Turbine lubricating oil intelligent operation and maintenance system, method, equipment and storage medium

Abstract

The application discloses a steam turbine lubricating oil intelligent operation and maintenance system, method and equipment and a storage medium, relates to the technical field of industrial digital intelligent operation and maintenance, and achieves the intelligent conversion from data to decision through the following steps: covering multi-dimensional parameters such as physical, chemical and morphological parameters of oil through a plurality of sensors of a data acquisition module; guaranteeing zero data loss and low delay through a data transmission module to solve the data fault problem of traditional manual meter reading; converting static parameters into dynamic trends, residual life and abnormal early warning through a hybrid model of a visual cloud operation and maintenance module to realize the intelligent conversion from data to decision; and automatically responding to instructions through an oil filtering execution module to complete the closed-loop processing of polluted oil to clean oil without manual intervention, so that the real-time monitoring and intelligent analysis of the operation state of the steam turbine are realized, the oil filtering system is automatically started when a set threshold is reached, the oil is kept in a normal state, and the maintenance cost is reduced and the operation and maintenance efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial digital intelligent operation and maintenance, in particular to a steam turbine lubricating oil intelligent operation and maintenance system, method, device and storage medium. BACKGROUND

[0002] With the popularization of the concepts of industry and intelligent manufacturing, the intelligentization and digitization of equipment operation and maintenance have become a key trend to improve the reliability and efficiency of industrial equipment. As core power equipment in the power and petrochemical industries, the stable operation of the lubricating oil system of a steam turbine is crucial to ensuring the service life of the equipment and improving production efficiency.

[0003] In related technologies, traditional steam turbine lubricating oil operation and maintenance mainly relies on manual inspection and periodic offline detection. However, the applicant realizes that this approach has obvious defects such as response lag, discontinuous data, and insufficient fault prediction capability. Moreover, the existing technical system does not have a perfect supervision of oil state information, and system data feedback is not timely. These problems lead to the contamination of steam turbine oil during operation due to factors such as moisture, wear particles, temperature, humidity, and external pollutants, which in turn causes equipment wear and accelerated aging, ultimately resulting in low work efficiency and increased failure rate of oil-using equipment, and even threatening the safe operation of the entire power system. SUMMARY

[0004] Therefore, the present application provides a steam turbine lubricating oil intelligent operation and maintenance system, method, device and storage medium, which mainly aims to solve the problems of response lag, discontinuous data, insufficient fault prediction capability, and other obvious defects in the prior art, which cause equipment wear and accelerated aging, ultimately resulting in low work efficiency and increased failure rate of oil-using equipment, and even threatening the safe operation of the entire power system.

[0005] According to a first aspect of the present application, a steam turbine lubricating oil intelligent operation and maintenance system is provided, which comprises a data acquisition module, a data transmission module, a visual oil cloud operation and maintenance module, and an oil filtration execution module.

[0006] The data acquisition module is configured to acquire state parameters of steam turbine lubricating oil and transmit the state parameters to the data transmission module.

[0007] The data transmission module is configured to store the state parameters and transmit them to the visual oil cloud operation and maintenance module.

[0008] The visual oil cloud operation and maintenance module is configured to perform data analysis on the state parameters by using a hybrid machine learning model to obtain an oil health index, visually display the oil health index, and generate a filter device start-stop instruction when detecting that the oil health index belongs to a second-level warning of a hierarchical warning mechanism, and transmit the filter device start-stop instruction to the filter execution module.

[0009] The filter execution module is configured to perform multistage filtering and constant temperature control on the steam turbine lubricating oil by using a filter assembly in response to the filter device start-stop instruction.

[0010] According to a second aspect of the present application, an intelligent operation and maintenance method for a steam turbine lubricating oil is provided, which includes the following steps:

[0011] Collecting state parameters of the steam turbine lubricating oil based on a data collection module;

[0012] Storing the state parameters and transmitting them to a visual oil cloud operation and maintenance module based on a data transmission module;

[0013] Performing data analysis on the state parameters by using a hybrid machine learning model based on the visual oil cloud operation and maintenance module to obtain an oil health index, and visually displaying the oil health index;

[0014] When detecting that the oil health index belongs to a second-level warning of a hierarchical warning mechanism based on the visual oil cloud operation and maintenance module, generating a filter device start-stop instruction and transmitting the filter device start-stop instruction to a filter execution module;

[0015] Performing multistage filtering and constant temperature control on the steam turbine lubricating oil by using a filter assembly based on the filter execution module in response to the filter device start-stop instruction.

[0016] According to a third aspect of the present application, a device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the system of any one of the first aspect when executing the computer program.

[0017] According to a fourth aspect of the present application, a storage medium is provided, which stores a computer program, and the computer program implements the steps of the system of any one of the first aspect when executed by a processor.

[0018] By means of the above technical solutions, the technical solutions provided by the embodiments of the present application have at least the following advantages:

[0019] The application provides a steam turbine lubricating oil intelligent operation and maintenance system, method, equipment and storage medium, and comprises a data acquisition module, a data transmission module, a visual oil cloud operation and maintenance module and an oil filtering execution module. Due to the fact that manual inspection relies on experience and data is delayed, the oil deterioration trend is difficult to capture in time, and equipment failure often breaks out suddenly, the data acquisition module covers multi-dimensional parameters such as oil physical, chemical and morphological parameters through multiple sensors; the data transmission module guarantees zero data loss and low delay, and solves the data fault problem of traditional manual meter reading; the mixed model of the visual cloud operation and maintenance module converts static parameters into dynamic trends, residual life and abnormal early warning, realizes intelligent conversion from data to decision, and the oil filtering execution module automatically responds to instructions, completes the closed-loop processing of contaminated oil to clean oil without manual intervention, thereby realizing real-time monitoring and intelligent analysis of the operation state of the steam turbine, and automatically starting the oil filtering system when the set threshold is reached, ensuring that the oil is in a normal state, thereby reducing maintenance cost and improving operation and maintenance efficiency.

