Thermal power plant equipment lubricating oil on-line monitoring system and method thereof

By using timed signal capture and statistical analysis technology in the online monitoring system of lubricant equipment in thermal power plant equipment, real-time monitoring and dynamic early warning of lubricant oil is achieved, solving the problems of untimely response and uneven resource allocation in the existing technology, and improving the The accuracy and maintenance efficiency of equipment failure prediction extends equipment life and reduces risks.

CN119001068APending Publication Date: 2024-11-22CHN ENERGY SHENHUA JIUJIANG POWER GENERATION CO LTD +2
View PDF 0 Cites 27 Cited by

Patent Information

Application Number
CN202411001699.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-25
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The existing online monitoring system for equipment lubricant oil in thermal power plants has shortcomings in real-time data processing and dynamic threshold adjustment, resulting in insufficient response to equipment abnormal status, lack of maintenance strategies and dynamic resource allocation for specific equipment adjustments, and uneven resource utilization , low maintenance efficiency, long-term trend forecasting fails to integrate sufficient historical data with oil real-time monitoring data, affecting the accuracy and reliability of the prediction results.

Method used

The parameter real-time monitoring module uses the timing signal capture technology to record continuous data flow, implement threshold monitoring and distinguish between normal and abnormal readings, and generate real-time oil monitoring data; the oil condition analysis module conducts statistical analysis to identify abnormal fluctuations, and the abnormal warning signal sending module automatically Trigger warning notification; the maintenance scheduling module updates the maintenance plan in real time and dynamically allocates resources; the health trend analysis module uses statistical methods to make long-term trend predictions, adjusts monitoring parameters and sensitivity, and forms a lubricant monitoring execution plan.

Benefits of technology

Significantly improves rapid response to abnormal states, optimizes maintenance window time, ensures operational continuity and efficiency, provides in-depth insights into equipment health, helps predict and prevent potential problems, extends equipment service life, and reduces non-instances The risk of planned downtime improves the operating efficiency and safety of thermal power plants.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119001068A_ABST
    Figure CN119001068A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of equipment monitoring, in particular to a thermal power plant equipment lubricating oil online monitoring system and method, and the system comprises a parameter real-time monitoring module, an oil condition analysis module, an abnormity early warning and transmitting module, a maintenance scheduling module, a health trend analysis module and an equipment performance strengthening module. In the invention, by implementing a timing signal capturing technology and threshold monitoring, the system improves the quick response capability to an abnormal state, and combines continuous data flow recording and oil real-time monitoring data analysis, capturing and classification of abnormal fluctuations, so that a maintenance team quickly takes preventive measures, the equipment failure risk is reduced, and the reliability of the system is improved. The real-time updated maintenance scheduling plan and the dynamic allocation of resources optimize the maintenance window time, improve the operation efficiency, deepen the understanding of the health condition of the equipment through the long-term trend prediction of the statistical method, facilitate the prevention of future problems, prolong the service life of the equipment, and improve the reliability of the equipment by finely adjusting the measurement interval and the sensitivity of the monitoring points. And the operation efficiency and safety of the thermal power plant are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of equipment monitoring, and in particular to an online monitoring system and method for lubricating oil of equipment in a thermal power plant. Background Art

[0002] The field of equipment monitoring technology focuses on real-time tracking and evaluation of the performance and health status of industrial equipment to prevent failures and optimize operations. The technology uses various sensors to collect equipment operation data, such as temperature, vibration, pressure, and chemical composition. Through continuous analysis of the data, the monitoring system can identify potential equipment failures and performance degradation, and achieve failure prediction and health management. Equipment monitoring not only improves the reliability and safety of equipment, but also significantly reduces maintenance costs and downtime, thereby improving the efficiency of the entire production process. Intelligent equipment monitoring combines data analysis, machine learning, and cloud computing technologies to enable remote real-time monitoring.

[0003] Among them, the online monitoring system for lubricating oil in thermal power plant equipment is a system designed for monitoring the lubricating oil status of various equipment (such as steam turbines and auxiliary machinery) in thermal power plants. The main purpose of this system is to ensure that the quality and performance of the lubricating oil meet the requirements of normal operation of the equipment. By continuously monitoring the physical and chemical properties of the lubricating oil (such as viscosity, acid value, moisture content and contaminant levels), lubricating oil degradation and contamination problems can be discovered in time. This system helps prevent equipment failures and damage caused by lubricating oil problems, thereby reducing the risk of unplanned downtime, extending equipment life, and improving the overall operating efficiency and safety of thermal power plants. By introducing intelligent algorithms and remote monitoring technologies, the system can also achieve optimized management and predictive maintenance of lubricating oil, which is an indispensable technical support for modern thermal power plants in the pursuit of efficient and safe production.

[0004] Although the existing online monitoring system for lubricating oil in thermal power plant equipment covers a wide range of sensor applications and data analysis, it has deficiencies in real-time data processing and dynamic adjustment of thresholds. The batch analysis of traditional data leads to an untimely response to abnormal equipment conditions, thereby increasing the risk of equipment failure. The lack of maintenance strategies and dynamic resource allocation for specific equipment adjustments leads to uneven resource utilization and low maintenance efficiency. The prediction of long-term trends fails to integrate sufficient historical data with real-time oil monitoring data, which limits the accuracy and reliability of the prediction results. Technical limitations cause thermal power plant equipment to continue to operate when no initial degradation signals are detected, increasing the overall risk and maintenance cost of the system. The lack of the ability to flexibly adjust monitoring parameters is also a key shortcoming of existing technologies, limiting the adaptability and predictive ability of the monitoring system in the face of complex and changing working environments. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an online monitoring system and method for lubricating oil of thermal power plant equipment.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme: an online monitoring system for lubricating oil of thermal power plant equipment comprises:

