Water erosion resisting method and system for low-pressure cylinder blade

Abstract

The invention provides a low-pressure cylinder blade water erosion resistance method and system, and the method comprises the steps: collecting the multi-dimensional operation parameters of unit load, main steam flow, low-pressure cylinder exhaust steam temperature and low-pressure cylinder exhaust steam pressure, and covering the main influence factors of the water erosion risk; inputting the operation parameters into a pre-constructed machine learning water erosion risk prediction model, outputting a quantized water erosion risk coefficient in advance by learning an association rule of each parameter in historical operation data and a water erosion risk, and when the water erosion risk coefficient is greater than a first risk threshold, generating a risk avoidance instruction set to predict the water erosion risk. Load is improved, steam expansion is optimized, the steam exhaust humidity of a condenser is improved, blade corrosion is reduced, and the problem that turbine low-pressure cylinder blades are prone to water erosion is solved.

Description

Technical Field

[0001] This invention relates to the field of steam turbine technology, and in particular to a method and system for preventing water erosion of low-pressure cylinder blades. Background Technology

[0002] In energy systems such as thermal power generation and nuclear power generation, the steam turbine, as the core power equipment that converts thermal energy into mechanical energy, directly determines the economy and safety of the entire energy system through its operating efficiency and reliability.

[0003] Among them, the low-pressure cylinder blades of the steam turbine are key components for steam expansion and work. They are in the humid steam environment of the exhaust area for a long time and are very susceptible to water erosion due to the impact of liquid water droplets in the steam. This can lead to damage to the blade surface and distortion of the blade profile, which can not only reduce the efficiency of the steam turbine, but also cause serious failures such as blade cracks and fractures, threatening the safe and stable operation of the unit. Summary of the Invention

[0004] This invention provides a water erosion resistant low-pressure cylinder blade to solve the problem of easy water erosion of low-pressure cylinder blades in steam turbines in the prior art.

[0005] On one hand, the present invention provides a method for preventing water erosion of low-pressure cylinder blades, comprising:

[0006] The sensor system deployed on the unit collects operating parameters including unit load, main steam flow, low-pressure cylinder exhaust temperature, and low-pressure cylinder exhaust pressure.

[0007] Input the operating parameters into the pre-built water erosion risk prediction model and output the water erosion risk coefficient; the water erosion prediction model is a prediction model built based on machine learning algorithms;

[0008] When the water erosion risk coefficient is greater than the first risk threshold, a risk avoidance instruction set is generated and executed; the risk avoidance instruction set includes: sending a load adjustment instruction to the load regulation system to increase the unit load within the dispatching permission range, and / or sending an operating parameter adjustment instruction to the circulating water control system to increase the condenser vacuum;

[0009] Within a preset time window after the execution of the risk avoidance instruction set, the operating parameters are re-collected and the water erosion risk coefficient is calculated;

[0010] If the water erosion risk coefficient is greater than the first risk threshold and the low-pressure cylinder exhaust temperature is greater than the preset safe temperature threshold, then a water spray decision logic is generated.

[0011] Optionally, operating parameters, including unit load, main steam flow, low-pressure cylinder exhaust temperature, and low-pressure cylinder exhaust pressure, are collected through a sensor system deployed on the unit.

[0012] The unit load and main steam flow, which characterize the unit's operating status, are obtained through a distributed control system.

[0013] The temperature distribution data of the low-pressure cylinder exhaust is collected by an array of temperature sensors installed on the low-pressure cylinder exhaust pipe.

[0014] The low-pressure cylinder exhaust pressure is collected by a pressure transmitter installed at the throat of the condenser.

[0015] Data quality verification and signal filtering are performed on the unit load, steam flow, low-pressure cylinder exhaust temperature distribution data, and low-pressure cylinder exhaust pressure data to remove abnormal data points and supplement missing values, thereby obtaining preprocessed operating parameters.

[0016] The preprocessed running parameters are timestamped and normalized to obtain the running parameters.

[0017] Optionally, constructing a water erosion risk prediction model includes:

[0018] Acquire historical operating data of the unit under various operating conditions;

[0019] The historical operational data is preprocessed and feature-engineered to generate a training sample set;

[0020] An initial water erosion analysis and prediction model is established, and the gradient boosting decision tree algorithm is used to train the initial water erosion analysis and prediction model with the training sample set to obtain the water erosion risk prediction model.

[0021] Optionally, when the water erosion risk coefficient is greater than a first risk threshold, a risk avoidance instruction set is generated and executed, including:

[0022] Calculate the extent to which the water erosion risk coefficient exceeds the first risk threshold, and calculate the load increase amount through the preset load-risk mapping relationship;

[0023] Obtain the load adjustment margin, constrain the load increase amount within the adjustment margin range, and generate a load adjustment command;

[0024] Based on the deviation between the condenser operating efficiency and the calibrated vacuum degree, the optimal operating parameters of the circulating water system are calculated;

[0025] Based on the optimized operating parameters, an operating parameter adjustment instruction is generated;

[0026] Integrate the load adjustment command and the operating parameter adjustment command to generate a risk avoidance command set;

[0027] The load adjustment command is sent to the load regulation system, and the operating parameter adjustment command is sent to the circulating water control system at the same time.

