Hydraulic oil health monitoring method and hydraulic oil health monitoring system

Abstract

The invention discloses a hydraulic oil health monitoring method. The method comprises the following steps: acquiring a real-time image of hydraulic oil flowing through a sampling pipeline by means of an imaging system, and detecting oil product parameters of the hydraulic oil by means of an oil product sensor; determining pollutant information based on the real-time image, wherein the pollutant information describes characteristics of pollutants in the hydraulic oil; and determining the current health condition of the hydraulic oil according to the pollutant information and the oil product parameters. According to the scheme, more comprehensive and accurate oil product parameters and particle pollutant information are obtained in a multi-sensor and multi-mode detection mode, so that the health state of the hydraulic oil can be determined more accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of construction machinery detection, and particularly relates to a method for monitoring the health of hydraulic oil and a corresponding hydraulic oil health monitoring system. Background Art

[0002] The hydraulic system is an important part of construction machinery. In the hydraulic system, hydraulic oil is used as the working fluid to transmit power or control the movement of various mechanical components. The working environment of construction machinery is usually very harsh. Particulate matters such as dust and debris will pollute the hydraulic system. Working environments such as high temperature and high pressure will cause the quality of the oil to decline. If left unchecked, it will eventually lead to the damage of hydraulic components such as pumps, cylinders, and valves. Therefore, it is necessary to monitor and evaluate the health status of the hydraulic oil in construction machinery in real time.

[0003] Currently, the detection means for the hydraulic oil of construction machinery generally realizes on-line contamination detection by installing a particle size sensor. The current mainstream particle size sensors mainly use the light extinction method for detection. However, the light extinction method cannot distinguish between bubbles and particles, so it will lead to inaccurate output of the contamination level, and it is also impossible to detect other parameters of the oil, such as viscosity, moisture, dielectric constant, etc.

[0004] Therefore, there is a need to further improve the existing hydraulic oil health monitoring methods and hydraulic oil health monitoring systems. Summary of the Invention

[0005] The present invention proposes a method for monitoring the health of hydraulic oil and a corresponding hydraulic oil health monitoring system, aiming to overcome one or more of the above technical problems and / or other technical problems in the prior art.

[0006] According to one aspect of the present invention, a method for monitoring the health of hydraulic oil is proposed. The method includes the following steps:

[0007] Obtaining a real-time image of the hydraulic oil flowing through the sampling pipeline by means of an imaging system and detecting the oil product parameters of the hydraulic oil by means of an oil product sensor;

[0008] Determining pollutant information based on the real-time image, where the pollutant information describes the characteristics of particulate pollutants in the hydraulic oil;

[0009] Determining the current health status of the hydraulic oil according to the pollutant information and the oil product parameters.

[0010] According to another aspect of the present invention, a hydraulic oil health monitoring system is proposed, which has a signal acquisition module and a signal processing module, wherein the signal acquisition module includes an oil sensor and an imaging system, the imaging system includes a light source and a camera relatively arranged on both sides of the sampling pipe path, and the signal processing module is signal-connected to the signal acquisition module and is designed to implement the method according to the present invention.

[0011] In the hydraulic oil health monitoring method proposed in the present invention, pollutant information and multiple parameters of hydraulic oil are obtained by means of an imaging system and an oil sensor. Pollutant information refers to the parameters of particulate pollutants in hydraulic oil, such as the number of particles, particle size, particle type, particle shape, etc. The pollutant information also includes the pollutant level determined by these parameters; the parameters of hydraulic oil include the temperature of hydraulic oil, the viscosity of hydraulic oil, the moisture in hydraulic oil, the dielectric constant of hydraulic oil, etc. By analyzing and evaluating the various parameters and pollutant information obtained, the health status of hydraulic oil can be more accurately evaluated based on multiple characteristic values. In addition, the method according to the present invention can also manage and predict the life cycle of hydraulic oil, visualize the entire life cycle of hydraulic oil, and promptly inform customers of the current health status of hydraulic oil, thereby informing users in advance when hydraulic oil needs to be replaced. The method of the present invention can also trace the source according to the particle characteristics, locate the key components of wear, so that users can intervene and maintain in advance when necessary. At the same time, by accumulating fault information, a fault information library of target components can be formed, which can facilitate R&D personnel to optimize component design in a targeted manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0013] Figure 1 is a flow chart of an implementation form of the method according to the present invention;