[0007] The parameter real-time monitoring module reads the lubricating oil data from the turbine, records the continuous data stream using the timing signal capture technology, implements threshold monitoring and distinguishes between normal and abnormal readings, and obtains real-time monitoring data of the oil;

[0008] The oil condition analysis module performs statistical analysis on the real-time monitoring data of the oil, identifies abnormal fluctuations in the data, distinguishes normal and excessive indicators through threshold comparison, and generates abnormal status data;

[0009] The abnormal warning signal sending module determines the warning level based on the abnormal status data according to preset rules, automatically triggers a signal to be sent to the maintenance team's mobile device, and generates warning notification information;

[0010] The maintenance scheduling module uses the warning notification information to evaluate maintenance resources, update the scheduling plan of the maintenance team in real time, dynamically allocate and adjust personnel and resources, and obtain the maintenance scheduling plan;

[0011] The health trend analysis module compares the maintenance scheduling plan with historical data, applies statistical methods to make long-term trend predictions, identifies continuous downward trends, and constructs trend analysis results;

[0012] The equipment performance enhancement module evaluates the effectiveness of the monitoring parameter settings based on the trend analysis results, adjusts the measurement interval and the sensitivity of the monitoring points by comparing historical and current data, and forms a lubricating oil monitoring execution plan.

[0013] As a further solution of the present invention, the real-time oil monitoring data includes viscosity level, moisture content, liquid density, and metal particle count; the abnormal status data includes detected abnormal points, degree of exceeding the standard, type of abnormality, and source of abnormality; the early warning notification information includes warning level, receiving device identification code, signal strength, and notification time; the maintenance scheduling plan includes personnel allocation list, resource utilization rate, planned execution date, and expected maintenance duration; the trend analysis results include long-term health trend charts, risk assessment levels, potential failure points, and prediction accuracy; the lubricating oil monitoring execution plan includes monitoring parameter fine-tuning, monitoring cycle optimization, key performance indicators, and data collection frequency.

[0014] As a further solution of the present invention, the parameter real-time monitoring module includes:

[0015] The data capture submodule reads the lubricating oil data from the steam turbine, including viscosity, moisture, density and metal wear particle indicators, collects data regularly through a timing trigger mechanism, transmits the data to the monitoring configuration, and obtains a continuous data stream record;

[0016] The threshold monitoring submodule receives data from the continuous data stream record, monitors multiple indicators in real time, and immediately adjusts the threshold when it is detected that the data indicator exceeds the preset threshold, analyzes the real-time data and determines the indicator exceeding the threshold, and generates the threshold monitoring result;

[0017] The data partition submodule processes the threshold monitoring results, analyzes normal and abnormal indicators, performs logical analysis and records abnormal data, marks key abnormal indicators and prepares data to form real-time oil monitoring data.

[0018] As a further solution of the present invention, the oil condition analysis module includes:

[0019] The abnormality detection submodule receives the real-time oil monitoring data of the steam turbine, uses the historical data of the equipment operation to establish a preliminary abnormality determination benchmark, compares the real-time monitoring data with the benchmark, adjusts the threshold of abnormality detection and matches the cyclically changing operating conditions, identifies abnormal fluctuations in the data, and obtains abnormal fluctuation identification results;

[0020] The data comparison submodule compares the data points with the preset multi-level thresholds based on the abnormal fluctuation identification results, performs threshold analysis on the abnormal data, marks the data points that exceed the normal operating range, and obtains the judgment result of the data exceeding the standard;

[0021] The marking and classification submodule marks and classifies the excessive data according to the excessive data determination result, allocates a processing strategy and an archiving path for each type of data, and constructs abnormal status data.

[0022] As a further solution of the present invention, the abnormal warning signaling module includes:

[0023] The warning level determination submodule classifies the abnormal data based on the abnormal state data, determines the warning level of the data according to the set threshold, classifies the warning categories, sorts the data according to the severity, adjusts the processing rules and subdivides the differentiated levels to obtain the warning level determination result;

[0024] The signal triggering submodule uses the warning level to determine the result, adjusts the signal sending mechanism according to the warning state, configures the signal priority and transmission path, performs state transmission without external dependence, and obtains the signal sending state;

[0025] The notification making submodule receives data from the signal sending status, formats the notification information and sends it to the designated device using the preset message template and sending rules, and generates the early warning notification information based on the internal message distribution logic.

[0026] As a further solution of the present invention, the maintenance scheduling module includes:

[0027] The resource assessment submodule analyzes the matching degree between the current maintenance demand and the existing resources based on the warning notification information, evaluates the demand and status of human resources, tools, and time resources through data comparison, identifies resource shortages and redundancies, and generates resource demand and supply assessment results;

[0028] The scheduling plan update submodule reconfigures the timing and priority of maintenance tasks based on the resource demand and supply evaluation results, analyzes the correspondence between tasks and resource configurations, adjusts the time schedule of maintenance windows, optimizes task processes and time management through dynamic updates, and generates scheduling plan optimization results;

[0029] The dynamic allocation submodule dynamically allocates maintenance personnel and resources based on the scheduling plan optimization results, reallocates manpower and materials according to the urgency of the task, adjusts resource distribution and matches task priority and time requirements, and obtains a maintenance scheduling plan.

[0030] As a further solution of the present invention, the health trend analysis module includes:

[0031] The data comparison and evaluation submodule compares each maintenance indicator based on the maintenance scheduling plan and historical maintenance data through a data integration process, analyzes the relationship between historical trends and current plans, identifies key change points and consistency, and generates maintenance effect comparison results;

[0032] The trend prediction submodule uses the random forest algorithm and statistical analysis tools based on the maintenance effect comparison results to perform trend line analysis on long-term data, identify and predict the development direction of key health indicators, and generate health trend predictions through trend analysis technology;

[0033] The trend result construction submodule utilizes the health trend prediction, integrates the trend data, points out the key risks and areas of health status, and forms the trend analysis results.

