High-pressure oil pump fault early warning method, system and device

Abstract

The invention discloses a high-pressure oil pump fault early warning method, system and device, and relates to the technical field of marine diesel engine fault diagnosis. The early warning method comprises the steps of obtaining sample data of a high-pressure oil pump in response to a stable-pressure working condition of the high-pressure oil pump; the sample data comprises the front end temperature and the rail pressure of the high-pressure oil pump; inputting the sample data into a pre-trained early warning model to obtain detection parameters; and generating early warning information in response to the fact that the detection parameter is greater than the first threshold. Therefore, by monitoring the front-end temperature and the rail pressure of the high-pressure oil pump and inputting the data into the pre-trained early warning model, online fault early warning of the high-pressure oil pump can be achieved, potential fault conditions can be found early, further deterioration of the faults is avoided, the downtime and the maintenance cost are reduced, and the reliability and the stability of the diesel engine are improved. And secondly, by generating the detection parameter and the first threshold value, when the detection parameter exceeds the first threshold value, the early warning information is generated, so that the accuracy of fault detection is improved, and the situations of false alarm and missing alarm are reduced.

Description

Technical Field

[0001] The present application relates to the technical field of marine diesel engine fault diagnosis, and in particular to a high-pressure oil pump fault early warning method, system and device. Background Art

[0002] As a fuel system for marine diesel engines, the high-pressure common rail system has the advantages of high injection pressure, injection pattern, injection timing, cyclic injection volume and other key characteristics of ships that can be adjusted flexibly. The high-pressure common rail system usually includes a high-pressure oil pump, a common rail pipe, a high-pressure oil pipe, an injector, an electronic control unit, various sensors and actuators. Among them, the high-pressure oil pump, as the core component of the high-pressure common rail system, compresses the metered low-pressure fuel to create high pressure, and then transports the high-pressure fuel to other components of the high-pressure common rail system through the common rail pipe.

[0003] However, current diesel engines are unable to provide online fault warning for the high-pressure fuel pump. Once the high-pressure fuel pump fails, the high-pressure common rail system will not be able to normally establish high pressure, which seriously affects the reliability of diesel engine operation. Summary of the invention

[0004] An embodiment of the present application provides a high-pressure oil pump fault warning method, which realizes online fault warning of the high-pressure oil pump by inputting the front-end temperature and rail pressure of the high-pressure oil pump into a warning model, thereby solving the problem that the high-pressure oil pump cannot be warned online in the prior art; an embodiment of the present application also provides a high-pressure oil pump fault warning system; an embodiment of the present application also provides a high-pressure oil pump fault warning device.

[0005] The present application provides a high-pressure oil pump fault warning method, comprising:

[0006] In response to the high-pressure oil pump being in a pressure-stabilizing condition, acquiring sample data of the high-pressure oil pump; the sample data including a front-end temperature and a rail pressure of the high-pressure oil pump;

[0007] Inputting the sample data into a pre-trained early warning model to obtain detection parameters;

[0008] In response to the detection parameter being greater than a first threshold, generating warning information.

[0009] In some embodiments, the early warning model is pre-trained based on the following steps:

[0010] Acquire a plurality of historical data; the historical data including the historical front-end temperature and the historical rail pressure generated by the high-pressure oil pump during the historical pressure-stabilizing operation;

[0011] Dividing the plurality of historical data into a training data set and a test data set according to a preset ratio;

[0012] Training the early warning model based on the training data set, a predetermined early warning model structure, a predetermined initial learning rate, and a predetermined maximum number of iterations;

[0013] Performing early warning model accuracy verification on the early warning model based on the training data set and the test data set;

[0014] In response to the early warning model failing the early warning model accuracy verification, the initial learning rate and the maximum number of iterations are re-determined, and the step of training the early warning model based on the training data set, the early warning model structure, the initial learning rate and the maximum number of iterations is returned until the early warning model passes the early warning model accuracy verification and the training of the early warning model is completed.

