A dynamic prediction monitoring system for an automobile engine

Abstract

The application discloses a kind of automobile engine dynamic prediction monitoring systems, comprising: data acquisition module, data preprocessing module, prediction module, monitoring module;Data acquisition module is used to obtain historical data, running standard data, running measured data;Data preprocessing module is used to process the data collected, is divided into different data sets, and corresponding database is constructed;Prediction module is used to predict whether failure is possible according to convolutional neural network and the data collected.The monitoring module judges whether the engine operation is normal;Display module is used to display the engine operating data, and the data judged as abnormal is marked.The application divides the running standard data and the running measured data obtained according to the engine working condition, and constructs the corresponding comparison database, which only needs to be compared with the data under the corresponding working condition to quickly judge whether a failure occurs and the corresponding fault type.

Description

Technical Field

[0001] This invention belongs to the field of engine condition monitoring, and in particular relates to a dynamic predictive monitoring system for automobile engines. Background Technology

[0002] When driving, people often cannot immediately know the true working condition of the car's engine. As a result, when the engine malfunctions in its early stages, the driver may not be able to deal with or resolve the problem in a timely manner, which can lead to the malfunction worsening and potentially causing damage to car parts or traffic accidents.

[0003] The automotive engine is a highly complex mechanical, electrical, and hydraulic system, composed of two major mechanisms (crankshaft and connecting rod mechanism, valve train) and five major systems (fuel supply system, cooling system, lubrication system, ignition system, and starting system). Its malfunctions are numerous and varied, and the relationships between malfunction phenomena and causes are complex, potentially involving one-to-many, many-to-one, or many-to-many relationships. Different malfunctions manifest in different ways. Quickly and accurately determining whether the engine is operating normally is essential to ensure timely risk mitigation for drivers. Therefore, a dynamic predictive monitoring system for automotive engines is proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a dynamic predictive monitoring system for automobile engines to solve the problems existing in the prior art.

[0005] To achieve the above objectives, the present invention provides a dynamic predictive monitoring system for an automotive engine, comprising:

[0006] Data acquisition module, data preprocessing module, prediction module, monitoring module, and display module;

[0007] The data acquisition module is used to acquire historical data, run standard data, and run measured data.

[0008] The data preprocessing module is used to process the collected data, divide it into different datasets, and build corresponding databases;

[0009] The prediction module is used to predict whether a fault may occur based on the convolutional neural network and the collected data; if so, the monitoring module determines the fault type.

[0010] The monitoring module determines whether the engine is operating normally based on the standard operating data and the actual operating data, combined with the real-time acquired data of the engine under test.

[0011] The display module is used to display the operating data of various parts of the engine and to mark data that is determined to be abnormal.

[0012] The data acquisition module is connected to the data preprocessing module, the data preprocessing module is connected to the prediction module and the monitoring module respectively, the display module is connected to the monitoring module, and the prediction module is connected to the monitoring module.

[0013] Optionally, the data acquisition module includes a first acquisition unit and a second acquisition unit;

[0014] The first acquisition unit is used to acquire standard operating data and measured operating data of the engine through multiple sensors. The measured operating data is the measurement data under different fault conditions of the engine. The multiple sensors include a vibration sensor, a temperature sensor, a pressure sensor, and a battery data acquisition device.

[0015] The second acquisition unit is used to acquire historical engine data.

[0016] Optionally, the data preprocessing module includes a data receiving unit, a first processing unit, and a second processing unit;

[0017] The data receiving unit is connected to the data acquisition module to acquire standard operating data, actual operating data, and historical data.

[0018] The first processing unit is used to encode and reconstruct the measured data during operation, calculate the variance of the data obtained by each sensor, remove data whose variance exceeds a preset range, and obtain valid measured data.

[0019] The second processing unit is used to divide the standard data and the valid measured data into datasets under different operating conditions according to the engine operating condition type, and to construct a comparison database for the standard data and the valid measured data respectively; and to divide the historical data into datasets for different stages according to the time series.

[0020] Optionally, the monitoring module includes a data acquisition unit, a data processing unit, and a status determination unit;

[0021] The data acquisition unit is used to acquire real-time measurement data from multiple sensors of the engine and to acquire data from the comparison database;

[0022] The data processing unit is used to extract features from the real-time measurement data and the data in the comparison database, respectively.

[0023] The status determination unit is used to determine the working status of the transmitter based on the feature extraction results. If the working status is normal, it will continue to monitor; if the working status is abnormal, it will send a signal to indicate the location of the fault.

[0024] Optionally, the data processing unit performs feature extraction on the data in the comparison database, dividing it into a standard comparison database and a measured comparison database, and extracts standard features and fault features.