[0028] Optionally, if the water erosion risk coefficient is greater than a first risk threshold and the low-pressure cylinder exhaust temperature is greater than a preset safe temperature threshold, then a water spray decision logic is generated, including:

[0029] Based on the water erosion risk coefficient and the extent to which the low-pressure cylinder exhaust temperature exceeds the safe temperature threshold, the initial water spray intensity benchmark value is calculated using a fuzzy control algorithm.

[0030] The rate of change of the low-pressure cylinder exhaust temperature is obtained, and combined with the initial water injection intensity reference value, the necessary water injection quantity is dynamically corrected and output through a proportional-integral-derivative control algorithm.

[0031] A water spraying control command is generated based on the required water spraying volume; the water spraying control command specifies that the water spraying cooling system operates in an intermittent pulse mode; wherein the water spraying duration of a single pulse and the pulse interval time are set in reverse correlation based on the water erosion risk coefficient.

[0032] Optionally, after generating the water spray control command based on the required water spray volume, the method further includes:

[0033] Monitor the trend of exhaust steam temperature change after water spraying;

[0034] The spray duration and the interval time are adjusted based on the rate of decrease of the exhaust steam temperature.

[0035] When the exhaust steam temperature drops below the safe threshold and remains stable, gradually reduce the water spray volume until water spraying stops.

[0036] Optional, also includes:

[0037] When the generator unit is detected to be operating in a low-pressure cylinder zero-output state, the first risk threshold is adjusted to the second risk threshold; the second risk threshold is less than the first risk threshold.

[0038] The condenser vacuum degree is added as a weighted input parameter to the water erosion risk prediction model.

[0039] When the water erosion risk coefficient is greater than the second risk threshold, the risk avoidance instruction set and the water spraying decision logic are generated simultaneously.

[0040] Optional, also includes:

[0041] When it is detected that the unit has entered the low-pressure cylinder zero-output operation state and the condenser vacuum is lower than the preset threshold, a blade pretreatment command is generated to perform wear-resistant strengthening pretreatment on the blades.

[0042] Optional, also includes:

[0043] The dynamic structural response signals of the blades are collected by vibration sensor arrays and acoustic emission sensors arranged on the last and second-last stage blades.

[0044] Time-frequency domain analysis was performed on the dynamic response signal of the structure to extract characteristic parameters of water erosion damage to the blades, including vibration mode frequency shift and acoustic emission impact count.

[0045] The feature parameters are input into the blade remaining life prediction model, and the cumulative damage and remaining life of the blade are output.

[0046] When the cumulative damage exceeds a preset damage threshold, the control parameters in the water spraying decision logic are adjusted.

[0047] When the remaining service life is lower than the service life warning threshold, the execution priority of the risk avoidance instruction set is increased.

[0048] On the other hand, the present invention also provides a low-pressure cylinder anti-water erosion system, comprising:

[0049] The data acquisition module is configured to collect operating parameters, including unit load, main steam flow, low-pressure cylinder exhaust temperature, and low-pressure cylinder exhaust pressure, through a sensor system deployed on the unit.

[0050] The risk prediction module, which is communicatively connected to the data acquisition module, is configured to receive the operating parameters and input them into a pre-built water erosion risk prediction model, and output the water erosion risk coefficient; the water erosion risk prediction model is a prediction model built based on machine learning algorithms.

[0051] The risk avoidance control module, which is communicatively connected to the risk prediction module, is configured to generate and execute a risk avoidance instruction set when the water erosion risk coefficient is greater than a first risk threshold. The risk avoidance instruction set includes: sending a load adjustment instruction to the load regulation system to increase the unit load within the dispatching permission range, and / or sending an operating parameter adjustment instruction to the circulating water control system to increase the condenser vacuum.

[0052] The effect evaluation module, which is communicatively connected to the data acquisition module and the risk avoidance control module, is configured to trigger the data acquisition module to re-acquire operating parameters and trigger the risk prediction module to recalculate the water erosion risk coefficient within a preset time window after the execution of the risk avoidance instruction set.

[0053] The water spraying decision module, which is communicatively connected to the effect evaluation module and the data acquisition module, is configured to generate water spraying decision logic if the recalculated water erosion risk coefficient is greater than the first risk threshold and the low-pressure cylinder exhaust temperature is greater than the preset safe temperature threshold.

[0054] On the other hand, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the water erosion prevention method for low-pressure cylinder blades as described above.

[0055] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the water erosion prevention method for low-pressure cylinder blades as described above.

[0056] On the other hand, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the low-pressure cylinder blade anti-water erosion method as described above.

[0057] This invention provides a method and system for preventing water erosion of low-pressure cylinder blades. The method collects multi-dimensional operating parameters such as unit load, main steam flow, low-pressure cylinder exhaust temperature, and low-pressure cylinder exhaust pressure, which can cover the main influencing factors of water erosion risk. The operating parameters are then input into a pre-constructed machine learning water erosion risk prediction model. By learning the correlation between each parameter and water erosion risk in historical operating data, a quantified water erosion risk coefficient is output in advance. When the water erosion risk coefficient is greater than a first risk threshold, a risk avoidance instruction set is generated to improve load, optimize steam expansion, increase condenser vacuum, reduce exhaust steam humidity, and reduce blade corrosion, thereby solving the problem of easy water erosion of turbine low-pressure cylinder blades. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0059] Figure 1 This is a schematic diagram of the water erosion prevention method for low-pressure cylinder blades provided in an embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of the anti-water erosion system for low-pressure cylinder blades provided in an embodiment of the present invention;

[0061] Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this 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 this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0063] Figure 1 This is a schematic diagram of the water erosion prevention method for low-pressure cylinder blades provided in an embodiment of the present invention.

