Active safety monitoring system and monitoring method for crankcase of diesel engine

Abstract

The invention discloses an active safety monitoring system and method for a crankcase of a diesel engine, and the system comprises a blow-by monitoring module which is used for monitoring the blow-by flow of the crankcase of the diesel engine; the oil gas monitoring module is used for monitoring the oil mist concentration of the engine crankcase; the local monitoring box is used for collecting and sending the blow-by flow, the oil mist concentration, the engine rotating speed and the engine torque; and the cloud server is used for determining the wear degree of a piston ring and a main bearing of the engine through a trained transfer learning model according to the blow-by flow, the oil mist concentration, the engine rotating speed and the engine torque sent by the machine-side monitoring box, so that the blow-by condition of the crankcase of the engine is accurately monitored, and the accuracy of the blow-by condition of the crankcase of the engine is improved. And the wear degrees of the piston ring and the main bearing of the engine are accurately identified through the transfer learning model, so that the health online state of the engine is monitored in real time.

Description

Technical Field

[0001] The present application relates to the technical field of diesel engine status monitoring, and in particular to a diesel engine crankcase active safety monitoring system and monitoring method. Background Art

[0002] As the power density of automotive diesel engines increases, the cylinder pressure and piston linear speed also increase significantly, resulting in an increase in the tensile pressure on the engine connecting rod, which greatly increases the risk of wear on the engine piston rings and bearings. Piston ring wear will lead to an increase in crankcase blowby, which may affect the engine's charging efficiency and reduce the compression ratio of the combustion chamber, resulting in a decrease in the engine's output power and efficiency. Bearing wear will increase the oil content of crankcase blowby, causing some of the oil to be carried away through the crankcase ventilation system, causing the engine oil to be consumed too quickly, requiring more frequent oil replenishment, resulting in increased vehicle operating costs.

[0003] Due to the tight chassis layout of diesel engine vehicles, the existing crankcase active safety system cannot be installed on current automotive diesel engines, resulting in the inability to identify bearing wear. Therefore, it is necessary to develop new monitoring methods to monitor the engine crankcase status, so as to accurately identify bearing wear. Summary of the invention

[0004] The present application provides a diesel engine crankcase active safety monitoring system and method, which can solve the technical problem in the prior art that diesel engine vehicles cannot identify bearing wear.

[0005] In a first aspect, an embodiment of the present application provides a diesel engine crankcase active safety monitoring system, the diesel engine crankcase active safety monitoring system comprising:

[0006] A blowby gas monitoring module, which is used to monitor the blowby gas flow rate of the engine crankcase;

[0007] An oil and gas monitoring module, which is used to monitor the oil mist concentration of the engine crankcase;

[0008] A machine-side monitoring box, which is used to collect the blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque, and send them;

[0009] The cloud server is used to determine the degree of wear of the engine piston ring and the main bearing according to the blowby gas flow, the oil mist concentration, the engine speed and the engine torque sent by the machine-side monitoring box through a trained transfer learning model.

[0010] In combination with the first aspect, in one implementation, the blowby gas monitoring module includes a blowby gas monitoring pipeline and a flow sensor;

[0011] The blowby gas monitoring pipeline includes a pressure-expanding pipe, a pressure-stabilizing chamber and a venturi tube connected in sequence, and the pressure-expanding pipe is connected to the engine crankcase ventilation interface;

[0012] The flow sensor is arranged in the throat of the venturi tube.

[0013] In one embodiment, the oil and gas monitoring module includes an oil mist concentration sensor;

[0014] The oil mist concentration sensor is arranged in the throat of the venturi tube.

[0015] In one embodiment, the machine-side monitoring box comprises:

[0016] A collection module, which is used to collect the blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque;

[0017] a processing module, which is used to package the blow-by gas flow rate, the oil mist concentration, the engine speed and the engine torque to generate packaged data;

[0018] A transmitting module, used for sending the packaged data to the cloud server;

[0019] A thermal insulation coating is used to wrap the collection module, the processing module and the emission module.

