Steam turbine speed regulating system monitoring method and related equipment

Abstract

The invention discloses a steam turbine speed regulating system monitoring method and related equipment, and aims to overcome the defect of high state misjudgment rate of a steam turbine speed regulating system in a variable working condition environment in the prior art. The comprehensive operation state information of the system is obtained by collecting multi-source operation parameters of the steam turbine speed regulating system in real time. And constructing a dynamic reference model by utilizing historical normal state data to adapt to complex operation characteristics of the steam turbine. And a real-time health index is obtained by calculating the deviation between the real-time parameter and the dynamic reference value, so that quantitative evaluation of the health state of the system is realized. And false alarms are effectively reduced through a grading alarm mechanism. Combined features of the health indexes and the multi-source operation parameters are combined, common fault types can be accurately identified, and the fault diagnosis capability is improved. According to the invention, the monitoring precision and reliability of the steam turbine speed regulation system in the variable working condition environment are obviously improved, and powerful support is provided for stable operation and maintenance of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of power generation equipment monitoring, and in particular to a steam turbine speed regulation system monitoring method and related equipment. Background Art

[0002] Currently, common monitoring technologies for steam turbine speed control systems primarily include fixed-threshold monitoring methods and simple fault alarm systems. These existing technologies determine whether the system is operating normally by setting upper and lower thresholds for key operating parameters, such as speed control oil pressure and hydraulic motor displacement. If a parameter exceeds the set threshold, the system triggers an alarm, alerting maintenance personnel to conduct an inspection. Furthermore, some systems include basic fault identification capabilities, enabling preliminary diagnosis of some common faults.

[0003] However, these existing technologies have exposed many deficiencies in practical applications. First, the monitoring method based on fixed thresholds cannot adapt to the complex operating conditions of the steam turbine under variable operating conditions. The operating conditions of the steam turbine will continue to change due to factors such as load changes and steam parameter fluctuations. It is difficult for fixed thresholds to accurately reflect the normal parameter range under different operating conditions, resulting in a high misjudgment rate under variable operating conditions. Secondly, the existing fault alarm system has a low level of intelligence. Although it can trigger an alarm, its ability to accurately identify the type of fault is limited, and it is unable to quickly and accurately locate the root cause of the fault. For example, it is difficult to effectively distinguish specific fault types such as servo valve jamming, oil leakage or sensor failure. These problems have seriously affected the stable operation and operation and maintenance efficiency of the steam turbine speed control system.

[0004] In summary, there is an urgent need for a steam turbine speed control system monitoring solution that can accurately adapt to changing operating conditions, effectively reduce the misjudgment rate, and have the ability to accurately identify faults. Summary of the Invention

[0005] The object of the present invention is to provide a method and related equipment for monitoring a steam turbine speed control system, so as to overcome the disadvantage of the prior art that the steam turbine speed control system has a high misjudgment rate under variable operating conditions.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a method for monitoring a steam turbine speed control system, comprising:

[0008] Real-time collection of multi-source operating parameters of the steam turbine speed control system, including speed control oil pressure, oil motor displacement, servo valve current, oil particle size, control command response delay, and valve action consistency deviation;

[0009] Build a dynamic benchmark model based on historical normal state data, and generate dynamic benchmark values ​​for each operating parameter in combination with real-time operating parameters;

[0010] Calculate the real-time health index based on the deviation between the real-time collected operating parameters and the corresponding dynamic benchmark values;

[0011] When the health index falls below the first threshold, a level one alarm is triggered, and a level two alarm is activated if the level continues to fall beyond the preset time.

[0012] By combining the health index and the combined characteristics of multi-source operating parameters, the fault types such as servo valve jamming, oil circuit leakage or sensor failure can be identified.

[0013] Real-time collection of multi-source operating parameters of the steam turbine speed control system, including speed control oil pressure, oil motor displacement, servo valve current, oil particle size, control command response delay and valve action consistency deviation. The oil particle size is collected online by a laser scattering particle counter with an accuracy of ≤0.1μm. The valve action consistency deviation is calculated by synchronously comparing the phase difference of the displacement curves of at least two control valves.

[0014] Real-time collection of multi-source operating parameters of the turbine speed control system, including speed control oil pressure, oil motor displacement, servo valve current, oil particle size, control command response delay and valve action consistency deviation, as well as obtaining the real-time temperature of the speed control system oil tank and dynamically compensating the speed control oil pressure according to the oil temperature-viscosity relationship curve.

[0015] A dynamic benchmark model is constructed based on historical normal state data, and the dynamic benchmark values ​​of various operating parameters are generated in combination with real-time operating parameters. The construction method of the dynamic benchmark model is as follows:

[0016] A time series neural network model is adopted, with load, main steam pressure and main steam temperature as the working condition input layer, and speed control oil pressure and oil motor displacement as the output layer.

[0017] The time series neural network is an LSTM model with a hidden layer structure of ≥32 hidden units and a sliding time window length of ≥60 sampling periods.

