Operation efficiency monitoring method for thermal generator set
By constructing a standard operating model and multi-stage fault diagnosis method for thermal generator sets, the accuracy and timeliness of thermal generator set operation efficiency monitoring are solved, and the rapid identification and effective correction of faults are achieved, and the reliability and automation of the system are improved.
Patent Information
- Application Number
- CN202510526451.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-24
- Publication Date
- 2025-08-15
AI Technical Summary
The prior art is difficult to quickly and accurately monitor the operating efficiency of thermal power generator sets, resulting in untimely fault detection, affecting system reliability and safety.
By building a standard operating model of thermal power generator sets, combining historical data and real-time monitoring data, a collection of evaluation values is generated, multi-level fault diagnosis is carried out, including preliminary fault type judgment and secondary verification, and correction instructions are generated to optimize monitoring strategies.
It improves the accuracy and timeliness of fault diagnosis, reduces misjudgment, dynamically adjusts monitoring frequency, improves the degree of automation and scientific decision-making of the system, and ensures the continuous and effective resolution of faults.
Smart Images
Figure CN120496291A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of operation monitoring of power generation sets, and in particular to an operation efficiency monitoring method for thermal power generation sets. Background Art
[0002] In the thermal power generation sector, improving the operating efficiency of generator sets is key to reducing power generation costs and improving competitiveness. With intensified competition in the energy market and rising demands for environmental protection, accurately monitoring operating efficiency and promptly detecting faults are crucial for optimizing thermal power generation performance. For example, improving operating efficiency can reduce fuel consumption, such as coal, and lower emissions of pollutants, such as carbon dioxide.
[0003] Thermal power generation units are complex, large-scale systems whose reliability and safety are directly linked to the stability of power supply. Any failure, if not promptly detected and addressed, can lead to power outages, severely impacting industrial production and residents' lives. Therefore, an effective operational efficiency monitoring method is needed to provide early warning of failures and ensure safe and reliable unit operation. Summary of the Invention
[0004] The purpose of this invention is to combine data analysis and machine learning technology to improve the timeliness and accuracy of monitoring and enhance the operation and maintenance efficiency of power plants.
[0005] To achieve the above object, the present invention provides a method for monitoring the operating efficiency of a thermal power generating set, comprising: Constructing a standard operation model of the thermal power generation unit based on the acquired historical operation data of the thermal power generation unit; Set monitoring time nodes and obtain monitoring data of thermal power generators at each monitoring time node; Generate evaluation value set A for each functional partition of thermal power generation unit based on monitoring data of thermal power generation unit at each monitoring time node; Combine the evaluation value set A and the standard operation model of the thermal power generating unit to generate the corresponding final fault type; Among them, the evaluation value set A, A={a1, a2…a i …a n}; ai is the evaluation value of the i-th functional partition of the thermal power generating unit, and n is the total number of functional partitions of the thermal power generating unit.
[0006] In some embodiments of the present invention, the step of constructing a standard operation model of a thermal power generating set based on the acquired historical operation data of the thermal power generating set includes: Based on the historical operating parameters of the thermal power generating unit, a historical data set of each functional partition of the thermal power generating unit is obtained; Obtain the historical data range of each functional zone based on the historical data set; Conduct hierarchical evaluation of the historical operating status of each functional zone to generate multiple evaluation levels; Divide the historical data range of each functional zone based on the evaluation level and generate the data interval corresponding to the evaluation level; Among them, multiple evaluation levels include: general level, good level and excellent level; The data interval corresponding to the excellent level of each functional partition is obtained to generate a standard operation model of the thermal power generating unit.
[0007] In some embodiments of the present invention, the step of obtaining monitoring data of the thermal power generating set at each monitoring time node further includes: Obtain the historical abnormal event set of the thermal power generating unit; Obtain the data feature set of each functional partition based on the historical abnormal event set; The data feature set includes: the maximum data value m1, the minimum data value m2 and the slope a1 of the data of the current functional partition during the process from the occurrence to the end of the abnormal event; Generate a first reference value E based on the data feature set; E=w1*(m2-m1)+w2*a1; Among them, w1 is the coefficient of the abnormal event data difference, and w2 is the coefficient of the slope a1; Setting time intervals of multiple monitoring time nodes based on the first reference value E of all functional partitions; Obtain monitoring data of thermal power generating units at various time intervals.