[0020] The above description is only a summary of the technical scheme of the application, in order to more clearly understand the technical means of the application, the specific embodiments of the application can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the application more obvious and easy to understand, the following specific embodiments of the application are described. BRIEF DESCRIPTION OF DRAWINGS

[0021] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not meant to limit the present application. Moreover, the same reference numerals in the attached drawings refer to the same or similar components. In the drawings:

[0022] Figure 1 A system architecture schematic diagram of a steam turbine lubricating oil intelligent operation and maintenance system provided by an embodiment of the application is shown;

[0023] Figure 2 A system architecture schematic diagram of a data acquisition module provided by an embodiment of the application is shown;

[0024] Figure 3 This paper illustrates a system architecture diagram of a data transmission module provided in an embodiment of this application.

[0025] Figure 4 This illustration shows a flowchart of an intelligent operation and maintenance method for turbine lubricating oil according to an embodiment of this application.

[0026] Figure 5 This paper illustrates a schematic flowchart of another intelligent operation and maintenance method for turbine lubricating oil provided in an embodiment of this application.

[0027] Figure 6 This illustration shows a structural schematic diagram of an intelligent operation and maintenance system for turbine lubricating oil provided in an embodiment of this application;

[0028] Figure 7 A schematic diagram of the device structure of an embodiment of this application is shown. Detailed Implementation

[0029] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0030] Furthermore, 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 number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0031] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0032] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0033] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the present application to those skilled in the art.

[0034] This application provides an intelligent operation and maintenance system for turbine lubricating oil, such as... Figure 1 As shown, the system includes a data acquisition module 101, a data transmission module 102, a visualized oil cloud operation and maintenance module 103, and an oil filtration execution module 104.

[0035] The data acquisition module 101 is used to collect the state parameters of the turbine lubricating oil and transmit these parameters to the data transmission module 102. Traditional methods rely on manual sampling and can only detect limited parameters such as viscosity and particle size, failing to capture the chemical changes and dynamic characteristics of the oil, making it difficult to detect hidden faults. Therefore, this application uses multiple sensors to collect multi-dimensional indicators of the oil in real time, achieving full-parameter, high-frequency, and comprehensive monitoring of the oil's state. This allows for the early detection of hidden fault precursors such as oil oxidation and additive loss, enabling maintenance to shift from passively waiting for faults to proactively identifying potential problems.

[0036] The data transmission module 102 is used to store status parameters and transmit them to the visualized oil cloud operation and maintenance module 103. Traditional methods lack reliable data transmission and storage mechanisms, and oil parameters are easily lost and difficult to trace. Therefore, this application uses a data exchanger and an oil cloud server to ensure zero data loss in the data collection, storage, and transmission process, and also enables long-term traceability of historical data, providing a data foundation for the full life cycle management of oil.

[0037] The visualized oil cloud operation and maintenance module 103 utilizes a hybrid machine learning model to analyze state parameters, obtain oil health indicators, and visualize these indicators. When an oil health indicator falls under the level-two warning mechanism (i.e., exceeding a preset upper limit but not exceeding a safety limit, or showing a continuously deteriorating trend, such as viscosity exceeding 1.2 times the new oil standard, particle size exceeding grade 8, or moisture exceeding 100 mg / L), the module generates start / stop commands for the oil filtration device. These commands are then transmitted to the oil filtration execution module 104, providing timely warnings and maintenance suggestions, thus improving operation and maintenance efficiency. Traditional methods rely on manual interpretation of oil reports, depending on experience, leading to significant discrepancies in conclusions among different maintenance personnel and an inability to predict future conditions. Therefore, the visualized oil cloud operation and maintenance module employs a hybrid machine learning model including LSTM, Random Forest, and SVM, replacing experience-based interpretation with algorithms. LSTM can predict future trends in oil parameters, Random Forest can quantify remaining lifespan, and SVM can identify abnormal patterns, transforming oil health diagnosis from subjective and ambiguous judgment to objective and accurate decision-making.

[0038] The oil filtration execution module 104 is used to respond to the start and stop commands of the oil filtration device, and performs multi-stage filtration and temperature control of the turbine lubricating oil through the oil filtration assembly. Traditional methods require manual coordination to start / stop the oil filtration / machine after detecting an oil abnormality, a time-consuming process that cannot be handled promptly. Therefore, this application constructs an automated closed loop including early warning, decision-making, and execution, reducing the anomaly response time from hours of manual intervention to minutes within the system, preventing minor anomalies from escalating into major failures and ensuring continuous equipment operation.

[0039] Specifically, such as Figure 2 As shown, the data acquisition module 101 includes a multi-parameter sensing unit 1011, a signal conversion unit 1012, and a transmission unit 1013.