[0025] Optionally, the state determination unit includes a first determination unit and a second determination unit;

[0026] The first determination unit is used to match the standard features with the real-time measurement data features. If the deviation value exceeds the preset range, the engine is determined to be malfunctioning. If it does not exceed the preset range, the real-time measurement data is stored.

[0027] The second determination unit is used to compare the real-time measurement data features with the fault features. If it matches a certain fault feature, it is determined to be the fault, and the real-time measurement data is stored in the corresponding comparison database according to the operating conditions.

[0028] Optionally, the prediction module includes a construction unit, a training unit, and a judgment unit;

[0029] The building unit is used to construct a convolutional neural network model;

[0030] The training unit divides the running data in the historical data into a training set and a test set; the training set is input into the convolutional neural network model for training until the loss function converges and the training ends, thus obtaining the optimal model.

[0031] The judgment unit inputs the test set into the optimal model to obtain the predicted probability, and then inputs the real-time measurement data into the optimal model for prediction to determine whether a fault may occur. If so, it feeds back to the monitoring module, which determines the fault type and adjusts the data acquisition frequency.

[0032] The technical effects of this invention are as follows:

[0033] This invention divides the acquired standard operating data and actual operating data according to the engine operating conditions and constructs a corresponding comparison database. By simply comparing the data with the data under the corresponding operating conditions, it is possible to quickly determine whether a fault has occurred and the corresponding fault type. By training a convolutional neural network with historical data, the probability of a fault can be predicted. Data on areas where faults may occur are subject to status determination by a monitoring module, reducing the possibility of missed fault detection and avoiding losses caused by faults occurring within the data acquisition interval. Attached Figure Description

[0034] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0035] Figure 1This is a schematic diagram of the system structure in an embodiment of the present invention. Detailed Implementation

[0036] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0037] Example 1

[0038] like Figure 1 As shown, this embodiment provides a dynamic predictive monitoring system for an automotive engine, including:

[0039] Data acquisition module, data preprocessing module, prediction module, monitoring module, and display module;

[0040] The data acquisition module is used to acquire historical data, run standard data, and run measured data;

[0041] The data preprocessing module is used to process the collected data, divide it into different datasets, and build corresponding databases;

[0042] The prediction module is used to predict whether a fault may occur based on the convolutional neural network and the collected data; if so, the monitoring module determines the fault type.

[0043] The monitoring module determines whether the engine is operating normally based on standard operating data and actual operating data, combined with real-time acquired data of the engine under test.

[0044] The display module is used to display the operating data of various parts of the engine and mark the data that is judged to be abnormal;

[0045] The data acquisition module is connected to the data preprocessing module, which in turn is connected to the prediction module and the monitoring module. The display module is connected to the monitoring module, and the prediction module is also connected to the monitoring module.

[0046] In some embodiments, the data acquisition module includes a first acquisition unit and a second acquisition unit;

[0047] The first acquisition unit is used to acquire standard operating data and measured operating data of the engine through multiple sensors. The measured operating data are the measurement data under different fault conditions of the engine. The multiple sensors include vibration sensors, temperature sensors, pressure sensors and battery data acquisition devices.

[0048] The second acquisition unit is used to acquire historical engine data.

[0049] In some embodiments, the data preprocessing module includes a data receiving unit, a first processing unit, and a second processing unit;

[0050] The data receiving unit is connected to the data acquisition module to acquire standard operating data, actual operating data, and historical data.

[0051] The first processing unit is used to encode and reconstruct the actual measurement data, calculate the fluctuation variance of the data acquired by each sensor, remove data with fluctuation variance exceeding the preset range, and obtain valid actual measurement data.

[0052] The second processing unit is used to divide the standard data and valid measured data into datasets under different operating conditions according to the engine operating condition type, and to construct comparison databases for the standard data and valid measured data respectively; and to divide the historical data into datasets for different stages according to the time series.

[0053] In some embodiments, the monitoring module includes a data acquisition unit, a data processing unit, and a status determination unit;

[0054] The data acquisition unit is used to acquire real-time measurement data from multiple sensors of the engine and to obtain data from the comparison database;

[0055] The data processing unit is used to extract features from the real-time measurement data and the data in the comparison database, respectively;

[0056] The status determination unit is used to determine the transmitter's working status based on the feature extraction results. If the working status is normal, it will continue to monitor; if the working status is abnormal, it will send a signal to indicate the fault location.

[0057] In some embodiments, the data processing unit performs feature extraction on the data in the comparison database, dividing it into a standard comparison database and a measured comparison database, and extracts standard features and fault features.