[0018] Combined with the health index and the combined characteristics of multi-source operating parameters, the fault type of servo valve jamming, oil leakage or sensor failure can be identified. Fault type diagnosis includes:

[0019] If the servo valve current rises by more than 10% and the displacement response delay is more than 300ms, the servo valve is diagnosed as stuck;

[0020] If the speed regulating oil pressure drop rate is >5% / min and the oil charge pump starting frequency increases by >50%, it is diagnosed as an oil circuit leakage.

[0021] It also includes generating predictive maintenance instructions when the average monthly decline rate of the health index exceeds a preset degradation rate, where the preset degradation rate is 5%, and the average monthly decline rate is calculated by linear regression fitting the health index trend line of the last 30 days.

[0022] In a second aspect, the present invention provides a steam turbine speed control system monitoring system, comprising:

[0023] The parameter acquisition module is used to collect multi-source operating parameters of the steam turbine speed control system in real time, including speed control oil pressure, oil motor displacement, servo valve current, oil particle size, control command response delay, and valve action consistency deviation;

[0024] Dynamic benchmark module, used to build a dynamic benchmark model based on historical normal state data, and generate dynamic benchmark values ​​for various operating parameters in combination with real-time operating parameters;

[0025] A real-time index calculation module is used to calculate the real-time health index based on the deviation between the real-time collected operating parameters and the corresponding dynamic reference values;

[0026] An alarm response module is used to trigger a first-level alarm when the health index falls below a first threshold, and to activate a second-level alarm when the threshold lasts for more than a preset time;

[0027] The fault identification module is used to identify the fault type of servo valve jamming, oil circuit leakage or sensor failure by combining the health index and the combined characteristics of multi-source operating parameters.

[0028] In a third aspect, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the steam turbine speed control system monitoring method as described above are implemented.

[0029] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, which implements the steps of the steam turbine speed control system monitoring method as described above when executed by a processor.

[0030] Compared with the prior art, the present invention has the following beneficial technical effects:

[0031] In its first aspect, the present invention provides a method for monitoring a steam turbine speed control system. This method collects multiple operating parameters from the turbine speed control system in real time, covering key indicators such as speed control oil pressure, motor displacement, servo valve current, oil particle size, control command response delay, and valve operation consistency deviation. The method also ensures data acquisition accuracy and reliability through the use of a laser scattering particle counter and displacement curve phase difference calculation. For baseline value generation, a dynamic baseline model based on historical normal state data is used, combined with real-time operating parameters, to generate dynamic baseline values ​​for each operating parameter. This model fully considers the operating characteristics of the steam turbine under different operating conditions and accurately reflects the normal state range of the system under the current operating conditions. This effectively addresses the problem of fixed thresholds in the prior art, which are difficult to adapt to changing operating conditions. By calculating the deviation between the real-time collected operating parameters and the dynamic baseline values, and further calculating a real-time health index, a real-time quantitative assessment of the system's health status is achieved, accurately reflecting real-time changes in the system's status and providing a solid basis for subsequent alarm triggering and fault diagnosis. The alarm mechanism design incorporates both primary and secondary alarm mechanisms. This hierarchical alarm mechanism avoids false alarms caused by short-term fluctuations, significantly improving alarm accuracy and reliability. Combining the combined features of health index and multi-source operating parameters, fault diagnosis based on feature combination can quickly locate the root cause of the fault and avoid misjudgment caused by limited fault identification capabilities.

[0032] In the second aspect, the present invention provides a steam turbine speed control system monitoring system. Through the collaborative work of various modules, it realizes the optimization of the entire process from data acquisition, dynamic benchmark generation, health index calculation, alarm response to fault diagnosis. The parameter acquisition module collects the multi-source operating parameters of the speed control system in real time. The comprehensive data collection lays the foundation for subsequent dynamic analysis, ensuring that the system can monitor the operating status of the steam turbine speed control system from multiple dimensions. The dynamic benchmark module can adapt to the changes in operating conditions at different times, reflect the normal range of system operating parameters under different operating conditions, and more accurately match the actual operating conditions of the steam turbine, avoiding the problem of misjudgment caused by changes in operating conditions. The real-time index calculation module calculates the health index based on the deviation, reflects the health status of the system in real time, maintains a stable monitoring effect under dynamic conditions, and avoids frequent false alarms due to changes in operating conditions. The alarm response module effectively reduces false alarms caused by short-term fluctuations or transient anomalies, ensures the accuracy of the alarm signal, and helps operation and maintenance personnel to more accurately judge the system status. The fault identification module uses a comprehensive diagnostic method based on multiple features, which improves the accuracy of fault diagnosis, reduces misjudgments caused by changes in a single parameter, and avoids blind inspections of incorrect fault types by operation and maintenance personnel.

[0033] In a third aspect, the present invention provides a computer device that can efficiently implement the steps of the method of the present invention by executing a specific computer program through a processor. When performing data processing tasks, the computer device can accurately perform numerical calculations and logical judgments, avoiding errors caused by human factors; at the same time, since the computer program has a high degree of stability and reliability, the accuracy and consistency of the data processing results can be ensured.

