An oil fluid monitoring method, system, electronic device and storage medium

Abstract

The present invention relates to the field of data processing, and in particular to an oil fluid monitoring method, system, electronic device and storage medium. The method includes: obtaining in real time the oil fluid monitoring data collected by an online monitoring platform; mapping the oil fluid monitoring data into the monitoring nodes corresponding to a virtual monitoring space according to the production links to which different equipment components belong, so as to obtain an oil fluid monitoring network; the virtual monitoring space is constructed based on the type and production link to which the target equipment belongs; based on the historical operation data and the oil fluid monitoring data of the target equipment, simulating the future change trends of each monitoring node in the oil fluid monitoring network to obtain real-time prediction data; based on the real-time prediction data, constructing a visualization model of the oil fluid monitoring network so that a user can monitor in real time the operation conditions of each production link in the target equipment. This method can realize the full-equipment automatic monitoring of the oil fluid and improve the monitoring efficiency.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an oil fluid monitoring method, system, electronic device and storage medium. Background Art

[0002] With the in-depth development of intelligent manufacturing, the predictive maintenance technology for the firmware part of manufacturing equipment has been gradually improved. In order to further detect faults, there is an urgent need for the detection and diagnosis of oil fluid.

[0003] Existing oil fluid monitoring often adopts manual monitoring methods, which require technical personnel familiar with the operation status of the equipment to conduct manual inspections. During the existing oil fluid monitoring process, technical personnel mainly carry out their work through regular inspections, going to the equipment site to take oil fluid samples for testing at fixed cycles such as daily or weekly. The monitoring cycle is long and the response speed is slow. In addition to using professional equipment, they also use empirical judgment methods to preliminarily judge the state of the oil fluid by observing the color and turbidity of the oil fluid, smelling its odor, and observing its viscosity. And further use test papers to detect the pH value of the oil fluid and use simple tools such as simple viscometers to measure the fluidity of the oil fluid to assist in the detection. Obviously, the existing oil fluid monitoring methods have low monitoring efficiency and slow fault diagnosis response, seriously affecting the operation efficiency of each production link. Therefore, there is an urgent need to design a new technical solution to overcome at least one of the above technical problems. Summary of the Invention

[0004] In view of the technical problems existing in the existing oil fluid monitoring technology, the present invention provides an oil fluid monitoring method, system, electronic device and storage medium to solve the technical problems of low monitoring efficiency and slow fault diagnosis response in the existing oil fluid monitoring technology, which seriously affect the operation efficiency of each production link, and realize the full-equipment automatic monitoring of oil fluid and improve the monitoring efficiency.

[0005] In a first aspect, an embodiment of the present invention provides an oil fluid monitoring method, which includes:

[0006] Obtain in real time the oil fluid monitoring data collected by the online monitoring platform; the online monitoring platform at least includes: multi-source sensors deployed in the target equipment and an oil fluid data monitoring station connected to the multi-source sensors; the multi-source sensors at least include: an oil fluid temperature sensor, a pressure sensor, and a particle counter;

[0007] Map the oil fluid monitoring data to the monitoring nodes corresponding to the virtual monitoring space according to the production links to which different equipment components belong, to obtain an oil fluid monitoring network; the virtual monitoring space is constructed based on the type and production link to which the target equipment belongs;

[0008] Based on the historical operation data of the target device and the oil fluid monitoring data, simulate the future change trends of each monitoring node in the oil fluid monitoring network to obtain the real-time prediction data corresponding to the oil fluid monitoring network; the real-time prediction data at least includes: the change characteristics of the oil fluid data of each monitoring node;

[0009] Based on the real-time prediction data, construct a visualization model of the oil fluid monitoring network so that the user can monitor the operation conditions of each production link in the target device in real time.

[0010] In a second aspect, an embodiment of the present invention provides an oil fluid monitoring system, which includes the following units, where,

[0011] An acquisition unit, configured to acquire in real time the oil fluid monitoring data collected by an online monitoring platform; the online monitoring platform at least includes: multi-source sensors deployed in the target device and an oil fluid data monitoring console connected to the multi-source sensors; the multi-source sensors at least include: an oil fluid temperature sensor, a pressure sensor, and a particle counter;

[0012] A construction unit, configured to map the oil fluid monitoring data into the monitoring nodes corresponding to the virtual monitoring space according to the production links to which different device components belong, to obtain an oil fluid monitoring network; the virtual monitoring space is constructed based on the type and production links of the target device;

[0013] A prediction unit, configured to simulate the future change trends of each monitoring node in the oil fluid monitoring network based on the historical operation data of the target device and the oil fluid monitoring data to obtain the real-time prediction data corresponding to the oil fluid monitoring network; the real-time prediction data at least includes: the change characteristics of the oil fluid data of each monitoring node;

[0014] A display unit, configured to construct a visualization model of the oil fluid monitoring network based on the real-time prediction data so that the user can monitor the operation conditions of each production link in the target device in real time.

[0015] In a third aspect, an embodiment of the present invention provides an electronic device, which includes:

[0016] At least one processor, a memory, and an input / output unit;

[0017] Wherein, the memory is used to store a computer program, and the processor is used to call the computer program stored in the memory to execute the oil fluid monitoring method in the first aspect.

[0018] In a fourth aspect, a computer-readable storage medium is provided, which includes instructions. When the instructions run on a computer, the computer is made to execute the oil fluid monitoring method in the first aspect.

[0019] The beneficial effects of the present invention are as follows: A method, system, electronic device and storage medium for oil fluid monitoring are provided. In this technical solution, oil fluid monitoring data collected by an online monitoring platform is obtained in real time; the online monitoring platform at least includes: multi-source sensors deployed in a target device and an oil fluid data monitoring console connected to the multi-source sensors; the multi-source sensors at least include: an oil fluid temperature sensor, a pressure sensor, and a particle counter; according to the production links to which different device components belong, the oil fluid monitoring data is mapped into monitoring nodes corresponding to a virtual monitoring space to obtain an oil fluid monitoring network; the virtual monitoring space is constructed based on the type and production link of the target device; based on the historical operation data of the target device and the oil fluid monitoring data, the future change trends of each monitoring node in the oil fluid monitoring network are simulated to obtain real-time prediction data corresponding to the oil fluid monitoring network; the real-time prediction data at least includes: the change characteristics of the oil fluid data of each monitoring node; based on the real-time prediction data, a visualization model of the oil fluid monitoring network is constructed to enable a user to monitor the operation conditions of each production link in the target device in real time. This technical solution can solve the technical problems in traditional oil fluid monitoring technologies, such as low monitoring efficiency, slow fault diagnosis response, and seriously affecting the operation efficiency of each production link, realize full-device automatic monitoring of oil fluid, improve monitoring efficiency and production efficiency, and assist in improving the fault diagnosis response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flowchart of a method for oil fluid monitoring according to an embodiment of the present invention;

[0021] Figure 2 is a structural schematic diagram of a system for oil fluid monitoring according to an embodiment of the present invention;

[0022] Figure 3 is a structural schematic diagram of an electronic device according to an embodiment of the present invention;

[0023] Figure 4 is a structural schematic diagram of a medium device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0025] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined.

[0026] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to implement and use the present invention. In the following description, details are set forth for purposes of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be implemented without the use of these specific details. In other instances, well-known structures and processes are not described in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0027] An embodiment of the present invention provides an oil fluid monitoring method, system, electronic device, and storage medium. In this technical solution, oil fluid monitoring data collected by an online monitoring platform is obtained in real time; the online monitoring platform at least includes: a multi-source sensor deployed in a target device and an oil fluid data monitoring console connected to the multi-source sensor; the multi-source sensor at least includes: an oil fluid temperature sensor, a pressure sensor, and a particle counter; according to the production links to which different device components belong, the oil fluid monitoring data is mapped to monitoring nodes corresponding to a virtual monitoring space to obtain an oil fluid monitoring network; the virtual monitoring space is constructed based on the type and production link of the target device; based on the historical operation data of the target device and the oil fluid monitoring data, the future change trends of each monitoring node in the oil fluid monitoring network are simulated to obtain real-time prediction data corresponding to the oil fluid monitoring network; the real-time prediction data at least includes: the change characteristics of the oil fluid data of each monitoring node; based on the real-time prediction data, a visualization model of the oil fluid monitoring network is constructed to enable a user to monitor the operation conditions of each production link in the target device in real time.