[0014] Figure 2 is a partial schematic diagram of a hydraulic oil health monitoring system according to the present invention;

[0015] Figure 3 It is a schematic diagram of the architecture of a preferred implementation form of the hydraulic oil health monitoring system according to the present invention. DETAILED DESCRIPTION

[0016] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. In the figures, for the sake of clarity, the dimensions of some elements may be exaggerated or distorted. In the figures, like reference numerals denote like or similar structures and thus their detailed description will be omitted.

[0017] In addition, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of the present invention. However, those skilled in the art will realize that the technical solutions of the present invention may be practiced without one or more of the specific details, or other methods, elements, etc. may be employed. In other instances, well-known structures, methods, or operations are not shown or described in detail to avoid obscuring aspects of the present invention.

[0018] In this application, "hydraulic oil" should be understood as the hydraulic oil that circulates in the hydraulic circuit, which is a mixture of hydraulic oil that may contain one or more contaminants such as particulate contaminants, moisture, air bubbles, etc. For simplicity, in this application, the mixture of hydraulic oil that may contain one or more contaminants such as particulate contaminants, moisture, air bubbles, etc. is simply referred to as "hydraulic oil".

[0019] Figure 1 is a flowchart of a preferred implementation form of the hydraulic oil health monitoring method according to the present invention. The method includes the following steps:

[0020] S1: Obtain a real-time image of the hydraulic oil flowing through the sampling pipeline by means of a camera and detect the oil product parameters of the hydraulic oil by means of an oil product sensor;

[0021] S2: Determine the contaminant information based on the real-time image, and this contaminant information describes the characteristics of the particulate contaminants in the hydraulic oil;

[0022] S3: Determine the current health status of the hydraulic oil according to the contaminant information and the oil product parameters.

[0023] The sampling pipeline is connected to the hydraulic oil circulation pipeline. That is to say, at least part of the hydraulic oil will flow through the sampling pipeline during the circulation process. Generally speaking, one sampling pipeline can achieve the sampling function. However, it is also possible to consider setting up multiple sampling pipelines. Multiple sampling pipelines can be considered to be set at different positions of the hydraulic oil circulation pipeline, so that the hydraulic oil can be sampled at different positions of the entire hydraulic oil circulation pipeline, thereby taking into account the influence of factors such as different flow rates, pressures, and position heights of the hydraulic oil. For example, there may be more large particles at positions with slower flow rates and lower positions. Imaging systems are arranged on both radial sides of each sampling pipeline, specifically a light source and a camera respectively arranged oppositely. The real-time image of the hydraulic oil flowing through the sampling pipeline can be collected through the camera. Here, the light source is preferably an LED light source, and the camera is preferably a micro camera. The optoelectronic signals collected by the micro camera are processed by the signal processing module and converted into corresponding image data.

[0024] Determine the pollutant information based on the real-time image, and this pollutant information describes the characteristics of the particulate pollutants in the hydraulic oil. The image information collected by the camera is further processed by the signal processing module. The so-called further processing can include, for example, preprocessing such as noise reduction and filtering, and also includes image recognition, such as recognizing particulate pollutants and bubbles in the image. The pollutant information of the particulate pollutants contained in the hydraulic oil can be determined through image recognition. The pollutant information includes, for example, the size, quantity, type, shape, etc. of the particulate pollutants, as well as the pollutant level (ISO 4406:2021) determined based on these characteristics.

[0025] Common oil sensors can be, for example, capacitive, resistive, or acoustic wave sensors. Of course, other appropriate sensors such as optoelectronic sensors can also be considered. The oil parameters collected by the oil sensor include, for example, parameters such as the temperature, viscosity, moisture content, and dielectric constant of the hydraulic oil. The current health status of the hydraulic oil can be determined based on the pollutant information and the oil parameters. The correspondence between the pollutant information and the oil parameters and the health status of the hydraulic oil can be determined through an algorithm model, such as a neural network algorithm. Of course, it is also possible to consider using a table or a family of characteristic curves, etc. to determine the health status of the hydraulic oil.