[0008] In some embodiments of the present invention, when setting the time intervals of multiple monitoring time nodes based on the first reference value E of all functional partitions, the method further includes: The E values of all functional partitions were clustered and divided into different categories; Set different time intervals for different categories of functional zones; Calculate the standard deviation σ of the first reference value E of the current category functional zoning E and the deviation from the mean μ E ; Combined with the standard deviation σ of the first reference value E E and the deviation from the mean μ E Set the monitoring time interval for the current category functional partition; If E>μ E +2*σ E or E<μ E -2*σ E , then the monitoring time interval of the current category functional zone is n1.
[0009] In some embodiments of the present invention, generating the evaluation value set A for each functional partition of the thermal power generating set further includes: Obtain monitoring data of each monitoring time node of the current functional partition and generate operating efficiency reference values of each monitoring time node of the current functional partition; Calculate the efficiency change value and efficiency mean of the current functional partition by combining the operating efficiency reference values of each time node of the current functional partition; Obtain the efficiency change values and efficiency mean values of all functional partitions to generate the evaluation value set A.
[0010] In some embodiments of the present invention, when generating the corresponding final fault type by combining the evaluation value set A and the standard operation model of the thermal power generating set, the method further includes: Combined with whether the efficiency mean of each functional zone is within the data interval corresponding to the standard operation model of the thermal power generating unit; If the efficiency mean of the current functional partition is not within the data interval corresponding to the standard operation model of the thermal power generating set, a first fault detection is performed on the current functional partition based on the efficiency change value and the efficiency mean of the current functional partition to generate a preliminary fault type of the current functional partition; Obtain preliminary fault types for all functional partitions; Perform secondary verification based on the correlation between the preliminary fault types of each functional partition to generate the final fault type; Generate corrective instructions based on the resulting fault type.
[0011] In some embodiments of the present invention, the generating of the preliminary fault type of the current functional partition further includes: Obtain the efficiency reference value of each fault type in historical fault data; Construct a fault type characteristic diagram based on the efficiency value and efficiency change value of the current fault type; The fault type characteristic graph comprises a plurality of sequentially connected nodes; Set the efficiency reference value difference between nodes based on historical fault data; Generate a preliminary fault type judgment model by integrating all fault type characteristic diagrams; Generate an efficiency characteristic diagram of the current functional partition based on the efficiency reference value difference, the efficiency change value of the current functional partition, and the efficiency mean; The preliminary fault type of the current functional partition is generated by combining the efficiency characteristic diagram of the current functional partition and the preliminary fault type judgment model.
[0012] In some embodiments of the present invention, the secondary verification based on the correlation between the preliminary fault types of each functional partition further includes: Calculate the impact of the current functional partition fault type on the corresponding efficiency reference values of other functional partitions in sequence; Calculate the mean impact value of each functional zone; Combine the efficiency reference value and the impact value mean of the current functional partition to generate the predicted repair value of the current functional partition; Determine whether the predicted repair value of the current functional partition belongs to the corresponding data interval; If yes, then the current preliminary fault type is judged to be correct; If it does not, an alarm message is generated.
[0013] In some embodiments of the present invention, the generating of the correction instruction based on the final fault type further includes: executing correction instructions based on the final fault type; Monitor and correct the efficiency reference values of each functional partition during the instruction execution process; The correction instruction is adjusted by comparing the efficiency reference value of each functional partition during the execution process of the correction instruction with the preset value.
[0014] Compared with the prior art, the method for monitoring the operating efficiency of a thermal power generation unit provided by the embodiment of the present invention has the following advantages: By dynamically adjusting the monitoring frequency, unnecessary data collection and processing can be reduced, thereby improving system efficiency. The multi-level fault diagnosis mechanism can more accurately identify the type of fault and reduce misjudgments. Comprehensively evaluate the system status: Taking into account the correlation between various functional partitions, it is possible to comprehensively evaluate the system operation status. Optimizing the monitoring strategy based on historical abnormal events can detect potential problems more quickly. By continuously updating the standard operating model and correction instructions, the system can adapt to new operating conditions and failure modes. From data collection to fault diagnosis to correction instruction generation, the entire process is highly automated, reducing human intervention. Analytical decisions are made based on large amounts of historical data and real-time data to improve the scientific nature and reliability of decisions. By continuously monitoring the effects of correction instructions, a closed-loop control system is formed to ensure that faults are continuously and effectively resolved. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 The present invention provides a flowchart of a method for monitoring the operating efficiency of a thermal power generating set. DETAILED DESCRIPTION
[0016] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following embodiments are used to illustrate the present invention but are not intended to limit the scope of the present invention.