[0015] In some embodiments, before the step of dividing the plurality of historical data into a training data set and a test data set according to a preset ratio, the step further includes:

[0016] Removing data missing points in the historical data;

[0017] Removing the over-range data and over-alarm value data in the historical data;

[0018] The historical data is converted into dimensionless numerical values.

[0019] In some embodiments, the step of verifying the accuracy of the early warning model based on the training data set and the test data set includes:

[0020] Obtaining the target neuron of the early warning model;

[0021] Acquire a second threshold based on the target neuron and the training data set;

[0022] Inputting the test data set into the early warning model, and obtaining a second distance data set;

[0023] Acquire the proportion of abnormal samples in the second distance data set based on the second threshold;

[0024] In response to the abnormal sample ratio being greater than or equal to a preset third threshold, outputting a result that the early warning model fails the early warning model accuracy verification;

[0025] In response to the abnormal sample ratio being less than the third threshold, a result that the early warning model passes the early warning model accuracy verification is output.

[0026] In some embodiments, the step of acquiring a second threshold based on the target neuron and the training data set includes:

[0027] Acquire a first distance data set, wherein the first distance data set includes a plurality of distance samples, and the distance samples are acquired based on the training data set and the target neuron;

[0028] Dividing the plurality of distance samples in the first distance data set into a plurality of normal samples and a plurality of abnormal samples;

[0029] The second threshold is obtained based on the normal sample and the abnormal sample.

[0030] In some embodiments, the step of dividing the plurality of distance samples in the first distance data set into a plurality of normal samples and a plurality of abnormal samples comprises:

[0031] Determining a normal ratio of the first distance data set; wherein the normal ratio is configured to be less than a second threshold;

[0032] The plurality of distance samples are arranged from large to small, and the plurality of distance samples are divided into a plurality of normal samples and a plurality of abnormal samples based on the normal ratio.

[0033] In some embodiments, in response to the high-pressure oil pump being in a stable pressure condition, the step of inputting the sample data into a pre-trained early warning model to obtain detection parameters includes:

[0034] Obtaining the minimum distance between the sample data and the target neuron in the early warning model;

[0035] The detection parameter is acquired based on the minimum distance and the second threshold.

[0036] In some embodiments, in response to the high-pressure oil pump being in a pressure-stabilizing condition, the step of acquiring sample data of the high-pressure oil pump comprises:

[0037] acquiring the rail pressure and the front end temperature;

[0038] In response to a change in the rail pressure, obtaining a monotonicity of the front end temperature;

[0039] In response to a change in the monotonicity of the front end temperature, determining that the high-pressure oil pump is in the pressure-stabilizing condition;

[0040] The rail pressure and the front end temperature when the high-pressure oil pump is in the pressure-stabilizing condition are determined as the sample data.

[0041] In some embodiments, it also includes:

[0042] In response to the high-pressure oil pump running for a preset time, acquiring the sample data within the preset time;

[0043] The early warning model is retrained based on the sample data.

[0044] In some embodiments, the first threshold is 1 or 0.9; the normal ratio is 95% or 99%; and the third threshold is 90%.

[0045] Accordingly, the embodiment of the present application further provides a high-pressure oil pump failure warning system, comprising:

[0046] A first acquisition module is configured to acquire real-time data sample data of the high-pressure oil pump in response to the high-pressure oil pump being in a pressure-stabilizing condition; the real-time data sample data includes a front-end temperature and a rail pressure of the high-pressure oil pump;

[0047] A second acquisition module is configured to input the real-time data sample data into a pre-trained early warning model to obtain detection parameters;

[0048] The early warning module is configured to generate early warning information in response to the detection parameter being greater than a first threshold.

[0049] Accordingly, the embodiment of the present application further provides a high-pressure oil pump failure warning device, comprising:

[0050] A host computer, wherein the host computer is used to execute the high-pressure oil pump failure warning method as described in any one of the above embodiments.