[0026] According to the health status of the hydraulic oil, the hydraulic oil can be divided into a healthy state, a sub-healthy state, and a dangerous state. Thus, for example, a hydraulic oil health indicator can be provided to the user on the display interface of the construction machinery. For example, on the display screen in the cab of the construction machinery, the healthy state is represented by green, the sub-healthy state is represented by yellow, and the dangerous state is represented by red. An alarm device, such as an alarm light, can also be set in the cab of the construction machinery, and this alarm device can visually present the current state of the hydraulic oil to the user in different colors and / or sounds.

[0027] The recognition and processing of images can be achieved with the help of artificial intelligence algorithms, such as neural networks and / or deep learning. There are already various artificial intelligence algorithms for image recognition, such as convolutional neural network algorithms (CNN), region-based convolutional neural network (RCNN), YOLO (You Only Look Once) algorithm, SSD (Single Shot Multi Box Detector) algorithm, etc. Herein, preferably, the convolutional neural network algorithm (CNN) can be used to process the real-time image of hydraulic oil. The convolutional neural network algorithm (CNN) is a mainstream algorithm in the field of image recognition and is very mature and reliable. A convolutional neural network usually consists of three parts: a convolutional layer, a pooling layer, and a fully connected layer. The convolutional layer is responsible for extracting local and global features in the image; the pooling layer is used to significantly reduce the number of parameters, that is, dimensionality reduction; the fully connected layer is used to process the "compressed image information" and output the result. In the current application scenario, the main steps of the convolutional neural network algorithm (CNN) include image preprocessing, binarization, feature extraction, classification, pollution level mapping, etc.

[0028] Furthermore, the whole life cycle management of hydraulic oil can also be carried out according to the pollutant information, oil product parameters, and the current health status of the hydraulic oil. The whole life cycle management of hydraulic oil includes the prediction of the remaining life of hydraulic oil and / or the tracing of pollutants and the fault diagnosis of components based on the pollutant information of the hydraulic oil.

[0029] The prediction of the remaining life of hydraulic oil can be achieved, for example, through a hydraulic oil remaining life prediction model. The hydraulic oil remaining life prediction model is preferably an algorithm model based on long short-term memory neural network (LSTM, Long Short-Term Memory). The long short-term memory neural network (LSTM) belongs to a type of recurrent neural network (RNN) and is suitable for processing and predicting important events with relatively long intervals and delays in time series. The monitoring of hydraulic oil is a long-term accumulation process and is more dependent on long-term data. Therefore, it is very beneficial to use the long short-term memory neural network (LSTM) here. The input parameters of the hydraulic oil remaining life prediction model include, for example, the pollutant information, oil product parameters, the current health status of the hydraulic oil, the time of the last hydraulic oil change, etc. For a specific type of hydraulic oil, the theoretical life or the oil change guidance time interval is usually known. Using the theoretical life or the oil change guidance time interval of this specific type of hydraulic oil as the input parameter of the hydraulic oil remaining life prediction model is particularly beneficial for the prediction of the remaining life of hydraulic oil. To improve the accuracy of the prediction of the remaining life of hydraulic oil, the hydraulic oil remaining life prediction model can be combined with a deep learning algorithm (DL).

[0030] The sources of contaminants in hydraulic oil mainly include residual contaminants inside the system, contaminants invading the system from the outside, and contaminants generated inside the system. Residual contaminants inside the system mainly refer to the contaminants existing in the hydraulic system before filling the hydraulic oil, mainly the superfluous substances that were not cleared cleanly and remained inside the system during the manufacturing process of hydraulic components, such as molding sand of castings, welding slag, chips, iron sheets, cotton yarn heads, peeling flakes of paint coatings, seal fragments, dust, etc. Contaminants invading the system from the outside refer to the contaminants invading the system from outside the hydraulic system during the assembly, use, and maintenance of the system. For example, impurities such as air, moisture, dust, and sand grains in the environment may enter the hydraulic oil through gaps such as the breather hole of the fuel tank, the piston rod, and the filling port. In addition, when the hydraulic system is working, friction occurs during the operation of parts, and the resulting residual rust, wear particles, corrosion products, etc. may enter the hydraulic oil and cause pollution.

