such as wavelet transform and empirical mode decomposition, among which wavelet transform denoising removes noise by decomposing and reconstructing the signal in the wavelet domain; after denoising, the noise components in the acoustic data will be further reduced, and the clarity and readability of the signal will be significantly improved; providing more accurate and reliable data support for subsequent analysis and diagnosis.

[0069] In an optional embodiment of the present invention, step 13 may include:

[0070] Step 131, extracting characteristic parameters of the fuel injection state from the pre-processed acoustic wave data to obtain key characteristic parameters of the fuel injection state;

[0071] Step 132, performing statistical analysis on the key characteristic parameters of the injection state to obtain a characteristic vector;

[0072] Step 133, normalize the feature vector to obtain feature data.

[0073] In this embodiment, feature extraction is performed on the pre-processed acoustic wave data, including frequency analysis, time-frequency analysis, etc., to obtain key parameters that can characterize the working state of the injector and the controller; the specific process includes:

[0074] (1) Frequency characteristics analysis: Analyze the frequency distribution of acoustic wave data and extract parameters such as energy or power spectrum density in specific frequency bands. These parameters are directly related to the injection status (such as injection pressure, injection rate, etc.);

[0075] (2) Time domain characteristic analysis: Analyze the waveform characteristics of the acoustic wave data in the time domain, such as peak value, valley value, average value, variance and other statistical parameters. These parameters can reflect the fluctuations in the injection process;

[0076] (3) Envelope detection: By performing envelope detection on the acoustic wave data, the envelope characteristics of the injection signal are extracted, which helps to identify the start and end times of the injection and the duration of the injection process;

[0077] (4) Correlation analysis: Analyze the correlation between the acoustic wave data and other known injection state parameters (such as engine speed, fuel injection amount, etc.) to extract characteristic parameters with strong correlation;

[0078] After the characteristic parameters are extracted, they need to be further statistically analyzed to construct a characteristic vector that can characterize the injection state. The specific process includes:

[0079] (1) Data cleaning: remove outliers or invalid data to ensure data accuracy and reliability;

[0080] (2) Feature selection: Screening based on the importance or relevance of feature parameters to retain the most representative features of the injection status;

[0081] (3) Feature dimensionality reduction: If the number of feature parameters is too large, dimensionality reduction can be performed through methods such as principal component analysis and linear discriminant analysis to reduce computational complexity and improve model performance;

[0082] (4) Feature combination: The filtered feature parameters are combined according to certain rules or algorithms to form a feature vector; a feature vector is a high-dimensional vector, each dimension of which corresponds to a specific feature parameter;

[0083] Through statistical analysis and processing, a feature vector that can fully and accurately characterize the injection state can be obtained;

[0084] In order to eliminate the dimensional differences and numerical range differences between different feature parameters, the feature vectors need to be normalized to improve the generalization ability and stability of the model.

[0085] In an optional embodiment of the present invention, step 14 may include:

[0086] Step 141, input the characteristic data into the fault diagnosis model for processing to obtain fault diagnosis data; wherein the fault diagnosis model processes the characteristic data through an activation function to obtain an intermediate result, and classifies the intermediate result to obtain engine injection fault diagnosis data.

[0087] In this embodiment, the extracted characteristic parameters are compared and analyzed with the preset normal mode or fault mode database, and the potential faults of the injector, controller and wiring harness are automatically identified by using machine learning or pattern recognition algorithm; the system can distinguish different types of faults, such as injector blockage, controller signal abnormality, poor wiring harness contact, etc., and give corresponding fault diagnosis reports; the specific process includes:

[0088] The extracted feature data is input into a pre-trained fault diagnosis model. This model can use a neural network model, such as a convolutional neural network, a recurrent neural network, or a multi-layer perceptron; these models have the ability to process complex data relationships and can extract useful information from the input feature data;

[0089] Inside the model, feature data is first processed through a series of hidden layers; in each layer, the feature data is multiplied by a weight matrix and a bias term is added, and then a nonlinear transformation is performed through an activation function; the role of the activation function is to introduce nonlinear factors so that the model can learn and fit complex nonlinear relationships;