[0020] In one embodiment, the cloud server is further used for:

[0021] Performing chi-square test, information gain ratio and correlation coefficient test on the received packaged data, with the goal of minimum redundancy of correlation coefficient and maximum correlation between information gain ratio and chi-square test result, to identify target data in the packaged data;

[0022] Performing a simplification dimensionality reduction reconstruction on the target data to generate dimensionality reduction target data;

[0023] The outliers in the dimension reduction target data are identified by a spatial clustering algorithm, and the outliers are filtered to obtain the final target data.

[0024] In one embodiment, the cloud server is further used for:

[0025] The blowby gas flow, the oil mist concentration, the engine speed and the engine torque in the final target data are input into the trained transfer learning model to obtain the wear degree of the engine piston ring and the main bearing.

[0026] In one embodiment, the cloud server is further used for:

[0027] The engine is respectively used as source domain auxiliary data for a plurality of operating condition data of different wear degrees of the engine piston ring and the main bearing under the first operating condition, the second operating condition and the third operating condition, the wear degree of the engine piston ring and the main bearing, the corresponding blowby gas flow rate and the oil mist concentration, wherein the operating condition data include the engine load rate and the engine speed, and the wear degree of the engine piston ring and the main bearing includes no wear, slight wear, moderate wear and severe wear;

[0028] The engine under the fourth working condition, multiple working condition data of different wear degrees of the engine piston ring and the main bearing, the wear degree of the engine piston ring and the main bearing, the corresponding blowby gas flow rate and oil mist concentration are used as target domain data;

[0029] All the source domain auxiliary data and M data in the target domain data are used as training data, and all the target domain data are used as test data to train the transfer learning model to obtain the trained transfer learning model.

[0030] In a second aspect, an embodiment of the present application provides a diesel engine crankcase active safety monitoring method implemented by a diesel engine crankcase active safety monitoring system according to any one of the above, the diesel engine crankcase active safety monitoring method comprising:

[0031] Monitor the blowby gas flow rate of the engine crankcase through the blowby gas monitoring module;

[0032] Monitoring the oil mist concentration of the engine crankcase through an oil and gas monitoring module;

[0033] The blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque are collected through a machine-side monitoring box and sent;

[0034] The wear degree of the engine piston ring and the main bearing is determined by the cloud server based on the blowby gas flow, the oil mist concentration, the engine speed and the engine torque sent by the machine-side monitoring box through the trained transfer learning model.

[0035] In combination with the second aspect, in one implementation, the machine-side monitoring box includes a collection module, a processing module, a transmission module, and a thermal insulation coating, and the blowby gas flow rate, the oil mist concentration, the engine speed, and the engine torque are collected and sent by the machine-side monitoring box, including:

[0036] collecting the blow-by gas flow rate, the oil mist concentration, the engine speed and the engine torque through the collection module;

[0037] Packaging the blow-by gas flow rate, the oil mist concentration, the engine speed and the engine torque through the processing module to generate packaged data;

[0038] Sending the packaged data to the cloud server through the transmitting module;

[0039] The collection module, the processing module and the emission module are wrapped by the thermal insulation coating.

[0040] In one embodiment, before determining the wear degree of the engine piston ring and the main bearing through the trained transfer learning model according to the blowby gas flow, the oil mist concentration, the engine speed and the engine torque sent by the engine side monitoring box through the cloud server, the further comprising:

[0041] Performing chi-square test, information gain ratio and correlation coefficient test on the received packaged data through the cloud server, with the goal of minimizing redundancy of correlation coefficient and maximizing correlation between information gain ratio and chi-square test result, to identify target data in the packaged data;

[0042] Performing a simplification dimensionality reduction reconstruction on the target data to generate dimensionality reduction target data;

[0043] The outliers in the dimension reduction target data are identified by a spatial clustering algorithm, and the outliers are filtered to obtain the final target data.