[0028] In the embodiments of the present invention, first, the oil fluid monitoring data is obtained in real time through an online monitoring platform. Multi-source sensors (such as oil fluid temperature sensors, pressure sensors, particle counters, etc.) can continuously collect data, and the data can be transmitted to the oil fluid data monitoring console in real time. This real-time collection and transmission method greatly improves the timeliness and frequency of data acquisition compared with the traditional manual regular data collection, enabling the oil fluid state information to be grasped at the first time. According to the production links to which different equipment components belong, the oil fluid monitoring data is mapped to the corresponding monitoring nodes in the virtual monitoring space, and an oil fluid monitoring network is automatically constructed. This process does not require manual collation and association of data, reducing the time and energy costs of manual operations, realizing the rapid and effective integration of data, and thus improving the overall monitoring efficiency.

[0029] Second, based on the historical operation data of the target equipment and the oil fluid monitoring data obtained in real time, the future change trends of each monitoring node in the oil fluid monitoring network are simulated to obtain real-time prediction data, which includes the oil fluid data change characteristics of each monitoring node. In this way, potential changes and abnormal trends in the oil fluid state can be discovered in advance, and early warnings can be issued before the failure occurs or is in its infancy, rather than waiting until the failure has significantly affected the operation of the equipment before diagnosis. A visualization model of the oil fluid monitoring network is constructed, and users can intuitively see the operation conditions of each production link in the target equipment. When an abnormality occurs, users can quickly locate the monitoring node where the problem lies, and combined with the real-time prediction data and the visualization display, more quickly judge the cause of the failure and the possible scope of influence, so as to make a faster response and take corresponding maintenance or adjustment measures.

[0030] Third, multi-source sensors are deployed in the target equipment, covering the monitoring of multiple key parameters such as oil fluid temperature, pressure, and particles, realizing the comprehensive monitoring of the oil fluid state. And these sensors automatically collect data without manual intervention, ensuring the continuity and stability of monitoring. From the data collection, transmission, to the construction of the oil fluid monitoring network, the generation of real-time prediction data, and the construction of the visualization model, the whole process has achieved automation. Reducing the influence of human factors on the monitoring results, improving the accuracy and reliability of monitoring, and truly realizing the full-equipment automatic monitoring of the oil fluid. Through timely fault warnings and rapid fault diagnosis responses, it is possible to handle the equipment before or at the initial stage of the failure, avoiding the situation where the equipment is shut down for a long time for maintenance due to sudden failures. This enables the equipment to maintain a high operating time, thereby improving the overall production efficiency. Since the oil fluid state of each production link can be monitored in real time, production managers can optimize and adjust the production links according to the monitoring data, such as adjusting the operating parameters of the equipment, replacing the oil fluid, etc., to ensure the stability and efficiency of the production process and further improve the production efficiency.

[0031] In summary, through various innovations and improvements, this technical solution effectively solves the problems existing in traditional oil fluid monitoring technologies and has significant beneficial effects in terms of monitoring efficiency, fault diagnosis response speed, automation level, and production efficiency.

[0032] The oil fluid monitoring solution provided by the embodiments of the present invention can also be executed by an electronic device, which can be a server, a server cluster, or a cloud server. The electronic device can also be a terminal device such as a mobile phone, a computer, a tablet computer, a wearable device, or a dedicated device (such as a dedicated terminal device with an oil fluid monitoring system). The above-mentioned chips introduced in the above embodiments can also be installed in these electronic devices. Alternatively, a service program for executing the oil fluid monitoring solution can also be installed in these electronic devices.

[0033] Figure 1 It is a schematic flowchart of an oil fluid monitoring method provided by an embodiment of the present invention. As Figure 1 shown, the method includes the following steps:

[0034] 101. Obtain the oil fluid monitoring data collected by the online monitoring platform in real time;

[0035] 102. Map the oil fluid monitoring data to the monitoring nodes corresponding to the virtual monitoring space according to the production links to which different device components belong, to obtain an oil fluid monitoring network;

[0036] 103. Based on the historical operation data of the target device and the oil fluid monitoring data, simulate the future change trends of each monitoring node in the oil fluid monitoring network to obtain the real-time prediction data corresponding to the oil fluid monitoring network;

[0037] 104. Based on the real-time prediction data, construct a visualization model of the oil fluid monitoring network to enable the user to monitor the operation conditions of each production link in the target device in real time.

[0038] In the embodiments of the present invention, the online monitoring platform at least includes: multi-source sensors deployed in the target device, and an oil fluid data monitoring console connected to the multi-source sensors. The multi-source sensors at least include: an oil fluid temperature sensor, a pressure sensor, and a particle counter.

[0039] The oil temperature is one of the important parameters reflecting the working state of the oil. The function of the oil temperature sensor is to measure the temperature of the oil in the target device in real time. During the operation of the device, an appropriate oil temperature is crucial for ensuring the lubrication performance of the oil and the normal operation of the device. If the oil temperature is too high, it may cause the viscosity of the oil to decrease, the lubrication effect to deteriorate, and increase the wear between the device components; if the temperature is too low, the viscosity of the oil increases, the fluidity becomes poor, which will also affect the startup and operation efficiency of the device. Through the oil temperature sensor, the oil temperature data can be obtained in a timely manner and transmitted to the oil data monitoring console for further analysis and processing.

[0040] The pressure sensor is used to monitor the pressure of the oil in the target device. In equipment such as hydraulic systems, the oil pressure is one of the key factors driving the operation of the device. The pressure sensor can detect the pressure values of the oil at different positions in real time. When abnormal pressure fluctuations occur, such as too high pressure may cause problems such as pipeline rupture and seal damage, and too low pressure may affect the power output of the device, resulting in the device being unable to work properly. The pressure sensor transmits the monitored pressure data to the oil data monitoring console, which helps to detect and solve pressure-related problems in a timely manner and ensure the normal operation of the device.

[0041] The particle counter is mainly used to detect the content of solid particle impurities in the oil. The impurities in the oil will cause wear and damage to the components of the device, affecting the service life and performance of the device. The particle counter can accurately measure the particles of different sizes and quantities in the oil. By analyzing these data, the degree of oil contamination can be understood. For example, when the particle counter detects a sudden increase in the particle content in the oil, it may mean that there are problems such as increased wear or seal failure inside the device, and it is necessary to check and maintain in a timely manner. The multi-source sensors work together to comprehensively monitor the state of the oil from different angles, providing rich and accurate data support for subsequent analysis and decision-making.

[0042] The oil data monitoring console is the connection center of the multi-source sensors. It receives the oil monitoring data transmitted from multi-source sensors such as oil temperature sensors, pressure sensors, and particle counters. The monitoring console stores, processes, and analyzes these data. On the one hand, the data can be displayed in an intuitive way to facilitate the operator to view the state of the oil in real time; on the other hand, the monitoring console can also analyze the data according to preset rules and algorithms to judge whether the oil state is normal and issue an alarm in a timely manner when abnormalities are found. In addition, the oil data monitoring console can also perform data interaction with other systems, such as uploading the data to a higher-level management system for comprehensive management and decision-making.

[0043] In summary, the multi-source sensors and the oil fluid data monitoring console in the online monitoring platform cooperate with each other to achieve real-time and comprehensive monitoring of the oil fluid status in the target equipment, providing a strong guarantee for the reliable operation and maintenance of the equipment.

[0044] As an optional embodiment, in 101, the oil fluid monitoring data collected by the online monitoring platform is obtained in real time. A variety of types of sensors are deployed at key parts of the target equipment, such as oil fluid temperature sensors, pressure sensors, particle counters, etc. These sensors should be reasonably arranged according to the structure of the equipment and the flow path of the oil fluid to ensure that relevant parameters of the oil fluid at different positions and states can be accurately collected. For example, pressure sensors are installed at key positions such as the outlet of the oil pump and the inlet and outlet of the oil cylinder in the hydraulic system to monitor the pressure change of the oil fluid in real time; temperature sensors and particle counters are installed inside the oil tank or in the oil fluid circulation pipeline to obtain the temperature and impurity content data of the oil fluid.

[0045] In some examples, the sensors are connected to the oil fluid data monitoring console using wired communication methods such as cables. For example, using industrial Ethernet cables can provide high-speed and stable data transmission, ensuring that the data collected by the sensors can be transmitted to the monitoring console in a timely and accurate manner. For some scenarios with high requirements for data transmission stability, wired transmission is a reliable option. In some environments where wiring is not suitable, wireless communication technologies such as Wi-Fi, Bluetooth, LoRa, etc. can be used. The sensors send the collected data to a nearby wireless access point through a wireless module, and then the access point transmits the data to the oil fluid data monitoring console. The wireless transmission method has the advantages of flexible installation and easy expansion, and can adapt to complex equipment layouts and environmental conditions.