[0051] Compared with the prior art, a high-pressure oil pump fault warning method in an embodiment of the present application includes: in response to the high-pressure oil pump being in a stable pressure condition, obtaining sample data of the high-pressure oil pump; the sample data includes the front-end temperature and rail pressure of the high-pressure oil pump; inputting the sample data into a pre-trained warning model to obtain detection parameters; in response to the detection parameters being greater than a first threshold, generating warning information. In this way, by monitoring the front-end temperature and rail pressure of the high-pressure oil pump and inputting the data into the pre-trained warning model, online fault warning of the high-pressure oil pump can be achieved, potential fault conditions can be discovered early, the fault can be avoided from further deterioration, downtime and maintenance costs can be reduced, and the reliability and stability of the diesel engine can be improved. Secondly, the pre-trained warning model can generate detection parameters based on the input front-end temperature and rail pressure data, and then judge the detection parameters by setting a first threshold. When the detection parameters exceed the first threshold, a warning message is generated, thereby improving the accuracy of fault detection and reducing false alarms and missed alarms.

[0052] It can be understood that, compared with the prior art, the high-pressure oil pump fault warning system provided in the embodiment of the present application has all the technical features and beneficial effects of the above-mentioned high-pressure oil pump fault warning method, which will not be repeated here.

[0053] It can be understood that, compared with the prior art, the high-pressure oil pump fault warning device provided in the embodiment of the present application has all the technical features and beneficial effects of the above-mentioned high-pressure oil pump fault warning method, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0055] Figure 1 A first flow chart of a high-pressure oil pump failure warning method provided in an embodiment of the present application;

[0056] Figure 2 A second flow chart of a high-pressure oil pump failure warning method provided in an embodiment of the present application;

[0057] Figure 3 A schematic diagram of a training process of a warning model in a high-pressure oil pump fault warning method provided in an embodiment of the present application;

[0058] Figure 4 A schematic diagram of the structure of a high-pressure oil pump failure warning system provided in an embodiment of the present application;

[0059] Figure 5 A schematic structural diagram of a high-pressure oil pump fault warning device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0060] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0061] The present application embodiment provides a high-pressure oil pump failure warning method, please refer to Figure 1 and Figure 2 , Figure 1 A schematic diagram of a high-pressure oil pump fault warning method provided in an embodiment of the present application is shown; Figure 2 The second flow chart of a high-pressure oil pump failure warning method provided by an embodiment of the present application is illustrated. The first embodiment of the present application provides a high-pressure oil pump failure warning method, including the following steps 101 to 103.

[0062] Step 101, in response to the high-pressure oil pump being in a pressure-stabilizing condition, obtain sample data of the high-pressure oil pump; the sample data includes the front-end temperature and rail pressure of the high-pressure oil pump. Specifically, the embodiment of the present application analyzes the performance of the high-pressure oil pump by collecting sample data of the high-pressure oil pump, and evaluates the efficiency and stability of the high-pressure oil pump by monitoring the changes in the front-end temperature and rail pressure, and monitors and diagnoses the operating status of the high-pressure oil pump. If the sample data shows that the temperature is too high or the rail pressure fluctuates greatly, it may be necessary to adjust or optimize the operating parameters of the high-pressure oil pump to improve its performance and reliability.

[0063] Step 102: Input the sample data into the pre-trained early warning model to obtain detection parameters. Specifically, the sample data is input into the pre-trained early warning model to obtain detection parameters, and the operating status of the high-pressure oil pump can be reflected through the detection parameters.

[0064] Step 103, in response to the detection parameter being greater than the first threshold, generate warning information. Specifically, the first threshold is a preset value, which is used to compare with the detection parameter to determine whether the detection parameter is normal or abnormal. When the detection parameter is abnormal, the generated warning information can respond and handle the possible fault of the high-pressure oil pump in time to reduce the potential risks and losses of the high-pressure oil pump, and avoid the high-pressure oil pump from continuing to operate under abnormal conditions, thereby protecting the safety and reliability of the diesel engine.