[0058] In some embodiments, the state determination unit includes a first determination unit and a second determination unit;

[0059] The first determination unit is used to match the standard features with the real-time measurement data features. If the deviation value exceeds the preset range, the engine is determined to be abnormal. If it does not exceed the preset range, the real-time measurement data is stored.

[0060] The second determination unit is used to compare the real-time measurement data characteristics with the fault characteristics. If it matches a certain fault characteristic, it is determined to be the fault, and the real-time measurement data is stored in the corresponding comparison database according to the operating conditions.

[0061] In some embodiments, the prediction module includes a construction unit, a training unit, and a judgment unit;

[0062] Building units are used to construct convolutional neural network models;

[0063] The training unit uses historical running data to divide the training set and the test set; the training set is input into the convolutional neural network model for training until the loss function converges and the training ends, thus obtaining the optimal model.

[0064] The judgment unit inputs the test set into the optimal model to obtain the predicted probability, and then inputs the real-time measurement data into the optimal model for prediction to determine whether a fault may occur. If so, it feeds back to the monitoring module, which determines the fault type and adjusts the data acquisition frequency.

[0065] The above description is merely a preferred 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 dynamic predictive monitoring system for an automotive engine, characterized in that, include: Data acquisition module, data preprocessing module, prediction module, monitoring module, and display module; The data acquisition module is used to acquire historical data, run standard data, and run measured data. The data preprocessing module is used to process the collected data, divide it into different datasets, and build corresponding databases; The prediction module is used to predict whether a failure may occur based on the convolutional neural network and the collected data. If so, the fault type is determined through the monitoring module; The monitoring module determines whether the engine is operating normally based on the standard operating data and the actual operating data, combined with the real-time acquired data of the engine under test. The display module is used to display the operating data of various parts of the engine and to mark data that is determined to be abnormal. The data acquisition module is connected to the data preprocessing module, the data preprocessing module is connected to the prediction module and the monitoring module respectively, the display module is connected to the monitoring module, and the prediction module is connected to the monitoring module. The data acquisition module includes a first acquisition unit and a second acquisition unit; The first acquisition unit is used to acquire standard operating data and measured operating data of the engine through multiple sensors. The measured operating data is the measurement data under different fault conditions of the engine. The multiple sensors include a vibration sensor, a temperature sensor, a pressure sensor, and a battery data acquisition device. The second acquisition unit is used to acquire historical engine data; The data preprocessing module includes a data receiving unit, a first processing unit, and a second processing unit; The data receiving unit is connected to the data acquisition module to acquire standard operating data, actual operating data, and historical data. The first processing unit is used to encode and reconstruct the measured data during operation, calculate the variance of the data obtained by each sensor, remove data whose variance exceeds a preset range, and obtain valid measured data. The second processing unit is used to divide the standard data and the valid measured data into datasets under different operating conditions according to the engine operating condition type, and to construct comparison databases for the standard data and the valid measured data respectively; and to divide the historical data into datasets for different stages according to the time series. The monitoring module includes a data acquisition unit, a data processing unit, and a status determination unit; The data acquisition unit is used to acquire real-time measurement data from multiple sensors of the engine and to acquire data from the comparison database; The data processing unit is used to extract features from the real-time measurement data and the data in the comparison database, respectively. The state determination unit is used to determine the engine working status based on the feature extraction results. If the working status is normal, it will continue to monitor; if the working status is abnormal, it will issue a signal to indicate the abnormal location. The data processing unit extracts features from the data in the comparison database; wherein, the comparison database is divided into a standard comparison database and a measured comparison database; the data processing unit extracts features from the standard comparison database and the measured comparison database respectively to obtain standard features and fault features; The state determination unit includes a first determination unit and a second determination unit; The first determination unit is used to match the standard features with the real-time measurement data features. If the deviation value exceeds the preset range, the engine is determined to be malfunctioning. If it does not exceed the preset range, the real-time measurement data is stored. The second determination unit is used to compare the real-time measurement data features with the fault features. If it matches a certain fault feature, it is determined to be the fault, and the real-time measurement data is stored in the corresponding comparison database according to the operating conditions.

2. The automotive engine dynamic predictive monitoring system according to claim 1, characterized in that, The prediction module includes a construction unit, a training unit, and a judgment unit; The building unit is used to construct a convolutional neural network model; The training unit divides the running data in the historical data into a training set and a test set; the training set is input into the convolutional neural network model for training until the loss function converges and the training ends, thus obtaining the optimal model. The judgment unit inputs the test set into the optimal model to obtain the predicted probability, and then inputs the real-time measurement data into the optimal model for prediction to determine whether a fault may occur. If so, it feeds back to the monitoring module, which determines the fault type and adjusts the data acquisition frequency.

Images