[0046] Furthermore, the oil fluid data monitoring console is provided with a special data acquisition module to collect the data of the connected sensors at certain time intervals (such as every second, every minute, etc.). The collected data is first stored in the local storage device (such as a hard disk, a solid-state drive, etc.) of the monitoring console for subsequent processing and analysis. At the same time, in order to ensure the security and integrity of the data, a data backup mechanism can be set to regularly back up important data to an external storage device or a cloud storage platform. The monitoring console transmits the collected oil fluid monitoring data to the superior system or the relevant application platform in real time through a network interface (such as an Ethernet interface, a wireless communication module, etc.). For example, the data is sent to the enterprise's production management system or the equipment operation and maintenance platform, enabling relevant personnel to obtain the oil fluid status information in a timely manner. At the same time, the monitoring console will continuously update the data stored locally to ensure the real-time and accuracy of the data.

[0047] Through the above solution, it is possible to achieve real-time collection and transmission of oil monitoring data, enabling operators and relevant management personnel to timely understand the current state of the oil, providing timely data support for equipment operation monitoring and fault diagnosis. For example, when the oil temperature suddenly rises or the pressure fluctuates abnormally, it can be detected and corresponding measures can be taken immediately to avoid the expansion of equipment failures caused by oil problems. The reasonable deployment of sensors and the stable data transmission network ensure that the collected data can accurately reflect the actual state of the oil. Multi-source sensors monitor the oil from different angles, verifying and complementing each other, improving the reliability of the data. For example, the combination of data from particle counters and pressure sensors can more accurately determine whether there are problems such as wear or blockage inside the equipment. After the real-time obtained oil monitoring data is transmitted to the superior system or application platform, relevant personnel can intuitively view the change trend and historical records of the oil data through the visualization interface. This helps to evaluate the operation status of the equipment, formulate maintenance plans, and optimize the production process. For example, according to the pollution degree and service time of the oil, reasonably arrange the oil change time, reduce the maintenance cost and downtime of the equipment, and improve production efficiency. Whether it is the adoption of wired transmission or wireless transmission methods, it can adapt to different equipment layouts and environmental conditions. In some large industrial equipment or complex production workshops, the flexibility of wireless transmission makes data collection more convenient; while the stability of wired transmission ensures reliable data transmission at key positions, ensuring the normal operation of the entire online monitoring platform.

[0048] Further optionally, the multi-source sensors respectively monitor different equipment components related to the oil circuit in the target equipment. Further, a moving track is provided between the multi-source sensors, and the shape and size of the moving track are obtained based on the equipment structure deployment of the target equipment.

[0049] In the target equipment, the oil circuit is a crucial component, which is responsible for providing functions such as lubrication, cooling, and power transmission for various components of the equipment. There are many equipment components related to the oil circuit. For example, in a hydraulic system, the oil pump is a key component that converts mechanical energy into oil pressure energy, the oil cylinder converts the oil pressure energy into mechanical energy to achieve the movement of the equipment, and various valves (such as directional control valves, relief valves, etc.) are used to control the flow direction, pressure, and flow rate of the oil. In addition, there are auxiliary components such as pipelines and filters, which together constitute the oil circuit.

[0050] Multi-source sensors (such as oil temperature sensors, pressure sensors, particle counters, etc.) respectively conduct targeted monitoring on different equipment components in the oil circuit according to their respective functional characteristics. For example, an oil temperature sensor can be installed at the outlet of the oil pump to monitor the temperature change of the oil after the oil pump works. Since the oil pump generates heat due to friction and other reasons during operation, if the temperature is too high, it may affect the performance and lifespan of the oil pump; a pressure sensor can be installed at the inlet and outlet of the oil cylinder to monitor the pressure situation of the oil cylinder in real time to determine whether the working state of the oil cylinder is normal; a particle counter can be installed at the front and rear ends of the filter. By comparing the particle content in the oil before and after the filter, the filtering effect of the filter and the pollution degree of the oil can be evaluated. Through this division of labor and cooperation, multi-source sensors can comprehensively and accurately obtain the operation status information of each equipment component in the oil circuit.

[0051] In the above steps, the purpose of setting the moving track is to enable the multi-source sensors to move flexibly on the target equipment, so as to achieve more comprehensive monitoring of different positions and different equipment components. Compared with sensors with fixed installation, movable sensors can adjust the monitoring position according to actual needs, improving the flexibility and coverage of monitoring. For example, when detailed monitoring of a specific component of the equipment is required, the sensor can move along the moving track to the vicinity of the component to obtain more accurate data.

[0052] It is worth noting that the shape and size of the moving track are carefully designed and deployed according to the equipment structure of the target equipment. This means that the shape of the track needs to adapt to the external contour and internal structure of the equipment so that the sensor can move smoothly to each key part. For example, if the target equipment is a large cylindrical mechanical equipment, the moving track may be designed as a ring and arranged around the cylindrical surface of the equipment, enabling the sensor to move in the circumferential direction; if the internal structure of the equipment is complex and there are multiple components at different heights and positions that need to be monitored, the moving track may be designed in a multi-layer and multi-branch shape to meet the moving requirements of the sensor at different spatial positions. The size of the track also needs to consider the size and moving requirements of the sensor to ensure that the sensor can move smoothly and without interference with other components of the equipment.

[0053] In practical applications, the moving track can be implemented in various ways. For example, mechanical structures such as slide rails and guide rails can be used. The slide rail can be linear for enabling the sensor to move in a straight line direction; it can also be curved to adapt to the special shape of the equipment. The guide rail can be made of rigid materials to ensure the stability and accuracy of the sensor during movement. In addition, a driving device (such as a motor, gear, etc.) can be equipped to control the movement of the sensor on the track to achieve automated monitoring.

[0054] Through the targeted monitoring of equipment components related to the oil circuit by multi-source sensors and the reasonable deployment of mobile tracks, the operating status information of the oil circuit in the target equipment can be obtained more comprehensively and flexibly, providing strong support for equipment maintenance, fault diagnosis, and performance optimization.

[0055] As an optional embodiment, after simulating the future change trends of each monitoring node in the oil monitoring network based on the historical operation data of the target equipment to obtain the real-time prediction data corresponding to the oil monitoring network in 103, the real-time prediction data can also be input into an error monitoring model to judge the error trend and obtain the corresponding error prediction result; according to the error prediction result, update the target deployment position of the multi-source sensors; control the multi-source sensors to adjust their positions on the mobile track so that the multi-source sensors move from the current position to the target deployment position, and correct the equipment error of the multi-source sensors.

[0056] Specifically, the error monitoring model is constructed based on certain algorithms and historical data. After inputting the real-time prediction data of the oil monitoring network into the model, the model will compare and analyze the prediction data with the actual situation (which can be historical accurate data or other reliable reference data) according to preset rules and algorithms, and judge whether there are errors in the prediction data and the change trend of the errors. For example, the model can calculate the deviation values between the prediction data and the historical real data, and analyze the change of these deviation values over time to determine the error trend.

[0057] Analyze the cause of the error according to the error prediction result obtained by the error monitoring model. If it is judged that the error is caused by the improper position of the sensor, that is, the data collection at some key parts is inaccurate, the target deployment position of the multi-source sensors will be re-determined according to factors such as the structural characteristics of the equipment, the oil flow characteristics, and the monitoring requirements. For example, if it is found that the measurement data of the oil temperature sensor at a certain position deviates greatly from the actual situation, it may be because this position cannot accurately reflect the real temperature of the oil, and its target deployment position will be adjusted to a position that can more accurately measure the oil temperature.

[0058] Using the mobile track and the corresponding control device, control the multi-source sensors to move on the mobile track according to the updated target deployment position. Through precise control algorithms and driving devices, the sensors can accurately move to the target deployment position, thereby changing the measurement position of the sensors, reducing the measurement error caused by improper position, and realizing the correction of the equipment error of the multi-source sensors.