[0040] The multi-parameter sensing unit 1011 is connected to the signal conversion unit 1012 and is used to acquire flow images of turbine lubricating oil through a visual sensor. The visual sensor can complete real-time dynamic image capture within 2-5 seconds under oil flow conditions, accurately detecting multi-dimensional indicators such as particle size distribution, quantity concentration, and moisture contamination. Subsequently, a hybrid machine learning model can intelligently classify particle morphology characteristics according to standards such as ISO / NAS1638 and ASTM, effectively distinguishing particle types such as cutting wear, fatigue wear, dust, and contamination. The fluorescence spectrum signal of turbine lubricating oil is acquired through a nanosensor. Nanoscale quantum dot fluorescent markers (such as CdSe / ZnS quantum dots) are pre-introduced into the lubricating oil to track the oil aging path in real time through fluorescence spectroscopy. Different aging products in the oil (such as aldehydes and ketones generated by oxidation and nitrogen-containing compounds generated by nitration) are utilized under specific excitation. The fluorescence characteristics at a specific wavelength are used to collect the fluorescence spectrum signal of the oil during use. Subsequently, characteristic fluorescence peaks are extracted and compared with an established fluorescence database of aging products. Combined with the trend of fluorescence intensity changes over time, the formation rate of various aging products (such as oxidation and nitration products) is determined, thereby identifying the path and stage of oil degradation from initial aging. This enables visualized monitoring of the molecular-level oil degradation process, overcoming the limitations of traditional sensors that only monitor macroscopic parameters. A viscosity sensor collects the shear resistance signal of the turbine lubricating oil; a contaminant concentration signal is collected by a contaminant sensor; a particle counter collects the particle size distribution and number concentration signals; a moisture sensor collects the free water and dissolved water content signals; and a temperature sensor collects the real-time temperature signal of the turbine lubricating oil. The multi-parameter sensing unit 1011 solves the problems of difficult particle tracing, difficult quantification of aging, and incomplete parameters in traditional oil monitoring by using a synergistic design of visual sensor for dynamic particle capture, nanosensor for molecular-level aging tracking, and multiple conventional sensors to supplement basic parameters.

[0041] The signal conversion unit 1012 is connected to the transmission unit 1013 and is used to perform digital-to-analog conversion on fluorescence spectral signals, shear resistance signals, pollutant concentration signals, particle size distribution and number concentration signals, free water content and dissolved water content signals, and real-time temperature signals to obtain state parameters. Since the signal types output by different sensors are diverse, direct transmission without processing can easily lead to data ambiguity and packet loss. Therefore, this application performs digital-to-analog conversion and standardized packaging on multi-source heterogeneous signals to provide low-noise, highly consistent input for downstream modules, thereby improving the accuracy of model analysis.

[0042] The transmission unit 1013 is used to transmit status parameters to the data transmission module 102. Traditional methods involve manual sampling and data entry, which suffers from high latency and is prone to errors, making it unsuitable for real-time operation and maintenance decision-making. This application achieves data upload in seconds through 5G / WIFI wireless transmission technology.

[0043] Specifically, such as Figure 3 As shown, the data transmission module 102 includes a data exchange 1021 and an oil cloud server 1022.

[0044] The data exchange 1021 is connected to the oil cloud server 1022 to receive the status parameters transmitted by the data acquisition module 101, perform verification and formatting on the status parameters, and transmit the processed status parameters to the oil cloud server 1022 to reduce the error rate of model analysis.

[0045] The oil cloud server 1022 stores the processed status parameters in the oil status database and transmits the processed status parameters to the visualized oil cloud operation and maintenance module 103.

[0046] Specifically, the visualized oil cloud operation and maintenance module 103 is used to input state parameters into the LSTM neural network of a hybrid machine learning model for time series prediction, mining the long-term dynamic correlations of oil parameters, such as the coupled changes in viscosity and temperature, and operating time, to obtain time series prediction curves. The random forest algorithm of the hybrid machine learning model is used to assess the remaining life of the state parameters, weighting multi-dimensional features including particle distribution, water emulsification degree, and viscosity change rate to obtain the remaining life of the oil. The state parameters are then input into the support vector machine of the hybrid machine learning model for anomaly pattern recognition, obtaining anomaly classification results, identifying minor anomaly patterns, and improving early warning accuracy. The time series prediction curves, remaining life of the oil, and anomaly classification results are used as oil health indicators, and these indicators are visualized using interactive visualization components. Traditional oil analysis relies on single tools such as manual trend chart reading and simple threshold alarms, which suffer from problems such as inaccurate trend prediction, vague life assessment, and crude anomaly identification. Therefore, the visualized oil cloud operation and maintenance module 103 uses the future trend output by LSTM to intuitively display the direction of oil degradation and provide early warning of potential risks; it uses the remaining life output by random forest to accurately guide planned oil changes, avoiding premature oil changes that waste resources or delayed oil changes that damage equipment; and it uses SVM to identify anomaly types, associate them with equipment components, and directly point to the fault, shortening the operation and maintenance response time.

[0047] Specifically, the visualized oil cloud operation and maintenance module 103 is used to generate operation and maintenance attention prompts when the oil health indicators are detected to be at the first level of the graded early warning mechanism, that is, when the oil health indicators are slightly higher than the preset baseline or the trend fluctuations are abnormal but do not significantly exceed the safety limit value, and to visualize the operation and maintenance attention prompts.

[0048] When the oil health indicators are detected to be at level three of the graded early warning mechanism—meaning the oil health indicators exceed safety limits or show a drastic abnormal trend (such as a sudden increase in abrasive particles)—an equipment shutdown protection command is generated and sent to the turbine unit to stop operation, triggering the equipment shutdown protection without manual intervention. The visualized oil cloud operation and maintenance module 103 can also achieve remote monitoring and intelligent diagnosis through a cloud platform, supporting remote expert intervention and improving the accuracy and timeliness of fault diagnosis.

