A method and related device for early warning of failure of key equipment in a thermal power plant

Abstract

The application discloses a power plant key equipment fault early warning method and related device, comprising: acquiring multi-dimensional data of the power plant; inputting the multi-dimensional data of the power plant into a hybrid early warning model to obtain a comprehensive early warning index S; and performing power plant key equipment fault early warning according to the comprehensive early warning index S and a preset dynamic threshold. The method and related device can improve the advance and accuracy of early warning.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of power systems and information technology, and relates to a method and related device for early warning of faults in key equipment of thermal power plants. Background Technology

[0002] The operating environment of key equipment in thermal power plants is complex, characterized by high temperature, high pressure, and high vibration. Fault early warning systems have always been a technical challenge in the power industry. Existing technologies have the following main shortcomings: First, single-parameter monitoring has limitations. Traditional methods mostly rely on whether a single parameter, such as temperature or pressure, exceeds the standard to trigger an alarm (for example, an alarm is triggered when the temperature of a steam turbine bearing exceeds 85°C). However, equipment failures are often caused by changes in multiple parameters simultaneously, and relying on a single parameter for alarms can easily lead to a delayed response. For instance, a boiler feed pump in a thermal power plant experienced a decrease in efficiency due to scale buildup on the impeller. Initially, only the outlet pressure fluctuated slightly (not reaching the alarm value), but it wasn't until 12 hours later that a sudden drop in flow rate occurred. By then, the pipeline had already been cavitated, and the repair cost was 30% higher than if addressed earlier.

[0003] Secondly, data sources are too fragmented. Current monitoring systems (like SIS and MIS) are mostly independent. SIS systems primarily monitor real-time operational data, while MIS systems record equipment maintenance-related information. The data from these two systems is not well integrated, making it difficult to detect early signs of malfunctions from past maintenance records. For example, a turbine in one unit experienced excessive vibration. Because the record of changing the lubricating oil three months prior (when the oil quality was substandard) wasn't considered, it was mistakenly treated as bearing wear and repaired in the wrong direction.

[0004] Third, the early warning models have poor adaptability. Traditional statistical models (such as regression analysis) have difficulty dealing with the nonlinear and time-varying characteristics of equipment operation, especially under load fluctuations (such as a 20% increase in electricity load during the morning peak) or sudden environmental changes (such as a sudden increase in humidity of 15%), the model accuracy will decrease significantly, and the false alarm rate can reach more than 25%.

[0005] Fourth, threshold values ​​are fixed. Existing systems often use factory-set, fixed thresholds that don't account for equipment aging. For example, the normal vibration range of a 10-year-old steam turbine can differ by up to 15% from that of a new unit. Fixed thresholds lead to frequent false alarms from older equipment (3-5 times per month). Maintenance personnel are prone to "early warning fatigue".

[0006] Therefore, there is an urgent need for a fault early warning method that integrates multi-source data, adaptively adjusts models and thresholds, in order to improve the early warning lead time and accuracy. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and related device for early warning of faults in key equipment of thermal power plants. This method and related device can improve the advance warning and accuracy.

[0008] To achieve the above objectives, this invention discloses a method for early warning of faults in key equipment of thermal power plants, comprising: Acquire multi-dimensional data from thermal power plants; The multi-dimensional data of the thermal power plant is input into the hybrid early warning model to obtain the comprehensive early warning index S. Fault warnings for key equipment in thermal power plants are generated based on the comprehensive early warning index S and preset dynamic thresholds.

[0009] A further improvement of the power plant key equipment fault early warning method described in this invention is as follows: Furthermore, the multi-dimensional data of the thermal power plant includes equipment-level data, system-level data, management-level data, and environmental-level data. The equipment-level data includes vibration signals, temperature data, pressure and flow data, and lubricating oil parameters. The system-level data includes real-time operating parameters and historical trend data of the unit. The management-level data includes equipment ledgers, maintenance records, and historical faults. The environmental-level data includes temperature data, humidity data, and dust concentration around the unit.