[0017] In the description of the present invention, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying 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.
[0018] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.
[0019] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.
[0020] Example 1: The embodiment of the present invention provides a method for monitoring the operating efficiency of a thermal power generating set. Figure 1 Shown, including: Constructing a standard operation model of the thermal power generation unit based on the acquired historical operation data of the thermal power generation unit; Set monitoring time nodes and obtain monitoring data of thermal power generators at each monitoring time node; Generate evaluation value set A for each functional partition of thermal power generation unit based on monitoring data of thermal power generation unit at each monitoring time node; Combine the evaluation value set A and the standard operation model of the thermal power generating unit to generate the corresponding final fault type; Among them, the evaluation value set A, A={a1, a2…a i …a n}; ai is the evaluation value of the i-th functional partition of the thermal power generating unit, and n is the total number of functional partitions of the thermal power generating unit.
[0021] Example 2: The method of constructing a standard operation model of a thermal power generating set based on the acquired historical operation data of the thermal power generating set includes: Based on the historical operating parameters of the thermal power generating unit, a historical data set of each functional partition of the thermal power generating unit is obtained; Obtain the historical data range of each functional zone based on the historical data set; Conduct hierarchical evaluation of the historical operating status of each functional zone to generate multiple evaluation levels; Divide the historical data range of each functional zone based on the evaluation level and generate the data interval corresponding to the evaluation level; Among them, multiple evaluation levels include: general level, good level and excellent level; The data interval corresponding to the excellent level of each functional partition is obtained to generate a standard operation model of the thermal power generating unit.
[0022] In this embodiment, the functional zones include but are not limited to: boiler efficiency zones, exhaust gas temperature zones, turbine heat rate zones, condenser vacuum zones, and feed water temperature control zones.
[0023] Boiler Efficiency and Exhaust Temperature: Improving boiler efficiency generally results in lower exhaust temperatures, as more heat is absorbed and used to generate steam rather than being lost in the flue gases. However, excessively low exhaust temperatures can lead to condensation or corrosion on heating surfaces, so a balance must be struck between efficiency and equipment safety.
[0024] Steam turbine heat rate and condenser vacuum: A higher condenser vacuum can reduce turbine back pressure, allowing more energy to be extracted from the turbine, thereby reducing heat rate. However, increasing vacuum requires increasing cooling water flow, which increases pump power consumption, so a trade-off must be made.
[0025] Feedwater temperature and boiler efficiency: Increasing feedwater temperature reduces the energy required to bring water to boiling point, thereby improving boiler efficiency. Feedwater temperature is affected by the condensate return temperature and is related to the condenser's operating conditions.
[0026] Flue gas composition (CO, O2 content) and other parameters: CO and O2 content not only affect combustion efficiency but also impact the boiler's heating surface. Excessive O2 content can lead to increased NOx formation, while high CO content indicates incomplete combustion. These factors can reduce efficiency and affect other parameters.
[0027] Turbine efficiency and boiler operation: Turbine efficiency is affected by inlet steam parameters (such as pressure and temperature), which are determined by boiler operating conditions. Turbine efficiency also affects the operating conditions of the condenser, which in turn affects the circulating water system.
[0028] Generator efficiency and overall system: Although generator efficiency is relatively independent, it directly affects the final net output power. Improving generator efficiency can increase electrical energy output for the same mechanical power.
[0029] The interactions between these parameters form a complex system that requires coordinated control to achieve overall optimization. For example, flue gas composition can be optimized by adjusting combustion conditions, boiler and turbine requirements can be balanced by adjusting the feedwater system, and turbine back pressure can be improved by optimizing condenser operation.
[0030] In actual operation, factors such as equipment limitations, environmental conditions, and load requirements must be considered, and various parameters must be dynamically adjusted to achieve the optimal balance. This complex interaction requires operators to possess extensive experience and expertise, as well as the use of advanced automated control systems for refined management.