[0064] like Figure 1 As shown, the water erosion prevention method for low-pressure cylinder blades provided in this embodiment of the invention mainly includes the following steps:

[0065] 101. Through the sensor system deployed on the unit, operating parameters including unit load, main steam flow, low-pressure cylinder exhaust temperature and low-pressure cylinder exhaust pressure are collected.

[0066] The sensor system includes various types of sensors, such as load sensors for accurately measuring unit load, steam flow sensors for acquiring main steam flow data, temperature sensors for monitoring low-pressure cylinder exhaust temperature, and pressure sensors for collecting low-pressure cylinder exhaust pressure information. The sensors work together to ensure that the collected operating parameters are comprehensive and accurate.

[0067] Specifically, through a sensor system deployed on the unit, operating parameters including unit load, main steam flow, low-pressure cylinder exhaust temperature, and low-pressure cylinder exhaust pressure are collected, including:

[0068] The unit load and main steam flow, which characterize the unit's operating status, are obtained through a distributed control system.

[0069] Temperature distribution data of low-pressure cylinder exhaust is collected by an array of temperature sensors installed on the low-pressure cylinder exhaust pipe.

[0070] The low-pressure cylinder exhaust pressure is collected by a pressure transmitter installed at the throat of the condenser.

[0071] Data quality verification and signal filtering are performed on the unit load, steam flow, low-pressure cylinder exhaust temperature distribution data, and low-pressure cylinder exhaust pressure. Abnormal data points are removed and missing values ​​are added to obtain preprocessed operating parameters.

[0072] The preprocessed runtime parameters are timestamped and normalized to obtain the runtime parameters.

[0073] Specifically, when collecting unit load, main steam flow, low-pressure cylinder exhaust temperature, and low-pressure cylinder exhaust pressure through the sensor system deployed on the unit, the unit load and main steam flow data, which characterize the unit's operating conditions, are first directly collected through the unit's distributed control system. Secondly, temperature distribution data at the low-pressure cylinder exhaust port is collected in real time through a temperature sensor array evenly distributed along the circumference of the low-pressure cylinder exhaust pipe. Simultaneously, the low-pressure cylinder exhaust pressure is collected through a pressure transmitter located at the condenser throat and connected to the low-pressure cylinder exhaust port.

[0074] Subsequently, the aforementioned multi-dimensional raw data underwent data quality verification and signal filtering. For example, an outlier detection algorithm based on the 3σ criterion was used to remove outlier data points exceeding the preset confidence interval, and missing data values ​​were supplemented using linear interpolation to obtain preprocessed operating parameters. Finally, the preprocessed operating parameters were timestamped to ensure a one-to-one correspondence between parameters of each dimension at the same time node. The Min-Max normalization method was used to uniformly map parameters of different physical quantities to the [0,1] interval, ultimately forming standardized operating parameters that meet the input format requirements of the water erosion risk prediction model.

[0075] 102. Input the running parameters into the pre-built water erosion risk prediction model and output the water erosion risk coefficient.

[0076] The water erosion risk prediction model is built based on machine learning algorithms. Trained from a large amount of historical operating data, the model accurately captures the complex nonlinear relationship between unit operating parameters and water erosion risk. By inputting standardized operating parameters in real time, the model can quickly output the water erosion risk coefficient under the current operating conditions, providing operators with intuitive risk assessment results.

[0077] The construction of a water erosion risk prediction model includes:

[0078] Acquire historical operating data of the unit under various operating conditions;

[0079] Historical operational data is preprocessed and feature-engineered to generate a training sample set;

[0080] An initial water erosion analysis and prediction model was established. The gradient boosting decision tree algorithm was used to train the initial water erosion analysis and prediction model with a training sample set to obtain a water erosion risk prediction model.

[0081] Specifically, when constructing a water erosion risk prediction model, it is first necessary to collect historical operating data of the unit under various operating conditions, including rated load, low load, variable load, and extreme operating conditions. This historical operating data includes multi-dimensional operating parameters related to water erosion risk, such as unit load, main steam flow, low-pressure cylinder exhaust temperature, and low-pressure cylinder exhaust pressure. It also includes data on the degree of water erosion damage to the low-pressure cylinder blades under the corresponding operating conditions. This water erosion damage detection data includes the blade surface wear depth and the percentage of corrosion area. Then, the water erosion damage detection data is converted into water erosion risk labels, with values ​​ranging from [0,1]. A higher water erosion risk value indicates a higher risk.

[0082] Subsequently, the historical operating data was preprocessed in accordance with the operating parameters. Then, Pearson correlation analysis was used to screen out key features that were strongly correlated with the water erosion risk label, and redundant features were removed to reduce the computational complexity of the model. At the same time, the key features were standardized, and finally a training sample set with feature vectors corresponding one-to-one with the water erosion risk label was generated.

[0083] After obtaining the training sample set, an initial water erosion analysis and prediction model based on a lightweight gradient boosting decision tree algorithm is established. Hyperparameters including learning rate, maximum tree depth, and number of iterations are set. The learning rate is set to 0.05, the maximum tree depth is set to 8, and the number of iterations is set to 100. The initial model is trained using the training sample set with a 5-fold cross-validation method. The model parameters are optimized by minimizing the mean squared error loss function. Finally, a water erosion risk prediction model with quantitative prediction capability is obtained, and its output is a quantified water erosion risk coefficient. The value range is [0,1].