[0065] See also Figure 3 , Figure 3The schematic diagram of the training process of the warning model in a high-pressure oil pump fault warning method provided by an embodiment of the present application is illustrated; in some embodiments, the warning model is pre-trained based on the following steps: obtaining multiple historical data; the historical data includes the historical front-end temperature and historical rail pressure generated by the high-pressure oil pump in the historical pressure-stabilizing condition; dividing the multiple historical data into a training data set and a test data set according to a preset ratio; training the warning model based on the training data set, a predetermined warning model structure, a predetermined initial learning rate and a predetermined maximum number of iterations; verifying the accuracy of the warning model based on the training data set and the test data set; in response to the warning model failing to pass the warning model accuracy verification, re-determining the initial learning rate and the maximum number of iterations, and returning to the step of training the warning model based on the training data set, the warning model structure, the initial learning rate and the maximum number of iterations, until the warning model passes the warning model accuracy verification, and the training of the warning model is completed. Specifically, the preset ratio of the training data set: the test data set can be, but is not limited to, 5:5, 6:4, 7:3 or 8:2. In this way, by using the training data set, the early warning model can learn the historical operation status and operation trend of the high-pressure oil pump through multiple historical data, thereby improving the accuracy of the early warning model and enabling the early warning model to better predict potential failures or abnormal conditions of the high-pressure oil pump. In addition, if the early warning model fails the early warning model accuracy verification, it is necessary to re-determine the initial learning rate and the maximum number of iterations to optimize the training parameters of the early warning model so that it can better adapt to the operation of the high-pressure oil pump. Then, by repeatedly verifying the accuracy of the early warning model, adjusting the training parameters and retraining the model, the accuracy and reliability of the prediction model can be improved.

[0066] In some embodiments, before the step of dividing the plurality of historical data into a training data set and a test data set according to a preset ratio, the step further includes: removing missing data points in the historical data; removing over-range data and over-alarm value data in the historical data; and converting the historical data into dimensionless values. Specifically, since some data points may not be recorded in the historical data, data missing points may appear, and these data missing points may have a negative impact on the training and testing of the early warning model. Therefore, it is necessary to remove the missing data points to improve the accuracy of the early warning model. Secondly, due to sensor failure, data collection errors or other reasons, there may also be abnormal values ​​in the historical data, that is, data points that exceed the normal range, which will also affect the training and testing results of the prediction model. Therefore, it is also necessary to remove over-range data and over-alarm value data to improve the reliability of the early warning model. For example, over-range data refers to values ​​that exceed the range of sensor measurement. For example, the range of the rail pressure sensor is 0-3000 bar. If the rail pressure parameter exceeds 3000 bar, the record is judged as an outlier sample. Parameters far exceeding the alarm value refer to parameters that exceed the alarm value by more than a certain multiple. For example, the normal upper limit of rail pressure is about 1800 bar. Records exceeding 1 times, that is, exceeding 3600 bar, are judged as outlier samples. In addition, historical data includes front-end temperature and rail pressure, which have different dimensions. In the early warning model, the difference in dimensions may cause the performance of the early warning model to deteriorate. Therefore, it is necessary to convert historical data into dimensionless values ​​to eliminate the dimension difference between front-end temperature and rail pressure and improve the training and testing accuracy of the early warning model.

[0067] In the embodiment of the present application, the historical data is converted into dimensionless values ​​based on the following formula.

[0068]

[0069] Among them, x z-score is a dimensionless value, x i is the historical data, x min is the minimum value in the historical data, x max This is the maximum value in the historical data.

[0070] In some embodiments, the step of verifying the accuracy of the early warning model based on the training data set and the test data set includes: obtaining the target neuron of the early warning model; obtaining the second threshold value based on the target neuron and the training data set; inputting the test data set into the early warning model and obtaining the second distance data set; obtaining the abnormal sample ratio in the second distance data set based on the second threshold value; in response to the abnormal sample ratio being greater than or equal to the preset third threshold value, outputting the result that the early warning model fails the early warning model accuracy verification; in response to the abnormal sample ratio being less than the third threshold value, outputting the result that the early warning model passes the early warning model accuracy verification. Specifically, the structure of the early warning model of the present application is a SOM network, and the target neuron refers to the best neuron in the SOM network. In this way, the training results of the early warning model are evaluated through the early warning model accuracy verification to improve the accuracy and adjustability of the early warning model.