[0034] In a fourth aspect, the present invention provides a computer-readable storage medium. By programming the steps of the method of the present invention into a computer program and storing it on a computer-readable storage medium, users can easily load these programs onto any compatible computer device and execute them without rewriting or converting the code, thereby greatly improving the convenience and flexibility of program execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 Schematic diagram of a method for monitoring a steam turbine speed control system according to an embodiment of the present invention.

[0036] Figure 2 Schematic diagram of a steam turbine speed control system monitoring system in an embodiment of the present invention. DETAILED DESCRIPTION

[0037] In the field of steam turbine speed control system monitoring, currently common technical approaches include fixed-threshold monitoring methods and simple fault alarm systems. These existing technologies determine whether the system is operating normally by setting upper and lower thresholds on key operating parameters, such as speed control oil pressure and hydraulic motor displacement. If a parameter exceeds the set threshold, the system triggers an alarm, prompting operations and maintenance personnel to conduct an inspection. At the same time, some systems also have basic fault identification capabilities, enabling preliminary diagnosis of some common faults.

[0038] However, these existing technologies have exposed many deficiencies in practical applications. First, the monitoring method based on fixed thresholds cannot adapt to the complex operating conditions of the steam turbine under variable operating conditions. The operating conditions of the steam turbine will continue to change due to factors such as load changes and steam parameter fluctuations. It is difficult for fixed thresholds to accurately reflect the normal parameter range under different operating conditions, resulting in a high misjudgment rate under variable operating conditions. Secondly, the existing fault alarm system has a low level of intelligence. Although it can trigger an alarm, its ability to accurately identify the type of fault is limited, and it is unable to quickly and accurately locate the root cause of the fault. For example, it is difficult to effectively distinguish specific fault types such as servo valve jamming, oil leakage or sensor failure. These problems have seriously affected the stable operation and operation and maintenance efficiency of the steam turbine speed control system.

[0039] In summary, with the continuous development of the electric power industry, the market and the industry have put forward higher requirements for the accuracy and intelligence level of steam turbine speed control system monitoring technology. There is an urgent need for a new steam turbine speed control system monitoring technology that can accurately adapt to changing operating conditions, effectively reduce the misjudgment rate and has accurate fault identification capabilities to meet the industry's new demands for safe, stable and efficient operation of steam turbine speed control systems. Therefore, how to overcome the shortcomings of existing steam turbine speed control system monitoring technology under changing operating conditions has become a key issue that needs to be urgently solved by technical personnel in this field. It is based on this background that the present invention is proposed, and aims to provide a steam turbine speed control system monitoring method and related equipment to overcome the existing technical difficulties.

[0040] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0041] Reference Figure 1 As shown, a specific implementation of the method for monitoring a steam turbine speed control system provided by the present invention includes:

[0042] S1, real-time collection of multi-source operating parameters of the steam turbine speed control system, including speed control oil pressure, oil motor displacement, servo valve current, oil particle size, control command response delay and valve action consistency deviation;

[0043] S2, build a dynamic benchmark model based on historical normal state data, and generate dynamic benchmark values ​​for each operating parameter in combination with real-time operating parameters;

[0044] S3, calculating the real-time health index based on the deviation between the real-time collected operating parameters and the corresponding dynamic baseline values;

[0045] S4: When the health index falls below the first threshold, a first-level alarm is triggered, and a second-level alarm is activated after the duration exceeds the preset time;

[0046] S5, combining the health index and the combined characteristics of multi-source operating parameters to identify the fault type of servo valve jamming, oil circuit leakage or sensor failure.

[0047] Specifically, S1 collects multiple operating parameters of the turbine speed control system in real time, including speed control oil pressure, oil motor displacement, servo valve current, oil particle size, control command response delay, and valve operation consistency deviation. Speed ​​control oil pressure is a key indicator of the proper operation of the turbine speed control system, and changes in oil pressure directly affect the turbine's speed control performance. Oil motor displacement reflects the opening changes of the turbine's speed control valve and is crucial for controlling turbine speed and power. Servo valve current is closely related to the servo valve's operating status. As a key actuator in the speed control system, changes in its current can reflect the servo valve's actuation status and potential faults such as jamming. Oil particle size reflects the cleanliness of the oil in the speed control system. Excessive particulate matter in the oil can cause wear and jamming of precision components such as the servo valve, impacting the reliability of the speed control system. Control command response delay reflects the speed of the speed control system's execution of control commands. Excessive response delay can lead to untimely turbine speed control, affecting stable unit operation. The valve action consistency deviation is used to evaluate whether the actions of the various valves in the turbine are synchronized and coordinated. If the deviation is too large, it may cause problems such as turbine vibration and power fluctuation.