[0084] After the water erosion risk prediction model is built, the operating parameters are fed into the model. The water erosion risk coefficient output by the model reflects the degree of water erosion risk to the low-pressure cylinder blades under the current operating conditions. Based on the specific value of the water erosion risk coefficient, it can be divided into different risk levels, such as low risk (0-0.3), medium risk (0.3-0.7), and high risk (0.7-1). When the risk level is low, the current operating parameters can be maintained, but regular monitoring is required. If the risk level is medium, the operating parameters need to be adjusted appropriately, such as reducing steam flow and increasing blade surface temperature, to reduce the risk of water erosion. When the risk level is high, operation must be stopped immediately, the blades thoroughly inspected, and the blades replaced if necessary, to prevent severe water erosion from damaging the blades and affecting the normal operation of the entire low-pressure cylinder and even the entire unit.

[0085] 103. When the water erosion risk coefficient is greater than the first risk threshold, generate and execute a risk avoidance instruction set.

[0086] The risk avoidance instruction set includes sending load adjustment instructions to the load regulation system to increase the unit load within the dispatching allowance, and / or sending operating parameter adjustment instructions to the circulating water control system to increase the condenser vacuum. By increasing the unit load and / or increasing the condenser vacuum, the work capacity of steam in the low-pressure cylinder can be increased, and the humidity of steam in the low-pressure cylinder can be reduced, thereby reducing the water erosion effect of steam on the blades.

[0087] Specifically, when the water erosion risk coefficient exceeds the first risk threshold, a risk avoidance instruction set is generated and executed, including:

[0088] Calculate the extent to which the water erosion risk coefficient exceeds the first risk threshold, and calculate the load increase amount through the preset load-risk mapping relationship;

[0089] Obtain the load adjustment margin, constrain the load increase within the adjustment margin range, and generate a load adjustment command;

[0090] Based on the deviation between the condenser operating efficiency and the calibrated vacuum degree, the optimal operating parameters of the circulating water system are calculated;

[0091] Based on optimized operating parameters, generate operating parameter adjustment instructions;

[0092] Integrate load adjustment commands and operating parameter adjustment commands to generate a risk avoidance command set;

[0093] It sends load adjustment commands to the load regulation system and operating parameter adjustment commands to the circulating water control system.

[0094] When the water erosion risk coefficient output by the water erosion risk prediction model is greater than the preset first risk threshold, the extent by which the water erosion risk coefficient exceeds the first risk threshold is first calculated. Based on the preset load-risk mapping relationship obtained by fitting the historical operating data of the unit, that is, the positive correlation between the risk excess and the load increase, the required load increase is calculated.

[0095] Subsequently, the load adjustment margin within the current dispatch allowable range is obtained through the unit's distributed control system. This margin is the difference between the unit's current actual load and the maximum allowable load. The calculated load increase is then constrained within the load adjustment margin. If the calculated load increase exceeds the load adjustment margin, the maximum value of the load adjustment margin is taken, and a load adjustment command that meets the dispatch requirements is generated.

[0096] Meanwhile, the condenser operation efficiency data is obtained through the condenser monitoring module, the deviation between the actual vacuum degree and the design calibration vacuum degree is calculated, and the optimized operating parameters of the circulating water system that can make the condenser vacuum degree approach the calibration value are derived. The optimized operating parameters include the circulating water pump speed, circulating water flow rate and cooling tower fan frequency.

[0097] After obtaining the optimized operating parameters, operating parameter adjustment instructions for the circulating water control system are generated based on the optimized operating parameters; then, the load adjustment instructions verified by constraints are integrated with the operating parameter adjustment instructions corresponding to the optimized operating parameters of the circulating water system to form a complete risk avoidance instruction set.

[0098] Finally, the load adjustment command is sent to the unit load regulation system, and the operating parameter adjustment command is sent to the circulating water control system. The low-pressure cylinder operating conditions are optimized through the coordinated regulation of the two systems to reduce the risk of water erosion.

[0099] 104. Within the preset time window after the risk avoidance instruction set is executed, re-collect operating parameters and calculate the water erosion risk coefficient;

[0100] Among them, re-collecting operating parameters and calculating the water erosion risk coefficient can compare and analyze the calculated water erosion risk coefficient with the preset safety threshold. If the water erosion risk coefficient is still higher than the safety threshold, the relationship between the water erosion risk coefficient and the first risk threshold can be further determined so as to take corresponding adjustment measures.

[0101] 105. If the water erosion risk coefficient is greater than the first risk threshold and the low-pressure cylinder exhaust temperature is greater than the preset safe temperature threshold, then the water spray decision logic is generated.

[0102] Specifically, by generating water spraying decision logic, the required water spraying volume and duration can be calculated based on the current water erosion risk coefficient and low-pressure cylinder exhaust temperature, ensuring that water erosion risk is reduced while avoiding other operational problems caused by excessive water spraying. Simultaneously, the water spraying decision logic must also consider the operating status of the circulating water system to ensure that the water spraying operation does not adversely affect the stability of the circulating water system.