[0071] In some embodiments, the step of obtaining a second threshold based on a target neuron and a training data set includes: obtaining a first distance data set, the first distance data set including a plurality of distance samples, the distance samples being obtained based on the training data set and the target neuron; dividing the plurality of distance samples in the first distance data set into a plurality of normal samples and a plurality of abnormal samples; and obtaining a second threshold based on the normal samples and the abnormal samples. In this way, by dividing the samples in the first distance data set into normal samples and abnormal samples, the present application obtains the second threshold according to the characteristics and distribution of the actual data, thereby characterizing the normal ratio of normal samples and abnormal samples, preparing for the verification of the subsequent test data set, and improving the accuracy of the early warning model.

[0072] In some embodiments, the step of dividing a plurality of distance samples in a first distance data set into a plurality of normal samples and a plurality of abnormal samples comprises: determining a normal ratio of the first distance data set; wherein the normal ratio is configured to be less than a second threshold; arranging the plurality of distance samples from large to small, and dividing the plurality of distance samples into a plurality of normal samples and a plurality of abnormal samples based on the normal ratio. Specifically, the present application first determines that the normal ratio p is 95% or 99%, and preliminarily regards the distance samples with a larger value (1-p) ratio in the distance data set as abnormal samples, and the distance samples with a smaller value p ratio as normal samples, and then determines the second threshold according to the maximum value in the normal sample and the minimum value in the abnormal sample to ensure that the normal ratio is less than the second threshold, for example, the average value of the maximum value in the normal sample and the minimum value in the abnormal sample is taken as the second threshold. In this way, the present application reasonably divides the distance samples into normal and abnormal samples by determining the normal ratio, thereby ensuring the accuracy of the early warning model in identifying normal samples and improving the reliability of the early warning model. Secondly, by arranging the distance samples from large to small and dividing them into normal samples and abnormal samples according to the normal ratio, the abnormal samples can be better distinguished, thereby reducing the false alarm rate and improving the accuracy and credibility of the early warning model. At the same time, by dividing the distance samples into normal samples and abnormal samples, information about the sample distribution is obtained, which helps to further analyze and understand the characteristics of the distance data set and the response of the early warning model to different samples, so as to improve the design and optimization of the early warning model.

[0073] In some embodiments, the first threshold is 1 or 0.9; the normal ratio is 95% or 99%; and the third threshold is 90%.

[0074] In some embodiments, the distance samples are obtained based on the following formula:

[0075]

[0076] Among them, D i is the distance sample, a i and b i is the historical data, n1 and n2 are the target neurons.

[0077] In some embodiments, in response to the high-pressure oil pump being in a stable pressure condition, the sample data is input into a pre-trained early warning model, and the step of obtaining the detection parameter includes: obtaining the minimum distance between the sample data and the target neuron in the early warning model; obtaining the detection parameter based on the minimum distance and the second threshold. Specifically, when the detection parameter is greater than the first threshold, that is, the distance between the t-th sample data and the target neuron (i.e., the best neuron) is greater than the maximum distance between the historical data in the training data set and the corresponding target neuron, it is judged that the current data may be abnormal data, and the first threshold can generally be set to 1.0 or 0.9. When the detection parameter is greater than 1, it means that the distance between the t-th sample data and the target neuron is greater than the maximum distance between the training data and the target neuron, that is, the t-th sample data is more different from the training data, and may be abnormal data. The larger the detection parameter, the greater the abnormality of the t-th sample data. In actual applications, there may be individual unstable working condition data in the training data. Setting the threshold to 1 will cause some abnormal data to not be discovered in time. Therefore, a threshold slightly less than 1 can also be selected, such as 0.9, 0.8 or others, to provide the accuracy of the judgment, which can be adjusted according to the usage in engineering applications.