[0048] When collecting oil particle size, a laser scattering particle counter is used for online data collection, achieving an accuracy of ≤0.1μm. A laser scattering particle counter transmits a laser beam through the oil sample. When particulate impurities in the oil scatter the laser light, a detector receives the scattered light signal. Based on the intensity and quantity of the scattered light, the detector calculates the size and distribution of the particles in the oil, thereby achieving precise measurement of oil particle size. This method enables real-time and continuous monitoring of oil quality, promptly identifying oil contamination issues, and providing a critical basis for speed control system maintenance and troubleshooting.

[0049] The calculation of valve action consistency deviation is achieved by synchronously comparing the phase difference of the displacement curves of at least two control valves. During turbine operation, the displacement curves of each control valve should maintain a certain degree of synchronization to ensure smooth operation of the turbine. By installing displacement sensors at locations such as the oil motor in the speed control system, the displacement signals of each control valve are collected in real time. These displacement signals are then synchronously analyzed to calculate the phase difference between them. If the phase difference exceeds the set allowable range, it indicates that there is inconsistency in valve action, which may be caused by mechanical jamming, hydraulic system failure, or abnormal control signal problems, and timely inspection and treatment are required.

[0050] Preferably, in this specific embodiment, this step also includes obtaining the real-time temperature of the oil tank of the speed control system, and dynamically compensating the speed control oil pressure according to the oil temperature-viscosity relationship curve. Temperature changes will affect the viscosity of the oil, and thus affect the speed control oil pressure. By obtaining the real-time temperature of the oil tank and dynamically compensating the speed control oil pressure according to the oil temperature-viscosity relationship curve, the actual operating status of the speed control system can be more accurately reflected, and the accuracy and reliability of monitoring can be improved. The oil temperature-viscosity relationship curve is pre-established based on the characteristics of the oil, and it describes the quantitative relationship between oil temperature and oil viscosity. In actual applications, by measuring the oil tank temperature in real time and combining it with the relationship curve, the viscosity change of the oil at the current oil temperature can be calculated, thereby making corresponding compensation for the speed control oil pressure, eliminating the interference of oil temperature changes on the oil pressure monitoring results, and making the monitoring data more authentic and representative.

[0051] Specifically, in S2, a dynamic baseline model is constructed based on historical normal state data. This data, combined with real-time operating parameters, generates dynamic baseline values ​​for various operating parameters. The first step in building the dynamic baseline model is to collect historical normal state data. This data covers the normal operating parameters of the steam turbine speed control system under different operating conditions, including speed control oil pressure and hydraulic motor displacement. To ensure data accuracy and completeness, the data must be cleaned and preprocessed to remove outliers and noise.

[0052] The dynamic benchmark model of this specific embodiment includes an operating condition input layer and an output layer. The operating condition input layer selects load, main steam pressure, and main steam temperature as input parameters. These parameters comprehensively reflect the operating conditions of the steam turbine and provide rich operating condition information for the model. The output layer selects speed control oil pressure and oil motor displacement as output parameters. These two parameters are core indicators of the steam turbine speed control system and directly reflect the operating status of the speed control system.

[0053] Preferably, the time series neural network in this specific embodiment is a long short-term memory network model (LSTM), which is a special recursive neural network (RNN) that can effectively process and predict long-term dependencies in time series data. In the monitoring of the steam turbine speed control system, the LSTM model is selected because it has the advantages of being able to memorize long-term information and handle complex changes in working conditions. LSTM can selectively retain and update long-term information through its unique gating mechanism (input gate, forget gate, and output gate), avoiding the problem of gradient disappearance or explosion in traditional RNN. The operating conditions of the steam turbine speed control system are complex and changeable. The LSTM model can capture the dynamic characteristics of these changes and adapt to the operating status under different working conditions.

[0054] The number of hidden units in the hidden layer structure of the LSTM model in this specific embodiment is ≥32, which means that the model has sufficient computing power and memory capacity to handle complex operating condition changes and the relationship between operating parameters. More hidden units can enhance the model's expressiveness, enabling it to learn more complex patterns and features. The sliding time window length is ≥60 sampling periods, ensuring that the model can consider historical data for a sufficiently long period of time to more accurately predict current and future operating parameters. A longer time window helps capture trends and cyclical characteristics of parameter changes.

[0055] After model training is complete, real-time operating parameters are input into the dynamic benchmark model, which then outputs dynamic benchmark values ​​for the corresponding operating parameters. These dynamic benchmark values ​​reflect the expected normal values ​​for each operating parameter of the steam turbine speed control system under the current operating conditions. By comparing the real-time collected operating parameters with the dynamic benchmark values, a real-time health index can be calculated, enabling real-time monitoring and assessment of the speed control system's operating status.