[0059] Therefore, through error monitoring and correction, the errors in real-time prediction data can be effectively reduced, enabling the prediction data to more accurately reflect the future change trends of each monitoring node in the oil fluid monitoring network, providing a more reliable reference for the operation and maintenance of equipment, helping to detect potential equipment failures in advance, and improving the reliability and safety of equipment. Dynamically adjusting the deployment positions of sensors according to the error prediction results makes the sensor layout more reasonable, enabling more comprehensive and accurate collection of key data during equipment operation, improving the monitoring ability and efficiency of the oil fluid monitoring system, and reducing monitoring blind spots and inaccurate data caused by unreasonable sensor layout. Correcting the equipment errors of multi-source sensors ensures the accuracy of sensor measurement data and makes the monitoring of equipment operating status more reliable. Timely discovery and handling of potential equipment problems avoid misjudgment or missed judgment caused by sensor errors, thereby improving the stability and operating efficiency of equipment and reducing equipment downtime and maintenance costs. This solution can dynamically adjust the positions of sensors according to the error conditions occurring during equipment operation, enabling the oil fluid monitoring system to better adapt to different operating states and working condition changes of equipment, improving the flexibility and adaptability of the system, and ensuring the long-term stable operation ability of the system.

[0060] As an optional embodiment, in the above steps, updating the target deployment positions of multi-source sensors according to the error prediction results to correct the equipment errors of multi-source sensors can be implemented as follows:

[0061] First, construct the maximum equipment error function corresponding to the i-th monitoring node in the oil fluid monitoring network; where the maximum equipment error function is expressed as the following formula:

[0062]

[0063] Where, represents the maximum equipment error function corresponding to the i-th monitoring node, represents the real-time oil fluid state value of the i-th monitoring node, represents the real-time prediction data of the i-th monitoring node, represents the oil fluid monitoring data of the i-th monitoring node, represents the average distance between the i-th monitoring node and other surrounding monitoring nodes, m is the total number of nodes, and are the weight parameters configured for the production link corresponding to the i-th monitoring node;

[0064] Furthermore, based on the output value of the maximum equipment error function, perform position information conversion on the current position of the i-th monitoring node to obtain the target deployment position of the sensor corresponding to the i-th monitoring node.

[0065] The above-mentioned maximum device error function comprehensively considers multiple key factors to evaluate the device error of monitoring nodes. It includes the real-time oil fluid state value, real-time prediction data, and oil fluid monitoring data of the i-th monitoring node. Through the comparative analysis of these data, the error situation of the current measurement data of this monitoring node can be more comprehensively and accurately reflected. It should be noted that the real-time oil fluid state value is a numerical value related to certain physical or chemical properties of the oil fluid detected in real time by specific sensor devices. For example, the measured values of specific parameters such as oil fluid viscosity, impurity content, and temperature, or the numerical values obtained by combining these values, can be used as a manifestation of the real-time oil fluid state. The specific type of numerical value needs to be determined according to the actual monitoring equipment, sensor type, and monitoring scenario. At the same time, the average distance between the i-th monitoring node and other surrounding monitoring nodes is also introduced into the function. This factor takes into account the spatial position relationship of the monitoring nodes in the oil fluid monitoring network. Monitoring nodes in different positions may be affected differently by the surrounding environment and other nodes. By incorporating the distance factor, the error source can be more reasonably evaluated. In addition, the weight parameters configured in the production link can perform weighted adjustment on the error evaluation according to the importance of different production links, making the error evaluation more in line with actual needs and improving the accuracy of the error evaluation.

[0066] Based on the output value of the maximum device error function, the position information of the monitoring node is converted for its current position, and the target deployment position of the sensor can be determined in a scientific manner. The output value of this function intuitively reflects the error situation of the monitoring node at the current position. By analyzing and processing the output value, the position with the minimum error can be found as the target deployment position. This position adjustment method based on quantitative analysis is more objective and accurate than the traditional empirical adjustment, and can effectively reduce the device error caused by improper sensor position and improve the reliability of the sensor measurement data.

[0067] In the above solution, by accurately correcting the device errors of multi-source sensors, each monitoring node in the oil monitoring network can collect and transmit data more accurately. This helps to improve the performance of the entire oil monitoring network, enabling the network to more timely and accurately reflect the true state of the oil in the target device. For example, in terms of equipment fault warning, more accurate monitoring data can detect abnormal conditions during equipment operation earlier, providing more sufficient time for equipment maintenance personnel to conduct fault troubleshooting and handling, thereby reducing the losses caused by equipment failures and improving the reliability and production efficiency of the equipment. This solution can adapt to the operation changes of the target device under different working conditions. Since the maximum device error function comprehensively considers various factors and can be adjusted according to the weights of different production links, when the operation conditions of the equipment change, it can still accurately evaluate the error situation of the monitoring nodes and accordingly adjust the positions of the sensors. This adaptive ability enables the oil monitoring system to better cope with complex and changeable operating environments, ensuring the stability and effectiveness of the monitoring system. Conducting error evaluation and position adjustment based on data reflects the characteristics of data-driven intelligent optimization. By analyzing and processing a large amount of real-time oil state values, prediction data, and monitoring data, the deployment positions of the sensors can be continuously optimized, realizing the self-optimization and improvement of the oil monitoring system. With the continuous accumulation of equipment operation data, the optimization effect of this solution will be continuously enhanced, further improving the performance and reliability of the oil monitoring system.

[0068] As an optional embodiment, in 102, according to the production links to which different equipment components belong, map the oil monitoring data to the monitoring nodes corresponding to the virtual monitoring space to obtain an oil monitoring network; the virtual monitoring space is constructed based on the type and production links of the target device.

[0069] Specifically, first conduct a detailed analysis of the type of the target device. For example, the device is an industrial machine, an aero-engine, an automotive power system, etc. Different types of devices have different structural characteristics, working principles, and oil circulation systems. Taking industrial machinery as an example, it may include large machine tools, mining equipment, etc., and their oil systems may involve multiple lubrication points, hydraulic systems, and other different parts.

[0070] For example, for the target device, sort out its various production links. For industrial machinery, the production links may include stages such as equipment startup, normal operation, load change, shutdown, etc., as well as the operation processes of each subsystem (such as the lubrication system, hydraulic system). In each production link, determine the equipment components related to the oil, such as oil pumps, oil cylinders, filters, etc.

[0071] Based on the analysis results of the target device type and production processes, a virtual monitoring space is constructed. This space can be a 3D model or a 2D model based on logical relationships. In the 3D model, according to the actual structural layout of the devices, each device component related to the oil fluid is positioned in the virtual space; in the 2D model, the relationships between each production process and device components can be shown in the form of a flow chart or a network diagram. For example, for a complex industrial device, a virtual device model can be created using 3D modeling software, and each component of the oil fluid system is accurately placed in the corresponding position to form an intuitive virtual monitoring space.

[0072] In the virtual monitoring space, according to the production processes to which different device components belong, corresponding monitoring nodes are determined. Each monitoring node corresponds to one or more device components related to oil fluid monitoring. For example, a monitoring node is set at the outlet of the oil pump in the lubrication system to monitor parameters such as the pressure and temperature of the oil fluid output by the oil pump; monitoring nodes are respectively set at the inlet and outlet of the oil cylinder to monitor the pressure change and flow rate of the oil fluid when the oil cylinder is working. The setting of the monitoring nodes should be considered to comprehensively and accurately reflect the oil fluid state of the device components.

[0073] Multi-source sensors (such as oil fluid temperature sensors, pressure sensors, particle counters, etc.) collect the monitoring data of the oil fluid in the target device in real time. These data are mapped to the corresponding monitoring nodes in the virtual monitoring space according to the production processes to which different device components belong. For example, the temperature data collected by the oil fluid temperature sensor is mapped to the corresponding monitoring node in the virtual monitoring space according to the device component corresponding to its installation position. In this way, each monitoring node carries the oil fluid monitoring data related to this node, thus constructing an oil fluid monitoring network.

[0074] In this way, by mapping the oil fluid monitoring data to the monitoring nodes in the virtual monitoring space to form an oil fluid monitoring network, the oil fluid states of various production links in the target device can be presented in an intuitive manner. Operators can quickly understand the real-time parameters of the oil fluid, such as temperature, pressure, impurity content, etc., in different equipment components and different production links by viewing the virtual monitoring space, without having to search for the specific sensor positions and data in the complex equipment structure, which improves the efficiency of data viewing and analysis. The oil fluid monitoring network clearly shows the relationships between the various monitoring nodes and the data flow. When a fault or anomaly occurs in the equipment, operators can quickly locate the possible positions and causes of the fault based on the data changes of the relevant monitoring nodes in the oil fluid monitoring network. For example, if the oil fluid pressure at a certain monitoring node suddenly drops, by viewing the oil fluid monitoring network, the connection relationship between this node and other nodes can be analyzed to determine whether it is a pipeline leak, a pump failure, or a problem with other components, which helps to carry out fault diagnosis and repair in a timely manner and reduce the equipment downtime. Based on the comprehensive and accurate data provided by the oil fluid monitoring network, production managers can conduct in-depth analysis of the operation conditions of the equipment. By analyzing the oil fluid states of different production links and different equipment components, the operation parameters of the equipment can be optimized, the production process can be adjusted, and the operation efficiency and reliability of the equipment can be improved. For example, according to the temperature and pressure data of the oil fluid, the load of the equipment can be reasonably adjusted to avoid equipment damage caused by overheating or excessive pressure of the oil fluid. Since the virtual monitoring space is constructed based on the type of the target device and the production links, this solution can adapt to different types and structures of equipment. Whether it is a simple mechanical equipment or a complex industrial system, the effective monitoring and management of the oil fluid state can be realized by constructing the corresponding virtual monitoring space and oil fluid monitoring network, which has strong versatility and scalability.