[0031] Example 3: When acquiring the monitoring data of the thermal power generating set at each monitoring time node, the method further includes: Obtain the historical abnormal event set of the thermal power generating unit; Obtain the data feature set of each functional partition based on the historical abnormal event set; The data feature set includes: the maximum data value m1, the minimum data value m2 and the slope a1 of the data of the current functional partition during the process from the occurrence to the end of the abnormal event; Generate a first reference value E based on the data feature set; E=w1*(m2-m1)+w2*a1; Among them, w1 is the coefficient of the abnormal event data difference, and w2 is the coefficient of the slope a1; Setting time intervals of multiple monitoring time nodes based on the first reference value E of all functional partitions; Obtain monitoring data of thermal power generating units at various time intervals.
[0032] In this embodiment, a set of historical abnormal events is obtained from the historical operation record database of the thermal power generator. These records may include past fault alarm records, records of sudden equipment performance degradation, and records of other abnormal operating conditions. For example, detailed information about an event such as a sudden drop in power generation or an abnormal temperature increase of a device may be recorded, including the time the event occurred and the duration of the event.
[0033] The acquired historical abnormal event sets are sorted and categorized according to different functional zones or fault types. For example, all abnormal events related to the combustion system are grouped into one category, and those related to the steam generation system into another. This facilitates subsequent analysis of each functional zone.
[0034] For each functional zone, the data range from the occurrence to the end of the abnormal event is determined in the historical abnormal event set. For example, for the power generation zone, during an abnormal event of a sudden drop in power generation, the maximum data value m1 and minimum data value m2 of the power data are determined, as well as the slope a1 of the power change curve over time. This slope a1 reflects the speed of power decline or increase.
[0035] The difference between the maximum data value m1 and the minimum data value m2 (m2 - m1) reflects the fluctuation amplitude of the functional partition during an abnormal event, while the slope a1 reflects the development trend of the abnormal event. For example, in the case of an abnormal temperature increase, a larger slope means that the temperature rise is faster, which may indicate a more serious fault potential.
[0036] The determination of coefficients w1 and w2 requires consideration based on practical circumstances and experience. w1, as the coefficient for the data difference of anomaly events, may be set based on historical data analysis. For example, if extensive historical data analysis reveals that data differences have a significant impact on faults, w1 can be set to a relatively large value. w2, as the coefficient for slope a1, is similarly set based on the slope's importance to fault diagnosis.
[0037] The time interval is set based on the first reference value E of all functional zones. This is because the E value comprehensively reflects the characteristics of each functional zone under abnormal conditions. By analyzing the E value, the abnormal risk level of different functional zones can be determined, thereby reasonably arranging the monitoring time interval.
[0038] Acquire monitoring data for thermal power generators at set intervals. For example, if the interval for a functional zone is set to 10 minutes, relevant operating data for that zone, such as temperature, pressure, and power, will be acquired every 10 minutes.
[0039] Embodiment 4: When the time intervals of multiple monitoring time nodes are set based on the first reference value E of all functional zones, the method further includes: The E values of all functional partitions were clustered and divided into different categories; Set different time intervals for different categories of functional zones; Calculate the standard deviation σ of the first reference value E of the current category functional zoning E and the deviation from the mean μ E ; Combined with the standard deviation σ of the first reference value E E and the deviation from the mean μ E Set the monitoring time interval for the current category functional partition; If E>μ E +2*σ Eor E<μ E -2*σ E , then the monitoring time interval of the current category functional zone is n1.
[0040] In this embodiment, a suitable cluster analysis method, such as the K-Means clustering algorithm, is selected. The E values of all functional zones are used as input data for cluster analysis, which classifies them into different categories. For example, based on the size and distribution of the E values, the functional zones may be divided into high-risk, medium-risk, and low-risk categories.
[0041] Different categories represent different abnormal risk characteristics. Functional areas in the high-risk category may have larger E values, which means that there are large fluctuations or obvious trend changes in abnormal events, and closer monitoring is required.
[0042] Different time intervals are set for different functional zones. High-risk functional zones require more frequent monitoring due to their high risk of abnormalities, so shorter time intervals are set. Low-risk functional zones can have longer time intervals set. For example, the time interval for high-risk zones can be set to 5 minutes, for medium-risk zones to 10 minutes, and for low-risk zones to 15 minutes.
[0043] For each category, calculate the standard deviation σ of the first reference value E E and deviation from the mean μE. Standard deviation σ E It reflects the dispersion of E values within the category, which deviates from the mean value μ E Indicates the degree of deviation of the E value of a functional zone from the average value of the category. For example, in the high-risk category, σ is obtained by calculating E = 5, μ E = 30.