[0090] In the fault diagnosis model, commonly used activation functions include Rectified Linear Unit (ReLU), sigmoid, etc. Among them, the ReLU function is usually used in the hidden layer, which can accelerate the training process of the model and reduce the problem of gradient disappearance; while the sigmoid function can be used in the output layer to map the output value to a specific range (such as between 0 and 1);

[0091] When the feature data is processed by the activation function, a series of intermediate results will be obtained; these intermediate results represent the state of the feature data after nonlinear transformation within the model;

[0092] After obtaining the intermediate results, the model will classify them. In the context of engine injection fault diagnosis, classification usually refers to mapping the intermediate results to different fault categories. These fault categories include injector clogging, fuel pump failure, fuel filter clogging, etc.

[0093] The classification process can be achieved through another fully connected layer, also called the output layer or classification layer, in which each neuron corresponds to a fault category. The intermediate result is multiplied by the weight matrix of the layer, and the bias term is added. It is then normalized through an activation function such as the softmax function to obtain the probability distribution of each fault category.

[0094] Finally, the model will select the fault category with the highest probability as the final fault diagnosis result based on the probability distribution; this result is output as fault diagnosis data to guide subsequent repair and maintenance work.

[0095] A specific embodiment of an engine fuel injection fault diagnosis method provided by an embodiment of the present invention is:

[0096] Step 1, obtaining at least one acoustic wave signal data generated during the engine fuel injection process;

[0097] Install a high-sensitivity microphone array near the engine spark plug to capture the sound wave signal generated during the engine injection process;

[0098] Step 2, preprocessing the at least one acoustic wave signal data to obtain preprocessed acoustic wave data;

[0099] After obtaining the sound wave signal data, the sound wave signal data is filtered and denoised in turn; the filter type can be selected as a low-pass filter or a band-pass filter; the denoising process can use advanced denoising techniques such as wavelet transform and empirical mode decomposition; after the above processing, the noise component in the sound wave data will be further reduced, and the clarity and readability of the signal will be significantly improved.

[0100] Step 3, performing feature extraction processing on the pre-processed sound wave data to obtain feature data;

[0101] Specifically, the process of performing feature extraction processing on the preprocessed sound wave data includes:

[0102] (1) Extracting characteristic parameters, the characteristic parameters include:

[0103] 1) Time domain features: Calculate the statistical features of the sound wave signal, such as mean, variance, kurtosis and skewness, to obtain the basic information of the signal;

[0104] 2) Frequency domain features: The sound wave signal is converted into the frequency domain through fast Fourier transform to extract the main frequency components, spectrum energy and frequency bandwidth;

[0105] 3) Time-frequency domain characteristics: Use wavelet transform or short-time Fourier transform to analyze the time-frequency characteristics of the signal and extract the frequency distribution information of different time periods;

[0106] (2) Feature selection: Use correlation analysis and principal component analysis to select feature parameters that are highly correlated with injection failures, reduce redundant features, and improve the efficiency and accuracy of subsequent models;

[0107] (3) Feature vector construction: The extracted feature parameters are combined into a feature vector for subsequent fault diagnosis model input.

[0108] Step 4: performing identification and diagnosis processing on the characteristic data to obtain engine fuel injection fault diagnosis data.

[0109] Specifically, the process of identifying and diagnosing the characteristic data includes:

[0110] (1) Model selection: Select an appropriate fault diagnosis model, such as support vector machine (SVM), random forest, deep neural network (DNN), etc. The specific selection depends on the complexity and availability of feature data;

[0111] (2) Model training: Use historical fault data and normal data to train the selected model. The training data should include different types of injection faults and their corresponding acoustic feature vectors. During the training process, use cross-validation and other techniques to evaluate the performance of the model and adjust the model parameters to improve accuracy.

[0112] (3) Fault diagnosis: The feature vector constructed in step 3 is input into the trained fault diagnosis model. The model processes the input through an activation function, such as ReLU or sigmoid, to obtain an intermediate result.