[0049] Specifically, the oil filtration module 104 is used to open the oil pump and inlet valve, close the return valve, and drive the turbine lubricating oil through the primary and secondary filtration components for staged filtration. Simultaneously, the electric heater is activated, and the oil temperature is collected in real time by a temperature sensor. Based on the temperature control feedback unit, the heating power is dynamically adjusted according to the oil temperature to maintain the viscosity of the turbine lubricating oil within a preset optimal range. When the cleanliness and viscosity of the turbine lubricating oil meet the set thresholds, the return valve is opened and the oil pump is shut off to ensure the oil is in a normal state and prevent turbine malfunctions. Multi-stage filtration removes large particles and micron-sized particles sequentially through primary and secondary filtration, significantly reducing contamination and preventing wear caused by the recycling of abrasive particles and contaminants. The electric heater maintains the oil within the optimal viscosity range through precise temperature control, preventing decreased fluidity due to low temperatures or accelerated oil aging due to high temperatures. Combined with the temperature control feedback mechanism, dynamic adjustment and constant temperature control are achieved through the temperature control feedback unit, ensuring continuous lubrication of the equipment under clean and stable conditions.

[0050] Specifically, the temperature control feedback unit increases the heating power by a preset ratio when the detected oil temperature is below the target lower limit; and decreases the heating power in a stepped manner when the detected oil temperature is above the target upper limit. Traditional temperature control relies on fixed power heating, which suffers from large oil temperature fluctuations, high energy consumption, and inability to adapt to complex operating conditions. Therefore, a dynamic power adjustment algorithm is used to achieve low-temperature compensation, i.e., when the oil temperature is below the target lower limit, the heating power is increased by a preset ratio to quickly raise the temperature to the optimal range; and over-temperature suppression, i.e., when the oil temperature is above the target upper limit, the power is reduced in a stepped manner, combined with air cooling, to prevent the oil from overheating and aging.

[0051] Optionally, a microencapsulated self-healing agent (such as molybdenum disulfide nanoparticles) can be pre-added to the lubricating oil. The multi-parameter sensing unit 1011 monitors indicators such as sudden viscosity drops, increased abrasive particle concentration, and abnormal oil film thickness to determine when there is a risk of oil film rupture. Combining electromagnetic field control technology, it guides the microcapsules to deposit directionally into the wear area, and uses ultrasonic waves to trigger the release of the repair agent from the microcapsules, achieving real-time repair. This process works synergistically with the oil filtration system, allowing the intelligent repair system to quickly intervene and remedy sudden risks, improving system stability and oil lifespan. Through self-healing lubricating film generation technology, self-repair can be completed within one hour of an oil film rupture warning, avoiding downtime losses, reducing bearing wear by 40%, and extending equipment overhaul cycles.

[0052] The intelligent operation and maintenance system for steam turbine lubricating oil relies on the computing power of servers to provide services to users. The servers can be independent servers or servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0053] This application provides an intelligent operation and maintenance system for turbine lubricating oil. Compared with the prior art, this application includes a data acquisition module, a data transmission module, a visualized oil cloud operation and maintenance module, and an oil filtration execution module. The data acquisition module is used to collect the state parameters of the turbine lubricating oil and transmit the state parameters to the data transmission module. The data transmission module is used to store the state parameters and transmit them to the visualized oil cloud operation and maintenance module. The visualized oil cloud operation and maintenance module is used to analyze the state parameters using a hybrid machine learning model to obtain oil health indicators, visualize the oil health indicators, and generate an oil filtration device start / stop command when the oil health indicator is detected to be a level 2 warning in the graded early warning mechanism, and transmit the oil filtration device start / stop command to the oil filtration execution module. The oil filtration execution module is used to respond to the oil filtration device start / stop command and perform multi-stage filtration and constant temperature control operations on the turbine lubricating oil through the oil filtration components. Because manual inspections rely on experience and suffer from data lag, it is difficult to capture oil deterioration trends in a timely manner, and equipment failures often occur suddenly. Therefore, this application uses a multi-sensor data acquisition module to cover multiple dimensions of oil parameters, including physical, chemical, and morphological parameters; a data transmission module ensures zero data loss and low latency, solving the data gap problem of traditional manual meter reading; and a hybrid model of a visualization cloud operation and maintenance module transforms static parameters into dynamic trends, remaining lifespan, and anomaly warnings, achieving intelligent transformation from data to decision-making. The oil filtration execution module automatically responds to commands, completing a closed-loop process from contaminated oil to clean oil without manual intervention, thereby achieving real-time monitoring and intelligent analysis of the turbine's operating status. When a set threshold is reached, the oil filtration system is automatically activated to ensure the oil is in a normal state, thereby reducing maintenance costs and improving operation and maintenance efficiency.

[0054] Furthermore, as Figure 1 In a specific implementation of the method, this application provides an intelligent operation and maintenance method for turbine lubricating oil, such as... Figure 4 As shown, the method includes:

[0055] 201. Collect the state parameters of the turbine lubricating oil based on the data acquisition module.

[0056] In this embodiment, traditional manual sampling and monitoring can only acquire limited static parameters such as viscosity and total particle count, failing to capture the dynamic characteristics and chemical changes of the oil, making it difficult to detect hidden faults in a timely manner. Therefore, by using multiple types of sensors to collaboratively collect data, covering all dimensions of the oil's physical, chemical, and morphological indicators, visual sensors capture oil flow images, enabling analysis of particle movement trajectories and morphologies to identify wear types; nanosensors collect fluorescence spectra, monitoring additive loss and oil oxidation paths; and sensors for viscosity, contamination, moisture, and temperature collect basic parameters, ensuring comprehensive knowledge of the oil's condition and solving the fundamental problems of incomplete, inaccurate, and untimely data in traditional oil maintenance.