[0044] Example 5: When generating the evaluation value set A of each functional partition of the thermal power generating set, the method further includes: Obtain monitoring data of each monitoring time node of the current functional partition and generate operating efficiency reference values of each monitoring time node of the current functional partition; Calculate the efficiency change value and efficiency mean of the current functional partition by combining the operating efficiency reference values of each time node of the current functional partition; Obtain the efficiency change values and efficiency mean values of all functional partitions to generate the evaluation value set A.
[0045] In this embodiment, a specific algorithm is used to generate an operating efficiency reference value for the monitoring data acquired at each monitoring time point in the current functional zone. For example, for the power generation functional zone, the operating efficiency reference value for each monitoring time point can be calculated using the formula (power generation / fuel consumption) based on data such as power generation and fuel consumption.
[0046] Other factors that may affect operating efficiency may also need to be considered, such as ambient temperature, equipment load, etc. If the ambient temperature is low, it may affect the performance of some equipment and thus affect the operating efficiency. These factors can be corrected when calculating the operating efficiency reference value.
[0047] The efficiency change value is calculated by combining the operating efficiency reference values at each time point in the current functional zone. For example, the efficiency change value can be calculated by subtracting the operating efficiency reference values at adjacent time points. If the operating efficiency reference value at time point t1 is e1 and the operating efficiency reference value at time point t2 is e2, then the efficiency change value Δe = e2 - e1.
[0048] To calculate the mean efficiency, add the operating efficiency reference values for all time nodes and divide by the total number of time nodes. For example, if there are n time nodes with operating efficiency reference values e1, e2, …, en, the mean efficiency is ē = (e1 + e2 + … + en) / n.
[0049] Obtain the efficiency change values and efficiency mean values of all functional partitions, and combine them to generate an evaluation value set A. For example, for a thermal power generating unit with three functional partitions (power generation, steam generation, and cooling), calculate their efficiency change values and efficiency mean values respectively, and then form an evaluation value set A in a certain order (such as the importance or numbering of the functional partitions), A = {a1, a2, a3}, where a1 is the evaluation value of the power generation functional partition (including its efficiency change value and efficiency mean information), a2 is the evaluation value of the steam generation functional partition, and a3 is the evaluation value of the cooling functional partition.
[0050] Example 6: When the corresponding final fault type is generated by combining the evaluation value set A and the standard operation model of the thermal power generator set, the method further includes: Combined with whether the efficiency mean of each functional zone is within the data interval corresponding to the standard operation model of the thermal power generating unit; If the efficiency mean of the current functional partition is not within the data interval corresponding to the standard operation model of the thermal power generating set, a first fault detection is performed on the current functional partition based on the efficiency change value and the efficiency mean of the current functional partition to generate a preliminary fault type of the current functional partition; Obtain preliminary fault types for all functional partitions; Perform secondary verification based on the correlation between the preliminary fault types of each functional partition to generate the final fault type; Generate corrective instructions based on the resulting fault type.
[0051] In this embodiment, the efficiency mean of each functional zone is determined to be within the data interval corresponding to the standard operating model of the thermal power generation unit. For example, for the power generation functional zone, the efficiency mean data interval in its standard operating model is [lower limit value, upper limit value]. If the actual efficiency mean of the functional zone is lower than the lower limit value or higher than the upper limit value, the preliminary fault detection process is triggered.
[0052] Fault detection based on efficiency variation and mean value If the mean efficiency value of the current functional zone is not within the data interval corresponding to the standard operating model of the thermal power generator set, a first fault detection is performed on the current functional zone based on the efficiency change value and the efficiency mean value of the current functional zone to generate a preliminary fault type for the current functional zone. For example, if the mean efficiency value of the power generation functional zone is lower than the lower limit and the efficiency change value continues to decrease, it may be preliminarily determined that the power generation efficiency has decreased due to aging or failure of the power generation equipment.
[0053] Obtain preliminary fault types for all functional partitions The above preliminary fault detection is performed on each functional zone to obtain preliminary fault types for all functional zones. For example, in addition to the power generation zone's preliminary fault diagnosis of equipment aging, the steam generation zone's preliminary fault diagnosis of pipe blockage, and the cooling zone's preliminary fault diagnosis of coolant shortage, etc.