[0113] Use the softmax function to classify the intermediate results and output the probability distribution of various types of faults;

[0114] (4) Fault identification: Based on the output probability distribution, the category with the highest probability is selected as the fault diagnosis result to form the final engine injection fault diagnosis data;

[0115] (5) Result output: The diagnostic results are compared with the preset fault types to generate a detailed fault report, including the fault type, possible causes, and recommended maintenance measures.

[0116] The engine fuel injection fault diagnosis method proposed in the present invention provides a piston aircraft engine fuel injection fault diagnosis method based on audio signal analysis. It monitors and diagnoses the faults of the engine fuel injection system in real time and accurately through non-invasive means. It not only realizes the real-time monitoring and accurate diagnosis of the working status of the fuel injection system, but also has a fault warning function, which provides a strong technical guarantee for the safe operation and efficient maintenance of aircraft engines.

[0117] like Figure 2 As shown, the embodiment of the present invention further provides an engine fuel injection fault diagnosis device 20, comprising:

[0118] An acquisition module 21, used to acquire at least one type of acoustic wave signal data generated during the engine fuel injection process;

[0119] The processing module 22 is used to preprocess the at least one acoustic wave signal data to obtain preprocessed acoustic wave data; perform feature extraction processing on the preprocessed acoustic wave data to obtain feature data;

[0120] The generating module 23 is used to perform identification and diagnosis processing on the characteristic data to obtain engine injection fault diagnosis data.

[0121] Optionally, the acquisition module 21 includes:

[0122] At least one kind of sound wave signal data generated during the engine fuel injection process is acquired through a microphone array and a data acquisition card installed at the fuel injection component position of the engine.

[0123] Optionally, the processing module 22 includes:

[0124] Performing filtering processing on the at least one sound wave signal data to obtain sound wave data after filtering processing;

[0125] The filtered sound wave data is subjected to denoising to obtain pre-processed sound wave data.

[0126] Optionally, the processing module 22 further includes:

[0127] Extracting characteristic parameters of the fuel injection state from the pre-processed acoustic wave data to obtain key characteristic parameters of the fuel injection state;

[0128] Performing statistical analysis on key characteristic parameters of the fuel injection state to obtain a characteristic vector;

[0129] The feature vector is normalized to obtain feature data.

[0130] Optionally, the generating module 23 includes:

[0131] The characteristic data is input into a fault diagnosis model for processing to obtain fault diagnosis data; wherein the fault diagnosis model processes the characteristic data through an activation function to obtain an intermediate result, and the intermediate result is classified to obtain engine injection fault diagnosis data.

[0132] Optionally, the fault diagnosis model processes the characteristic data through an activation function to obtain an intermediate result, and classifies the intermediate result to obtain the engine fuel injection fault diagnosis data, including:

[0133] The fault diagnosis model uses the activation function:

[0134]

[0135] Process the feature data to obtain intermediate results;

[0136] pass:

[0137]

[0138] Normalizing the intermediate results to obtain engine fuel injection fault diagnosis data;

[0139] Among them, y is the intermediate result vector, x is the feature data, P(y i ) is the predicted probability of the i-th category, z i is the i-th element of the intermediate result vector, i and j are positive integers and the range of i and j is [0, K], K=C-1, and C is the dimension of y.

[0140] Optionally, the engine fuel injection fault diagnosis device 20 further includes:

[0141] The display module 24 is used to output an engine fuel injection fault diagnosis report, wherein the engine fuel injection fault diagnosis report includes at least one of a sound wave waveform diagram, a spectrum analysis diagram, and a fault diagnosis result.

[0142] It should be noted that the device is a device corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0143] like Figure 3 As shown, the embodiment of the present invention further provides a computing device 30, including a processor 31, a memory 32, and a program or instruction stored in the memory 32 and executable on the processor 31. When the program or instruction is executed by the processor 31, each process of the above-mentioned engine fuel injection fault diagnosis method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. It should be noted that the computing device in the embodiment of the present invention includes the above-mentioned mobile electronic device and non-mobile electronic device.

[0144] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0145] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0146] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0147] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0148] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0149] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0150] In addition, it should be noted that in the apparatus and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. Moreover, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order, and some steps can be performed in parallel or independently of each other. For those of ordinary skill in the art, it is understood that all or any steps or components of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or a network of computing devices in hardware, firmware, software or a combination thereof, which can be achieved by those of ordinary skill in the art using their basic programming skills after reading the description of the present invention.