[0057] 202. The status parameters are stored and transmitted to the visualized oil cloud operation and maintenance module based on the data transmission module.

[0058] In this embodiment of the application, traditional data transmission relies on manual copying and simple network transmission, which has problems such as high packet loss rate, high latency and no verification mechanism. Therefore, reliable verification and low latency transmission are achieved by using a data exchange and an oil cloud server, which not only ensures low latency for real-time applications, but also reduces long-term storage costs.

[0059] 203. Based on the visualization oil cloud operation and maintenance module, a hybrid machine learning model is used to analyze the status parameters to obtain oil health indicators, and the oil health indicators are visualized.

[0060] In this embodiment, a hybrid machine learning model including LSTM, random forest, and SVM is used to achieve full-dimensional coverage of oil health diagnosis, transforming oil analysis from single-dimensional judgment to three-dimensional diagnosis, covering all scenarios of trends, lifespan, and anomalies, and improving diagnostic accuracy.

[0061] 204. When the oil health indicator is detected by the visual oil cloud operation and maintenance module to be a level 2 warning under the graded early warning mechanism, the oil filter device start / stop command is generated and transmitted to the oil filter execution module.

[0062] In this embodiment, traditional operation and maintenance (O&M) systems shut down even for minor anomalies such as slight oil contamination, resulting in significant production losses; and failure to intervene in major anomalies in a timely manner leads to equipment damage. When the cloud-based visual oil O&M module detects that the oil health indicators are within a preset range that requires intervention but not shutdown, an oil filtration command is triggered, which avoids overreacting to minor issues and ensuring timely handling of problems.

[0063] 205. Based on the oil filtration execution module responding to the start and stop commands of the oil filtration device, the oil filtration assembly performs multi-stage filtration and constant temperature control operations on the turbine lubricating oil.

[0064] In this embodiment, automatic start-stop control, full-process automation, and real-time status feedback are achieved through a closed loop of command response, automatic execution, and status feedback, reducing manpower input, shortening oil filtration response delay, improving oil viscosity stability, and reducing energy consumption.

[0065] This application provides an intelligent operation and maintenance method for turbine lubricating oil. Compared with the prior art, this application collects the state parameters of the turbine lubricating oil based on a data acquisition module, stores the state parameters based on a data transmission module, and transmits them to a visualized oil cloud operation and maintenance module. The visualized oil cloud operation and maintenance module uses a hybrid machine learning model to analyze the state parameters and obtain oil health indicators. The oil health indicators are then visualized. When the visualized oil cloud operation and maintenance module detects that the oil health indicators belong to the second-level warning of the graded warning mechanism, it generates a start / stop command for the oil filtration device and transmits the start / stop command to the oil filtration execution module. Based on the response of the oil filtration device start / stop command, the oil filtration component performs multi-stage filtration and constant temperature control operations on the turbine lubricating oil. Because manual inspections rely on experience and suffer from data lag, it is difficult to capture oil deterioration trends in a timely manner, and equipment failures often occur suddenly. Therefore, the data acquisition module uses multiple sensors to cover multiple dimensions of oil parameters, including physical, chemical, and morphological parameters. The data transmission module ensures zero data loss and low latency, solving the data gap problem of traditional manual meter reading. The hybrid model of the visualization cloud operation and maintenance module transforms static parameters into dynamic trends, remaining lifespan, and anomaly warnings, realizing intelligent transformation from data to decision. The oil filtration execution module automatically responds to commands, completing the closed-loop treatment from contaminated oil to clean oil without manual intervention. This enables real-time monitoring and intelligent analysis of the turbine's operating status, and automatically starts the oil filtration system when a set threshold is reached, ensuring that the oil is in a normal state, thereby reducing maintenance costs and improving operation and maintenance efficiency.

[0066] Furthermore, as a refinement and extension of the specific implementation methods of the above embodiments, and in order to fully illustrate the specific implementation process of this embodiment, this application provides another intelligent operation and maintenance method for turbine lubricating oil, such as... Figure 5 As shown, the method includes:

[0067] 301. The multi-parameter sensing unit based on the data acquisition module acquires fluorescence spectrum signals, shear resistance signals, pollutant concentration signals, particle size distribution and number concentration signals, free water content and dissolved water content signals, and real-time temperature signals.

[0068] In this embodiment, based on the multi-parameter sensing unit of the data acquisition module, flow images of the turbine lubricating oil are acquired through a visual sensor. By capturing these flow images, the trajectory and morphology of particles are analyzed to identify cutting wear and fatigue wear particles. The fluorescence spectrum signal of the turbine lubricating oil is acquired through a nanosensor to monitor additive loss and oil oxidation paths. The shear resistance signal of the turbine lubricating oil is acquired through a viscosity sensor. The contaminant concentration signal of the turbine lubricating oil is acquired through a contamination sensor. The particle size distribution and number concentration signals of the turbine lubricating oil are acquired through a particle counter. The free water content and dissolved water content signals of the turbine lubricating oil are acquired through a moisture sensor. The real-time temperature signal of the turbine lubricating oil is acquired through a temperature sensor. Traditional sensor-acquired data suffers from limitations such as single-dimensionality and high noise, making it difficult for downstream analysis models to distinguish particle types or quantify aging stages. Therefore, multi-sensor collaborative acquisition covers all physical, chemical, and dynamic dimensions of the oil, improving the fault detection rate.