[0075] As an alternative embodiment, in 103, based on the historical operation data of the target device and the oil fluid monitoring data, simulating the future change trends of each monitoring node in the oil fluid monitoring network to obtain the real-time prediction data corresponding to the oil fluid monitoring network can be implemented as follows:

[0076] Obtain the multi-source monitoring data of each monitoring node in the oil fluid monitoring data as input features;

[0077] Extract the key information features in the input features through the hybrid diagnosis unit to identify the potential change features of each monitoring node in the oil fluid monitoring network; the potential change features at least include: temperature change features, impurity content features, viscosity change features, oil fluid level features;

[0078] Input the potential change features into a virtual operation model constructed based on the historical operation data to obtain the oil fluid data change features corresponding to each monitoring node; the oil fluid data change features at least include: temperature change prediction features, impurity content change prediction features, viscosity change prediction features, and oil fluid level change prediction features.

[0079] In the above solution, by analyzing and processing the multi-source monitoring data of each monitoring node in the oil fluid monitoring network, the future state changes of the oil fluid are predicted, including the change trends in aspects such as temperature, impurity content, viscosity, and oil fluid level, providing forward-looking information for the maintenance and management of the equipment. Identifying the potential change features of each monitoring node in the oil fluid monitoring network can detect potential problems before obvious equipment failures occur, so as to take corresponding measures to avoid the occurrence of equipment failures, reduce equipment downtime and maintenance costs. According to the predicted oil fluid data change features, optimize the operation parameters and maintenance plan of the equipment, improve the operation efficiency and reliability of the equipment, and extend the service life of the equipment.

[0080] Specifically, the multi-source monitoring data of each monitoring node in the oil fluid monitoring data are used as input features. These data contain information about the oil fluid in different aspects, such as the data collected by temperature sensors, impurity sensors, viscosity sensors, and oil fluid level sensors, providing a basis for subsequent analysis and prediction.

[0081] Use a hybrid diagnostic unit to process the input features and extract the key information features. The hybrid diagnostic unit may combine various signal processing and feature extraction methods, such as time-domain analysis, frequency-domain analysis, wavelet transform, etc., to mine the potential features related to the oil fluid state change from the multi-source monitoring data, such as temperature change features, impurity content features, viscosity change features, oil fluid level features, etc.

[0082] Construct a virtual operation model based on the historical operation data of the target equipment. This model can simulate the operation conditions of the equipment under different working conditions and the change rules of the oil fluid state. Input the extracted potential change features into the virtual operation model. The model predicts the future changes of the oil fluid data according to the rules and trends contained in the historical data, and outputs the oil fluid data change prediction features corresponding to each monitoring node, such as temperature change prediction features, impurity content change prediction features, viscosity change prediction features, oil fluid level change prediction features, etc.

[0083] Taking a large industrial compressor as an example, there are multiple monitoring nodes in its oil monitoring network, which are distributed in parts such as oil pumps, oil coolers, and bearings. The multi-source monitoring data of each monitoring node includes data such as the oil temperature at the outlet of the oil pump, the oil temperature and flow rate at the inlet and outlet of the oil cooler, and the oil impurity content and viscosity at the bearing. The hybrid diagnosis unit analyzes these data and extracts potential change features such as the rising trend feature of the oil temperature at the outlet of the oil pump over a period of time, the change feature of the oil temperature difference between the inlet and outlet of the oil cooler, and the fluctuation feature of the oil impurity content at the bearing. These potential change features are input into a virtual operation model constructed based on the historical operation data of the compressor, and the model predicts change prediction features of oil data such as the oil temperature at the outlet of the oil pump may continue to rise by a certain degree, the oil temperature difference between the inlet and outlet of the oil cooler may decrease, and the oil impurity content at the bearing may exceed the normal range in the future period of time.

[0084] In practical applications, LSTM can effectively process time series data, capture long-term dependencies in the data, and is suitable for analyzing data with time series characteristics such as oil monitoring data, and modeling and predicting features that change over time such as temperature and impurity content. CNN can be used to extract spatial and local features from multi-source monitoring data. For example, it extracts features from the relationships between data of monitoring nodes at different positions, and then inputs the extracted features into LSTM for time series analysis and prediction. This combination method can make full use of the spatial and time information of the data and improve the prediction effect. VAE can learn the latent distribution of the data, map multi-source monitoring data to a low-dimensional latent space, extract more representative latent change features, and can also be used to generate prediction data to predict the future states of each monitoring node in the oil monitoring network.

[0085] Thus, through the fusion of multi-source monitoring data and the extraction of key information features, it is possible to more comprehensively and accurately reflect the actual state and change trend of the oil, thereby improving the prediction accuracy of oil data changes and providing a more reliable basis for equipment maintenance. Identifying potential change features in advance and predicting the change trend of oil data enables the system to issue early warnings before equipment failures occur, giving maintenance personnel sufficient time to prepare and take measures, and effectively reducing the losses caused by equipment failures. It provides support for the intelligent operation and maintenance of equipment. According to the prediction results, the operation parameters of the equipment can be automatically adjusted or a reasonable maintenance plan can be formulated, improving the automation and intelligent level of equipment operation and maintenance.

[0086] Further optionally, assume that the hybrid diagnosis unit is deployed with a coupled prediction model jointly trained based on the oil contamination data, equipment wear data, and performance degradation data in the historical operation data. Based on the above assumption, in 103, extracting the key information features in the input features through the hybrid diagnosis unit to identify the potential change features of each monitoring node in the oil monitoring network can be implemented as follows:

[0087] Input the input features into the hybrid diagnosis unit, and identify candidate features associated with at least one dimension of oil contamination, equipment wear, and performance degradation from the multi-source monitoring data of the input features; use the following association scoring function to perform dynamic threshold screening on each candidate feature to obtain the potential change features of each monitoring node in the oil monitoring network.

[0088] Among them, the association scoring function is expressed as: 。

[0089] In the above function, represents the scoring performance value corresponding to the i-th monitoring node, is the similarity coefficient between the i-th monitoring node and its surrounding nodes, is the historical data dependence coefficient between the i-th monitoring node and its surrounding nodes, is the business relevance coefficient between the i-th monitoring node and its surrounding nodes, 、 、 、 、 are the weight parameter values of each evaluation item corresponding to the i-th monitoring node respectively.

[0090] In the embodiments of the present invention, potential change features related to key dimensions such as oil contamination, equipment wear, and performance degradation are accurately extracted from multi-source input features, providing more targeted and effective information for subsequent analysis and prediction of the oil monitoring network. By considering various association factors between monitoring nodes and their surrounding nodes, screening and evaluating each candidate feature, the state and potential change trends of each monitoring node in the oil monitoring network can be comprehensively and accurately grasped, improving the evaluation ability of the overall system operation status. It provides a more accurate basis for equipment fault prediction and diagnosis, helps to detect potential problems of the equipment in advance, so as to take timely and effective maintenance measures, reduce the equipment fault risk, and ensure the normal operation of the equipment.

[0091] Specifically, the input features are input into the hybrid diagnosis unit, and the coupled prediction model trained based on historical operation data is used to identify candidate features related to at least one dimension among oil contamination, equipment wear, and performance degradation from multi-source monitoring data. This is based on the model's learning and understanding of the features related to these dimensions in historical data, and it can find potential features that may be related to the input data according to the input data. An association scoring function is used to evaluate each candidate feature. This function comprehensively considers the similarity coefficient between the i-th monitoring node and its surrounding nodes, the historical data dependence coefficient, the business relevance coefficient, and the weight parameter values of each evaluation item. In this way, a quantitative score is given to each candidate feature, and the features with higher scores are selected as potential change features according to the set dynamic threshold. The similarity coefficient reflects the similarity degree between nodes in the current state, the historical data dependence coefficient reflects the correlation degree of historical data between nodes, the business relevance coefficient represents the mutual relationship between nodes in the business process, and the weight parameter values are used to adjust the importance of each factor in the evaluation.