[0010] Furthermore, the hybrid early warning model is constructed using a dual-model and attention fusion architecture, wherein the dual models include an LSTM sub-model and a random forest sub-model.

[0011] Furthermore, the process of providing fault warnings for key equipment in thermal power plants based on the comprehensive early warning index S and preset dynamic thresholds is as follows: Based on the comprehensive early warning index S and the preset dynamic threshold, a graded early warning method is adopted to provide early warning of faults in key equipment of thermal power plants.

[0012] Furthermore, it also includes displaying fault warning results.

[0013] This invention discloses a fault early warning system for key equipment in thermal power plants, comprising: The acquisition module is used to acquire multi-dimensional data from thermal power plants. The calculation module is used to input the multi-dimensional data of the thermal power plant into the hybrid early warning model to obtain the comprehensive early warning index S; The early warning module is used to provide early warning of faults in key equipment of thermal power plants based on the comprehensive early warning index S and preset dynamic thresholds.

[0014] A further improvement of the power plant critical equipment fault early warning system described in this invention is as follows: Furthermore, the multi-dimensional data of the thermal power plant includes equipment-level data, system-level data, management-level data, and environmental-level data. The equipment-level data includes vibration signals, temperature data, pressure and flow data, and lubricating oil parameters. The system-level data includes real-time operating parameters and historical trend data of the unit. The management-level data includes equipment ledgers, maintenance records, and historical faults. The environmental-level data includes temperature data, humidity data, and dust concentration around the unit.

[0015] Furthermore, the hybrid early warning model is constructed using a dual-model and attention fusion architecture, wherein the dual models include an LSTM sub-model and a random forest sub-model.

[0016] The present invention discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the power plant critical equipment fault early warning method.

[0017] The present invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the power plant critical equipment fault early warning method.

[0018] The present invention has the following beneficial effects: In practical operation, the fault early warning method and related device for key equipment in thermal power plants described in this invention input multi-dimensional data of the thermal power plant into a hybrid early warning model to obtain a comprehensive early warning index S. Based on the comprehensive early warning index S and a preset dynamic threshold, fault early warnings for key equipment in the thermal power plant are issued to improve the lead time and accuracy of the early warnings. It should be noted that this invention combines industrial internet, big data analysis, and artificial intelligence technologies to construct an early warning system covering the entire lifecycle of equipment, breaking through the limitations of traditional monitoring methods and providing intelligent decision support for the operation and maintenance of thermal power plant equipment. This represents a cutting-edge technology in the field of power equipment condition monitoring and fault diagnosis. Attached Figure Description

[0019] The accompanying drawings, which form part of this specification, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0022] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0023] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this invention generally indicates that the preceding and following objects have an "or" relationship.

[0024] It should be understood that although terms such as first, second, third, etc., may be used in the embodiments of the present invention to describe the preset range, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from one another. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0025] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0027] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0028] Example 1 The method for early warning of faults in key equipment of thermal power plants according to the present invention includes the following steps: 1) Obtain multi-dimensional data from thermal power plants, including equipment-level data, system-level data, management-level data, and environmental-level data; Specifically, 11) the device layer data includes: Vibration signal: Install piezoelectric accelerometers on the bearing housing and motor end cover of rotating equipment (steam turbine, fan), with a sampling frequency of 1kHz and a range of ±50g, and record the vibration acceleration of the X, Y, and Z axes.

[0029] Temperature data: Thermocouple sensor (Type K, temperature range -200~1300℃) was used. Collect bearing temperature and winding temperature at 1s interval.

[0030] Pressure and flow data: Differential pressure transmitters are installed at the inlet and outlet of the pipeline, with a measurement range of 0-10MPa and an accuracy of 0.1% FS, to record real-time pressure and flow.