[0044] The embodiment of the present application provides a diesel engine crankcase active safety monitoring system and method, the diesel engine crankcase active safety monitoring system comprising: a blowby monitoring module, which is used to monitor the blowby gas flow rate of the engine crankcase; an oil and gas monitoring module, which is used to monitor the oil mist concentration of the engine crankcase; an engine-side monitoring box, which is used to collect the blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque, and send them; a cloud server, which is used to determine the degree of wear of the engine piston ring and the main bearing according to the blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque sent by the engine-side monitoring box through a trained transfer learning model, thereby realizing accurate monitoring of the engine crankcase blowby condition, and accurately identifying the degree of wear of the engine piston ring and the main bearing through the transfer learning model, thereby realizing real-time monitoring of the healthy online status of the engine. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a schematic diagram of the architecture of the diesel engine crankcase active safety monitoring system of this application;

[0046] Figure 2 This is a schematic diagram of the installation positions of the flow sensor and the oil mist concentration sensor;

[0047] Figure 3 This is a functional module diagram of the machine-side monitoring box for this application;

[0048] Figure 4Schematic diagram of the training process of the transfer learning model;

[0049] Figure 5 This is a flow chart of an embodiment of a method for active safety monitoring of a diesel engine crankcase of the present application. DETAILED DESCRIPTION

[0050] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described 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, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0051] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0052] In a first aspect, an embodiment of the present application provides a diesel engine crankcase active safety monitoring system, which can be used in various types of engine vehicles such as diesel engines or gasoline engines.

[0053] In one embodiment, referring to Figure 1 , Figure 1 This is a schematic diagram of the architecture of the diesel engine crankcase active safety monitoring system of this application. Figure 1 As shown, the diesel engine crankcase active safety monitoring system includes:

[0054] A blowby gas monitoring module, which is used to monitor the blowby gas flow rate of the engine crankcase;

[0055] The oil and gas monitoring module is used to monitor the oil mist concentration of the engine crankcase.

[0056] A machine-side monitoring box, which is used to collect the blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque, and send them;

[0057] The cloud server is used to determine the degree of wear of the engine main bearing according to the blowby gas flow, the oil mist concentration, the engine speed and the engine torque sent by the machine-side monitoring box and through a trained transfer learning model.

[0058] In this embodiment, the blowby gas monitoring module and the oil and gas monitoring module are used to accurately monitor the blowby gas flow rate and oil mist concentration of the engine crankcase, and the monitored blowby gas flow rate and oil mist concentration as well as the engine speed and engine torque are used as the input of the trained transfer learning model. The trained transfer learning model can accurately and efficiently identify the degree of wear of the engine piston ring and main bearing. And the calculation process of determining the degree of wear of the engine piston ring and main bearing through the trained transfer learning model is carried out in the cloud server, which can improve the computing efficiency and effectively reduce the data processing pressure on the vehicle side. And the diesel engine crankcase active safety monitoring system provided in this application adopts a modular design, which improves the interchangeability and versatility of the system.

[0059] In one embodiment, if Figure 2 As shown, the blowby monitoring module includes a blowby monitoring pipeline and a flow sensor; the blowby monitoring pipeline includes a diffuser, a pressure stabilizing chamber and a venturi tube connected in sequence, and the diffuser is connected to the engine crankcase ventilation interface; the flow sensor is arranged in the throat of the venturi tube.

[0060] The oil and gas monitoring module includes an oil mist concentration sensor, and the oil mist concentration sensor is arranged in the throat of the venturi tube.

[0061] It is worth mentioning that the wide-pressure tube can increase the airflow pressure, and the airflow is stabilized through the pressure stabilizing chamber. The flow sensor and the oil mist concentration sensor are arranged in the throat of the venturi tube. When the airflow passes through the throat part of the venturi tube, the airflow velocity increases and the pressure decreases, thereby realizing accurate monitoring of the blowby airflow and fog concentration of the engine crankcase.

[0062] The working principle of the oil mist concentration sensor is mainly based on the principle of light scattering. The oil mist concentration sensor detects the intensity of light of a specific wavelength to identify the type and density of reflective materials and measure the oil mist concentration. The oil mist concentration sensor uses the phenomenon that when light passes through the sample gas, it is reflected in all directions by impurity particles suspended in the gas, forming a light scattering phenomenon. By detecting the intensity of this scattered light, the type and density of reflective materials can be identified, thereby achieving accurate identification of the oil mist concentration.