[0034] As a further solution of the present invention, the formula of the random forest algorithm is as follows:

[0035]

[0036] Among them, I′(N) is the improved information entropy, C is the number of categories, and p iis the probability of selecting the i-th category in node N, λ is the regularization parameter, D is the number of newly introduced feature factors, and w j is the weight of the jth feature factor, v j is the value of the jth characteristic factor.

[0037] As a further solution of the present invention, the equipment performance enhancement module includes:

[0038] The parameter effect evaluation submodule analyzes the difference between the current monitoring parameters and the historical data based on the trend analysis results, evaluates the accuracy of the monitoring parameters' impact on the equipment performance, checks whether the parameters can capture key trends, determines the parameters that need to be adjusted through quantitative analysis, and obtains the parameter adjustment demand results;

[0039] The monitoring adjustment submodule adjusts the measurement interval and the sensitivity of the monitoring point based on the parameter adjustment demand result, monitors the effect of the parameter adjustment in real time, and generates a monitoring strategy optimization plan;

[0040] The execution plan building submodule utilizes the monitoring strategy optimization plan to refine lubricating oil monitoring, sets operating parameters for each monitoring point, verifies the effectiveness of the strategy through simulation tests, and forms a lubricating oil monitoring execution plan.

[0041] A method for online monitoring of lubricating oil for thermal power plant equipment, which is performed based on the above-mentioned online monitoring system for lubricating oil for thermal power plant equipment, comprises the following steps:

[0042] S1: Read lubricating oil data from the turbine, including viscosity, moisture, density and metal wear particle indicators, automatically identify and record normal and abnormal readings of each indicator based on preset thresholds, and obtain basic monitoring data;

[0043] S2: analyzing the basic monitoring data, identifying data points that deviate from the norm by data comparison analysis, classifying the data points according to abnormal types, including viscosity and metal particle concentration that exceed the normal range, and generating abnormal diagnosis results;

[0044] S3: using the abnormal diagnosis result to determine the warning level of the lubricating oil, automatically sending a customized warning signal to the communication equipment of the maintenance team to convey the potential maintenance needs and generate a warning notification output;

[0045] S4: Based on the warning notification output, dynamically evaluate the current maintenance team and resource configuration, adjust the maintenance schedule as needed, optimize resource allocation and shorten the response time and maintenance window, and generate a maintenance strategy plan;

[0046] S5: Utilize the maintenance strategy plan and historical maintenance data to perform trend analysis and predict the performance of lubricating oil and related equipment in future time periods, adjust monitoring parameters, and form a lubricating oil monitoring execution plan.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are:

[0048] In the present invention, by implementing timing signal capture technology and threshold monitoring, the ability to respond quickly to abnormal conditions is significantly improved. This continuous data stream recording combined with real-time oil monitoring data analysis allows the system to more accurately capture and classify abnormal fluctuations, so that the maintenance team can quickly take preventive measures to avoid potential equipment failures. Real-time updated maintenance scheduling plans and dynamic allocation of personnel and resources further optimize maintenance window time, ensure the continuity and efficiency of operations, and long-term trend prediction using statistical methods provides in-depth insights into the health status of equipment, helping to predict and prevent possible problems in the future. This method not only extends the service life of the equipment, but also reduces the risk of unplanned downtime. By adjusting the measurement interval and the sensitivity of the monitoring point, key performance indicators can be more finely regulated and captured, further improving the operating efficiency and safety of the entire thermal power plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a system flow chart of the present invention;

[0050] Figure 2 It is a schematic diagram of the system framework of the present invention;

[0051] Figure 3 This is a flow chart of the parameter real-time monitoring module in the present invention;

[0052] Figure 4 It is a flow chart of the oil condition analysis module in the present invention;

[0053] Figure 5 This is a flow chart of the abnormal warning signaling module in the present invention;

[0054] Figure 6 It is a flowchart of the maintenance scheduling module in the present invention;

[0055] Figure 7 This is a flow chart of the health trend analysis module in the present invention;

[0056] Figure 8 It is a flow chart of the equipment performance enhancement module in the present invention;

[0057] Fig. 9 It is a schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0060] Embodiment 1

[0061] See also Figure 1 , 2 The present invention provides a technical solution: an online monitoring system for lubricating oil of thermal power plant equipment includes:

[0062] The parameter real-time monitoring module reads the lubricating oil data from the turbine, including viscosity, moisture, density and metal wear particle indicators, uses the timing signal capture technology to record the continuous data stream, implements threshold monitoring and distinguishes between normal and abnormal readings, and obtains real-time monitoring data of the oil;

[0063] The oil condition analysis module performs statistical analysis on the real-time monitoring data of oil, identifies abnormal fluctuations in the data, distinguishes normal and excessive indicators through threshold comparison, marks and classifies the excessive data, and generates abnormal status data;

[0064] The abnormal warning signal sending module determines the warning level based on the abnormal status data according to the preset rules, automatically triggers the signal to be sent to the maintenance team's mobile device, and generates warning notification information;

[0065] The maintenance scheduling module uses early warning notification information to evaluate maintenance resources, update the scheduling plan of the maintenance team in real time, dynamically allocate and adjust personnel and resources, optimize maintenance window time, and obtain maintenance scheduling plans;

[0066] The health trend analysis module compares the maintenance scheduling plan with historical data, applies statistical methods to make long-term trend forecasts, identifies continuous downward trends, and constructs trend analysis results;

[0067] The equipment performance enhancement module evaluates the effectiveness of monitoring parameter settings based on trend analysis results. By comparing historical and current data, it adjusts measurement intervals and monitoring point sensitivity, optimizes strategies to capture key indicators, and forms a lubricant monitoring execution plan.