[0069] 302. The signal conversion unit based on the data acquisition module performs digital-to-analog conversion on fluorescence spectrum signal, shear resistance signal, pollutant concentration signal, particle size distribution and number concentration signal, free water content and dissolved water content signal, and real-time temperature signal to obtain the state parameters of the turbine lubricating oil.

[0070] In this embodiment, the signal conversion unit of the data acquisition module performs digital-to-analog conversion on fluorescence spectral signals, shear resistance signals, contaminant concentration signals, particle size distribution and number concentration signals, free water content and dissolved water content signals, and real-time temperature signals to obtain state parameters. This converts the raw physical signals collected by the sensors into standardized state parameters that can be recognized and analyzed by the system, which can solve the problem that multi-source heterogeneous signals are difficult to use directly for intelligent analysis and ensure the accuracy of oil health index analysis.

[0071] 303. The transmission unit based on the data acquisition module transmits the status parameters to the data transmission module.

[0072] In this embodiment, the state parameters are transmitted to the data transmission module based on the transmission unit of the data acquisition module, ensuring that the oil state data is transmitted without delay or loss, and providing real-time and reliable data support for subsequent data verification, storage and intelligent analysis of the visualization cloud operation and maintenance module.

[0073] 304. The status parameters are stored and transmitted to the visualized oil cloud operation and maintenance module based on the data transmission module.

[0074] In this embodiment, a data exchange based on the data transmission module receives status parameters transmitted by the data acquisition module, verifies and formats the status parameters, eliminates errors, standardizes the format, and ensures data quality. The oil cloud server based on the data transmission module stores the processed status parameters in the oil status database and transmits them to the visualized oil cloud operation and maintenance module. This not only preserves complete data for the entire lifecycle analysis of oil but also provides real-time, standardized input for hybrid machine learning models, ensuring reliable data support for intelligent operation and maintenance functions such as oil health diagnosis and graded early warning.

[0075] 305. Based on the visualization oil cloud operation and maintenance module, a hybrid machine learning model is used to analyze the status parameters to obtain oil health indicators, and the oil health indicators are visualized.

[0076] In this embodiment, based on the visualized oil cloud operation and maintenance module, the state parameters are input into the LSTM neural network of the hybrid machine learning model for time series prediction to obtain the time series prediction curve, thereby realizing the prediction of oil condition trends. The random forest algorithm of the hybrid machine learning model is used to quantify the state parameters in multiple dimensions to obtain the remaining lifespan of the oil. The state parameters are input into the support vector machine of the hybrid machine learning model for anomaly pattern recognition to locate the fault type and obtain the anomaly classification result. The time series prediction curve, the remaining lifespan of the oil, and the anomaly classification result are used as oil health indicators, and the oil health indicators are visualized using a visual interactive component. This provides operation and maintenance personnel with a comprehensive basis for trend prediction, lifespan management, and anomaly early warning, while also lowering the threshold for data analysis and significantly improving the foresight and accuracy of operation and maintenance.

[0077] 306. When the oil health indicator is detected by the visual oil cloud operation and maintenance module to be a level 2 warning under the graded early warning mechanism, the oil filter device start / stop command is generated and transmitted to the oil filter execution module.

[0078] In this embodiment, a tiered early warning response mechanism is used to achieve precise, automated, and differentiated handling of oil abnormalities. Specifically, when the visualized oil cloud operation and maintenance module detects that the oil health indicators belong to the second-level early warning of the tiered early warning mechanism, a start / stop command for the oil filtration device is generated and transmitted to the oil filtration execution module. This allows for timely repair of issues such as oil cleanliness, preventing the fault from escalating and ensuring continuous operation of the equipment.

[0079] 307. Based on the oil filtration execution module responding to the start and stop commands of the oil filtration device, the oil filtration assembly performs multi-stage filtration and constant temperature control operations on the turbine lubricating oil.

[0080] In this embodiment, the oil pump and inlet valve are activated by the oil filtration module, while the return valve is closed. This drives the turbine lubricating oil through a primary and secondary filtration assembly for staged filtration. Simultaneously, an electric heater is activated, and the oil temperature is collected in real time by a temperature sensor. This two-stage filtration process intercepts impurities, improving oil cleanliness, ensuring lubricating oil quality, and reducing equipment wear risks. Next, the heating power is dynamically adjusted based on the oil temperature by the temperature control feedback unit to maintain the turbine lubricating oil viscosity within a preset optimal range. This prevents temperature fluctuations from affecting lubrication performance, precisely maintaining stable oil viscosity to meet the different operating conditions of the turbine. Specifically, when the temperature control feedback unit detects that the oil temperature is below the target lower limit, the heating power is increased according to a preset ratio; when the temperature control feedback unit detects that the oil temperature is above the target upper limit, the heating power is reduced in a stepped manner.

[0081] When the oil filtration module detects that the cleanliness and viscosity of the turbine lubricating oil meet the set thresholds, it opens the return oil valve and shuts down the oil pump, avoiding excessive oil filtration and wasting energy. The entire process is automated, reducing manual intervention and improving maintenance efficiency. Based on this process, the lead time for fault warnings can be significantly increased from the traditional 24 hours to 72-120 hours, effectively preventing equipment failures. Furthermore, the improved oil detection sensitivity allows for the detection of finer particulate contamination, ensuring oil cleanliness and significantly reducing maintenance costs. Through adaptive maintenance strategies, unnecessary over-maintenance and downtime losses are minimized.