[0092] Suppose in an oil monitoring network of a large chemical production equipment, there are multiple monitoring nodes distributed in different parts of the oil circuit. The hybrid diagnosis unit receives multi-source monitoring data from each monitoring node, such as oil temperature, oil pressure, oil composition, etc. Based on the coupled prediction model, some candidate features are identified. For example, the change in the metal particle content in the oil at a certain monitoring node may be related to equipment wear, and the change in the impurity concentration in the oil may be related to oil contamination. For the identified candidate features, they are evaluated according to the association scoring function. For example, if the similarity coefficient of a monitoring node with its surrounding nodes is relatively high, it means that its state is relatively similar to that of the surrounding nodes; if the historical data dependence coefficient is also relatively high, it indicates that its historical data has a strong correlation with the surrounding nodes; if the business relevance coefficient is moderate, considering these factors and the weight parameter values of each evaluation item, the scoring performance value corresponding to this monitoring node is calculated. If this value exceeds the dynamic threshold, the corresponding candidate feature is determined as a potential change feature. For example, it is determined that the rapid increase in the metal particle content in the oil is an important potential change feature, which may indicate that the equipment has excessive wear problems.

[0093] Therefore, through the combination of the coupling prediction model and the correlation scoring function, potential change features related to key dimensions can be screened out more accurately, irrelevant or redundant features can be removed, the quality and effectiveness of the features are improved, and a more reliable data basis is provided for subsequent analysis and prediction. The dynamic threshold screening mechanism can adaptively adjust the screening criteria according to different monitoring scenarios and data characteristics, enabling the system to better adapt to the complex and changing actual operating environment, and improving the flexibility and adaptability of the system. More accurate extraction of potential change features helps to diagnose the operating state of the equipment more precisely, can detect problems in aspects such as oil contamination degree, equipment wear, and performance degradation of the equipment earlier and more accurately, improves the accuracy and reliability of fault diagnosis, and reduces the situations of misjudgment and missed judgment.

[0094] Further optionally, during the construction process of the coupling prediction model, a large amount of historical operation data can be collected, including oil contamination data, equipment wear data, performance degradation data, etc. Perform preprocessing operations such as data cleaning, denoising, and normalization on this data to ensure the quality and consistency of the data and provide a good data foundation for model training. Extract features related to oil contamination, equipment wear, and performance degradation from the original data, such as the impurity content in the oil, the size and quantity of metal particles, the wear amount of key components of the equipment, the change rate of equipment performance indicators, etc. Statistical methods, signal processing techniques, etc. can be used for feature extraction and selection to determine the features that have an important impact on model prediction. Select a suitable machine learning or deep learning model architecture, such as deep neural network (DNN), recurrent neural network (RNN) and its variants (such as LSTM, GRU), etc. According to the characteristics of the data and the requirements of the prediction task, design the architecture of the model and determine parameters such as the number of network layers, the number of neurons, and the connection method. For example, a multi-input multi-output architecture can be adopted, taking different types of features as different input channels and simultaneously outputting prediction results in multiple dimensions such as oil contamination, equipment wear, and performance degradation. Use the preprocessed data to train the model. By adjusting the parameters of the model, the model can learn the mapping relationship between the input features and the output targets. Adopt a suitable loss function to measure the difference between the model prediction results and the true values, such as mean square error (MSE), mean absolute error (MAE), etc. Use optimization algorithms (such as stochastic gradient descent, Adagrad, Adadelta, Adam, etc.) to minimize the loss function and update the parameters of the model. During the training process, techniques such as cross-validation and early stopping can be adopted to prevent the model from overfitting and improve the generalization ability of the model. Use the test data set to evaluate the trained model, calculate evaluation metrics such as accuracy, recall rate, F1 value, etc. of the model on the test set, and evaluate the performance of the model. According to the evaluation results, adjust and optimize the model, such as adjusting the hyperparameters of the model, adding or reducing features, improving the model architecture, etc., until the model reaches satisfactory performance metrics. Deploy the trained coupling prediction model to the hybrid diagnosis unit for actual feature recognition and prediction tasks. During the actual operation process, as new data is continuously generated, collect and update the data regularly, and retrain and optimize the model to ensure that the model can adapt to the changes in the equipment operation state and continuously provide accurate prediction and diagnosis results.

[0095] As an optional embodiment, in 104, based on the real-time prediction data, a visualization model of the oil monitoring network is constructed so that the user can monitor the oil production situation of each production link in the target equipment in real time, which can be realized as:

[0096] Obtain the key production links to be monitored; the key production links at least include: the core link of oil production of the target equipment, the high-load operation link, and the locations of easily worn parts; perform feature fusion processing on the corresponding oil data change characteristics in the real-time prediction data to obtain the oil production change characteristics corresponding to the key production links; based on the real-time prediction data and the oil production change characteristics, display a visualization model of the oil monitoring network to the user, so that the user can monitor the oil production situation of each production link in the target equipment in real time.

[0097] Specifically, by clarifying the key production links to be monitored, such as the core link of oil production, the high-load operation link, and the locations of easily worn parts, the monitoring focus is concentrated on the parts that have a greater impact on the equipment operation status and oil quality, improving the pertinence and effectiveness of monitoring, and avoiding wasting resources on secondary links. Perform feature fusion processing on the oil data change characteristics in the real-time prediction data, integrate information from multiple aspects, and obtain more comprehensive and accurate oil production change characteristics corresponding to the key production links, providing a more valuable decision-making basis for the user. Build a visualization model based on the processed data, and display the relevant information of the oil monitoring network to the user in an intuitive way, enabling the user to understand the oil production situation of each production link in the target equipment in real time and clearly, facilitating the timely discovery of problems and taking corresponding measures.

[0098] In the foregoing steps, according to the characteristics and operating rules of the target equipment, determine the production links that play a key role in the equipment operation and oil quality. For example, the core link of oil production is the basis for the normal operation of the equipment, oil problems are more likely to occur in the high-load operation link, and the oil state at the locations of easily worn parts directly affects the component life. These links are the key points of monitoring. For the oil data change characteristics corresponding to the key production links in the real-time prediction data, adopt appropriate feature fusion algorithms (such as weighted fusion, model-based fusion, etc.) to integrate multiple dimensions of features (such as temperature change characteristics, impurity content change characteristics, viscosity change characteristics, etc.), eliminate information redundancy, and extract more representative and comprehensive oil production change characteristics to more accurately reflect the actual situation of the key production links. Use data visualization techniques (such as drawing charts, graphs, building 3D models, etc.) to present the real-time prediction data and the oil production change characteristics obtained through feature fusion processing to the user in an intuitive way. For example, represent the changes in oil parameters through the changes in visual elements such as color, shape, and size, enabling the user to quickly understand and analyze the data.

[0099] Suppose the target device is a large industrial steam turbine. The core links of oil production include the oil supply and purification links where components such as the main oil pump and oil filter are located; high-load operation links such as the high-pressure cylinder part of the steam turbine, where the lubrication and cooling requirements of the oil are higher under high-load conditions; the locations of easily worn components such as the bearing part, and the wear of the bearing is closely related to the state of the oil. In the real-time prediction data, there are characteristics such as the change characteristics of the oil temperature, impurity content, and viscosity at the bearing part. Through feature fusion processing, these features are comprehensively analyzed. For example, a higher weight is assigned to the temperature change characteristic (because temperature has a greater impact on bearing wear), and a comprehensive oil production change characteristic is obtained, such as the comprehensive state index of the oil at the bearing. A visualization model of the oil monitoring network is constructed to display the structure of the steam turbine in a three-dimensional model and mark out the key production links. For the bearing part, different colors are used to represent the high and low of the oil temperature, and the thickness of the lines is used to represent the amount of oil impurities. Users can see the comprehensive state of the oil at the bearing and the oil production situation of other key production links in real time through this visualization model.

[0100] Thus, by focusing on the key production links, it avoids non-discriminatory monitoring of all links, saves monitoring resources and time, improves the monitoring efficiency, and enables users to obtain important information faster. Feature fusion processing provides a more comprehensive and accurate oil production change characteristic, enabling users to make decisions (such as equipment maintenance plans, oil change times, etc.) based on more reliable information, reducing the possibility of misjudgment and missed judgment, and improving the accuracy of decisions. The visualization model displays data in an intuitive and easy-to-understand way, reduces the difficulty for users to understand and analyze data, enhances the user experience, enables users to more conveniently monitor the oil production situation of the equipment in real time, and promptly discovers potential problems.