[0063] In some optional embodiments, such as Figure 3As shown, the machine-side monitoring box includes: a collection module, which is used to collect the blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque; a processing module, which is used to package the blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque to generate packaged data; a transmitting module, which is used to send the packaged data to the cloud server; a thermal insulation coating, which is used to wrap the collection module, the processing module and the transmitting module. The machine-side monitoring box also includes a radiator for dissipating heat for each module in the machine-side monitoring box.

[0064] It is worth noting that the thermal insulation coating is used to isolate the heat radiation of the engine. By wrapping the acquisition module, processing module and transmitting module, it can effectively prevent the high temperature gas of the engine from damaging the electronic components in the machine-side monitoring box. The acquisition module includes an acquisition board, which is used to collect the blowby flow signal collected by the flow sensor, the oil mist concentration signal collected by the oil mist concentration sensor, and the engine speed and torque signal collected by the whole vehicle, and send them to the processing module through the TCP / IP protocol. The processing module in this embodiment can use the NI 9033 chassis, build the underlying algorithm based on FPGA to process the signal in real time, and package the signal and send it to the cloud server through the transmitting module.

[0065] It is worth noting that before determining the degree of wear of the engine piston rings and main bearings through the trained transfer learning model, during the engine development stage, fault simulation tests can be conducted to test different piston ring and bearing wear conditions, and the engine piston ring and bearing wear data, as well as the corresponding blowby gas flow rate and oil mist concentration can be recorded. The transfer learning model can be trained using this data to obtain a trained transfer learning model.

[0066] Specifically, when training the transfer learning model, the cloud server is also used to: use multiple operating condition data of the engine piston ring and main bearing with different degrees of wear under the first operating condition, the second operating condition and the third operating condition, the degree of wear of the engine piston ring and the main bearing, the corresponding blowby gas flow rate and oil mist concentration as source domain auxiliary data, wherein the operating condition data include engine load rate and engine speed, and the degree of wear of the engine piston ring and the main bearing include no wear, slight wear, moderate wear and severe wear; use multiple operating condition data of the engine piston ring and main bearing with different degrees of wear under the fourth operating condition, the degree of wear of the engine piston ring and the main bearing, the corresponding blowby gas flow rate and oil mist concentration as target domain data; use all of the source domain auxiliary data and M data of the target domain data as training data, and use all of the target domain data as test data to train the transfer learning model to obtain the trained transfer learning model.

[0067] Exemplarily, in this embodiment, the first engine operating condition includes 20% engine load and a speed of 1500r / min, the second engine operating condition includes 40% engine load and a speed of 1500r / min, the third engine operating condition includes 80% engine load and a speed of 1500r / min, and the fourth engine operating condition includes 60% engine load and a speed of 1500r / min. The wear degree of the engine piston ring and the main bearing includes 4 levels, namely no wear, slight wear, moderate wear and severe wear, wherein the wear degree can be set according to the wear percentage of the engine piston ring and the main bearing. M can be set according to demand, and the value of M in this embodiment is 100.

[0068] In this embodiment, the first, second and third working conditions of the engine are combined with the wear degree of the piston ring and main bearing of the 4th gear engine and the corresponding blowby gas flow rate and oil mist concentration as the source domain auxiliary data. The fourth working condition of the engine is combined with the wear degree of the piston ring and main bearing of the 4th gear engine and the corresponding blowby gas flow rate and oil mist concentration as the target domain data. The data division of the source domain auxiliary data and the target domain data is shown in Table 1.

[0069] Table 1 Division table of source domain auxiliary data and target domain data

[0070]

[0071] Among them, the number of source domain auxiliary data is 6000, 3 engine operating conditions and 4 wear gears have a total of 12 situations, and 500 data for each situation. The number of target domain data is 400, 1 engine operating condition and 4 wear gears have a total of 4 situations, and 100 data for each situation. The training set of the transfer learning model in this embodiment includes 6000 source domain auxiliary data and 100 target domain data randomly selected from 400 target domain data, a total of 6100 data. The training set of the transfer learning model includes 400 target domain data. The transfer learning model is trained by training set and test set. The data of the test set is the crankcase signal feature data of 60% load of the engine, and the data of the training set is the data of 20% load, 40% load and 80% load of the engine plus a very small amount of 60% load data. If the diagnosis success rate of the transfer learning model reaches the preset success rate threshold, it means that the diagnosis model can realize the transfer of knowledge between different loads, and the training of the transfer learning model is completed.