[0103] Specifically, if the water erosion risk coefficient is greater than the first risk threshold and the low-pressure cylinder exhaust temperature is greater than the preset safe temperature threshold, then a water spray decision logic is generated, including:

[0104] Based on the water erosion risk coefficient and the extent to which the low-pressure cylinder exhaust temperature exceeds the safe temperature threshold, the initial water spray intensity benchmark value is calculated using a fuzzy control algorithm.

[0105] The rate of change of the low-pressure cylinder exhaust temperature is obtained, and combined with the initial water injection intensity reference value, the necessary water injection quantity is dynamically corrected and output through the proportional-integral-derivative control algorithm.

[0106] Generate water spray control commands based on the necessary water volume.

[0107] The water spray control command specifies that the water spray cooling system operates in an intermittent pulse mode; the duration of a single pulse and the pulse interval are set in reverse correlation based on the water erosion risk coefficient.

[0108] Among these, water erosion risk factors exceeding the first risk threshold and low-pressure cylinder exhaust temperature exceeding the preset safe temperature threshold often occur under extreme operating conditions of the unit or in abnormal scenarios of critical auxiliary systems. For example, when the unit encounters an extreme variable load condition of rapid load recovery after a sudden load drop, the main steam flow fluctuates drastically in a short period of time, causing the steam expansion process in the low-pressure cylinder to become disordered. This results in both an increased proportion of wet steam causing the water erosion risk factor to exceed the standard and a sudden change in steam flow rate causing an abnormal rise in exhaust temperature. Alternatively, when the condenser cooling system experiences malfunctions such as partial blockage of the cooling tower packing or insufficient output of the circulating water pump, the condenser vacuum decreases, leading not only to an increase in low-pressure cylinder exhaust pressure and exhaust humidity but also to exhaust temperature exceeding the safe range due to insufficient cooling of the exhaust.

[0109] When the water erosion risk coefficient exceeds the first risk threshold and the low-pressure cylinder exhaust temperature exceeds the preset safety temperature threshold, the initial water spray intensity benchmark value is calculated first based on the extent to which the water erosion risk coefficient exceeds the first risk threshold and the extent to which the low-pressure cylinder exhaust temperature exceeds the safety temperature threshold, using a preset fuzzy control algorithm.

[0110] Subsequently, by monitoring the rate of change of the low-pressure cylinder exhaust temperature and incorporating the rate of change of temperature as a feedback parameter into the proportional-integral-derivative control algorithm, if the temperature continues to rise, the water injection volume is increased proportionally; if the temperature begins to fall, the water injection volume is smoothly adjusted through the integral term to avoid sudden changes in the water injection volume that could cause fluctuations in the operating conditions. Finally, the necessary water injection volume is output to adapt to the current operating conditions.

[0111] The water spray control command is generated based on the necessary water spray volume. The water spray control command explicitly specifies that the water spray desuperheating system operates in intermittent pulse mode. The water spray duration of a single pulse and the pulse interval are set inversely based on the water erosion risk coefficient. That is, the higher the water erosion risk coefficient, the longer the water spray duration of a single pulse and the shorter the pulse interval, and vice versa. Pulsed water spray ensures precise control of steam humidity and avoids sudden changes in low-pressure cylinder pressure caused by continuous water spray, thus ensuring the stability of unit operation.

[0112] In some embodiments, after generating the water spray control command according to the necessary water spray volume, the method further includes:

[0113] Monitor the trend of exhaust steam temperature change after water spraying;

[0114] Adjust the spray duration and interval based on the rate of decrease in exhaust steam temperature;

[0115] When the exhaust steam temperature drops below the safe threshold and remains stable, gradually reduce the water spray volume until water spraying stops.

[0116] Specifically, after generating water spray control commands based on the necessary water spray volume and controlling the water spray desuperheating system to operate in intermittent pulse mode, the system also continuously monitors the temperature change trend of the low-pressure cylinder exhaust after water spraying through a temperature sensor array arranged circumferentially along the low-pressure cylinder exhaust pipe.

[0117] The water spray parameters are dynamically adjusted based on the collected exhaust steam temperature drop rate. For example, a temperature drop of 3°C per minute is considered a rapid drop, a drop of 1-2°C is considered a gradual drop, and a drop of less than 1°C is considered a slow drop. If the drop rate is fast, it indicates that the current water spray intensity is suitable for the operating conditions, and the duration of a single pulse and the pulse interval can be appropriately shortened to avoid excessive water spraying causing a sudden drop in exhaust steam temperature. If the drop rate is gradual, the current water spray duration and interval remain unchanged. If the drop rate is slow, it indicates that the water spray intensity is insufficient, and the water spray duration needs to be extended and the pulse interval shortened to improve humidity control efficiency. When the exhaust steam temperature is monitored to drop below the preset safe temperature threshold and continues to operate stably for 3-5 minutes, the necessary water spray volume is gradually reduced by 10%-20% each time, and the pulse water spray duration and interval are adjusted simultaneously until the water spray volume drops to the minimum threshold and then water spraying is stopped. This ensures that the risk of water erosion is completely eliminated and avoids operating condition fluctuations caused by sudden stoppage of spraying, thus ensuring the stability of unit operation.

[0118] In some embodiments, the low-pressure cylinder blade anti-water erosion method provided by the present invention further includes:

[0119] When the generator unit is detected to be operating in a low-pressure cylinder zero-output state, the first risk threshold is adjusted to the second risk threshold. The second risk threshold is lower than the first risk threshold.