[0054] Embodiment 7: When generating the preliminary fault type of the current functional partition, the method further includes: Obtain the efficiency reference value of each fault type in historical fault data; Construct a fault type characteristic diagram based on the efficiency value and efficiency change value of the current fault type; The fault type characteristic graph comprises a plurality of sequentially connected nodes; Set the efficiency reference value difference between nodes based on historical fault data; Generate a preliminary fault type judgment model by integrating all fault type characteristic diagrams; Generate an efficiency characteristic diagram of the current functional partition based on the efficiency reference value difference, the efficiency change value of the current functional partition, and the efficiency mean; The preliminary fault type of the current functional partition is generated by combining the efficiency characteristic diagram of the current functional partition and the preliminary fault type judgment model.
[0055] In this embodiment, the efficiency mean of each functional zone is first compared with the corresponding data interval in the standard operating model of the thermal power generation unit. For example, for the power generation functional zone of the thermal power generation unit, the efficiency mean data interval in the standard operating model is [lower limit value, upper limit value]. If the actual efficiency mean of the functional zone is lower than the lower limit value or higher than the upper limit value, the preliminary fault detection process is triggered.
[0056] This judgment based on data intervals is based on the fact that the data intervals in the standard operation model represent the reasonable range of the efficiency mean when the functional division is operating normally. Once this range is exceeded, there may be a potential fault risk.
[0057] Efficiency reference values for each fault type are extracted from historical fault data. This data includes relevant data for each functional zone during past thermal power generator failures. For example, for overheating faults, the efficiency changes of the power generation functional zone during overheating are recorded, including efficiency values at different time points. This data is then compiled into efficiency reference values.
[0058] A fault type characteristic graph is constructed based on the efficiency value and efficiency change value of the current fault type. Assuming the current fault type is efficiency fluctuation, the efficiency value is plotted on the vertical axis and time on the horizontal axis. The change in efficiency value over time forms a series of connected nodes. For example, if efficiency first decreases and then increases within a certain time period, the efficiency values at these different time points are connected to form the characteristic graph for this fault type.
[0059] Use historical fault data to set the reference efficiency difference between nodes. For example, based on the development patterns of similar fault types in the past, determine the reasonable range of efficiency changes between adjacent nodes. If the fault trend is continuously decreasing, the reference efficiency difference between adjacent nodes should gradually decrease.
[0060] A preliminary fault type judgment model is generated by integrating all fault type feature maps. This model integrates various possible fault types and their corresponding feature maps. By analyzing and summarizing the feature maps of different fault types, it can judge the possible fault type based on the relevant data of the current functional partition input.
[0061] The efficiency characteristic diagram of the current functional zone is generated based on the efficiency reference value difference, the efficiency change value of the current functional zone, and the efficiency mean. For example, the efficiency characteristic diagram of the current power generation functional zone is drawn based on the efficiency mean value of the current power generation functional zone, the efficiency change value over a period of time, and the previously set efficiency reference value difference between nodes.
[0062] The preliminary fault type for the current functional partition is generated by combining the efficiency characteristic graph of the current functional partition with the preliminary fault type judgment model. The efficiency characteristic graph of the current functional partition is matched with the various fault type characteristic graphs in the preliminary fault type judgment model to find the most similar fault type characteristic graph, thereby determining the preliminary fault type. For example, if the efficiency characteristic graph of the current functional partition is most similar to the bearing wear fault type characteristic graph in the preliminary fault type judgment model, then the preliminary fault type for the current functional partition is determined to be bearing wear.
[0063] Embodiment 8: The secondary verification based on the correlation between the preliminary fault types of each functional partition further includes: Calculate the impact of the current functional partition fault type on the corresponding efficiency reference values of other functional partitions in sequence; Calculate the mean impact value of each functional zone; Combine the efficiency reference value and the impact value mean of the current functional partition to generate the predicted repair value of the current functional partition; Determine whether the predicted repair value of the current functional partition belongs to the corresponding data interval; If yes, then the current preliminary fault type is judged to be correct; If it does not, an alarm message is generated.