[0078] Specifically, the present application obtains the detection parameters based on the following formula.

[0079]

[0080] Among them, τ t is the detection parameter, r t is the minimum distance, and w is the second threshold.

[0081] In some embodiments, in response to the high-pressure oil pump being in a stable pressure condition, the step of obtaining sample data of the high-pressure oil pump includes: obtaining rail pressure and front-end temperature; in response to changes in rail pressure, obtaining the monotonicity of the front-end temperature; in response to changes in the monotonicity of the front-end temperature, judging that the high-pressure oil pump is in a stable pressure condition; determining the rail pressure and front-end temperature when the high-pressure oil pump is in a stable pressure condition as sample data. For example, when the rail pressure is rising, the front-end temperature is also rising, which is a variable condition; when the front-end temperature changes from rising to falling, it is judged that the current variable condition stage is over, and the moment when the rising turns to falling is the moment when the variable condition ends, and the steady-state condition is entered. In the embodiment of the present application, by obtaining the two parameters of rail pressure and front-end temperature, key information on the working state of the high-pressure oil pump can be obtained, and by observing the monotonic changes of the front-end temperature, it can be judged whether the high-pressure oil pump is in a stable pressure condition, thereby screening out sample data that meets the requirements, and providing an accurate data basis for subsequent analysis and modeling.

[0082] In some embodiments, it also includes: in response to the preset time of the high-pressure oil pump, obtaining sample data within the preset time; retraining the early warning model based on the sample data. Specifically, after the high-pressure oil pump runs for a preset time, all steady-state operating data records of the high-pressure oil pump and a certain proportion of steady-state operating data of other high-pressure oil pumps of the same model are regularly selected. The preset time can be determined according to the actual situation, with reference to the running-in period of the high-pressure oil pump of this model, such as 100 hours, and the regular duration can also be determined according to the actual situation, such as 50h. Then, the early warning model (i.e., SOM network) is retrained according to the training steps of the early warning model, and when dividing the training samples and the test samples, the proportion of training data and test data in the data of the high-pressure oil pump and other high-pressure oil pumps of the same model is kept consistent. When the early warning model training is completed, it is compared with the abnormal proportion of the original test data. If the abnormal proportion of the test data of the current early warning model is less than the abnormal proportion of the original test data, the SOM network model, the third threshold, the initial learning rate and the maximum number of iterations are updated. If the abnormal proportion of the test data of the current SOM network is not less than the abnormal proportion of the original test data, the SOM network model, the third threshold, the initial learning rate and the maximum number of iterations are not updated. In this way, by regularly selecting the steady-state operating data records of the high-pressure oil pump and the data of other high-pressure oil pumps of the same model, the data set of the early warning model can be kept updated so that it can adapt to changes in actual conditions. Secondly, retraining the early warning model can optimize the performance of the early warning model by using new sample data. By comparing the abnormal proportion with the original test data, it is decided whether to update the parameters and thresholds of the early warning model, so that the early warning model has better adaptability and can adapt to different data distributions and changes.

[0083] A high-pressure oil pump fault warning method of an embodiment of the present application includes: in response to the high-pressure oil pump being in a stable pressure condition, obtaining sample data of the high-pressure oil pump; the sample data includes the front-end temperature and rail pressure of the high-pressure oil pump; inputting the sample data into a pre-trained warning model to obtain detection parameters; in response to the detection parameters being greater than a first threshold, generating warning information. In this way, by monitoring the front-end temperature and rail pressure of the high-pressure oil pump and inputting the data into the pre-trained warning model, online fault warning of the high-pressure oil pump can be achieved, potential fault conditions can be discovered early, the fault can be avoided from further deterioration, downtime and maintenance costs can be reduced, and the reliability and stability of the diesel engine can be improved. Secondly, the pre-trained warning model can generate detection parameters based on the input front-end temperature and rail pressure data, and then judge the detection parameters by setting a first threshold. When the detection parameters exceed the first threshold, a warning message is generated, thereby improving the accuracy of fault detection and reducing false alarms and missed alarms.