[0120] Add condenser vacuum degree as a weighted input parameter to the water erosion risk prediction model.

[0121] When the water erosion risk coefficient is greater than the second risk threshold, a risk avoidance instruction set and a water spraying decision logic are generated simultaneously.

[0122] Specifically, when the distributed control system of the unit identifies that the unit has entered a zero-output operation state of the low-pressure cylinder due to factors such as increasing the amount of steam extracted for heating and the demand for peak shaving of the power grid, only a small amount of cooling steam is introduced into the low-pressure cylinder at this time. The steam flow rate drops sharply, the expansion process is unstable, and the fluctuation range of exhaust steam humidity is significantly greater than that under normal operating conditions.

[0123] Since the risk of blade water erosion is more sensitive to parameter changes when the low-pressure cylinder is operating at zero output, the first risk threshold is adjusted to a second risk threshold with a smaller value, so as to achieve early warning of risk by reducing the threshold.

[0124] Meanwhile, since the influence of condenser vacuum on exhaust humidity increases when the low-pressure cylinder is at zero output, adding condenser vacuum as a weighted input parameter to the water erosion risk prediction model can enhance the model's accuracy in capturing risk changes under the zero-output operating state of the low-pressure cylinder. When the risk coefficient output by the water erosion risk prediction model is greater than the second risk threshold, the step-by-step protection of first avoiding the command and then spraying water is no longer executed. Instead, a risk avoidance command set and water spraying decision logic are generated simultaneously. On the one hand, load fine-tuning commands are sent to the load regulation system and vacuum optimization commands are sent to the circulating water control system. On the other hand, pulse water spraying control commands are generated according to the fuzzy PID algorithm. Through the coordinated action of active operating condition regulation and precise water spraying protection, the risk development is quickly suppressed.

[0125] In some embodiments, the low-pressure cylinder blade anti-water erosion method provided by the present invention further includes:

[0126] When it is detected that the unit has entered the low-pressure cylinder zero-output operation state and the condenser vacuum is lower than the preset threshold, a blade pretreatment command is generated to perform wear-resistant strengthening pretreatment on the blades.

[0127] Specifically, when the distributed control system of the unit detects that the unit has entered a zero-output operation state of the low-pressure cylinder, and the pressure transmitter at the condenser throat detects that the condenser vacuum is lower than the preset vacuum threshold (for example, the preset vacuum threshold is 90 kPa), the condenser vacuum is lower than 90 kPa, and only a small amount of cooling steam is introduced into the low-pressure cylinder. The steam flow rate is extremely low and the humidity is easily saturated. Coupled with the increase in exhaust pressure caused by insufficient condenser vacuum, the high-risk areas of blade water erosion will be directly subjected to the continuous impact of high-humidity steam. The response speed of conventional dynamic protection measures is difficult to match the rate of risk escalation. Therefore, a blade pretreatment command is generated to perform wear-resistant strengthening pretreatment on the blades.

[0128] Specifically, the blade pretreatment instructions involve strengthening the areas of the blade prone to water erosion. For example, plasma spraying technology can be used to quickly apply a 0.1-0.3 mm thick WC-CoCr alloy coating to the blade tip and leading edge surface, or a nickel-based wear-resistant alloy layer can be clad onto key parts of the blade's back arc using laser cladding. By improving the blade's surface hardness and wear resistance, a physical protective barrier can be constructed.

[0129] In some embodiments, the low-pressure cylinder blade anti-water erosion method provided by the present invention further includes:

[0130] The dynamic structural response signals of the blades are collected by an array of vibration sensors and acoustic emission sensors arranged on the last and second-to-last stage blades.

[0131] The system utilizes an array of vibration sensors spaced along the height of the final and second-to-last stage blades, along with acoustic emission sensors installed at points prone to water erosion on the leading edge of the blades, to collect dynamic structural response signals during blade operation. These dynamic response signals include vibration acceleration signals and acoustic emission pulse signals.

[0132] Time-frequency domain analysis was performed on the dynamic response signal of the structure to extract characteristic parameters of water erosion damage to the blades.

[0133] Among them, wavelet packet transform can be used to denoise and decompose the vibration signal to perform time-frequency domain analysis on the collected structural dynamic response signal, extract the frequency offset of the first to third natural vibration modes of the blade, and count the acoustic emission impacts per unit time through the acoustic emission signal threshold triggering algorithm. The frequency offset and the acoustic emission impact count are used together as characteristic parameters to characterize the water erosion damage of the blade.

[0134] The feature parameters are input into the blade remaining life prediction model, which outputs the cumulative damage and remaining life of the blade.

[0135] The extracted feature parameters are input into a pre-trained blade remaining life prediction model. This model is built based on a long short-term memory network combined with Miner's linear cumulative damage theory. By learning historical damage and lifespan correlation data, the model outputs the current cumulative damage level and remaining lifespan of the blade.

[0136] When the cumulative damage exceeds the preset damage threshold, adjust the control parameters in the water spray decision logic.

[0137] When the remaining service life is lower than the service life warning threshold, the execution priority of the risk avoidance instruction set is increased.