[0064] In this embodiment, the impact on the remaining functional zones is calculated, sequentially calculating the impact of the current functional zone's fault type on the corresponding efficiency reference values of the remaining functional zones. For example, if the power generation zone's preliminary fault type is determined to be an overheating fault, the impact of this overheating fault on the efficiency reference values of other related functional zones, such as the steam transmission zone and the cooling zone, is analyzed. This impact value can be calculated by establishing a relevant mathematical model or empirical relationships based on historical data.
[0065] Calculate the mean impact value for each functional zone. Average the impact values of each functional zone affected by the current functional zone fault type. For example, if three related functional zones are affected by the overheating fault in the power generation zone, and their impact values are [impact value 1, impact value 2, impact value 3], then the mean impact value = (impact value 1 + impact value 2 + impact value 3) / 3.
[0066] The predicted repair value for the current functional zone is generated by combining the efficiency reference value and the mean impact value of the current functional zone. For example, the efficiency reference value of the current power generation functional zone is a specific value, and the calculated mean impact value is added to obtain the predicted repair value.
[0067] Determine whether the predicted repair value falls within the corresponding data interval. This corresponding data interval is set based on the standard for normal operation of the thermal power generator. If the predicted repair value falls within this interval, the current preliminary fault type is correct. If not, an alarm is generated, indicating that a more complex fault situation may exist or that the preliminary fault type is incorrect.
[0068] Embodiment 9: When generating a correction instruction based on the final fault type, the method further includes: executing correction instructions based on the final fault type; Monitor and correct the efficiency reference values of each functional partition during the instruction execution process; The correction instruction is adjusted by comparing the efficiency reference value of each functional partition during the execution process of the correction instruction with the preset value.
[0069] In this embodiment, the corrective instructions are generated based on the final fault type. For example, if the final fault type is bearing wear in the power generation functional area, the corrective instructions may include operations such as replacing the bearing or adjusting the bearing lubrication system.
[0070] Execute the generated correction instructions and perform corresponding maintenance or adjustment operations on the thermal power generating unit.
[0071] During the execution of the correction instructions, the efficiency reference values of each functional zone are continuously monitored. For example, the efficiency values of the power generation zone and steam transmission zone are obtained in real time to understand the effect of the correction instructions.
[0072] Corrective instructions are adjusted by comparing the efficiency reference values of each functional zone during execution with preset values. If the efficiency reference value of a functional zone does not meet the preset value, the execution method or parameters of the correction instruction may need to be adjusted. For example, if the efficiency of the power generation functional zone remains below the preset normal operating efficiency after executing the correction instruction, further inspection of bearing replacement or adjustment of lubrication system parameters may be necessary.
[0073] Finally, it should be noted that it is apparent that those skilled in the art may make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, to the extent such modifications and variations fall within the scope of the present invention and its equivalents, the present invention is intended to include such modifications and variations.
[0074] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.
Claims
1. A method for monitoring the operating efficiency of a thermal power generating unit, characterized in that: include: Constructing a standard operation model of the thermal power generation unit based on the acquired historical operation data of the thermal power generation unit; Set monitoring time nodes and obtain monitoring data of thermal power generators at each monitoring time node; Generate evaluation value set A for each functional partition of thermal power generation unit based on monitoring data of thermal power generation unit at each monitoring time node; Combine the evaluation value set A and the standard operation model of the thermal power generating unit to generate the corresponding final fault type; Among them, the evaluation value set A, A={a1, a2…a i …a n }; ai is the evaluation value of the i-th functional partition of the thermal power generating unit, and n is the total number of functional partitions of the thermal power generating unit.
2. The method for monitoring the operating efficiency of a thermal power generating set according to claim 1, wherein: The method of constructing a standard operation model of a thermal power generating set based on the acquired historical operation data of the thermal power generating set includes: Based on the historical operating parameters of the thermal power generating unit, a historical data set of each functional partition of the thermal power generating unit is obtained; Obtain the historical data range of each functional zone based on the historical data set; Conduct hierarchical evaluation of the historical operating status of each functional zone to generate multiple evaluation levels; Divide the historical data range of each functional zone based on the evaluation level and generate the data interval corresponding to the evaluation level; Among them, multiple evaluation levels include: general level, good level and excellent level; The data interval corresponding to the excellent level of each functional partition is obtained to generate a standard operation model of the thermal power generating unit.