[0068] Real-time oil monitoring data includes viscosity level, moisture content, liquid density, and metal particle count. Abnormal status data includes detected abnormal points, degree of exceeding the standard, type of abnormality, and source of abnormality. Early warning notification information includes warning level, receiving device identification code, signal strength, and notification time. The maintenance scheduling plan includes personnel allocation list, resource utilization rate, planned execution date, and expected maintenance time. Trend analysis results include long-term health trend charts, risk assessment levels, potential failure points, and prediction accuracy. The lubricant monitoring execution plan includes monitoring parameter fine-tuning, monitoring cycle optimization, key performance indicators, and data collection frequency.

[0069] Specifically, if Figure 2 , 3 As shown, the parameter real-time monitoring module includes:

[0070] The data capture submodule reads the lubricating oil data from the steam turbine, including viscosity, moisture, density and metal wear particle indicators, collects data regularly through a timing trigger mechanism, and transmits the data to the monitoring configuration. The execution process of obtaining continuous data stream records is as follows;

[0071] The specific operations performed during the lubricant parameter measurement process include the use of sensors to continuously monitor viscosity, moisture, density and metal wear particle indicators. The timing trigger mechanism obtains data from the turbine control system according to the set collection frequency, such as once an hour. The data is first subjected to preliminary filtering to exclude obvious abnormal values ​​or erroneous readings. The data is transmitted to the database of the central monitoring system through a secure network connection to form a continuous data stream record for further analysis. The weighted average calculation of density and viscosity is used to estimate the overall quality index of the lubricant. The formula is:

[0072]

[0073] Among them, ρ represents density, η represents viscosity, and w 1 and w 2 are the weight coefficients of the two indicators respectively, and Ma represents the weighted average, which is used to adjust the influence of each indicator in the overall quality assessment.

[0074] The threshold monitoring submodule receives data from continuous data stream records and monitors multiple indicators in real time. When it is detected that the data indicator exceeds the preset threshold, the threshold is adjusted immediately. The execution process of generating the threshold monitoring result is as follows through real-time data analysis and determining the indicator exceeding the standard item;

[0075] Receive real-time data from continuous data stream records and analyze it to determine whether any indicator exceeds the preset threshold. The judgment process is based on the normal operating range set for each indicator. For example, viscosity should not exceed the operating range recommended by the equipment manufacturer. Once an indicator exceeds the threshold, the system will automatically adjust the monitoring threshold of the indicator and record it for subsequent analysis and adjustment. It is used to dynamically adjust the threshold and increase the feedback coefficient to respond more sensitively to changes in indicators. The formula:

[0076] T new =T old +k·(xT old )

[0077] Among them, T new and T old Represent the new and old thresholds respectively, x is the current reading, and k is the adjustment factor, which is used to adjust the sensitivity of the threshold based on the recent data trend.

[0078] The data partition submodule processes the threshold monitoring results, analyzes normal and abnormal indicators, performs logical analysis and records abnormal data, marks key abnormal indicators and prepares data, and forms the execution process of real-time oil monitoring data as follows;

[0079] Process the received threshold monitoring results, and distinguish which data belongs to the normal range and which belongs to the abnormal range through the set logical rules. For abnormal data, conduct more in-depth logical analysis, such as calculating the degree of deviation from the normal range, and record and mark it. Key abnormal indicators will be highlighted and prepared for further analysis or reporting. The abnormal scoring mechanism is introduced to calculate the influence of abnormal indicators by weight according to the degree of deviation. The formula is:

[0080]

[0081] Where Sa represents the anomaly score, x i is the current indicator reading, μ i is the normal value of the indicator, w i is the weight coefficient of each indicator, which is used to indicate the importance of the indicator in the overall evaluation.

[0082] Specifically, if Figure 2 , 4 As shown in the figure, the oil condition analysis module includes:

[0083] The abnormality detection submodule receives the real-time oil monitoring data of the steam turbine, uses the historical data of the equipment operation to establish a preliminary abnormality judgment benchmark, compares the real-time monitoring data with the benchmark, adjusts the threshold of abnormality detection and matches the cyclically changing operating conditions, identifies abnormal fluctuations in the data, and obtains the abnormal fluctuation identification results. The execution process is as follows;

[0084] Receive real-time monitoring data and compare it with the statistical benchmark in the historical data set to establish the standard for abnormal judgment. Real-time monitoring data is obtained in real time through sensors and compared with historical operating data stored in the database, such as parameters such as oil viscosity and density. Based on the comparison results, the module adjusts the threshold of abnormal detection to adapt to the cyclic changes of operating conditions. The detection of abnormal fluctuations is achieved by calculating the deviation between real-time data and historical average values, and dynamically adjusting the threshold T by applying the formula a :

[0085] T b =T hist ·(1+r·σ)

[0086] Among them, T hist represents the historical average threshold, σ is the standard deviation, which measures the volatility of historical data, and r is the adjustment factor, which is used to adjust the threshold according to the latest operating environment. This dynamic adjustment helps the system more accurately identify true anomalies rather than routine fluctuations.

[0087] The data comparison submodule compares the data points with the preset multi-level thresholds based on the abnormal fluctuation identification results, performs threshold analysis on the abnormal data, marks the data points that exceed the normal operating range, and obtains the execution process of the excessive data judgment result as follows;

[0088] Using the abnormal fluctuation identification results for further analysis, the module uses a preset multi-level threshold system to evaluate each real-time data point to determine whether there is a situation that exceeds the normal operating parameters. When a data point exceeds the preset threshold, the module marks the data point as abnormal. The formula used in the process is used to calculate the degree of abnormality of the data point and decide whether to mark it. The formula is:

[0089]

[0090] Among them, x is the value of the current data point, μ is the preset normal operating value, θ is the threshold range, and Eb represents the degree of abnormality. A value greater than 1 indicates an obvious abnormal state. This calculation helps the module accurately identify and mark those data points that require further analysis.