[0072] The transfer learning model in this embodiment is the TrAdaBoost model. When the TrAdaBoost model is used in the diesel engine crankcase condition monitoring and diagnosis, the source domain auxiliary data and a small amount of target domain data are used as training sets, and then the target domain data is used as a training set. It is imported into the final classifier to obtain the diagnosis result, which solves the problem of low diagnosis accuracy caused by different data distribution characteristics due to variable working conditions. The training process is as follows: Figure 4Its impact on the training process, the specific steps are as follows:

[0073] In the first step, the input of the model includes the training set T and the test set S. The training set T consists of two parts: the auxiliary training set Ta and the source domain training set Tb. The auxiliary training set Ta is 100 target domain data randomly selected from 400 target domain data, and the source domain training set Tb is 6000 source domain auxiliary data.

[0074] The second step is to calculate the initial weight w:

[0075]

[0076] Among them, β is the learning machine weight, N is the maximum number of iterations, n is the number of samples in the auxiliary training set Ta, and m is the number of samples in the source domain training set Tb.

[0077] Step 3: Set the time step t = 1, 2, ..., N, and set the weight The output prediction value ht can be calculated, and the error is:

[0078]

[0079] Among them, c(x i ) is the true value, and x is the sample in the test set S.

[0080] Step 4: During the calculation process, set β≤1 / 2;

[0081]

[0082] Among them, β t is the learning machine weight for each iteration.

[0083] Step 5: Calculate new training weights:

[0084]

[0085] Step 6: Calculate the final classifier:

[0086]

[0087] The transfer learning is trained in sequence to obtain a trained transfer learning model, and the trained transfer learning model is placed in the cloud server for subsequent identification of the wear degree of the engine piston ring and main bearing.

[0088] As a preferred embodiment, the cloud server is used to determine the degree of wear of the engine main bearing according to the blowby flow rate, the oil mist concentration, the engine speed and the engine torque sent by the machine-side monitoring box and through a trained transfer learning model. It is also used to: perform chi-square test, information gain ratio and correlation coefficient test on the received packaged data, identify target data in the packaged data with the goal of minimum redundancy of correlation coefficient and maximum correlation between information gain ratio and chi-square test result; perform dimensionality reduction and reconstruction on the target data to generate dimensionality reduction target data; identify outliers in the dimensionality reduction target data through a spatial clustering algorithm, and filter the outliers to obtain final target data.

[0089] Exemplarily, in this embodiment, the algorithms used for simplification and dimensionality reduction reconstruction of target data include linear discriminant analysis LDA (Linear Discriminant Analysis) dimensionality reduction, principal component analysis PAC (Principal Component Analysis) dimensionality reduction or autoencoding dimensionality reduction. In this embodiment, the spatial clustering algorithm includes the K-Means algorithm or the density-based noise application spatial clustering DBSCAN algorithm. Since the data collected in real time and the data in the laboratory environment are very different, in order to extract useful data, the chi-square test, information gain ratio and correlation coefficient test are used to identify useful target data; then the target data is simplified through the reduction algorithm, and the singular values ​​are identified and filtered through DBSCAN to obtain the final target data, so as to ensure the accuracy of the output of the engine piston ring and main bearing wear degree by inputting the data into the transfer learning model.

[0090] The cloud server is also used to input the blowby gas flow, the oil mist concentration, the engine speed and the engine torque in the final target data into the trained transfer learning model to obtain the engine main bearing fault degree output by the trained transfer learning model.

[0091] In some optional embodiments, the cloud service is also used to send the obtained engine piston ring and main bearing wear levels to a preset terminal and vehicle end via notification information, so that the user can know the engine status in time and perform engine maintenance.