[0031] Lubricating oil parameters: Viscosity (kinematic viscosity at 40℃) and moisture content (ppm) are collected by an online oil monitoring sensor at a sampling interval of 10 min.

[0032] 12) System-level data: Data at the unit level is obtained through the SIS system interface, including: Real-time operating parameters: unit load (MW), main steam temperature (°C), furnace pressure (Pa), etc., updated every 5 seconds.

[0033] Historical trend data: Stores parameter change curves for the past 3 years, and supports querying by hourly, daily, and monthly granularity.

[0034] 13) Management Data: Integrating with third-party systems (ERP, CMMS) to obtain full lifecycle information: Equipment records: manufacturer, model and specifications, design life (e.g., steam turbine design life of 30 years), and installation date.

[0035] Maintenance records: time of each maintenance, parts replaced (e.g., bearing model 7310 was replaced in May 2023), maintenance personnel, and acceptance results.

[0036] Historical faults: time of occurrence, phenomenon (e.g., "vibration value suddenly increased to 15mm / s in August 2022"), cause analysis, and handling plan.

[0037] 14) Environmental layer data: Deploy environmental sensors around the equipment to collect humidity (range 0-100% RH), dust concentration (0-100mg / m³), and ambient temperature (-30~70℃) at 1min intervals. This data is used to analyze the impact of environmental factors on the equipment (e.g., high humidity can lead to a decrease in the insulation of electrical components).

[0038] 15) The collected multidimensional data is transmitted to the edge computing gateway via industrial Ethernet for local caching (storing the data of the last 24 hours), and then uploaded to the cloud database via 5G private network to ensure data transmission latency ≤50ms.

[0039] 2) Feature extraction is performed on the multi-dimensional data of the thermal power plant; The specific process of step 2) is as follows: 21) Outlier Handling: An improved Isolation Forest algorithm is used to handle outliers. The specific process is as follows: 211) Constructing a training set: Select 10,000 data points from when the equipment is running normally as samples.

[0040] 212) Improved isolated tree: Introduce time series window (window size = 30 time points) to assign higher path length (anomaly score) to data with continuous anomalies within the window.

[0041] 213) Anomaly Repair: For marked outliers, fill them with a weighted average of the five nearest time points (with the weight decreasing as the distance increases) to avoid time series breaks caused by direct deletion.

[0042] 214) Missing value handling: For missing values ​​caused by sensor offline (such as communication interruption), if the missing duration is ≤10min, linear interpolation is used to fill the missing value; if the missing duration is >10min, the data from the same period of the same type of equipment is called to fill the missing value (such as the historical data of the #1 steam turbine for #2 steam turbine) to ensure data integrity.

[0043] 22) Feature extraction: 221) Vibration signal characteristics: Perform EMD decomposition on the vibration signal sampled at 1kHz to obtain 5th-8th order IMF components (lower order IMF reflects high frequency vibration, and higher order IMF reflects low frequency trend). Calculate the energy ratio of each component (Σ energy of the component / total energy) and kurtosis value (measures the peak characteristics of the signal; kurtosis value > 5 when bearing is worn).

[0044] 222) Operating parameter characteristics: Calculate the 24-hour moving average (smoothing short-term fluctuations), volatility ((maximum value - minimum value) / mean), and diurnal difference (the difference between the mean values ​​during the day and night) for parameters such as temperature and pressure.

[0045] 223) Environmental characteristics: Calculate the cross-characteristics of environmental parameters and equipment parameters (such as the Pearson coefficient of humidity and insulation resistance).

[0046] The final result is a 128-dimensional feature vector, which includes 32 dimensions of vibration features, 64 dimensions of operating parameter features, 16 dimensions of environmental features, and 16 dimensions of maintenance features (such as the number of days since the last maintenance and the number of historical faults).