[0031] Preferably, the contaminants can be traced back and the fault diagnosis of components can be carried out according to the contaminant information. The traceability can be carried out, for example, through a traceability model trained with empirical data or experimental data. The traceability model is preferably based on the convolutional neural network algorithm (CNN). By analyzing information such as the size, type, quantity, and shape of particulate contaminants in the traceability model, it can be inferred where the contaminants come from, that is, the traceability. The worn components can be located through the contaminant traceability.

[0032] It is also beneficial to judge the wear degree of multiple components in the hydraulic system according to the traceability of contaminants to carry out the fault diagnosis of the components, predict the service life of the relevant components, and give maintenance or replacement prompts when necessary. For example, it is known that a certain type of particulate contaminant in the hydraulic oil is mainly generated by the wear of the meshing gears in the oil pump. If the quantity of this type of particulate contaminant in the hydraulic oil reaches a certain threshold or the average size of this type of particulate contaminant exceeds a certain threshold, it means that the gears are severely worn and need to be repaired or replaced.

[0033] In addition, the wear information of the components in the hydraulic system and the information of its change over time are stored in the database. That is to say, the wear information or fault information of the target components during the entire life cycle from new use to failure or fault will be recorded in the database. By accumulating the fault or wear information of these components, a fault / wear information library of the target components can be formed, which can facilitate the R & D personnel to optimize the component design targeted. For example, if it is concluded based on big data analysis that the wear rate of a certain component is significantly higher than that of other components in the hydraulic system, it means that in future R & D designs, the component should be optimized to improve its wear resistance and service life.

[0034] Preferably, the predicted remaining life of the hydraulic oil and / or the currently evaluated health state of the hydraulic oil are sent to a mobile terminal or a remote control terminal and presented to the user in a visual form or an active warning is given. This is particularly advantageous for driverless construction machinery or construction machinery that can be remotely controlled. For example, a remote operator can view the currently evaluated health state and the predicted remaining life of the hydraulic oil in a visual form on a mobile terminal or a remote control terminal, such as a smart phone, a PAD or a remote management device for driverless construction machinery.

[0035] Figure 2 A partial view of the hydraulic oil health monitoring system according to the present invention is schematically shown, specifically, a schematic diagram of a signal acquisition module for sampling the hydraulic oil. In Figure 2 , the sampling pipeline 1 is shown by an arrow, where the arrow indicates the flow direction of the hydraulic oil. Air bubbles 2 and other particulate contaminants 3 are contained in the hydraulic oil. For clarity, only one air bubble 2 and one particulate contaminant 3 are schematically shown in the figure respectively, but obviously more air bubbles and particulate contaminants may exist in actual situations. The sampling pipeline 1 is designed as a pipeline that is at least partially transparent, which may have the shape of a cylindrical pipeline or may be designed into a suitable shape such as a cuboid. At least part of the hydraulic oil will flow through the sampling pipeline during the circulation process. It is also possible to consider arranging multiple sampling pipelines at multiple positions.