[0084] Accordingly, see Figure 4 , Figure 4A structural schematic diagram of a high-pressure oil pump fault warning system provided in an embodiment of the present application is illustrated; an embodiment of the present application provides a high-pressure oil pump fault warning system, comprising a first acquisition module, a second acquisition module and a warning module; wherein the first acquisition module is configured to acquire real-time data sample data of the high-pressure oil pump in response to the high-pressure oil pump being in a stable pressure condition; the real-time data sample data includes the front-end temperature and rail pressure of the high-pressure oil pump; the second acquisition module is configured to input the real-time data sample data into a pre-trained warning model to acquire detection parameters; the warning module is configured to generate warning information in response to the detection parameters being greater than a first threshold.

[0085] It can be understood that, compared with the prior art, the high-pressure oil pump fault warning system provided in the embodiment of the present application has all the technical features and beneficial effects of the above-mentioned high-pressure oil pump fault warning method, which will not be repeated here.

[0086] Accordingly, see Figure 5 , Figure 5 The schematic diagram of the structure of a high-pressure oil pump fault warning device provided by an embodiment of the present application is illustrated. The embodiment of the present application provides a high-pressure oil pump fault warning device, including: a host computer, the host computer is used to execute a high-pressure oil pump fault warning method such as any one of the above embodiments. Among them, the host computer is a control system host computer. In addition, the warning device also includes a human-computer interaction unit and a signal acquisition and processing unit. Among them, the human-computer interaction unit is used to realize the interaction between the user and the warning device, such as setting warning parameters, viewing warning information, and performing fault diagnosis. The human-computer interaction unit may include devices such as a display screen, a keyboard, a mouse or a touch screen, so that the user can easily input instructions, view results and perform operations. The signal acquisition and processing unit is used to collect and process real-time data of key parameters such as the front-end temperature and rail pressure of the high-pressure oil pump, and perform signal processing, such as filtering, amplification, digitization, etc. The signal acquisition and processing unit can also pre-process the collected data, such as calibration, normalization, etc., to ensure the accuracy and reliability of the data. Through the signal acquisition and processing unit, the warning device can obtain high-quality data for subsequent fault prediction and warning analysis.

[0087] It can be understood that, compared with the prior art, the high-pressure oil pump fault warning device provided in the embodiment of the present application has all the technical features and beneficial effects of the above-mentioned high-pressure oil pump fault warning method, which will not be repeated here.

[0088] The above is a detailed introduction to a high-pressure oil pump fault warning method, system and device provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the technical solution and its core idea of ​​the present application. Ordinary technicians in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some of the technical features therein with equivalents; and these modifications or replacements do not make the essence of the corresponding technical solution deviate from the scope of the technical solution of the embodiments of the present application.

Claims

1. A high-pressure oil pump fault early warning method, characterized in that: include: In response to the high-pressure oil pump being in a pressure-stabilizing condition, acquiring sample data of the high-pressure oil pump; the sample data including a front-end temperature and a rail pressure of the high-pressure oil pump; Inputting the sample data into a pre-trained early warning model to obtain detection parameters; In response to the detection parameter being greater than a first threshold, generating warning information.

2. The high-pressure oil pump failure warning method according to claim 1, characterized in that: The early warning model is pre-trained based on the following steps: Acquire a plurality of historical data; the historical data including the historical front-end temperature and the historical rail pressure generated by the high-pressure oil pump during the historical pressure-stabilizing operation; Dividing the plurality of historical data into a training data set and a test data set according to a preset ratio; Training the early warning model based on the training data set, a predetermined early warning model structure, a predetermined initial learning rate, and a predetermined maximum number of iterations; Performing early warning model accuracy verification on the early warning model based on the test data set; In response to the early warning model failing the early warning model accuracy verification, the initial learning rate and the maximum number of iterations are re-determined, and the step of training the early warning model based on the training data set, the early warning model structure, the initial learning rate and the maximum number of iterations is returned until the early warning model passes the early warning model accuracy verification and the training of the early warning model is completed.