[0047] 3) A hybrid early warning model is constructed using a dual-model and attention fusion architecture, wherein the dual models include an LSTM sub-model and a random forest sub-model; 31) The LSTM sub-model is specifically as follows: Network structure: Input layer (128-dimensional features) → 2-layer LSTM (64 hidden units per layer) → Dropout layer (to prevent overfitting, dropout rate = 0.2) → Fully connected layer → Output layer (P1, range 0-1).

[0048] Training process: Using time series data from the past 3 years as samples, the Adam optimizer (learning rate = 0.001) was used, and the loss function was mean squared error (MSE).

[0049] Trend anomaly judgment: Input the feature sequence of the past 72 hours and predict the parameter value of the next 12 hours. When the deviation between the actual value and the predicted value exceeds 5% (the deviation threshold for different parameters can be adjusted, such as the vibration deviation threshold = 8%), P1 will increase as the deviation increases (P1 = 0.6 when the deviation is 10% and P1 = 0.9 when the deviation is 20%).

[0050] 32) The random forest sub-model is specifically as follows: Model construction: Using 5,000 historical fault data (including 5 types of faults: bearing wear, impeller scaling, motor inter-turn short circuit, seal aging, and pipeline blockage) as training samples, 100 decision trees were constructed.

[0051] Feature importance: The importance of features is assessed by the Gini coefficient. For example, important features of bearing wear are vibration kurtosis (weight 25%) and lubricating oil viscosity (weight 18%).

[0052] Output P2: Input the current 128-dimensional features and output the probability of occurrence of 5 types of faults. Take the maximum value as P2 (reflecting the probability of the most likely fault type).

[0053] 33) The process of attention mechanism fusion is as follows: Dynamic weight allocation: Set basic weights based on equipment type (P1 weight = 0.6 for rotating equipment, P1 weight = 0.4 for stationary equipment), and then dynamically adjust based on real-time accuracy (e.g., if P1 issues 3 consecutive correct warnings, the weight increases by 2%).

[0054] The comprehensive early warning index S = α × P1 + (1-α) × P2, where α is the dynamic weight (range 0.4-0.8).

[0055] 4) Set dynamic thresholds; Achieve "aging-adaptive + feedback adjustment" of the threshold.

[0056] Calculate the health decay coefficient K: For example: A steam turbine has a design life of 20 years (7300 days), operates for 5 years (1825 days), and has λ=0.6. Then K=1-0.6×(1825 / 7300)=1-0.6×0.25=0.85.

[0057] Threshold adjustment: Real-time threshold S_th = initial threshold (0.7) × K, that is, the real-time threshold of the turbine = 0.7 × 0.85 = 0.595, reducing the warning threshold of old equipment.

[0058] Feedback and adjustment mechanism: Positive feedback: If the equipment has no faults for 3 consecutive months, it indicates that the threshold may be too high, and it should be increased by 5% (e.g., 0.7 → 0.735).

[0059] Negative feedback: When a false alarm occurs (no fault occurs within 24 hours after the warning), the associated parameters are traced through fault tree analysis (FTA) (e.g., the false alarm is caused by temperature sensor drift), and the threshold is lowered by 3% (0.7→0.679).

[0060] Adjust the constraints: set the upper and lower limits of the thresholds to 0.9 and 0.5 respectively to avoid over-adjustment that could cause the warning to fail.

[0061] 5) Visualization and handling of early warning results; 51) Construct a "three-level early warning + full-process response" mechanism: Grading standards: 511) Warning (0.5≤S<0.7): The parameter shows a slight abnormal trend. The warning is pushed to the operation and maintenance PC and you are prompted to strengthen monitoring.

[0062] 512) Alarm (0.7≤S<0.9): There is a relatively clear risk of failure. The alarm is pushed to the PC and mobile terminals and a checklist is generated (such as "Check bearing temperature and lubricating oil level").

[0063] 513) Emergency Alarm (S≥0.9): High risk of failure, triggering an audible and visual alarm (red flashing on the monitoring screen), pushing to the mobile phone of maintenance personnel (SMS + APP push), and associating spare parts information (e.g., "3 bearings of model 7310 in stock, which can meet the replacement needs").