[0036] On one side in the radial direction of the sampling pipeline 1, an LED light source 4 is provided, and this LED light source is preferably integrated on an LED board 5. On the LED board 5, there are circuits for supplying power to the LED light source 4 and controlling the LED light source 4. Opposite to the LED light source 4, on the other side in the radial direction of the sampling pipeline 1, a camera 6 is provided. The LED light source 4 provides auxiliary light. The camera 6 is designed as a micro camera integrated in a circuit board 7. The circuit board 7 is preferably an image signal processing circuit board including an Image Signal Processor (ISP). The image signal processing circuit board 7 converts the optoelectronic signals collected by the camera 6 into image data. At least one oil product sensor 8 is also provided at the sampling pipeline 1. Although only one oil product sensor 8 is schematically shown in the figure, actually, it is also possible to consider setting multiple oil product sensors with different functions or one or more oil product sensors integrated with multiple functions. The oil product sensor 8 is arranged to detect the oil product parameters of the hydraulic oil flowing through the sampling pipeline 1, such as parameters like the temperature, viscosity, moisture content, dielectric constant, etc. of the hydraulic oil. The detection end of the oil product sensor 8 especially extends into the sampling pipeline 1, and the oil product sensor 8 is also preferably integrated with the other end opposite to the detection end in the LED board 5 and powered by the LED board 5. The oil product sensor 8 is also preferably signal-connected to the image signal processing circuit board 7. That is to say, the signal of the oil product sensor 8 can be processed or pre-processed by the image signal processing circuit board 7. By integrating the oil product sensor 8 and the LED light source 4 on the LED board 5 and arranging the micro camera integratedly in the image signal processing circuit board 7, a compact layout is achieved, making power supply, signal transmission, and control more reliable. Advantageously, the aforementioned algorithms, such as the aforementioned image recognition algorithms, including image pre-processing, binarization, feature extraction, classification, and contamination level mapping algorithms, are built into the image signal processing circuit board 7.

[0037] Figure 3 The schematic diagram of the architecture of a preferred implementation form of the hydraulic oil health monitoring system according to the present invention is shown. Generally speaking, the hydraulic oil health monitoring system includes a signal acquisition module 12, a signal processing module 13, and a cloud platform 10. Although the cloud platform 12 and the programs deployed on the cloud platform are shown here, obviously, it is also possible to consider abandoning the cloud platform and adopting a local computing mode, that is, all programs are deployed locally, such as on the computing device of a construction machine and relevant calculations and storage are performed on the computing device of the construction machine. However, it should be pointed out that there are significant advantages in terms of computing power and storage space when using the cloud platform 12.

[0038] For the detailed structure of the signal acquisition module 12, reference can be made to, for example, Figure 2The illustrated embodiment mainly includes an oil sensor and an imaging system, where the imaging system includes a light source and a camera. The signal processing module 13 can be a microprocessor integrated in the image signal processing (ISP) circuit board 7, or a controller separately and independently configured from the signal acquisition module 12. Of course, it is also possible to consider using other computing devices on the construction machinery as the signal processing module. The signal processing module 13 is signal-connected to the signal acquisition module 12, specifically directly or indirectly signal-connected to the oil sensor and the camera of the signal acquisition module 12, obtains signals from the oil sensor and the camera, and processes the obtained signals, for example, processes them through the aforementioned built-in algorithms.

[0039] The signal processing module 13 is signal-connected to the control device 9 of the construction machinery through the CAN bus. In this construction machinery, there is a display device and / or an alarm light for indicating the health status of the hydraulic oil. The result processed by the signal processing module 13 can be transmitted to the control device 9 of the construction machinery through the CAN bus of the construction machinery, such as the control device in the cab. A display device can be set in the cab, and the real-time status of the hydraulic oil, the predicted remaining life, the predicted oil change time, etc. can be presented to the user through the display interface. Preferably, different health statuses of the hydraulic oil are represented by different colors. For example, a green color represents a healthy status, a yellow color represents a sub-healthy status, and a red color represents a dangerous status. An alarm device, such as an alarm light, can also be set in the cab of the construction machinery. This alarm device can visually present the current status of the hydraulic oil to the user in different colors and / or sounds.

[0040] On the other hand, the data from the signal acquisition module 12 and the signal processing module 13 can be transmitted to a remote server or a cloud platform 10 through a remote communication network, such as an Ethernet or a mobile communication network. These data are stored in the database 11 on the remote server or the cloud platform 10. A database 11, a hydraulic oil remaining life prediction model, and a traceability model for pollutants are deployed on the remote server or the cloud platform. Both the hydraulic oil remaining life prediction model and the traceability model are artificial intelligence analysis models based on neural network algorithms. For example, the hydraulic oil remaining life prediction model is an algorithm model based on a long short-term memory neural network, and the traceability model is an algorithm model based on a convolutional neural network. The traceability model can be trained with empirical data or experimental data.