[0151] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing device. The computing device can be a well-known general-purpose device. Therefore, the purpose of the present invention can also be achieved by simply providing a program product containing a program code for implementing a method or device. That is to say, such a program product also constitutes the present invention, and a storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any well-known storage medium or any storage medium developed in the future. It should also be pointed out that in the device and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order. Some steps can be performed in parallel or independently of each other.

[0152] The above are preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for diagnosing engine fuel injection faults, characterized in that: include: Acquiring at least one type of acoustic signal data generated during the engine fuel injection process; Preprocessing the at least one acoustic wave signal data to obtain preprocessed acoustic wave data; Performing feature extraction processing on the preprocessed sound wave data to obtain feature data; The characteristic data is identified and diagnosed to obtain engine fuel injection fault diagnosis data.

2. The engine fuel injection fault diagnosis method according to claim 1, characterized in that: The step of obtaining at least one acoustic wave signal data generated during the engine fuel injection process includes: At least one kind of sound wave signal data generated during the engine fuel injection process is acquired through a microphone array and a data acquisition card installed at the fuel injection component position of the engine.

3. The engine fuel injection fault diagnosis method according to claim 1, characterized in that: Preprocessing the at least one acoustic wave signal data to obtain preprocessed acoustic wave data includes: Performing filtering processing on the at least one sound wave signal data to obtain sound wave data after filtering processing; The filtered sound wave data is subjected to denoising to obtain pre-processed sound wave data.

4. The engine fuel injection fault diagnosis method according to claim 1, characterized in that: Performing feature extraction processing on the preprocessed sound wave data to obtain feature data includes: Extracting characteristic parameters of the fuel injection state from the pre-processed acoustic wave data to obtain key characteristic parameters of the fuel injection state; Performing statistical analysis on key characteristic parameters of the fuel injection state to obtain a characteristic vector; The feature vector is normalized to obtain feature data.

5. The engine fuel injection fault diagnosis method according to claim 1, characterized in that: The characteristic data is identified and diagnosed to obtain engine fuel injection fault diagnosis data, including: The characteristic data is input into a fault diagnosis model for processing to obtain fault diagnosis data; wherein the fault diagnosis model processes the characteristic data through an activation function to obtain an intermediate result, and the intermediate result is classified to obtain engine injection fault diagnosis data.

6. The engine fuel injection fault diagnosis method according to claim 1, characterized in that: The fault diagnosis model processes the characteristic data through an activation function to obtain an intermediate result, and classifies the intermediate result to obtain engine fuel injection fault diagnosis data, including: The fault diagnosis model uses the activation function: Process the feature data to obtain intermediate results; pass: Normalizing the intermediate results to obtain engine fuel injection fault diagnosis data; Among them, y is the intermediate result vector, x is the feature data, P(y i ) is the predicted probability of the i-th category, z i is the i-th element of the intermediate result vector, i and j are positive integers and the range of i and j is [0, K], K=C-1, and C is the dimension of y.

7. The engine fuel injection fault diagnosis method according to claim 1, characterized in that: Also includes: Outputting an engine fuel injection fault diagnosis report, wherein the engine fuel injection fault diagnosis report includes at least one of a sound wave waveform diagram, a spectrum analysis diagram, and a fault diagnosis result.

8. An engine fuel injection fault diagnosis device, characterized in that: include: An acquisition module, used to acquire at least one type of acoustic wave signal data generated during the engine fuel injection process; A processing module, used to preprocess the at least one acoustic wave signal data to obtain preprocessed acoustic wave data; perform feature extraction processing on the preprocessed acoustic wave data to obtain feature data; The generating module is used to perform identification and diagnosis processing on the characteristic data to obtain engine fuel injection fault diagnosis data.

9. A computing device, characterized in that include: A processor and a memory storing a computer program, wherein when the computer program is executed by the processor, the method according to any one of claims 1 to 7 is performed.

10. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 7.

Images