[0138] When the cumulative damage exceeds the preset damage threshold, the control parameters in the water spray decision logic are automatically adjusted. For example, if the preset damage threshold is 0.6, the initial water spray intensity benchmark value is increased by 10%-20%, and the pulse interval time is shortened by 3-5 seconds to reduce subsequent damage accumulation by strengthening water spray protection.

[0139] When the remaining service life is lower than the preset service life warning threshold, the execution priority of the risk avoidance instruction set is automatically increased. For example, when the remaining service life is less than 500 hours, the load adjustment instruction is responded to first within the scheduling permission range, and the circulating water control system performs vacuum optimization operation first to avoid the failure from being subjected to high water erosion risk due to insufficient remaining blade service life.

[0140] Based on the same general inventive concept, this invention also protects a water erosion prevention system for low-pressure cylinder blades. The water erosion prevention system for low-pressure cylinder blades provided by this invention will be described below. The water erosion prevention system for low-pressure cylinder blades described below can be referred to in correspondence with the water erosion prevention method for low-pressure cylinder blades described above.

[0141] like Figure 2 As shown, this embodiment of the invention also provides a low-pressure cylinder anti-water erosion system, comprising:

[0142] The data acquisition module 210 is configured to acquire operating parameters, including unit load, main steam flow, low-pressure cylinder exhaust temperature and low-pressure cylinder exhaust pressure, through a sensor system deployed on the unit.

[0143] The risk prediction module 220 is communicatively connected to the data acquisition module 210 and is configured to receive operating parameters and input them into a pre-built water erosion risk prediction model, and output water erosion risk coefficients; the water erosion risk prediction model is a prediction model built based on machine learning algorithms.

[0144] The risk avoidance control module 230, which is communicatively connected to the risk prediction module 220, is configured to generate and execute a risk avoidance instruction set when the water erosion risk coefficient is greater than a first risk threshold. The risk avoidance instruction set includes: sending a load adjustment instruction to the load regulation system to increase the unit load within the dispatching permission range, and / or sending an operating parameter adjustment instruction to the circulating water control system to increase the condenser vacuum.

[0145] The effect evaluation module 240 is communicatively connected to the data acquisition module 210 and the risk avoidance control module 230. It is configured to trigger the data acquisition module 210 to re-acquire the operating parameters and trigger the risk prediction module 220 to recalculate the water erosion risk coefficient within a preset time window after the execution of the risk avoidance instruction set.

[0146] The water spray decision module 250 is communicatively connected to the effect evaluation module 240 and the data acquisition module 210. It is configured to generate water spray decision logic if the recalculated water erosion risk coefficient is greater than the first risk threshold and the low-pressure cylinder exhaust temperature is greater than the preset safe temperature threshold.

[0147] Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention.

[0148] like Figure 3As shown, the electronic device may include a processor 310, a communication interface 320, a memory 330, and a communication bus 340. The processor 310, communication interface 320, and memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions from the memory 330 to execute a water erosion prevention method for low-pressure cylinder blades.

[0149] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0150] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the water erosion prevention method for low-pressure cylinder blades provided by the above methods.

[0151] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the water erosion prevention method for low-pressure cylinder blades provided by the methods described above.

[0152] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0153] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0154] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for preventing water erosion of low-pressure cylinder blades, characterized in that, include: The sensor system deployed on the unit collects operating parameters including unit load, main steam flow, low-pressure cylinder exhaust temperature, and low-pressure cylinder exhaust pressure. Input the operating parameters into the pre-built water erosion risk prediction model and output the water erosion risk coefficient; The water erosion risk prediction model is a prediction model built based on machine learning algorithms; When the water erosion risk coefficient is greater than the first risk threshold, a risk avoidance instruction set is generated and executed; the risk avoidance instruction set includes: sending a load adjustment instruction to the load regulation system to increase the unit load within the dispatching permission range, and / or sending an operating parameter adjustment instruction to the circulating water control system to increase the condenser vacuum; Within a preset time window after the execution of the risk avoidance instruction set, the operating parameters are re-collected and the water erosion risk coefficient is calculated; If the water erosion risk coefficient is greater than the first risk threshold and the low-pressure cylinder exhaust temperature is greater than the preset safe temperature threshold, then a water spray decision logic is generated.

2. The method for preventing water erosion of low-pressure cylinder blades according to claim 1, characterized in that, The unit's sensor system collects operating parameters including unit load, main steam flow, low-pressure cylinder exhaust temperature, and low-pressure cylinder exhaust pressure. The unit load and main steam flow, which characterize the unit's operating status, are obtained through a distributed control system. The temperature distribution data of the low-pressure cylinder exhaust is collected by an array of temperature sensors installed on the low-pressure cylinder exhaust pipe. The low-pressure cylinder exhaust pressure is collected by a pressure transmitter installed at the throat of the condenser. Data quality verification and signal filtering are performed on the unit load, steam flow, low-pressure cylinder exhaust temperature distribution data, and low-pressure cylinder exhaust pressure data to remove abnormal data points and supplement missing values, thereby obtaining preprocessed operating parameters. The preprocessed running parameters are timestamped and normalized to obtain the running parameters.

3. The method for preventing water erosion of low-pressure cylinder blades according to claim 1, characterized in that, The construction of a water erosion risk prediction model includes: Acquire historical operating data of the unit under various operating conditions; The historical operational data is preprocessed and feature-engineered to generate a training sample set; An initial water erosion analysis and prediction model is established, and the gradient boosting decision tree algorithm is used to train the initial water erosion analysis and prediction model with the training sample set to obtain the water erosion risk prediction model.