[0056] Specifically, S3 calculates a real-time health index based on the deviation between the real-time collected operating parameters and the corresponding dynamic baseline values. The calculation of the real-time health index (HI) is a key step in monitoring the turbine speed control system. It intuitively reflects the current health status of the system by quantifying the deviation between the actual value of the operating parameter and the dynamic baseline value. The actual value is the real-time operating parameter obtained from the sensor, while the baseline value is the ideal value calculated by the dynamic baseline model based on the real-time operating parameters. The deviation is calculated by calculating the absolute difference between the actual value and the baseline value. For multiple operating parameters, these deviations can be comprehensively considered, for example, using the Euclidean distance to measure the total deviation. The threshold is determined based on factors such as historical data, system characteristics, and safety margins, and is used to define the boundaries of the system's health status.

[0057] The formula for calculating the health index is:

[0058]

[0059] This formula limits the range of the health index to [0,1], where HI=1 indicates that the system is in an ideal health state and the actual value is completely consistent with the baseline value; HI=0 indicates that the deviation reaches the maximum range allowed by the threshold and the system may have serious problems; values ​​between 0 and 1 indicate that the system is in the transition zone between normal operation and failure state. The closer the value is to 1, the better the system status.

[0060] In practice, obtaining actual and baseline values ​​is a fundamental step in calculating the health index. Sensors collect the actual values ​​of various operating parameters in real time, while the dynamic baseline model generates corresponding baseline values ​​based on the current operating conditions. The deviation is then calculated and normalized by dividing it by a threshold to obtain a value within the range [0, 1]. Finally, the real-time health index is calculated according to the above formula.

[0061] The significance of the real-time health index lies in providing a visual, quantitative indicator of the turbine speed control system's operating status. It enables real-time monitoring of the system, helping operators understand the system's health status in real time. Furthermore, when the health index approaches or falls below a set threshold, it serves as a fault warning signal, prompting operators to promptly inspect and maintain the system. Furthermore, through long-term monitoring and trend analysis of the health index, system degradation trends can be predicted, allowing maintenance plans to be scheduled in advance and preventing damage to the system from unexpected failures. This approach not only improves system reliability but also enhances operational efficiency and reduces economic losses caused by downtime.

[0062] Specifically, in S4, when the health index falls below the first threshold, a level one alarm is triggered, and a level two alarm is activated after the duration exceeds the preset time; when the real-time health index (HI) falls below the set first threshold, the system will trigger a level one alarm. The first threshold is determined based on the fluctuation range of the health index under normal system operation and operating experience. It represents the starting point where the system may become abnormal. Once the health index reaches this threshold, it indicates that the system operation status may have deviated from the normal range and there is a potential risk of failure. At this point, the level one alarm is activated, reminding the operation and maintenance personnel to pay attention to the system status and conduct preliminary inspection and analysis.

[0063] To avoid unnecessary alarms caused by brief fluctuations in the health index, the present invention also incorporates a secondary alarm mechanism. If the health index remains below the first threshold for a period exceeding a preset duration, the system activates a secondary alarm. This preset duration is determined based on the system's operating characteristics and the patterns of fault development, providing a reasonable waiting time window to distinguish between occasional fluctuations and truly concerning fault conditions. The activation of the secondary alarm indicates that the system's health may have reached a more serious stage, requiring more urgent measures, such as immediate shutdown or more comprehensive inspection and repair of critical components.

[0064] Through this hierarchical alarm mechanism, the present invention effectively balances alarm sensitivity and accuracy. Level 1 alarms promptly signal health index anomalies, allowing operators to conduct preliminary investigations with sufficient lead time. Level 2 alarms, however, provide stronger warnings upon confirmation of a continued deterioration in system health, prompting the implementation of necessary emergency measures. This design avoids wasted resources due to false or frequent alarms while ensuring a timely response when actual system issues arise, safeguarding the safe and stable operation of the turbine speed control system.

[0065] Specifically, S5 combines the health index and the combined characteristics of multi-source operating parameters to identify the fault types of servo valve jamming, oil circuit leakage, or sensor failure. In this specific embodiment, two fault types are diagnosed:

[0066] If the servo valve current increases by >10% and the displacement response delay exceeds 300ms, the servo valve is diagnosed as stuck. A servo valve current increase exceeding 10% indicates that the servo valve requires a higher current to execute control commands, possibly due to mechanical obstruction between the valve core and the valve sleeve, or impurities hindering normal movement. Furthermore, if the hydraulic motor displacement response delay exceeds 300ms, this indicates that the servo valve is delayed in its operation and unable to respond to control commands in a timely manner, further supporting the diagnosis of a stuck servo valve.

[0067] If the speed control oil pressure drops by >5% / min and the charge pump activation frequency increases by >50%, a fuel line leak is diagnosed. If the speed control oil pressure drops by more than 5% per minute, this indicates a rapid and sustained loss of oil pressure within the system, which is often associated with a fuel line leak. Furthermore, if the charge pump activation frequency increases by more than 50%, this indicates that the system is frequently activating the charge pump to replenish oil to maintain normal oil pressure, a phenomenon that strongly suggests a fuel line leak.

[0068] Through this combined feature diagnosis mechanism based on health index and multi-source operating parameters, the present invention achieves accurate identification of key fault types in the turbine speed control system, helping operation and maintenance personnel to quickly locate the root cause of the problem and take effective maintenance measures.