[0082] 308. When the cloud-based visual oil health module detects that the oil health indicators belong to the first-level warning of the graded early warning mechanism, an operation and maintenance attention prompt is generated and the operation and maintenance attention prompt is displayed visually.

[0083] In this embodiment, when the cloud-based visual oil health module detects that the oil health indicator belongs to the first-level warning of the graded early warning mechanism, an operation and maintenance attention prompt is generated and visualized, so that operation and maintenance personnel can pay attention to the trend changes of the oil in advance and intervene in the early stage of the fault.

[0084] 309. When the cloud-based visual oil operation and maintenance module detects that the oil health indicators belong to the third level of the graded early warning mechanism, it generates an equipment shutdown protection command and sends the equipment shutdown protection command to the turbine unit to stop the turbine unit from operating.

[0085] In this embodiment, when the cloud-based visual oil maintenance module detects that the oil health indicators fall under the level three warning category of the tiered early warning mechanism, an equipment shutdown protection command is generated and sent to the turbine unit to stop its operation. This effectively prevents serious oil abnormalities from causing significant damage to the turbine unit, reducing equipment maintenance costs and downtime losses. This tiered approach balances the need for continuous equipment operation with fault prevention requirements, transforming maintenance response from passive manual judgment to automated, tiered system handling, significantly improving the intelligence level of oil maintenance and equipment protection efficiency.

[0086] Based on the above process, the structural schematic diagram of an intelligent operation and maintenance method for turbine lubricating oil proposed in this application embodiment is as follows:

[0087] like Figure 6 As shown, the lubricating oil from the turbine's main oil tank enters the oil filter module 9, passes through the inlet valve 1, inlet filter 2, oil pump 3, and electric heater assembly 4, and is purified sequentially through the primary filter assembly 5 and the secondary filter assembly 6. Then, it passes through the return oil filter 7 and return oil valve 8 to complete the circulation. In the data acquisition and sensing system, the visual sensor captures information such as oil flow images. The signal acquisition module and transmission module transmit multi-dimensional oil parameters to the server via the data exchange. The server synchronizes the processed data to the oil cloud, and finally, the data is stored, analyzed, and visualized on the visual oil cloud operation and maintenance platform, supporting oil health diagnosis, graded early warning, and intelligent operation and maintenance decision-making.

[0088] This application provides an intelligent operation and maintenance method for turbine lubricating oil. Compared with the prior art, this application collects the state parameters of the turbine lubricating oil based on a data acquisition module, stores the state parameters based on a data transmission module, and transmits them to a visualized oil cloud operation and maintenance module. The visualized oil cloud operation and maintenance module uses a hybrid machine learning model to analyze the state parameters and obtain oil health indicators. The oil health indicators are then visualized. When the visualized oil cloud operation and maintenance module detects that the oil health indicators belong to the second-level warning of the graded warning mechanism, it generates a start / stop command for the oil filtration device and transmits the start / stop command to the oil filtration execution module. Based on the response of the oil filtration device start / stop command, the oil filtration component performs multi-stage filtration and constant temperature control operations on the turbine lubricating oil. Because manual inspections rely on experience and suffer from data lag, it is difficult to capture oil deterioration trends in a timely manner, and equipment failures often occur suddenly. Therefore, the data acquisition module uses multiple sensors to cover multiple dimensions of oil parameters, including physical, chemical, and morphological parameters. The data transmission module ensures zero data loss and low latency, solving the data gap problem of traditional manual meter reading. The hybrid model of the visualization cloud operation and maintenance module transforms static parameters into dynamic trends, remaining lifespan, and anomaly warnings, realizing intelligent transformation from data to decision. The oil filtration execution module automatically responds to commands, completing the closed-loop treatment from contaminated oil to clean oil without manual intervention. This enables real-time monitoring and intelligent analysis of the turbine's operating status, and automatically starts the oil filtration system when a set threshold is reached, ensuring that the oil is in a normal state, thereby reducing maintenance costs and improving operation and maintenance efficiency.

[0089] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0090] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0091] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

[0092] In an exemplary embodiment, see Figure 7Furthermore, a computer device is provided, comprising a bus, a processor, a memory, and a communication interface. It may also include an input / output interface and a display device, wherein the various functional units can communicate with each other via the bus. The memory stores computer programs, and the processor executes the programs stored in the memory to perform the intelligent operation and maintenance system for turbine lubricating oil described in the above embodiments.

[0093] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the intelligent operation and maintenance system for turbine lubricating oil.

[0094] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented in hardware or by using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0095] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application.

[0096] Those skilled in the art will understand that the modules in the apparatus of the implementation scenario can be distributed within the apparatus of the implementation scenario as described, or they can be located in one or more apparatuses different from this implementation scenario, with corresponding changes. The modules of the above-described implementation scenario can be combined into one module, or they can be further divided into multiple sub-modules.

[0097] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of the implementation scenario.