[0064] 52) Visualization platform functions: 521) Real-time dashboard: Displays key parameter curves (such as the 24-hour trend of vibration values) and fault probability heatmaps (P2 values ​​for different fault types).

[0065] 522) Historical backtracking: Supports querying early warning records by time and device, and comparing the consistency between actual faults and early warnings.

[0066] 523) Decision support: Recommend handling solutions based on historical cases (e.g., "Similar vibration characteristics correspond to bearing wear in 2022, and it is recommended to stop the machine and replace it").

[0067] Example 2 This invention was piloted in a 300MW thermal power plant for six months, covering five key equipment categories including steam turbines and boiler feedwater pumps. The results showed that: Warning lead time: Increased from 2-4 hours with existing technology to 8-12 hours, with an average advance warning time of 9.3 hours.

[0068] False alarm rate: decreased from 25% to 9%, a reduction of 64%; false negative rate: decreased from 12% to 3%.

[0069] Economic benefits: Avoids two major downtimes (each resulting in a loss of approximately 1.5 million yuan), reduces unplanned maintenance time by 48 hours, and saves approximately 4 million yuan in costs annually.

[0070] It should be noted that this invention breaks down information silos by using multi-source data fusion, captures early signs of faults using intelligent algorithms, and adapts to equipment aging conditions through dynamic thresholds, thereby significantly improving the accuracy and timeliness of fault warnings for key equipment in thermal power plants and providing strong support for the safe and stable operation of the power system.

[0071] Example 3 The fault early warning system for key equipment in thermal power plants described in this invention includes: The acquisition module is used to acquire multi-dimensional data from thermal power plants. The calculation module is used to input the multi-dimensional data of the thermal power plant into the hybrid early warning model to obtain the comprehensive early warning index S; The early warning module is used to provide early warning of faults in key equipment of thermal power plants based on the comprehensive early warning index S and preset dynamic thresholds.

[0072] In this embodiment, the multi-dimensional data of the thermal power plant includes equipment layer data, system layer data, management layer data, and environmental layer data. The equipment layer data includes vibration signals, temperature data, pressure and flow data, and lubricating oil parameters. The system layer data includes real-time operating parameters and historical trend data of the unit. The management layer data includes equipment ledgers, maintenance records, and historical faults. The environmental layer data includes temperature data, humidity data, and dust concentration around the unit.

[0073] In this embodiment, the hybrid early warning model is constructed using a dual-model and attention fusion architecture, wherein the dual-model includes an LSTM sub-model and a random forest sub-model.

[0074] The module division in this embodiment is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in each embodiment of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0075] Example 4 A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of a method for early warning of faults in key equipment of a thermal power plant. For example, the method includes: acquiring multi-dimensional data of the thermal power plant; inputting the multi-dimensional data of the thermal power plant into a hybrid early warning model to obtain a comprehensive early warning index S; and performing fault early warning for key equipment of the thermal power plant based on the comprehensive early warning index S and a preset dynamic threshold. The multi-dimensional data of the thermal power plant includes equipment-level data, system-level data, management-level data, and environmental-level data. The equipment-level data includes vibration signals, temperature data, pressure and flow data, and lubricating oil parameters. The system-level data includes real-time operating parameters and historical trend data of the unit. The management-level data includes equipment ledgers, maintenance records, and historical faults. The environmental-level data includes temperature data, humidity data, and dust concentration around the unit. The memory may include main memory, such as high-speed random access memory (RAM), or non-volatile memory, such as at least one disk storage device. The processor, network interface, and memory are interconnected via an internal bus, which may be an industry-standard architecture bus, a peripheral component interconnection standard bus, or an extended industry-standard architecture bus. The bus can be categorized as an address bus, data bus, or control bus. The memory stores programs; specifically, the program may include program code, which includes computer operation instructions. The memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0076] Example 5 A computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of a method for generating and uploading trusted state alarm information. For example, the method includes: acquiring multi-dimensional data from a thermal power plant; inputting the multi-dimensional data into a hybrid early warning model to obtain a comprehensive early warning index S; and performing fault early warning for key equipment in the thermal power plant based on the comprehensive early warning index S and a preset dynamic threshold. The multi-dimensional data from the thermal power plant includes equipment-level data, system-level data, management-level data, and environmental-level data. The equipment-level data includes vibration signals, temperature data, pressure and flow data, and lubricating oil parameters. The system-level data includes real-time operating parameters and historical trend data of the unit. The management-level data includes equipment ledgers, maintenance records, and historical faults. The environmental-level data includes temperature data, humidity data, and dust concentration around the unit. Specifically, the computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. The volatile memory may include random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include read-only memory (ROM), hard disk, flash memory, optical disk, magnetic disk, etc.