[0092] The diesel engine crankcase active safety monitoring system provided in this embodiment adopts a modular design to divide the system into a blowby monitoring module, an oil and gas monitoring module, a machine-side monitoring box and a cloud server. The blowby monitoring module includes a variable pressure reduction port, a pressure stabilizing chamber, a venturi tube and a flow sensor arranged at the throat position of the venturi tube, which can realize accurate monitoring of the blowby flow of the engine crankcase. The oil and gas content detection module includes an oil mist concentration sensor installed at the throat position of the venturi tube, and the crankcase oil and gas state is identified by a photoelectric sensor; the machine-side test box is used to send information, and a heat insulation coating is provided to prevent high-temperature gas from damaging electronic components; the cloud service can be developed based on Nandou Cloud, and the transfer learning model based on the supervised learning algorithm analyzes and diagnoses the signals transmitted and packaged by the machine-side test box, and sends fault information of the wear degree of the engine piston ring and the main bearing to the user. The modular system design can improve the replaceability and versatility of the system, and calculate the wear degree of the engine main bearing in the cloud server, improve the computing efficiency, and reduce the data processing pressure on the vehicle end.

[0093] In a second aspect, an embodiment of the present application also provides a method for active safety monitoring of a diesel engine crankcase.

[0094] In one embodiment, referring to Figure 5 , Figure 5 This is the process intention of an embodiment of the diesel engine crankcase active safety monitoring method of this application. Figure 5 As shown, the diesel engine crankcase active safety monitoring method includes:

[0095] Monitor the blowby gas flow rate of the engine crankcase through the blowby gas monitoring module;

[0096] Monitoring the oil mist concentration of the engine crankcase through an oil and gas monitoring module;

[0097] The blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque are collected through a machine-side monitoring box and sent;

[0098] The wear degree of the engine piston ring and the main bearing is determined by the cloud server based on the blowby gas flow, the oil mist concentration, the engine speed and the engine torque sent by the machine-side monitoring box through the trained transfer learning model.

[0099] Further, in one embodiment, the blowby gas monitoring module includes a blowby gas monitoring pipeline and a flow sensor;

[0100] The blowby gas monitoring pipeline includes a pressure-expanding pipe, a pressure-stabilizing chamber and a venturi tube connected in sequence, and the pressure-expanding pipe is connected to the engine crankcase ventilation interface;

[0101] The flow sensor is arranged in the throat of the venturi tube.

[0102] Further, in one embodiment, the oil and gas monitoring module includes an oil mist concentration sensor;

[0103] The oil mist concentration sensor is arranged in the throat of the venturi tube.

[0104] Furthermore, in one embodiment, the machine-side monitoring box includes a collection module, a processing module, a transmission module, and a thermal insulation coating, and the blowby gas flow, the oil mist concentration, the engine speed, and the engine torque are collected and sent by the machine-side monitoring box, including:

[0105] collecting the blow-by gas flow rate, the oil mist concentration, the engine speed and the engine torque through the collection module;

[0106] Packaging the blow-by gas flow rate, the oil mist concentration, the engine speed and the engine torque through the processing module to generate packaged data;

[0107] Sending the packaged data to the cloud server through the transmitting module;

[0108] The collection module, the processing module and the emission module are wrapped by the thermal insulation coating.

[0109] Furthermore, in one embodiment, before determining the wear degree of the engine piston ring and the main bearing through the trained transfer learning model based on the blowby gas flow, the oil mist concentration, the engine speed and the engine torque sent by the engine side monitoring box through the cloud server, the further comprising:

[0110] Performing chi-square test, information gain ratio and correlation coefficient test on the received packaged data through the cloud server, with the goal of minimizing redundancy of correlation coefficient and maximizing correlation between information gain ratio and chi-square test result, to identify target data in the packaged data;

[0111] Performing a simplification dimensionality reduction reconstruction on the target data to generate dimensionality reduction target data;

[0112] The outliers in the dimension reduction target data are identified by a spatial clustering algorithm, and the outliers are filtered to obtain the final target data.