[0091] The marking and classification submodule marks and classifies the data exceeding the standard according to the judgment result of the data exceeding the standard, allocates processing strategies and archiving paths for each type of data, and constructs the execution process of abnormal status data as follows;

[0092] The data that has been identified as exceeding the standard is further marked and classified. This module classifies the data into different categories according to the degree of abnormality and assigns corresponding processing strategies and archiving paths to each category. These processing strategies include immediate maintenance scheduling, long-term equipment performance monitoring or data archiving, etc. The classification index is introduced to assign priorities and resources according to the degree of abnormality of the data point. The formula is:

[0093]

[0094] Among them, E i is the abnormal degree of each data point, P represents the processing priority threshold, and Cb is the classification index, which is used to determine the urgency of data processing and the archiving path. In this way, the system can effectively handle and respond to various abnormal situations.

[0095] Specifically, if Figure 2 , 5 As shown, the abnormal warning signaling module includes:

[0096] The warning level determination submodule classifies abnormal data based on abnormal status data, determines the warning level of the data according to the set threshold, classifies the warning categories, sorts the data according to the severity, adjusts the processing rules and subdivides the differentiated levels, and obtains the execution process of the warning level determination result as follows;

[0097] Receive abnormal status data, which has been marked and classified by the previous step. This module uses a multi-level warning system according to the set threshold and data severity. The data is divided into three warning levels: minor, medium and severe. Based on the severity of the data, the module dynamically adjusts the processing rules and subdivides the differentiated processing levels to quantify the severity of the data and determine the corresponding warning level. The formula is:

[0098]

[0099] Among them, S represents the severity score obtained from the abnormal status data, T is the threshold boundary of each warning level, and Lc represents the final warning level, which is expressed in numerical form, ranging from 1 (mild) to 3 (severe). This quantitative method helps the system to accurately control the response to warnings of different levels.

[0100] The signal trigger submodule uses the warning level to determine the result, adjusts the signal sending mechanism according to the warning status, configures the signal priority and transmission path, and performs state transmission without external dependencies. The execution process of the signal sending status is as follows;

[0101] The signal sending mechanism is adjusted using the warning level decision result. This module configures the signal priority and transmission path to ensure that high-priority warnings can be quickly transmitted. The sending of warning signals does not rely on external systems to ensure stability and timeliness in the internal network. The tuning mode takes into account the signal priority and warning level, and optimizes the signal transmission path and speed. The formula is:

[0102] Pc=10-2×L

[0103] Among them, Pc represents the priority of the signal, and L is obtained from the warning level determination result. This method ensures that higher-level warnings have higher transmission priority and are processed faster.

[0104] The notification production submodule receives data from the signal sending status, uses the preset message template and sending rules, formats the notification information and sends it to the specified device. Based on the internal message distribution logic, the execution process of generating the warning notification information is as follows;

[0105] Receive the data of signal sending status and format it using the preset message template. The message is sent to the designated device through the internal network according to the sending rules, such as the control room display or the operator's mobile device. The key to this process is to ensure the consistency and accuracy of the message format, which is used to adjust the urgency and content of the message to ensure that the receiver can understand and respond quickly. The formula is:

[0106] Uc=5×LR

[0107] Among them, Uc represents the urgency of the message, L is the warning level from the signal trigger submodule, and R is the response level of the message receiving device. This calculation method ensures that the urgency of the message matches the warning level and optimizes the message delivery process.

[0108] Specifically, if Figure 2 , 6 As shown, the maintenance scheduling module includes:

[0109] The resource assessment submodule analyzes the matching degree between the current maintenance demand and the existing resources based on the early warning notification information, evaluates the demand and status of human, tool, and time resources through data comparison, identifies resource shortages and redundancies, and generates the resource demand and supply assessment results. The execution process is as follows;

[0110] It involves receiving early warning notification information and analyzing the matching degree between the current maintenance needs and the existing resources based on the information. The submodule first quantifies the demand for human resources, tools and time resources and compares them with the existing resources. By calculating the resource matching degree, it identifies the places where resources are short or redundant, thereby providing data support for optimizing resource allocation. The process uses the following improved formula to evaluate resource matching degree:

[0111]

[0112] Among them, r i represents the existing resources of the i-th category, d i It represents the demand for the i-th type of resource, n is the total number of resource categories, Md represents the resource matching degree, and the smaller the value, the higher the matching degree. This formula helps to accurately identify which resources are in short supply and which are redundant.

[0113] The scheduling plan update submodule reconfigures the timing and priority of maintenance tasks based on the resource demand and supply evaluation results, analyzes the correspondence between tasks and resource configuration, adjusts the time schedule of maintenance windows, optimizes task processes and time management through dynamic updates, and generates the execution process of scheduling plan optimization results as follows;

[0114] The timing and priority of maintenance tasks are reconfigured based on the resource evaluation results. The submodule analyzes the correspondence between tasks and resources and adjusts the schedule of maintenance windows according to the actual availability of resources. It is used to dynamically adjust the priority and schedule of tasks and optimize the entire task process and time management. The formula is:

[0115] T d =T old +α×(δ-U)

[0116] Among them, T old and T d They represent the original and newly adjusted schedules respectively, α is the adjustment coefficient, δ represents the task urgency, and U is the difference between available resources and demand. This adjustment method ensures that the timing and priority of tasks match the actual available resources, thereby improving maintenance efficiency.