3. The method for monitoring the operating efficiency of a thermal power generating set according to claim 2, wherein: The acquisition of monitoring data of the thermal power generating set at each monitoring time node also includes: Obtain the historical abnormal event set of the thermal power generating unit; Obtain the data feature set of each functional partition based on the historical abnormal event set; The data feature set includes: the maximum data value m1, the minimum data value m2 and the slope a1 of the data of the current functional partition during the process from the occurrence to the end of the abnormal event; Generate a first reference value E based on the data feature set; E=w1*(m2-m1)+w2*a1; Among them, w1 is the coefficient of the abnormal event data difference, and w2 is the coefficient of the slope a1; Setting time intervals of multiple monitoring time nodes based on the first reference value E of all functional partitions; Obtain monitoring data of thermal power generating units at various time intervals.
4. The method for monitoring the operating efficiency of a thermal power generating set according to claim 3, wherein: When the time intervals of multiple monitoring time nodes are set based on the first reference value E of all functional partitions, the method further includes: The E values of all functional partitions were clustered and divided into different categories; Set different time intervals for different categories of functional zones; Calculate the standard deviation σ of the first reference value E of the current category functional zoning E and the deviation from the mean μ E ; Combined with the standard deviation σ of the first reference value E E and the deviation from the mean μ E Set the monitoring time interval for the current category functional partition; If E>μ E +2*σ E or E<μ E -2*σ E , then the monitoring time interval of the current category functional zone is n1.
5. The method for monitoring the operating efficiency of a thermal power generating set according to claim 4, wherein: When generating the evaluation value set A of each functional partition of the thermal power generating set, the following steps are also included: Obtain monitoring data of each monitoring time node of the current functional partition and generate operating efficiency reference values of each monitoring time node of the current functional partition; Calculate the efficiency change value and efficiency mean of the current functional partition by combining the operating efficiency reference values of each time node of the current functional partition; Obtain the efficiency change values and efficiency mean values of all functional partitions to generate the evaluation value set A.
6. The method for monitoring the operating efficiency of a thermal power generating set according to claim 5, wherein: When the corresponding final fault type is generated by combining the evaluation value set A and the standard operation model of the thermal power generating set, the method further includes: Combined with whether the efficiency mean of each functional zone is within the data interval corresponding to the standard operation model of the thermal power generating unit; If the efficiency mean of the current functional partition is not within the data interval corresponding to the standard operation model of the thermal power generating set, a first fault detection is performed on the current functional partition based on the efficiency change value and the efficiency mean of the current functional partition to generate a preliminary fault type of the current functional partition; Obtain preliminary fault types for all functional partitions; Perform secondary verification based on the correlation between the preliminary fault types of each functional partition to generate the final fault type; Generate corrective instructions based on the resulting fault type.
7. The method for monitoring the operating efficiency of a thermal power generating set according to claim 6, wherein: The generating of the preliminary fault type of the current functional partition further includes: Obtain the efficiency reference value of each fault type in historical fault data; Construct a fault type characteristic diagram based on the efficiency value and efficiency change value of the current fault type; The fault type characteristic graph comprises a plurality of sequentially connected nodes; Set the efficiency reference value difference between nodes based on historical fault data; Generate a preliminary fault type judgment model by integrating all fault type characteristic diagrams; Generate an efficiency characteristic diagram of the current functional partition based on the efficiency reference value difference, the efficiency change value of the current functional partition, and the efficiency mean; The preliminary fault type of the current functional partition is generated by combining the efficiency characteristic diagram of the current functional partition and the preliminary fault type judgment model.
8. The method for monitoring the operating efficiency of a thermal power generating set according to claim 7, wherein: The secondary verification based on the correlation between the preliminary fault types of each functional partition also includes: Calculate the impact of the current functional partition fault type on the corresponding efficiency reference values of other functional partitions in sequence; Calculate the mean impact value of each functional zone; Combine the efficiency reference value and the impact value mean of the current functional partition to generate the predicted repair value of the current functional partition; Determine whether the predicted repair value of the current functional partition belongs to the corresponding data interval; If yes, then the current preliminary fault type is judged to be correct; If it does not, an alarm message is generated.
9. The method for monitoring the operating efficiency of a thermal power generating set according to claim 8, wherein: When generating the correction instruction based on the final fault type, the method further includes: executing correction instructions based on the final fault type; Monitor and correct the efficiency reference values of each functional partition during the instruction execution process; The correction instruction is adjusted by comparing the efficiency reference value of each functional partition during the execution process of the correction instruction with the preset value.