4. The method for preventing water erosion of low-pressure cylinder blades according to claim 1, characterized in that, When the water erosion risk coefficient is greater than the first risk threshold, a risk avoidance instruction set is generated and executed, including: Calculate the extent to which the water erosion risk coefficient exceeds the first risk threshold, and calculate the load increase amount through the preset load-risk mapping relationship; Obtain the load adjustment margin, constrain the load increase amount within the adjustment margin range, and generate a load adjustment command; Based on the deviation between the condenser operating efficiency and the calibrated vacuum degree, the optimal operating parameters of the circulating water system are calculated; Based on the optimized operating parameters, an operating parameter adjustment instruction is generated; Integrate the load adjustment command and the operating parameter adjustment command to generate a risk avoidance command set; The load adjustment command is sent to the load regulation system, and the operating parameter adjustment command is sent to the circulating water control system at the same time.

5. The method for preventing water erosion of low-pressure cylinder blades according to claim 1, characterized in that, If the water erosion risk coefficient is greater than the first risk threshold and the low-pressure cylinder exhaust temperature is greater than the preset safe temperature threshold, then a water spray decision logic is generated, including: Based on the water erosion risk coefficient and the extent to which the low-pressure cylinder exhaust temperature exceeds the safe temperature threshold, the initial water spray intensity benchmark value is calculated using a fuzzy control algorithm. The rate of change of the low-pressure cylinder exhaust temperature is obtained, and combined with the initial water injection intensity reference value, the necessary water injection quantity is dynamically corrected and output through the proportional-integral-derivative control algorithm. A water spraying control command is generated based on the required water spraying volume; the water spraying control command specifies that the water spraying cooling system operates in an intermittent pulse mode; wherein the water spraying duration of a single pulse and the pulse interval time are set in reverse correlation based on the water erosion risk coefficient.

6. The method for preventing water erosion of low-pressure cylinder blades according to claim 5, characterized in that, After generating the water spray control command based on the required water spray volume, the method further includes: Monitor the trend of exhaust steam temperature change after water spraying; The spray duration and the interval time are adjusted based on the rate of decrease of the exhaust steam temperature. When the exhaust steam temperature drops below the safe threshold and remains stable, gradually reduce the water spray volume until water spraying stops.

7. The method for preventing water erosion of low-pressure cylinder blades according to claim 1, characterized in that, Also includes: When the generator unit is detected to be operating in a low-pressure cylinder zero-output state, the first risk threshold is adjusted to the second risk threshold. The second risk threshold is less than the first risk threshold; The condenser vacuum degree is added as a weighted input parameter to the water erosion risk prediction model. When the water erosion risk coefficient is greater than the second risk threshold, the risk avoidance instruction set and the water spraying decision logic are generated simultaneously.

8. The method for preventing water erosion of low-pressure cylinder blades according to claim 7, characterized in that, Also includes: When it is detected that the unit has entered the low-pressure cylinder zero-output operation state and the condenser vacuum is lower than the preset threshold, a blade pretreatment command is generated to perform wear-resistant strengthening pretreatment on the blades.

9. The method for preventing water erosion of low-pressure cylinder blades according to claim 1, characterized in that, Also includes: The dynamic structural response signals of the blades are collected by vibration sensor arrays and acoustic emission sensors arranged on the last and second-last stage blades. Time-frequency domain analysis was performed on the dynamic response signal of the structure to extract characteristic parameters of water erosion damage to the blades, including vibration mode frequency shift and acoustic emission impact count. The feature parameters are input into the blade remaining life prediction model, and the cumulative damage and remaining life of the blade are output. When the cumulative damage exceeds a preset damage threshold, the control parameters in the water spraying decision logic are adjusted. When the remaining service life is lower than the service life warning threshold, the execution priority of the risk avoidance instruction set is increased.

10. A low-pressure cylinder blade anti-water erosion system, characterized in that, include: The data acquisition module is configured to collect operating parameters, including unit load, main steam flow, low-pressure cylinder exhaust temperature, and low-pressure cylinder exhaust pressure, through a sensor system deployed on the unit. The risk prediction module, which is communicatively connected to the data acquisition module, is configured to receive the operating parameters and input them into a pre-built water erosion risk prediction model, and output the water erosion risk coefficient; the water erosion risk prediction model is a prediction model built based on machine learning algorithms. The risk avoidance control module, which is communicatively connected to the risk prediction module, is configured to generate and execute a risk avoidance instruction set when the water erosion risk coefficient is greater than a first risk threshold. The risk avoidance instruction set includes: sending a load adjustment instruction to the load regulation system to increase the unit load within the dispatching permission range, and / or sending an operating parameter adjustment instruction to the circulating water control system to increase the condenser vacuum. The effect evaluation module, which is communicatively connected to the data acquisition module and the risk avoidance control module, is configured to trigger the data acquisition module to re-acquire operating parameters and trigger the risk prediction module to recalculate the water erosion risk coefficient within a preset time window after the execution of the risk avoidance instruction set. The water spraying decision module, which is communicatively connected to the effect evaluation module and the data acquisition module, is configured to generate water spraying decision logic if the recalculated water erosion risk coefficient is greater than the first risk threshold and the low-pressure cylinder exhaust temperature is greater than the preset safe temperature threshold.

Images