[0113] Further, in one embodiment, the wear degree of the engine piston ring and the main bearing is determined by the cloud server according to the blowby gas flow, the oil mist concentration, the engine speed and the engine torque sent by the engine side monitoring box through the trained transfer learning model, and further includes:

[0114] The blowby gas flow, the oil mist concentration, the engine speed and the engine torque in the final target data are input into the trained transfer learning model to obtain the wear degree of the engine piston ring and the main bearing.

[0115] Furthermore, in one embodiment, the method further includes:

[0116] The cloud server is used to store multiple operating condition data of the engine piston ring and the main bearing at different wear degrees under the first operating condition, the second operating condition and the third operating condition, the wear degree of the engine piston ring and the main bearing, the corresponding blowby gas flow rate and the oil mist concentration as source domain auxiliary data, wherein the operating condition data includes the engine load rate and the engine speed, and the wear degree of the engine piston ring and the main bearing includes no wear, slight wear, moderate wear and severe wear;

[0117] The engine under the fourth working condition, multiple working condition data of different wear degrees of the engine piston ring and the main bearing, the wear degree of the engine piston ring and the main bearing, the corresponding blowby gas flow rate and oil mist concentration are used as target domain data;

[0118] All the source domain auxiliary data and M data in the target domain data are used as training data, and all the target domain data are used as test data to train the transfer learning model to obtain the trained transfer learning model.

[0119] Among them, the functional implementation of each module in the above-mentioned diesel engine crankcase active safety monitoring method corresponds to the steps in the above-mentioned diesel engine crankcase active safety monitoring method embodiment, and its functions and implementation processes are no longer repeated here.

[0120] It should be noted that the serial numbers of the above-mentioned embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.

[0121] The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices. The terms "first", "second" and "third" are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit "first", "second" and "third" to different types.

[0122] In the description of the embodiments of the present application, "exemplary", "for example" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary", "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "for example" or "for example" is intended to present related concepts in a specific way.

[0123] In the description of the embodiments of the present application, unless otherwise specified, “ / ” means or, for example, A / B can mean A or B; the “and / or” in the text is merely a description of the association relationship of associated objects, indicating that three relationships may exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, “multiple” refers to two or more than two.

[0124] In some processes described in the embodiments of the present application, multiple operations or steps that appear in a specific order are included, but it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present application or in parallel, and the sequence number of the operation is only used to distinguish the different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in sequence or in parallel, and these operations or steps may be combined.

[0125] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD) as described above, and includes a number of instructions for a terminal device to execute the methods described in each embodiment of the present application.

[0126] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A diesel engine crankcase active safety monitoring system, characterized in that: The diesel engine crankcase active safety monitoring system comprises: A blowby gas monitoring module, which is used to monitor the blowby gas flow rate of the engine crankcase; An oil and gas monitoring module, which is used to monitor the oil mist concentration of the engine crankcase; A machine-side monitoring box, which is used to collect the blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque, and send them; The cloud server is used to determine the degree of wear of the engine piston ring and the main bearing according to the blowby gas flow, the oil mist concentration, the engine speed and the engine torque sent by the machine-side monitoring box through a trained transfer learning model.

2. The diesel engine crankcase active safety monitoring system according to claim 1, characterized in that: The blowby gas monitoring module includes a blowby gas monitoring pipeline and a flow sensor; The blowby gas monitoring pipeline includes a pressure-expanding pipe, a pressure-stabilizing chamber and a venturi tube connected in sequence, and the pressure-expanding pipe is connected to the engine crankcase ventilation interface; The flow sensor is arranged in the throat of the venturi tube.

3. The diesel engine crankcase active safety monitoring system according to claim 1, characterized in that: The oil and gas monitoring module includes an oil mist concentration sensor; The oil mist concentration sensor is arranged in the throat of the venturi tube.

4. The diesel engine crankcase active safety monitoring system according to claim 1, characterized in that: The machine-side monitoring box comprises: A collection module, which is used to collect the blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque; a processing module, which is used to package the blow-by gas flow rate, the oil mist concentration, the engine speed and the engine torque to generate packaged data; A transmitting module, used for sending the packaged data to the cloud server; A thermal insulation coating is used to wrap the collection module, the processing module and the emission module.