3. The high-pressure oil pump failure warning method according to claim 2, characterized in that: Before the step of dividing the plurality of historical data into a training data set and a test data set according to a preset ratio, the method further includes: Removing data missing points in the historical data; Removing the over-range data and over-alarm value data in the historical data; The historical data are converted into dimensionless numerical values.

4. The high-pressure oil pump failure early warning method according to claim 2, characterized in that: The step of verifying the accuracy of the early warning model based on the training data set and the test data set includes: Obtaining the target neuron of the early warning model; Acquire a second threshold based on the target neuron and the training data set; Inputting the test data set into the early warning model, and obtaining a second distance data set; Acquire the proportion of abnormal samples in the second distance data set based on the second threshold; In response to the abnormal sample ratio being greater than or equal to a preset third threshold, outputting a result that the early warning model fails the early warning model accuracy verification; In response to the abnormal sample ratio being less than the third threshold, a result that the early warning model passes the early warning model accuracy verification is output.

5. The high-pressure oil pump failure warning method according to claim 4, characterized in that: The step of acquiring a second threshold based on the target neuron and the training data set includes: Acquire a first distance data set, wherein the first distance data set includes a plurality of distance samples, and the distance samples are acquired based on the training data set and the target neuron; Dividing the plurality of distance samples in the first distance data set into a plurality of normal samples and a plurality of abnormal samples; The second threshold is obtained based on the normal sample and the abnormal sample.

6. The high-pressure oil pump failure warning method according to claim 5, characterized in that: The step of dividing the plurality of distance samples in the first distance data set into a plurality of normal samples and a plurality of abnormal samples comprises: Determining a normal ratio of the first distance data set; wherein the normal ratio is configured to be less than a second threshold; The plurality of distance samples are arranged from large to small, and the plurality of distance samples are divided into a plurality of normal samples and a plurality of abnormal samples based on the normal ratio.

7. The high-pressure oil pump failure warning method according to claim 4, characterized in that: In response to the high-pressure oil pump being in a pressure-stabilizing condition, the step of inputting the sample data into a pre-trained early warning model to obtain detection parameters includes: Obtaining the minimum distance between the sample data and the target neuron in the early warning model; The detection parameter is acquired based on the minimum distance and the second threshold.

8. The high-pressure oil pump failure warning method according to claim 1, characterized in that: The step of acquiring sample data of the high-pressure oil pump in response to the high-pressure oil pump being in a pressure-stabilizing condition comprises: acquiring the rail pressure and the front end temperature; In response to a change in the rail pressure, obtaining a monotonicity of the front end temperature; In response to a change in the monotonicity of the front end temperature, determining that the high-pressure oil pump is in the pressure-stabilizing condition; The rail pressure and the front end temperature when the high-pressure oil pump is in the pressure-stabilizing condition are determined as the sample data.

9. The high-pressure oil pump failure early warning method according to claim 1, characterized in that: Also includes: In response to the high-pressure oil pump running for a preset time, acquiring the sample data within the preset time; The early warning model is retrained based on the sample data.

10. The high-pressure oil pump failure warning method according to claim 6, characterized in that: The first threshold is 1 or 0.9; the normal ratio is 95% or 99%; and the third threshold is 90%.

11. A high-pressure oil pump fault warning system, characterized in that: include: A first acquisition module is configured to acquire real-time data sample data of the high-pressure oil pump in response to the high-pressure oil pump being in a pressure-stabilizing condition; the real-time data sample data includes a front-end temperature and a rail pressure of the high-pressure oil pump; A second acquisition module is configured to input the real-time data sample data into a pre-trained early warning model to obtain detection parameters; The early warning module is configured to generate early warning information in response to the detection parameter being greater than a first threshold.

12. A high-pressure oil pump fault warning device, characterized in that: include: A host computer, wherein the host computer is used to execute the high-pressure oil pump failure warning method as described in any one of claims 1 to 10.

Images