[0069] In the specific implementation of the method for monitoring the steam turbine speed control system provided by the present invention, in order to further improve the reliability and maintenance efficiency of the system, it is preferably also included in the step of generating a predictive maintenance instruction when the average monthly decline rate of the health index exceeds the preset degradation rate, wherein the preset degradation rate is 5%, and the average monthly decline rate is calculated by linear regression fitting the health index trend line of the last 30 days. By monitoring the changing trend of the health index in the past 30 days, the system can identify potential signs of degradation in advance, thereby providing a scientific decision-making basis for preventive maintenance. Specifically, the system will use the linear regression method to fit the health index data of the last 30 days to calculate the average monthly decline rate of the health index. This process not only focuses on the immediate value of the health index, but also looks at its long-term changing trend to obtain more comprehensive system health status information.

[0070] The preset degradation rate set in this invention is 5%. When the calculated monthly average decline rate exceeds this preset value, the system will automatically trigger a predictive maintenance instruction. The core advantage of this mechanism lies in its foresight and proactive nature. It does not rely on obvious fault signals in the system, but instead issues early warnings based on the changing trends of the health index. This strategy can effectively reduce the risk of sudden failures and avoid production interruptions caused by equipment downtime, while optimizing maintenance plans and avoiding unnecessary maintenance work. By promptly addressing potential problems, equipment can operate in a healthier state, thereby extending its service life and reducing overall operating costs.

[0071] In summary, by introducing a predictive maintenance mechanism, the present invention can not only monitor the health status of the system in real time, but also predict and prevent possible degradation risks in advance, providing strong support for the stable operation and efficient maintenance of the turbine speed control system.

[0072] In order to make the steam turbine speed control system monitoring method provided by the present invention easier to understand, an implementation method is provided below in combination with a specific scenario to further illustrate this solution.

[0073] In a thermal power plant, the steam turbine is one of the core equipment, and the stable operation of its speed control system is crucial to ensuring power supply. To achieve accurate monitoring and fault diagnosis of the steam turbine speed control system, the power plant adopted the monitoring method provided by the present invention.

[0074] First, various sensors were installed at key points in the turbine speed control system to collect multi-source operating parameters in real time. These parameters include speed control oil pressure, motor displacement, servo valve current, oil particle size, control command response delay, and valve operation consistency deviation. Oil particle size was collected online using a high-precision laser scattering particle counter with an accuracy of ≤0.1μm. Valve operation consistency deviation was calculated by synchronously comparing the phase difference of the displacement curves of at least two control valves. Furthermore, the real-time temperature of the speed control system's oil tank was acquired, and dynamic compensation of the speed control oil pressure was performed based on the oil temperature-viscosity relationship curve to eliminate interference from oil temperature fluctuations on the monitoring results.

[0075] During the construction of the dynamic benchmark model, the system collected historical operating data of the turbine speed control system under normal conditions. This data includes operating conditions such as load, main steam pressure, and main steam temperature, as well as corresponding parameter values ​​such as speed control oil pressure and motor displacement. A time series neural network model (preferably an LSTM model) was trained using load, main steam pressure, and main steam temperature as the operating condition input layer, and speed control oil pressure and motor displacement as the output layer. In the hidden layer structure of the LSTM model, the number of hidden units was set to ≥32, and the sliding time window length was set to ≥60 sampling periods to ensure that the model fully learned the dynamic characteristics of the historical data.

[0076] During real-time monitoring, the system uses a trained dynamic benchmark model to generate dynamic baseline values ​​for each operating parameter based on the current operating parameters. Simultaneously, the system calculates the deviation between the real-time collected operating parameters and the dynamic baseline values, and uses the formula HI = 1 - ||actual value - baseline value|| / threshold value to derive a real-time health index. When the health index falls below the first threshold, the system immediately triggers a level 1 alarm, alerting on-site maintenance personnel to the system status. If the health index remains below the first threshold for longer than a preset period, the system triggers a level 2 alarm, prompting maintenance personnel to take emergency measures for inspection and repair.

[0077] In terms of fault diagnosis, the system combines the health index and the combined characteristics of multiple operating parameters to identify common fault types such as servo valve jams, oil leakage, and sensor failure. For example, if the servo valve current increases by more than 10% and the displacement response delay exceeds 300ms, the system diagnoses a servo valve jam. If the speed control oil pressure drops by more than 5% / min and the charge pump activation frequency increases by more than 50%, the system diagnoses an oil leakage. By accurately identifying these fault types, operations and maintenance personnel can quickly locate the root cause of the problem and implement targeted repairs.

[0078] To enable predictive maintenance, the system also analyzes long-term health index trends. Using linear regression, it fits the health index trend line over the past 30 days and calculates the average monthly decline rate. If this decline rate exceeds a preset 5% degradation rate, the system generates a predictive maintenance instruction, scheduling maintenance in advance to avoid unexpected equipment failures and reduce downtime and repair costs.