5. The diesel engine crankcase active safety monitoring system according to claim 4, characterized in that: The cloud server is also used for: Performing chi-square test, information gain ratio and correlation coefficient test on the received packaged data, with the goal of minimum redundancy of correlation coefficient and maximum correlation between information gain ratio and chi-square test result, to identify target data in the packaged data; Performing a simplification dimensionality reduction reconstruction on the target data to generate dimensionality reduction target data; The outliers in the dimension reduction target data are identified by a spatial clustering algorithm, and the outliers are filtered to obtain the final target data.

6. The diesel engine crankcase active safety monitoring system according to claim 5, characterized in that: The cloud server is also used for: The blowby gas flow, the oil mist concentration, the engine speed and the engine torque in the final target data are input into the trained transfer learning model to obtain the wear degree of the engine piston ring and the main bearing.

7. The diesel engine crankcase active safety monitoring system according to claim 1, characterized in that: The cloud server is also used for: The engine is respectively used as source domain auxiliary data for a plurality of operating condition data of different wear degrees of the engine piston ring and the main bearing under the first operating condition, the second operating condition and the third operating condition, the wear degree of the engine piston ring and the main bearing, the corresponding blowby gas flow rate and the oil mist concentration, wherein the operating condition data include the engine load rate and the engine speed, and the wear degree of the engine piston ring and the main bearing includes no wear, slight wear, moderate wear and severe wear; The engine under the fourth working condition, multiple working condition data of different wear degrees of the engine piston ring and the main bearing, the wear degree of the engine piston ring and the main bearing, the corresponding blowby gas flow rate and oil mist concentration are used as target domain data; All the source domain auxiliary data and M data in the target domain data are used as training data, and all the target domain data are used as test data to train the transfer learning model to obtain the trained transfer learning model.

8. A diesel engine crankcase active safety monitoring method implemented by the diesel engine crankcase active safety monitoring system according to any one of claims 1 to 7, characterized in that: The diesel engine crankcase active safety monitoring method comprises: Monitor the blowby gas flow rate of the engine crankcase through the blowby gas monitoring module; Monitoring the oil mist concentration of the engine crankcase through an oil and gas monitoring module; The blowby gas flow rate, the oil mist concentration, the engine speed and the engine torque are collected through a machine-side monitoring box and sent; The cloud server determines the degree of wear of the engine piston ring and the main bearing according to the blowby gas flow, the oil mist concentration, the engine speed and the engine torque sent by the machine-side monitoring box through the trained transfer learning model.

9. The diesel engine crankcase active safety monitoring method according to claim 8, characterized in that: The machine-side monitoring box includes a collection module, a processing module, a transmission module, and a thermal insulation coating. The blowby gas flow rate, the oil mist concentration, the engine speed, and the engine torque are collected and sent by the machine-side monitoring box, including: collecting the blow-by gas flow rate, the oil mist concentration, the engine speed and the engine torque through the collection module; Packaging the blow-by gas flow rate, the oil mist concentration, the engine speed and the engine torque through the processing module to generate packaged data; Sending the packaged data to the cloud server through the transmitting module; The collection module, the processing module and the emission module are wrapped by the thermal insulation coating.

10. The diesel engine crankcase active safety monitoring method according to claim 9, characterized in that: Before determining the wear degree of the engine piston ring and the main bearing through the trained transfer learning model based on the blowby gas flow, the oil mist concentration, the engine speed and the engine torque sent by the engine side monitoring box through the cloud server, the method further includes: Performing chi-square test, information gain ratio and correlation coefficient test on the received packaged data through the cloud server, with the goal of minimum redundancy of correlation coefficient and maximum correlation between information gain ratio and chi-square test result, identifying target data in the packaged data; Performing a simplification dimensionality reduction reconstruction on the target data to generate dimensionality reduction target data; The outliers in the dimension reduction target data are identified by a spatial clustering algorithm, and the outliers are filtered to obtain the final target data.

Images