[0101] In the embodiment of the present invention, the visualization model of the oil monitoring network is a tool for displaying information related to the oil monitoring network in an intuitive form such as graphs and charts. It can be presented in a two-dimensional or three-dimensional form based on the structure of the target device and the layout of the oil monitoring network. The positions of each monitoring node and the key production links are marked in the model, and at the same time, the real-time prediction data and the processed oil production change characteristics are displayed through different visual elements (such as colors, shapes, sizes, dynamic effects, etc.). For example, different colored lines are used to represent the flow path of the oil, different colored regions are used to represent the high and low of the oil parameters, and an animation effect is used to display the change trend of the oil state, etc. Users can view the information in the model in more detail through interactive operations (such as zooming in, zooming out, rotating, etc.), so as to realize the real-time monitoring and analysis of the oil production situation of each production link in the target device.

[0102] In the embodiments of the present invention, it is possible to solve the technical problems in traditional oil fluid monitoring technologies, such as low monitoring efficiency, slow fault diagnosis response, which seriously affect the operation efficiency of each production link, realize the full-equipment automatic monitoring of oil fluid, improve the monitoring efficiency and production efficiency, and assist in improving the fault diagnosis response speed.

[0103] In another embodiment of the present invention, an oil fluid monitoring system is further provided. Refer to Figure 2 as described, the oil fluid monitoring system includes the following units:

[0104] An acquisition unit, configured to obtain in real time the oil fluid monitoring data collected by an online monitoring platform; the online monitoring platform at least includes: multi-source sensors deployed in target equipment and an oil fluid data monitoring console connected to the multi-source sensors; the multi-source sensors at least include: an oil fluid temperature sensor, a pressure sensor, and a particle counter.

[0105] A construction unit, configured to map the oil fluid monitoring data to the monitoring nodes corresponding in a virtual monitoring space according to the production links to which different equipment components belong, to obtain an oil fluid monitoring network; the virtual monitoring space is constructed based on the type and production link to which the target equipment belongs.

[0106] A prediction unit, configured to simulate the future change trends of each monitoring node in the oil fluid monitoring network based on the historical operation data of the target equipment and the oil fluid monitoring data, to obtain real-time prediction data corresponding to the oil fluid monitoring network; the real-time prediction data at least includes: the change characteristics of the oil fluid data of each monitoring node.

[0107] A display unit, configured to construct a visualization model of the oil fluid monitoring network based on the real-time prediction data, so that a user can monitor in real time the operation conditions of each production link in the target equipment.

[0108] Further optionally, the multi-source sensors respectively monitor different equipment components related to the oil fluid circuit in the target equipment; a moving track is arranged between the multi-source sensors, and the shape and size of the moving track are deployed based on the equipment structure of the target equipment.

[0109] Further optionally, after the prediction unit simulates the future change trends of each monitoring node in the oil fluid monitoring network based on the historical operation data of the target equipment to obtain the real-time prediction data corresponding to the oil fluid monitoring network, the acquisition unit is further configured to:

[0110] Input the real-time prediction data into an error monitoring model to determine the error trend, so as to obtain a corresponding error prediction result; update the target deployment positions of the multi-source sensors according to the error prediction result; control the multi-source sensors to adjust their positions on the moving track, so that the multi-source sensors move from the current position to the target deployment positions, and correct the device errors of the multi-source sensors.

[0111] Further optionally, the acquisition unit, according to the error prediction result, updates the target deployment positions of the multi-source sensors to correct the device errors of the multi-source sensors, and is configured as:

[0112] Construct a maximized device error function corresponding to the i-th monitoring node in the oil fluid monitoring network; the maximized device error function is expressed by the following formula:

[0113] Where, represents the maximized device error function corresponding to the i-th monitoring node, represents the real-time oil fluid state value of the i-th monitoring node, represents the real-time prediction data of the i-th monitoring node, represents the oil fluid monitoring data of the i-th monitoring node, represents the average distance between the i-th monitoring node and other surrounding monitoring nodes, m is the total number of nodes, and are weight parameters configured for the production link corresponding to the i-th monitoring node;

[0114] Based on the output value of the maximized device error function, perform position information conversion on the current position of the i-th monitoring node to obtain the target deployment position of the sensor corresponding to the i-th monitoring node.

[0115] Further optionally, the prediction unit, based on the historical operation data of the target device and the oil fluid monitoring data, simulates the future change trends of each monitoring node in the oil fluid monitoring network to obtain the real-time prediction data corresponding to the oil fluid monitoring network, and is configured as:

[0116] Obtain the multi-source monitoring data of each monitoring node in the oil fluid monitoring data as input features;

[0117] Extract key information features from the input features through a hybrid diagnosis unit to identify potential change features of each monitoring node in the oil fluid monitoring network; the potential change features at least include: temperature change features, impurity content features, viscosity change features, oil fluid level features;

[0118] Input the potential change features into a virtual operation model constructed based on the historical operation data to obtain the oil fluid data change features corresponding to each monitoring node; the oil fluid data change features at least include: temperature change prediction features, impurity content change prediction features, viscosity change prediction features, and oil fluid level change prediction features.

[0119] Further optionally, the hybrid diagnosis unit is deployed with a coupled prediction model jointly trained based on the oil fluid contamination data, equipment wear data, and performance degradation data in the historical operation data;

[0120] The prediction unit extracts key information features from the input features through the hybrid diagnosis unit to identify the potential change features of each monitoring node in the oil fluid monitoring network, and is configured to:

[0121] Input the input features into the hybrid diagnosis unit, and identify candidate features associated with at least one dimension of oil fluid contamination, equipment wear, and performance degradation from the multi-source monitoring data of the input features;

[0122] Use the following association scoring function to perform dynamic threshold screening on each candidate feature to obtain the potential change features of each monitoring node in the oil fluid monitoring network;

[0123] Among them, the association scoring function is expressed as: ;

[0124] Among them, represents the scoring performance value corresponding to the i-th monitoring node, is the similarity coefficient between the i-th monitoring node and its surrounding nodes, is the historical data dependence coefficient between the i-th monitoring node and its surrounding nodes, is the business relevance coefficient between the i-th monitoring node and its surrounding nodes, 、 、 、 、 are the weight parameter values of each evaluation item corresponding to the i-th monitoring node respectively.

[0125] Further optionally, the display unit constructs a visualization model of the oil fluid monitoring network based on the real-time prediction data so that the user can monitor the oil fluid production situation of each production link in the target equipment in real time, and is configured to:

[0126] Obtain the key production links to be monitored; the key production links at least include: the core link of oil fluid production of the target equipment, the high-load operation link, and the locations of easily worn parts;

[0127] Perform feature fusion processing on the corresponding oil fluid data change characteristics in the real-time prediction data to obtain the oil fluid production change characteristics corresponding to the key production links;

[0128] Based on the real-time prediction data and the oil fluid production change characteristics, display a visualization model of the oil fluid monitoring network to the user, so that the user can monitor the oil fluid production situation of each production link in the target device in real time.

[0129] The system can implement various steps in the above method embodiments, which will not be elaborated here for the time being.

[0130] The system provided by the embodiments of the present invention can solve the technical problems in traditional oil fluid monitoring technologies, such as low monitoring efficiency, slow fault diagnosis response, which seriously affect the operation efficiency of each production link, realize the full-device automatic monitoring of oil fluid, improve the monitoring efficiency and production efficiency, and assist in improving the fault diagnosis response speed.

[0131] Please refer to Figure 3 , Figure 3 which is a schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. As Figure 3 shown, an embodiment of the present invention provides an electronic device 500, including a memory 510, a processor 520, and a computer program 511 stored on the memory 510 and executable on the processor 520. When the processor 520 executes the computer program 511, the following steps are implemented: obtain in real time the oil fluid monitoring data collected by the online monitoring platform; the online monitoring platform at least includes: multi-source sensors deployed in the target device and an oil fluid data monitoring console connected to the multi-source sensors; the multi-source sensors at least include: an oil fluid temperature sensor, a pressure sensor, and a particle counter; map the oil fluid monitoring data to the monitoring nodes corresponding to the virtual monitoring space according to the production links to which different device components belong to obtain an oil fluid monitoring network; the virtual monitoring space is constructed based on the type and production links of the target device; based on the historical operation data of the target device and the oil fluid monitoring data, simulate the future change trends of each monitoring node in the oil fluid monitoring network to obtain the real-time prediction data corresponding to the oil fluid monitoring network; the real-time prediction data at least includes: the oil fluid data change characteristics of each monitoring node; based on the real-time prediction data, construct a visualization model of the oil fluid monitoring network so that the user can monitor the operation situation of each production link in the target device in real time.