[0079] In this example, the monitoring method of the present invention significantly improved the operational stability of the steam turbine speed control system in this thermal power plant. Over the past six months, two emergency shutdowns potentially caused by oil leaks and one power fluctuation caused by a stuck servo valve have been successfully avoided. Furthermore, through the rational scheduling of predictive maintenance instructions, the average maintenance interval for the equipment has been extended by 15%, and maintenance costs have been reduced by approximately 10%. These achievements fully demonstrate the effectiveness and practicality of the present invention in practical applications.

[0080] By building a dynamic benchmark model and integrating it with real-time operating parameters, this solution accurately generates dynamic benchmark values ​​for each operating parameter, adapting to the complex operating conditions of the steam turbine under variable operating conditions. Compared to traditional monitoring methods based on fixed thresholds, this dynamic benchmark model accounts for the variability of the turbine's operating conditions and more accurately reflects the normal parameter range under different operating conditions, effectively reducing the risk of misjudgment.

[0081] Furthermore, by incorporating a health index calculation and a tiered alarm mechanism, this solution not only monitors the system's operating status in real time but also triggers a level one alarm when the health index falls below a first threshold, and a level two alarm if the level remains below a pre-set threshold for a period of time. This tiered alarm mechanism avoids unnecessary alerts caused by brief health index fluctuations, improving alarm accuracy and reliability.

[0082] Furthermore, by combining the health index with the combined characteristics of multiple operating parameters, this solution can accurately identify fault types such as servo valve jams, oil leakage, or sensor failure, enabling rapid fault location and diagnosis. This intelligent fault diagnosis capability significantly enhances the system's intelligence and avoids misjudgments caused by the limited fault identification capabilities of existing technologies.

[0083] Through the comprehensive application of these technical means, this solution overcomes the shortcomings of existing technologies in the high misjudgment rate of the turbine speed control system status under variable operating conditions, realizes accurate monitoring and fault diagnosis of the turbine speed control system, and ensures the stable operation and operation and maintenance efficiency of the system.

[0084] Reference Figure 2 As shown, another specific embodiment of the present invention provides a steam turbine speed control system monitoring system, comprising:

[0085] The parameter acquisition module is used to collect multi-source operating parameters of the steam turbine speed control system in real time, including speed control oil pressure, oil motor displacement, servo valve current, oil particle size, control command response delay, and valve action consistency deviation;

[0086] Dynamic benchmark module, used to build a dynamic benchmark model based on historical normal state data, and generate dynamic benchmark values ​​for various operating parameters in combination with real-time operating parameters;

[0087] A real-time index calculation module is used to calculate the real-time health index based on the deviation between the real-time collected operating parameters and the corresponding dynamic reference values;

[0088] An alarm response module is used to trigger a first-level alarm when the health index falls below a first threshold, and to activate a second-level alarm when the threshold lasts for more than a preset time;

[0089] The fault identification module is used to identify the fault type of servo valve jamming, oil circuit leakage or sensor failure by combining the health index and the combined characteristics of multi-source operating parameters.

[0090] A computer device is also provided in a specific embodiment of the present invention. Specifically, the computer device includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGAs), or other processors. GateArray, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, and are suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used to collect multi-source operating parameters of the turbine speed control system in real time, including speed control oil pressure, oil motor displacement, servo valve current, oil quality particle size, control instruction response delay and valve action consistency deviation; construct a dynamic benchmark model based on historical normal state data, and generate dynamic benchmark values ​​for each operating parameter in combination with real-time operating condition parameters; calculate a real-time health index based on the deviation between the real-time collected operating parameters and the corresponding dynamic benchmark values; trigger a first-level alarm when the health index is lower than a first threshold, and activate a second-level alarm after the duration exceeds a preset time; identify the fault type of servo valve jamming, oil circuit leakage or sensor failure by combining the health index and the combined characteristics of the multi-source operating parameters.

[0091] A storage medium is also provided in a specific embodiment of the present invention, specifically, a computer-readable storage medium, which is a memory device in a terminal device for storing programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and, of course, the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. One or more instructions stored in a computer-readable storage medium can be loaded and executed by the processor to implement the corresponding steps of the relevant methods in the above embodiments; one or more instructions in the computer-readable storage medium are loaded by the processor and execute the following steps: real-time collection of multi-source operating parameters of the turbine speed control system, including speed control oil pressure, oil motor displacement, servo valve current, oil quality particle size, control instruction response delay and valve action consistency deviation; building a dynamic benchmark model based on historical normal state data, and generating dynamic benchmark values ​​for each operating parameter in combination with real-time operating parameters; calculating a real-time health index based on the deviation between the real-time collected operating parameters and the corresponding dynamic benchmark values; triggering a first-level alarm when the health index is lower than a first threshold, and starting a second-level alarm after the duration exceeds a preset time; identifying the fault type of servo valve jamming, oil circuit leakage or sensor failure by combining the health index and the combined characteristics of the multi-source operating parameters.