[0041] Based on the data from the signal acquisition module 12 and the signal processing module 13, the remaining life of the hydraulic oil can be predicted through the hydraulic oil remaining life prediction model. The prediction based on the remaining life can timely remind the user to replace the hydraulic oil to avoid damage to components caused by excessive contamination of the hydraulic oil. Here, not only the pollutant information contained in the image of the hydraulic oil is considered, but also multiple parameters detected by the oil quality sensor are considered to evaluate the health status of the hydraulic oil and predict the remaining life of the hydraulic oil. On the one hand, through the multi-sensor and multi-modal detection method, higher detection accuracy can be obtained. On the other hand, based on the collected multi-modal, multi-feature, and multi-sensor information, the health status of the hydraulic oil can be more accurately evaluated and the remaining life of the hydraulic oil can be predicted. Preferably, the predicted remaining life and / or the evaluated health status are sent to the control device 9 of the construction machinery and displayed to the user in a visual form on the display interface of the construction machinery or actively give an early warning. The predicted remaining life and / or the evaluated health status can also be sent to a mobile terminal or a remote control terminal, such as the smart phone, PAD of the construction machinery driver or administrator, or the remote management device for driverless construction machinery and displayed to the user in a visual form or actively give an early warning.

[0042] All pollutant information, health status information of the hydraulic oil and their data changing with time are stored in the database. Different key components will generate different types of particles when worn. By deeply mining this particulate pollutant information and with the help of a neural network model, the wear status of these components can be indirectly obtained, thereby realizing pollutant traceability and positioning of worn components. The particulate pollutant information includes, for example, particle size, type, shape, quantity, etc. The information of the hydraulic oil in different construction machinery of the same type will be collected and stored. Thus, big data on the wear of components in construction machinery of the same type can be obtained. Based on this big data on component wear, each component can be diagnosed and analyzed, so as to optimize the design of easily damaged components during subsequent R & D to improve their service life; or redesign components that are very difficult to damage, for example, use cheaper materials and reduce their service life without affecting normal use to reduce costs.

[0043] Overall, according to the solution of the present invention, through multi-sensor and multi-modal detection methods, more comprehensive and accurate oil product parameters and particulate pollutant information are obtained. Based on this, the current health status of the hydraulic oil can be determined more accurately. In addition, with the help of an algorithm model based on Long Short-Term Memory (LSTM) neural network, using oil product parameters and particulate pollutant information as input parameters, the remaining life of the hydraulic oil can be reliably predicted, thereby enabling active warning and timely reminder to replace the hydraulic oil to avoid component damage. The pollutant information and health status information during the entire life cycle of the hydraulic oil are collected and saved, and presented in a visual form. On the other hand, through the traceability model, the worn components in the hydraulic system are located and / or the wear degree of the relevant components is judged based on the pollutant information for fault diagnosis of the components, predict the life of the relevant components and issue maintenance or replacement prompts when necessary. Preferably, the size, type, shape, quantity and other information of particulate pollutants can be analyzed through a traceability model based on CNN, trace the source and locate the position of the worn components, and the pollutant can be traced and the components can be fault diagnosed. In addition, all monitoring and predictions are updated in real time, so the latest true state of the hydraulic oil can be reflected and thus the whole life cycle management can be achieved.

[0044] Those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the content disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include known common general knowledge or conventional technical means in the technical field not disclosed by the present invention. The specification and examples are only to be considered as exemplary, and the true scope and spirit of the present invention are pointed out by the appended claims.

Claims

1. A method for health monitoring of hydraulic oil, the method comprising the following steps: Obtaining a real-time image of the hydraulic oil flowing through a sampling pipeline by means of an imaging system and detecting the oil product parameters of the hydraulic oil by means of an oil product sensor; Determining pollutant information based on the real-time image, the pollutant information describing the characteristics of pollutants in the hydraulic oil; Determining the current health status of the hydraulic oil according to the pollutant information and the oil product parameters.

2. The method according to claim 1, characterized in that The oil product parameters include the temperature, viscosity, moisture content and dielectric constant of the hydraulic oil; the pollutant information includes the size, type, quantity, shape of particulate pollutants and the pollutant level.