[0077] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0078] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0079] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0080] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0081] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and disclosure of the invention. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0082] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

[0083] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for early warning of faults in key equipment of a thermal power plant, characterized in that, include: Acquire multi-dimensional data from thermal power plants; The multi-dimensional data of the thermal power plant is input into the hybrid early warning model to obtain the comprehensive early warning index S. Fault warnings for key equipment in thermal power plants are generated based on the comprehensive early warning index S and preset dynamic thresholds.

2. The method for early warning of faults in key power plant equipment according to claim 1, characterized in that, The multi-dimensional data of the thermal power plant includes equipment-level data, system-level data, management-level data, and environmental-level data. The equipment-level data includes vibration signals, temperature data, pressure and flow data, and lubricating oil parameters. The system-level data includes real-time operating parameters and historical trend data of the units. The management-level data includes equipment ledgers, maintenance records, and historical faults. The environmental-level data includes temperature data, humidity data, and dust concentration around the units.

3. The method for early warning of faults in key power plant equipment according to claim 1, characterized in that, The hybrid early warning model is constructed using a dual-model and attention fusion architecture, wherein the dual models include an LSTM sub-model and a random forest sub-model.

4. The method for early warning of faults in key power plant equipment according to claim 1, characterized in that, The process of providing fault early warning for key equipment in thermal power plants based on the comprehensive early warning index S and the preset dynamic threshold is as follows: Based on the comprehensive early warning index S and the preset dynamic threshold, a graded early warning method is adopted to provide early warning of faults in key equipment of thermal power plants.

5. The method for early warning of faults in key power plant equipment according to claim 1, characterized in that, Also includes: Displays fault warning results.

6. A fault early warning system for key equipment in a thermal power plant, characterized in that, include: The acquisition module is used to acquire multi-dimensional data from thermal power plants. The calculation module is used to input the multi-dimensional data of the thermal power plant into the hybrid early warning model to obtain the comprehensive early warning index S; The early warning module is used to provide early warning of faults in key equipment of thermal power plants based on the comprehensive early warning index S and preset dynamic thresholds.

7. The power plant key equipment fault early warning system according to claim 6, characterized in that, The multi-dimensional data of the thermal power plant includes equipment-level data, system-level data, management-level data, and environmental-level data. The equipment-level data includes vibration signals, temperature data, pressure and flow data, and lubricating oil parameters. The system-level data includes real-time operating parameters and historical trend data of the units. The management-level data includes equipment ledgers, maintenance records, and historical faults. The environmental-level data includes temperature data, humidity data, and dust concentration around the units.

8. The power plant key equipment fault early warning system according to claim 6, characterized in that, The hybrid early warning model is constructed using a dual-model and attention fusion architecture, wherein the dual models include an LSTM sub-model and a random forest sub-model.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the power plant critical equipment fault early warning method as described in any one of claims 1-5.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the power plant critical equipment fault early warning method as described in any one of claims 1-5.

Images