[0117] The dynamic allocation submodule dynamically allocates maintenance personnel and resources based on the scheduling plan optimization results, reallocates manpower and materials according to the urgency of the task, adjusts resource distribution and matches task priority and time requirements, and obtains the execution process of the maintenance scheduling plan as follows;

[0118] Based on the optimization results of the scheduling plan, maintenance personnel and resources are dynamically deployed. According to the urgency of the task and resource evaluation, manpower and materials are reallocated. The formula used in this process helps adjust the resource distribution to ensure that resource allocation can meet the priority and time requirements of the task:

[0119]

[0120] Among them, Rd represents the total allocation score of resources, p i is the priority of the ith task, v iis the amount of resources allocated to the i-th task, and k is the total number of tasks, ensuring that high-priority tasks are supported with sufficient resources to effectively cope with maintenance challenges.

[0121] Specifically, if Figure 2 , 7 As shown, the health trend analysis module includes:

[0122] The data comparison and evaluation submodule is based on the maintenance scheduling plan and historical maintenance data. Through the data integration process, it compares each maintenance indicator, analyzes the relationship between historical trends and current plans, identifies key change points and consistency, and generates maintenance effect comparison results. The execution process is as follows;

[0123] Integrate the current maintenance scheduling plan with historical maintenance data, identify the key differences between historical trends and current plans through detailed comparison of each maintenance indicator, and use a comprehensive evaluation index in the analysis process to quantitatively compare the effects of maintenance activities at different time points in order to identify key nodes where efficiency improves or decreases. This is used to calculate the degree of difference in maintenance effects, taking into account the weighted impact of historical data and real-time evaluation of current data. The formula is:

[0124]

[0125] Among them, x i is the value of the i-th indicator in the current plan, h i is the corresponding historical indicator value, w i is the weight set for the importance of the indicator, n is the total number of indicators, and Df represents the difference index of the maintenance effect. In this way, the module can accurately locate the changing trend of the maintenance effect and provide more targeted improvement suggestions.

[0126] The trend prediction submodule uses the random forest algorithm and statistical analysis tools based on the maintenance effect comparison results to perform trend line analysis on long-term data, identify and predict the development direction of key health indicators, and generate the execution process of health trend prediction through trend analysis technology as follows;

[0127] By analyzing the maintenance effect comparison results, the improved data model is used to perform trend line analysis on long-term maintenance data. This module calculates the predicted trend of long-term health indicators and uses statistical methods to generate development forecasts for key health indicators. In order to improve the prediction accuracy, a nonlinear regression model is introduced. This model optimizes the fitting of trend lines by considering random factors and historical deviations. The formula is:

[0128]

[0129] Among them, yf is the predicted trend value, x is the time variable, β 0 , β1 , β 2 , …, β m is the model parameter f j (x,∈) is a function of the introduced random factors and historical deviations, including interactions and nonlinear effects. Through this complex model, the trend prediction submodule can more accurately predict and explain the changes in key health indicators.

[0130] The trend result construction submodule utilizes health trend prediction, integrates trend data, points out key risks and areas of health status, and forms the execution process of trend analysis results as follows;

[0131] Based on the trend prediction results, the obtained trend data is integrated and analyzed to clearly point out the key risk points and areas of concern for the health of the equipment. The module uses data clustering analysis to classify trend data into different risk levels to facilitate the identification and response to potential health problems.

[0132] The formula for the random forest algorithm is as follows:

[0133]

[0134] Among them, I′(N) is the improved information entropy, C is the number of categories, and p i is the probability of selecting the i-th category in node N, λ is the regularization parameter, D is the number of newly introduced feature factors, and w j is the weight of the jth feature factor, v j is the value of the jth characteristic factor.

[0135] The execution process is as follows:

[0136] Calculate the probability p of each category appearing in node N i , then calculate the traditional information entropy, and then introduce new characteristic factors, including node depth, number of node samples, standard deviation of node samples and purity index of nodes, and set weight w for each characteristic factor j , and calculate its weighted sum, multiply the weighted sum by the regularization parameter λ and add it to the traditional information entropy to form an improved information entropy formula to determine the weight w j Specific methods include optimizing with a validation dataset or adjusting using gradient-based methods to maximize the overall prediction performance.

[0137] Specifically, if Figure 2 , 8 As shown, the equipment performance enhancement module includes:

[0138] The parameter effect evaluation submodule analyzes the difference between the current monitoring parameters and the historical data based on the trend analysis results, evaluates the accuracy of the monitoring parameters' impact on the equipment performance, checks whether the parameters can capture key trends, and determines the parameters that need to be adjusted through quantitative analysis. The execution process of obtaining the parameter adjustment demand results is as follows;

[0139] Analyze the difference between the current monitoring parameters and historical data to evaluate the accuracy of the parameters' impact on equipment performance. The submodule first calculates the sensitivity and responsiveness of each monitoring parameter. By comparing the correlation between parameter changes and equipment performance changes, it evaluates whether each parameter can effectively capture the key trends of the equipment. This process uses an improved statistical analysis formula to quantify the effect of the parameters and the necessary adjustments. The formula is:

[0140]

[0141] Among them, p i is the current monitoring parameter value, h i is the corresponding historical average, n is the number of samples, and Δp represents the standard deviation of the parameter deviation, which is used to measure the consistency between the fluctuation of the monitoring parameters and the historical data. In this way, the submodule can identify the monitoring parameters that need to be adjusted to improve the accuracy of equipment performance monitoring.

[0142] The monitoring adjustment submodule adjusts the measurement interval and the sensitivity of the monitoring point based on the parameter adjustment demand results, monitors the effect of parameter adjustment in real time, and generates the execution process of the monitoring strategy optimization plan as follows;

[0143] According to the results of parameter adjustment requirements, the measurement interval and the sensitivity of the monitoring points are optimized. The submodule monitors the effect of parameter adjustment through real-time data analysis to ensure the real-time and effectiveness of the monitoring strategy. The submodule adopts a dynamic adjustment algorithm to optimize the monitoring strategy in an adaptive manner. The formula is:

[0144]

[0145] Among them, S new and S old represent the new and old sensitivity settings respectively, λ is the adjustment factor, Δp is the standard deviation of the parameter deviation obtained from the parameter effect evaluation submodule, and p target is the target parameter value, and this formula helps adjust the sensitivity of the monitoring point to match the actual status of the equipment operation.