3. The method according to claim 1 or 2, characterized in that, Dividing the hydraulic oil into a healthy state, a sub-healthy state and a dangerous state according to the current health status of the hydraulic oil, wherein the current health status of the hydraulic oil is presented to the user in different colors and / or sounds.

4. The method according to claim 1 or 2, characterized in that, The image information collected by the imaging system is further processed, and the further processing includes preprocessing in the form of noise reduction and filtering and image recognition.

5. The method according to claim 4, wherein The image recognition algorithm includes image preprocessing, binarization, feature extraction, classification, and pollution degree level mapping.

6. The method according to claim 1 or 2, characterized in that, The method further includes the step of performing full life cycle management of the hydraulic oil according to the pollutant information, the oil product parameters and the current health status of the hydraulic oil.

7. The method according to claim 6, wherein The full life cycle management of the hydraulic oil includes predicting the remaining life of the hydraulic oil and / or tracing the source of pollutants and diagnosing the faults of components based on the pollutant information of the hydraulic oil.

8. The method according to claim 7, characterized in that, Predicting the remaining life of the hydraulic oil through a hydraulic oil remaining life prediction model, wherein the hydraulic oil remaining life prediction model is an algorithm model based on a long short-term memory neural network.

9. The method according to claim 7, wherein Tracing the source of pollutants through a tracing model trained with empirical data or experimental data, wherein the tracing model is an algorithm model based on a convolutional neural network.

10. The method according to claim 7, wherein Locating the worn components in the hydraulic system according to the pollutant tracing and / or judging the wear degree of relevant components based on the pollutant information to diagnose the faults of the components, predicting the life of the relevant components and giving a maintenance or replacement prompt when necessary.

11. The method according to claim 7, wherein Sending the predicted remaining life of the hydraulic oil and / or the evaluated health status of the hydraulic oil to a mobile terminal or a remote control terminal and presenting it to the user in a visual form or giving an active warning.

12. A hydraulic oil health monitoring system, the hydraulic oil health monitoring system having a signal acquisition module and a signal processing module, wherein, The signal acquisition module includes an oil product sensor and an imaging system, characterized in that the imaging system includes a light source and a camera arranged oppositely on both sides in the radial direction of the sampling pipeline, and the signal processing module is signal-connected to the signal acquisition module and is designed to implement the method according to any one of claims 1 to 11.

13. The hydraulic oil health monitoring system according to claim 12, wherein The camera is designed as a micro camera integrated in an image signal processing circuit board, and the signal processing module is designed as a microprocessor integrated in the image signal processing circuit board.

14. The hydraulic oil health monitoring system according to claim 12, characterized in that, The camera is designed as a micro camera, and the signal processing module is designed as a microprocessor independently formed relative to the signal acquisition module.

15. The hydraulic oil health monitoring system according to claim 12, wherein, The signal processing module is signal-connected to the control device of the construction machinery through a CAN bus, and a display device and / or an alarm lamp for indicating the health status of the hydraulic oil are provided in the construction machinery.

16. The hydraulic oil health monitoring system according to any one of claims 12 to 15, characterized in that, The light source is an LED light source, wherein the oil product sensor and the LED light source are integrated in an LED board.

17. The hydraulic oil health monitoring system according to any one of claims 12 to 15, characterized in that, It further includes a cloud platform or a remote server, on which a database, a hydraulic oil remaining life prediction model and a traceability model for pollutants are deployed. The signal acquisition module and the signal processing module are communicatively connected to the cloud platform or the remote server via an Ethernet or a mobile communication network.

18. The hydraulic oil health monitoring system according to claim 17, characterized in that, The cloud platform or the remote server is communicatively connected to the control device of the construction machinery or the mobile terminal of the user.

19. The hydraulic oil health monitoring system according to claim 17, characterized in that, The worn parts in the hydraulic system are located through the traceability model and / or the wear degree of the relevant parts is judged based on the pollutant information to conduct fault diagnosis on the parts, predict the life of the relevant parts and issue a maintenance or replacement prompt when necessary.

Images