[0092] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0093] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0094] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0095] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0096] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, from all points of view, the embodiments should be regarded as illustrative and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes that fall within the meaning and range of equivalents of the claims are included in the present invention. Any reference signs in the claims should not be construed as limiting the claim to which they relate.

[0097] In addition, it should be understood that although this specification describes the embodiments, not every embodiment contains only one independent technical solution. This description is for clarity only. Those skilled in the art should consider the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for the purpose of illustrating the technical concept of the present invention and cannot be used to limit the scope of protection of the present invention. Any changes made based on the technical solution in accordance with the technical concept proposed by the present invention fall within the scope of protection of the claims of the present invention.

Claims

1. A method for monitoring a steam turbine speed control system, characterized in that: include: Real-time collection of multi-source operating parameters of the steam turbine speed control system, including speed control oil pressure, oil motor displacement, servo valve current, oil particle size, control command response delay, and valve action consistency deviation; Build a dynamic benchmark model based on historical normal state data, and generate dynamic benchmark values ​​for each operating parameter in combination with real-time operating parameters; Calculate the real-time health index based on the deviation between the real-time collected operating parameters and the corresponding dynamic benchmark values; When the health index falls below the first threshold, a level one alarm is triggered, and a level two alarm is activated if the level continues to fall beyond the preset time. By combining the health index and the combined characteristics of multi-source operating parameters, the fault types such as servo valve jamming, oil circuit leakage or sensor failure can be identified.

2. A method for monitoring a steam turbine speed control system according to claim 1, characterized in that: The multi-source operating parameters of the steam turbine speed control system are collected in real time, including speed control oil pressure, oil motor displacement, servo valve current, oil particle size, control instruction response delay and valve action consistency deviation, wherein the oil particle size is collected online by a laser scattering particle counter with an accuracy of ≤0.1μm, and the valve action consistency deviation is calculated by synchronously comparing the phase difference of the displacement curves of at least two control valves.

3. A method for monitoring a steam turbine speed control system according to claim 2, characterized in that: The real-time collection of multi-source operating parameters of the steam turbine speed control system includes speed control oil pressure, oil motor displacement, servo valve current, oil quality particle size, control instruction response delay and valve action consistency deviation, and also includes obtaining the real-time temperature of the speed control system oil tank and dynamically compensating the speed control oil pressure according to the oil temperature-viscosity relationship curve.

4. A method for monitoring a steam turbine speed control system according to claim 1, characterized in that: The dynamic benchmark model is constructed based on historical normal state data, and the dynamic benchmark values ​​of various operating parameters are generated in combination with real-time operating parameters. The construction method of the dynamic benchmark model is as follows: A time series neural network model is adopted, with load, main steam pressure and main steam temperature as the working condition input layer, and speed control oil pressure and oil motor displacement as the output layer.

5. A method for monitoring a steam turbine speed control system according to claim 4, characterized in that: The time series neural network is an LSTM model, the number of hidden units in its hidden layer structure is ≥32, and the length of the sliding time window is ≥60 sampling cycles.

6. A method for monitoring a steam turbine speed control system according to claim 1, characterized in that: The combined features of the health index and multi-source operating parameters are used to identify the fault types of servo valve jamming, oil leakage, or sensor failure, wherein the fault type diagnosis includes: If the servo valve current rises by more than 10% and the displacement response delay is more than 300ms, the servo valve is diagnosed as stuck; If the speed regulating oil pressure drop rate is >5% / min and the oil charge pump starting frequency increases by >50%, it is diagnosed as an oil circuit leakage.

7. A method for monitoring a steam turbine speed control system according to claim 1, characterized in that: It also includes generating predictive maintenance instructions when the average monthly decline rate of the health index exceeds a preset degradation rate, where the preset degradation rate is 5%, and the average monthly decline rate is calculated by linear regression fitting the health index trend line of the last 30 days.

8. A steam turbine speed control system monitoring system, characterized in that: include: The parameter acquisition module is used to collect multi-source operating parameters of the steam turbine speed control system in real time, including speed control oil pressure, oil motor displacement, servo valve current, oil particle size, control command response delay, and valve action consistency deviation; Dynamic benchmark module, used to build a dynamic benchmark model based on historical normal state data, and generate dynamic benchmark values ​​for various operating parameters in combination with real-time operating parameters; A real-time index calculation module is used to calculate the real-time health index based on the deviation between the real-time collected operating parameters and the corresponding dynamic reference values; An alarm response module is used to trigger a first-level alarm when the health index falls below a first threshold, and to activate a second-level alarm when the threshold lasts for more than a preset time; The fault identification module is used to identify the fault type of servo valve jamming, oil circuit leakage or sensor failure by combining the health index and the combined characteristics of multi-source operating parameters.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the steam turbine speed control system monitoring method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for monitoring a steam turbine speed control system according to any one of claims 1 to 7 are implemented.

Images