[0146] The execution plan building submodule uses the monitoring strategy optimization plan to refine the lubricant monitoring, and verifies the effectiveness of the strategy through simulation tests for the operating parameters set for each monitoring point, forming the execution process of the lubricant monitoring execution plan as follows;

[0147] Use the monitoring strategy optimization scheme to refine lubricant monitoring, conduct simulation tests on the operating parameters set for each monitoring point, and verify the effectiveness of the strategy. By simulating different operating conditions and parameter settings, the submodule tests the performance of the monitoring strategy in various situations to ensure its effectiveness and reliability. The formula is used to evaluate the optimization effect of the monitoring execution scheme:

[0148]

[0149] Among them, Eh is the efficiency evaluation value of the execution plan, O j is the expected response E of the jth monitoring point in the simulation test j is the actual response, w j is the weight of the monitoring point, and m is the number of monitoring points. This approach ensures that the lubricant monitoring strategy can be effectively implemented under various operating conditions.

[0150] A method for online monitoring of lubricating oil for thermal power plant equipment is provided. The method is based on the above-mentioned online monitoring system for lubricating oil for thermal power plant equipment and comprises the following steps:

[0151] S1: Read lubricating oil data from the turbine, including viscosity, moisture, density and metal wear particle indicators, automatically identify and record normal and abnormal readings of each indicator based on preset thresholds, and obtain basic monitoring data;

[0152] S2: Analyze basic monitoring data, use data comparison analysis to identify data points that deviate from the norm, classify the data points according to abnormal types, including viscosity and metal particle concentrations that exceed the normal range, and generate abnormal diagnosis results;

[0153] S3: Use abnormal diagnosis results to determine the warning level of lubricant oil, automatically send customized warning signals to the maintenance team’s communication equipment, convey potential maintenance needs, and generate warning notification outputs;

[0154] S4: Based on the early warning notification output, dynamically evaluate the current maintenance team and resource configuration, adjust the maintenance schedule as needed, optimize resource allocation and shorten the response time and maintenance window, and generate a maintenance strategy plan;

[0155] S5: Use maintenance strategy plans and historical maintenance data to conduct trend analysis and predict the performance of lubricants and related equipment in the future, adjust monitoring parameters, and form a lubricant monitoring execution plan.

[0156] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An online monitoring system for lubricating oil of thermal power plant equipment, characterized by: The system comprises: The parameter real-time monitoring module reads the lubricating oil data from the turbine, records the continuous data stream using the timing signal capture technology, implements threshold monitoring and distinguishes between normal and abnormal readings, and obtains real-time monitoring data of the oil; The oil condition analysis module performs statistical analysis on the real-time monitoring data of the oil, identifies abnormal fluctuations in the data, distinguishes normal and excessive indicators through threshold comparison, and generates abnormal status data; The abnormal warning signal sending module determines the warning level based on the abnormal status data according to preset rules, automatically triggers a signal to be sent to the maintenance team's mobile device, and generates warning notification information; The maintenance scheduling module uses the warning notification information to evaluate maintenance resources, update the scheduling plan of the maintenance team in real time, dynamically allocate and adjust personnel and resources, and obtain the maintenance scheduling plan; The health trend analysis module compares the maintenance scheduling plan with historical data, applies statistical methods to make long-term trend predictions, identifies continuous downward trends, and constructs trend analysis results; The equipment performance enhancement module evaluates the effectiveness of the monitoring parameter settings based on the trend analysis results, adjusts the measurement interval and the sensitivity of the monitoring points by comparing historical and current data, and forms a lubricating oil monitoring execution plan.

2. The on-line monitoring system for lubricating oil of thermal power plant equipment according to claim 1 is characterized by: The real-time oil monitoring data includes viscosity level, moisture content, liquid density, and metal particle count; the abnormal status data includes detected abnormal points, degree of exceeding the standard, type of abnormality, and source of abnormality; the early warning notification information includes warning level, receiving device identification code, signal strength, and notification time; the maintenance scheduling plan includes personnel allocation list, resource utilization rate, planned execution date, and expected maintenance duration; the trend analysis results include long-term health trend charts, risk assessment levels, potential failure points, and prediction accuracy; the lubricant monitoring execution plan includes monitoring parameter fine-tuning, monitoring cycle optimization, key performance indicators, and data collection frequency.

3. The on-line monitoring system for lubricating oil of thermal power plant equipment according to claim 1 is characterized by: The parameter real-time monitoring module includes: The data capture submodule reads the lubricating oil data from the steam turbine, including viscosity, moisture, density and metal wear particle indicators, collects data regularly through a timing trigger mechanism, transmits the data to the monitoring configuration, and obtains a continuous data stream record; The threshold monitoring submodule receives data from the continuous data stream record, monitors multiple indicators in real time, and immediately adjusts the threshold when it is detected that the data indicator exceeds the preset threshold, analyzes the real-time data and determines the indicator exceeding the threshold, and generates the threshold monitoring result; The data partition submodule processes the threshold monitoring results, analyzes normal and abnormal indicators, performs logical analysis and records abnormal data, marks key abnormal indicators and prepares data to form real-time oil monitoring data.

4. The on-line monitoring system for lubricating oil of thermal power plant equipment according to claim 1 is characterized in that: The oil condition analysis module comprises: The abnormality detection submodule receives the real-time oil monitoring data of the steam turbine, uses the historical data of the equipment operation to establish a preliminary abnormality determination benchmark, compares the real-time monitoring data with the benchmark, adjusts the threshold of abnormality detection and matches the cyclically changing operating conditions, identifies abnormal fluctuations in the data, and obtains abnormal fluctuation identification results; The data comparison submodule compares the data points with the preset multi-level thresholds based on the abnormal fluctuation identification results, performs threshold analysis on the abnormal data, marks the data points that exceed the normal operating range, and obtains the judgment result of the data exceeding the standard; The marking and classification submodule marks and classifies the excessive data according to the excessive data determination result, allocates a processing strategy and an archiving path for each type of data, and constructs abnormal status data.