[0132] Please refer to Figure 4 , Figure 4 which is a schematic diagram of an embodiment of a computer-readable storage medium provided by an embodiment of the present invention. As Figure 4As shown in the figure, this embodiment provides a computer-readable storage medium 600, on which a computer program 611 is stored. When the computer program 611 is executed by a processor, the following steps are implemented: obtaining in real time the oil fluid monitoring data collected by the online monitoring platform; the online monitoring platform at least includes: multi-source sensors deployed in the target device and an oil fluid data monitoring console connected to the multi-source sensors; the multi-source sensors at least include: an oil fluid temperature sensor, a pressure sensor, and a particle counter; mapping the oil fluid monitoring data to the monitoring nodes corresponding in the virtual monitoring space according to the production links to which different device components belong, to obtain an oil fluid monitoring network; the virtual monitoring space is constructed based on the type and production link to which the target device belongs; simulating the future change trends of each monitoring node in the oil fluid monitoring network based on the historical operation data of the target device and the oil fluid monitoring data, to obtain real-time prediction data corresponding to the oil fluid monitoring network; the real-time prediction data at least includes: the change characteristics of the oil fluid data of each monitoring node; constructing a visualization model of the oil fluid monitoring network based on the real-time prediction data, so that a user can monitor in real time the operation conditions of each production link in the target device.

[0133] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0134] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0135] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0136] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means that implement the function specified in one or more of the flows Figure 1 and / or boxes Figure 1 specified in one or more of the boxes.

[0137] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the function specified in one or more of the flows Figure 1 and / or boxes Figure 1 specified in one or more of the boxes.

[0138] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to cover the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0139] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. An oil monitoring method, characterized in that: The method is applied to oil monitoring scenarios, and the method includes: Acquire oil monitoring data collected by an online monitoring platform in real time; the online monitoring platform at least includes: a multi-source sensor deployed in the target device, and an oil data monitoring station connected to the multi-source sensor; the multi-source sensor at least includes: an oil temperature sensor, a pressure sensor, and a particle counter; According to the production links to which different equipment components belong, the oil monitoring data is mapped to the monitoring nodes corresponding to the virtual monitoring space to obtain an oil monitoring network; the virtual monitoring space is constructed based on the type of target equipment and the production link; Based on the historical operation data of the target device and the oil monitoring data, the future change trend of each monitoring node in the oil monitoring network is simulated to obtain the real-time prediction data corresponding to the oil monitoring network; the real-time prediction data at least includes: the oil data change characteristics of each monitoring node; Based on the real-time prediction data, a visualization model of the oil monitoring network is constructed so that the user can monitor the operation status of each production link in the target equipment in real time; Multi-source sensors monitor different equipment components related to the oil circuit in the target equipment respectively; A moving track is arranged between the multi-source sensors, and the shape and size of the moving track are obtained based on the device structure deployment of the target device; After simulating the future change trend of each monitoring node in the oil monitoring network based on the historical operation data of the target device to obtain the real-time prediction data corresponding to the oil monitoring network, the method further includes: Inputting the real-time prediction data into the error monitoring model to perform error trend judgment to obtain corresponding error prediction results; updating the target deployment position of the multi-source sensor according to the error prediction result; The multi-source sensor is controlled to adjust its position on the moving track so as to move the multi-source sensor from a current position to a target deployment position, thereby correcting the equipment error of the multi-source sensor.

2. The oil monitoring method according to claim 1, characterized in that: The updating of the target deployment position of the multi-source sensor according to the error prediction result to correct the device error of the multi-source sensor includes: Construct a maximized device error function corresponding to the i-th monitoring node in the oil monitoring network; wherein the maximized device error function is expressed as the following formula: ； in, represents the maximized device error function corresponding to the i-th monitoring node, represents the real-time oil status value of the i-th monitoring node, represents the real-time prediction data of the i-th monitoring node, represents the oil monitoring data of the i-th monitoring node, represents the average distance between the i-th monitoring node and other surrounding monitoring nodes, m is the total number of nodes, and The weight parameter configured for the production link corresponding to the i-th monitoring node; Based on the output value of the maximized device error function, the current position of the i-th monitoring node is converted into position information to obtain the target deployment position of the sensor corresponding to the i-th monitoring node.

3. The oil monitoring method according to claim 1, characterized in that: The method of simulating the future change trend of each monitoring node in the oil monitoring network based on the historical operation data of the target device and the oil monitoring data to obtain the real-time prediction data corresponding to the oil monitoring network includes: Acquire multi-source monitoring data of each monitoring node in the oil monitoring data as input features; Extracting key information features from the input features through a hybrid diagnostic unit to identify potential change features of each monitoring node in the oil monitoring network; the potential change features at least include: temperature change features, impurity content features, viscosity change features, and oil level features; The potential change characteristics are input into a virtual operation model constructed based on the historical operation data to obtain the oil data change characteristics corresponding to each monitoring node; the oil data change characteristics at least include: temperature change prediction characteristics, impurity content change prediction characteristics, viscosity change prediction characteristics, and oil level change prediction characteristics.

4. The oil monitoring method according to claim 3, characterized in that: The hybrid diagnosis unit is deployed with: a coupled prediction model obtained by jointly training the oil contamination data, the equipment wear data, and the performance degradation data in the historical operation data; The extracting key information features from the input features by the hybrid diagnosis unit to identify potential change features of each monitoring node in the oil monitoring network includes: Inputting the input features into the hybrid diagnostic unit, identifying candidate features associated with at least one dimension of oil contamination, equipment wear, and performance degradation from multi-source monitoring data of the input features; The following correlation scoring function is used to perform dynamic threshold screening on each candidate feature to obtain the potential change characteristics of each monitoring node in the oil monitoring network; Among them, the association score function is expressed as: ; in, represents the score performance value corresponding to the i-th monitoring node, is the similarity coefficient between the ith monitoring node and the surrounding nodes, is the historical data dependency coefficient between the ith monitoring node and the surrounding nodes, is the business correlation coefficient between the ith monitoring node and the surrounding nodes, , , , , are the weight parameter values ​​of each evaluation item corresponding to the i-th monitoring node.

5. The oil monitoring method according to claim 1, characterized in that: Based on the real-time prediction data, a visualization model of the oil monitoring network is constructed so that the user can monitor the oil production status of each production link in the target equipment in real time, including: Obtain the key production links to be monitored; the key production links at least include: the core link of oil production of the target equipment, the high-load operation link, and the location of the easily worn parts; Performing feature fusion processing on the oil data change features corresponding to the real-time prediction data to obtain the oil production change features corresponding to the key production links; Based on the real-time prediction data and the oil production change characteristics, a visualization model of the oil monitoring network is displayed to the user so that the user can monitor the oil production status of each production link in the target equipment in real time.

6. An oil monitoring system, implementing the method according to claims 1-5, characterized in that: The system comprises the following units, wherein: The acquisition unit is configured to acquire in real time the oil monitoring data collected by the online monitoring platform; the online monitoring platform at least includes: a multi-source sensor deployed in the target device, and an oil data monitoring station connected to the multi-source sensor; the multi-source sensor at least includes: an oil temperature sensor, a pressure sensor, and a particle counter; The construction unit is configured to map the oil monitoring data to the monitoring nodes corresponding to the virtual monitoring space according to the production links to which the different equipment components belong, so as to obtain an oil monitoring network; the virtual monitoring space is constructed based on the type and production link of the target equipment; The prediction unit is configured to simulate the future change trend of each monitoring node in the oil monitoring network based on the historical operation data of the target device and the oil monitoring data, so as to obtain the real-time prediction data corresponding to the oil monitoring network; the real-time prediction data at least includes: the oil data change characteristics of each monitoring node; The display unit is configured to construct a visualization model of the oil monitoring network based on the real-time prediction data, so that the user can monitor the operation status of each production link in the target equipment in real time.

7. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing the oil monitoring method described in any one of claims 1-5.

8. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores a computer software program, and when the computer software program is executed by the processor, the oil monitoring method according to any one of claims 1 to 5 is implemented.

Images