[0098] The above disclosures are only a few specific implementation scenarios of this application. However, this application is not limited to these. Any variations that can be conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. An intelligent operation and maintenance system for steam turbine lubricating oil, characterized in that, It includes a data acquisition module, a data transmission module, a visualized oil cloud operation and maintenance module, and an oil filtration execution module; The data acquisition module is used to collect the state parameters of the turbine lubricating oil and transmit the state parameters to the data transmission module. The data transmission module is used to store the status parameters and transmit them to the visualized oil cloud operation and maintenance module; The visualized oil cloud operation and maintenance module is used to perform data analysis on the status parameters using a hybrid machine learning model to obtain oil health indicators, visualize the oil health indicators, and when the oil health indicators are detected to belong to the second level of the graded early warning mechanism, generate an oil filtration device start / stop command and transmit the oil filtration device start / stop command to the oil filtration execution module. The oil filtration execution module is used to respond to the start and stop command of the oil filtration device and perform multi-stage filtration and constant temperature control on the turbine lubricating oil through the oil filtration assembly; The data acquisition module is used to identify the wear area where the turbine lubricating oil is at risk of oil film rupture when the detection determines that the oil film rupture is at risk. It then uses electromagnetic field control technology to guide the microencapsulated self-healing agent in the turbine lubricating oil to be deposited directionally into the wear area, and uses ultrasonic technology to trigger the microencapsulated self-healing agent to release the repair agent to repair the wear area.

2. The system according to claim 1, characterized in that, The data acquisition module includes a multi-parameter sensing unit, a signal conversion unit, and a transmission unit; The multi-parameter sensing unit is connected to the signal conversion unit and is used to acquire flow images of the turbine lubricating oil through a visual sensor, acquire fluorescence spectrum signals of the turbine lubricating oil through a nanosensor, acquire shear resistance signals of the turbine lubricating oil through a viscosity sensor, acquire contaminant concentration signals of the turbine lubricating oil through a contamination sensor, acquire particle size distribution and number concentration signals of the turbine lubricating oil through a particle counter, acquire free water content and dissolved water content signals of the turbine lubricating oil through a moisture sensor, and acquire real-time temperature signals of the turbine lubricating oil through a temperature sensor. The signal conversion unit is connected to the transmission unit and is used to perform digital-to-analog conversion on the fluorescence spectrum signal, the shear resistance signal, the pollutant concentration signal, the particle size distribution and number concentration signal, the free water content and dissolved water content signal, and the real-time temperature signal to obtain the state parameters; The transmission unit is used to transmit the status parameters to the data transmission module.

3. The system according to claim 1, characterized in that, The data transmission module includes a data exchanger and an oil cloud server; The data exchange is connected to the oil cloud server and is used to receive the status parameters transmitted by the data acquisition module, perform verification and formatting on the status parameters, and transmit the processed status parameters to the oil cloud server. The oil cloud server stores the processed status parameters in the oil status database and transmits the processed status parameters to the visualized oil cloud operation and maintenance module.

4. The system according to claim 1, characterized in that, The visualized oil cloud operation and maintenance module is used to input the state parameters into the LSTM neural network of the hybrid machine learning model for time series prediction, and obtain the time series prediction curve. The remaining life of the oil is obtained by using the random forest algorithm of the hybrid machine learning model to evaluate the state parameters. The state parameters are input into the support vector machine of the hybrid machine learning model to perform anomaly pattern recognition and obtain anomaly classification results. The time series prediction curve, the remaining life of the oil, and the anomaly classification results are used as the oil health indicators, and the oil health indicators are visualized using a visual interactive component.

5. The system according to claim 1, characterized in that, The visualized oil cloud operation and maintenance module is used to generate an operation and maintenance attention prompt when the oil health indicator is detected to be a first-level warning in the graded early warning mechanism, and to visualize the operation and maintenance attention prompt. When the oil health indicator is detected to be at level three of the graded early warning mechanism, an equipment shutdown protection command is generated and sent to the turbine unit to stop the turbine unit from operating.

6. The system according to claim 1, characterized in that, The oil filtration module is used to turn on the oil pump and the oil inlet valve, close the oil return valve, drive the turbine lubricating oil to pass through the primary filter component and the secondary filter component for staged filtration, and at the same time start the electric heater and collect the oil temperature in real time through the temperature sensor. Based on the temperature control feedback unit, the heating power is dynamically adjusted according to the oil temperature so that the viscosity of the turbine lubricating oil is maintained within a preset optimal range; When the cleanliness and viscosity of the turbine lubricating oil meet the set threshold, the return oil valve is opened and the oil pump is shut down.

7. The system according to claim 6, characterized in that, The temperature control feedback unit is used to increase the heating power according to a preset ratio when the oil temperature is detected to be lower than the target lower limit. When the oil temperature is detected to be higher than the target temperature limit, the heating power is reduced in a stepwise manner.

8. A method for intelligent operation and maintenance of steam turbine lubricating oil, characterized in that, include: The state parameters of the turbine lubricating oil are collected based on the data acquisition module; The status parameters are stored and transmitted to the visualized oil cloud operation and maintenance module based on the data transmission module; Based on the aforementioned visualized oil cloud operation and maintenance module, a hybrid machine learning model is used to perform data analysis on the status parameters to obtain oil health indicators, which are then visualized. When the visual oil cloud operation and maintenance module detects that the oil health indicator belongs to the second level of the graded early warning mechanism, it generates an oil filtration device start / stop command and transmits the oil filtration device start / stop command to the oil filtration execution module. Based on the oil filtration execution module responding to the start / stop command of the oil filtration device, the oil filtration assembly performs multi-stage filtration and constant temperature control on the turbine lubricating oil. When the data acquisition module detects that there is a risk of oil film rupture in the turbine lubricating oil, the wear area with the risk of oil film rupture is identified. Electromagnetic field control technology is used to guide the microencapsulated self-healing agent in the turbine lubricating oil to be deposited in the wear area. Ultrasonic technology is used to trigger the microencapsulated self-healing agent to release the repair agent to repair the wear area.

9. A device comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the system according to any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the system according to any one of claims 1 to 7.

Images