5. The on-line monitoring system for lubricating oil of thermal power plant equipment according to claim 1 is characterized by: The abnormal warning signaling module includes: The warning level determination submodule classifies the abnormal data based on the abnormal state data, determines the warning level of the data according to the set threshold, classifies the warning categories, sorts the data according to the severity, adjusts the processing rules and subdivides the differentiated levels to obtain the warning level determination result; The signal triggering submodule uses the warning level to determine the result, adjusts the signal sending mechanism according to the warning state, configures the signal priority and transmission path, performs state transmission without external dependence, and obtains the signal sending state; The notification making submodule receives data from the signal sending status, formats the notification information and sends it to the designated device using the preset message template and sending rules, and generates the early warning notification information based on the internal message distribution logic.

6. The on-line monitoring system for lubricating oil of thermal power plant equipment according to claim 1 is characterized by: The maintenance scheduling module includes: The resource assessment submodule analyzes the matching degree between the current maintenance demand and the existing resources based on the warning notification information, evaluates the demand and status of human resources, tools, and time resources through data comparison, identifies resource shortages and redundancies, and generates resource demand and supply assessment results; The scheduling plan update submodule reconfigures the timing and priority of maintenance tasks based on the resource demand and supply evaluation results, analyzes the correspondence between tasks and resource configurations, adjusts the time schedule of maintenance windows, optimizes task processes and time management through dynamic updates, and generates scheduling plan optimization results; The dynamic allocation submodule dynamically allocates maintenance personnel and resources based on the scheduling plan optimization results, reallocates manpower and materials according to the urgency of the task, adjusts resource distribution and matches task priority and time requirements, and obtains a maintenance scheduling plan.

7. The on-line monitoring system for lubricating oil of thermal power plant equipment according to claim 1 is characterized by: The health trend analysis module includes: The data comparison and evaluation submodule compares each maintenance indicator based on the maintenance scheduling plan and historical maintenance data through a data integration process, analyzes the relationship between historical trends and current plans, identifies key change points and consistency, and generates maintenance effect comparison results; The trend prediction submodule uses the random forest algorithm and statistical analysis tools based on the maintenance effect comparison results to perform trend line analysis on long-term data, identify and predict the development direction of key health indicators, and generate health trend predictions through trend analysis technology; The trend result construction submodule utilizes the health trend prediction, integrates the trend data, points out the key risks and areas of health status, and forms the trend analysis results.

8. The on-line monitoring system for lubricating oil of thermal power plant equipment according to claim 7 is characterized in that: The formula of the random forest algorithm is as follows: Among them, I′(N) is the improved information entropy, C is the number of categories, and p i is the probability of selecting the i-th category in node N, λ is the regularization parameter, D is the number of newly introduced feature factors, and w j is the weight of the jth feature factor, v j is the value of the jth characteristic factor.

9. The on-line monitoring system for lubricating oil of thermal power plant equipment according to claim 1, characterized in that: The equipment performance enhancement module includes: The parameter effect evaluation submodule analyzes the difference between the current monitoring parameters and the historical data based on the trend analysis results, evaluates the accuracy of the monitoring parameters' impact on the equipment performance, checks whether the parameters can capture key trends, determines the parameters that need to be adjusted through quantitative analysis, and obtains the parameter adjustment demand results; The monitoring adjustment submodule adjusts the measurement interval and the sensitivity of the monitoring point based on the parameter adjustment demand result, monitors the effect of the parameter adjustment in real time, and generates a monitoring strategy optimization plan; The execution plan building submodule utilizes the monitoring strategy optimization plan to refine lubricating oil monitoring, sets operating parameters for each monitoring point, verifies the effectiveness of the strategy through simulation tests, and forms a lubricating oil monitoring execution plan.

10. A method for online monitoring of lubricating oil in thermal power plant equipment, characterized in that: The on-line monitoring system for lubricating oil of thermal power plant equipment according to any one of claims 1 to 9 comprises the following steps: Read lubricating oil data from the turbine, including viscosity, moisture, density and metal wear particle indicators, automatically identify and record normal and abnormal readings of each indicator based on preset thresholds, and obtain basic monitoring data; Analyze the basic monitoring data, use data comparison analysis to identify data points that deviate from the norm, classify the data points according to abnormal types, including viscosity and metal particle concentration that exceed the normal range, and generate abnormal diagnosis results; Using the abnormal diagnosis results to determine the warning level of the lubricant, automatically sending a customized warning signal to the maintenance team's communication equipment to convey potential maintenance needs and generate a warning notification output; According to the output of the early warning notification, dynamically evaluate the current maintenance team and resource configuration, adjust the maintenance schedule as needed, optimize resource allocation and shorten the response time and maintenance window, and generate a maintenance strategy plan; The maintenance strategy plan and historical maintenance data are used to perform trend analysis and predict the performance of lubricants and related equipment in future time periods, adjust monitoring parameters, and form a lubricant monitoring execution plan.

Citation Information

Cited By

  • Working condition operation efficiency monitoring method and system for flexible equipment

    CN119197653A

  • Boiler water protective agent online monitoring system and method for energy conservation and emission reduction

    CN119335141A

  • Water conservancy project monitoring AI safety identification method and system

    CN119357878A

  • Artificial intelligence circuit board production equipment management system

    CN119477233A

  • Equipment wear diagnosis control system based on lubrication state monitoring technology

    CN119806042A