Water-hydrogen-hydrogen turbine generator rotor overheating early warning method and equipment

Abstract

The present invention provides a water-hydrogen-hydrogen steam turbine generator rotor overheating warning method and equipment, comprising first predicting the steam turbine generator rotor temperature, then predicting the degree of steam turbine generator rotor overheating based on the predicted temperature, and finally diagnosing and warning the steam turbine generator rotor overheating condition based on the steam turbine generator rotor overheating condition; the present invention solves the current problem of generator rotors having no temperature measuring points to monitor operating temperature, effectively combines the rotor heat exchange physical process and the temperature change law during load changes, and fully utilizes the advantages of big data for physical modeling and machine learning, in order to achieve accurate modeling and dynamic prediction of rotor overheating.

Description

Technical Field

[0001] The present invention relates to the technical field of generator rotor temperature prediction, and in particular to a water-hydrogen-hydrogen steam turbine generator rotor overheating early warning method and device. Background Art

[0002] Large steam turbine generators, as crucial power generation equipment, inevitably generate heat during normal service due to various losses (such as rotor loss, winding loss, or mechanical loss). If this heat cannot be promptly removed by the cooling system within the unit, the generator set will gradually accumulate and reach a high temperature that affects its normal operation. Therefore, temperature naturally becomes a critical data point that requires real-time monitoring during unit operation, providing data support for advanced temperature control of large steam turbine generators, determining maintenance cycles, and fault diagnosis and analysis.

[0003] Improper start-up and shutdown operations may cause excessive thermal stress. Forced cooling of the steam turbine under high temperature conditions may cause a large temperature difference between the inside and outside, resulting in severe thermal deformation. The rotor operates in a high temperature and high pressure environment, and the inherent defects in the material will cause deformation and cracks after long-term operation; the metal fatigue damage of the rotor material will be aggravated with the increase of thermal stress, which may cause the rotor to deform or even crack. The deformation or cracking of the rotor will affect the stability and safety of the unit operation, and may even cause serious safety accidents. The operating environment and operating procedures of the steam turbine generator set need to be strictly monitored to avoid overheating. How to judge the degree of overheating of the water-hydrogen-hydrogen steam turbine generator rotor and issue a timely warning is very important for power plant operation and maintenance personnel to formulate reasonable operation control and maintenance strategies.

[0004] Because the rotor winding itself lacks embedded temperature measurement points, the average rotor winding temperature calculated based on electromagnetic theory differs significantly from the actual rotor winding temperature, making it inaccurate for determining rotor overheating. While the current finite element analysis method for rotor winding temperature fields can calculate the specific temperature of a specific rotor winding location, it is limited by boundary conditions and the complexity and time-consuming nature of the calculations. Therefore, it remains a laboratory research method with no proven engineering applications. Summary of the Invention

[0005] The present invention proposes a water-hydrogen-hydrogen steam turbine generator rotor overheating early warning method, device and storage medium, which can solve at least one of the technical problems in the background technology.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A water-hydrogen-hydrogen steam turbine generator rotor overheating early warning method comprises first predicting the steam turbine generator rotor temperature, then predicting the steam turbine generator rotor overheating degree based on the predicted temperature, and finally diagnosing and warning the steam turbine generator rotor overheating condition based on the steam turbine generator rotor overheating condition;

[0008] The turbine generator rotor temperature prediction specifically includes:

[0009] Calculate the average rotor temperature under different operating conditions based on the excitation current, excitation voltage, rotor DC resistance, and inlet hydrogen temperature;

[0010] Based on the correlation calculation, the parameters affecting the rotor temperature are found, and the rotor static temperature model and dynamic temperature model are constructed.

[0011] Furthermore, the average rotor temperature under different working conditions is calculated based on the excitation current, excitation voltage, rotor DC resistance, and inlet hydrogen temperature; specifically,

[0012] The calculation formula of rotor average temperature is:

[0013] 0 x =KR-235

[0014] Where: R0 represents the resistance value when the rotor winding temperature is θ0;

[0015] R X Indicates the resistance value of the rotor at any time, which is R X =U F / I F , where: U F is the excitation voltage, I F is the excitation current.

[0016] Furthermore, based on the correlation calculation, the rotor temperature influencing parameters are found, and the rotor static temperature model and dynamic temperature model are built, which specifically include:

[0017] 1) Rotor static temperature model

[0018] According to the correlation degree of the physical parameters on the turbine generator rotor temperature, a temperature calculation formula was established. Using the no-load test data of the unit, a preliminary static temperature model was built:

[0019] Rotor temperature calculation formula: T t =A×I f 2 +B×U f +T0

[0020] Among them: A and B are constants for a certain generator;

[0021] If is the excitation current, U f is the excitation voltage, T0 is the inlet hydrogen temperature;

[0022] 2) Rotor dynamic temperature model

[0023] Based on the principles of electromagnetics and heat conduction, combined with the characteristics of different working conditions and the coupling relationship between the steam turbine generator and the cooling of the parts, the dynamic temperature model is built and trained.

[0024] Furthermore, based on the electromagnetic principle and heat conduction principle, combined with the characteristics of different working conditions and the coupling relationship between the turbine generator and the cooling of the parts, the dynamic temperature model is built and trained, including:

[0025] 2.1) Working condition model

[0026] According to the different active power and reactive power values ​​of the generator, representative working conditions are named, including:

[0027] "Generator phase leading" represents the operating condition when the reactive power is less than 0;

[0028] "Generator heavy load operation" refers to the operating condition when the active power is greater than 300MW;

[0029] “Generator normal operating condition 210” represents the operating condition when the active power is less than 210MW;

[0030] "Generator normal operating condition 210-240" represents the operating condition when the active power is within the range of 210 to 240 MW;

[0031] 2.2) Electromagnetic principle modeling

[0032] Based on electromagnetic principles, a dynamic model was created according to different operating conditions of the generator. The explicit function algorithm was tried to select appropriate coefficients for different operating conditions and optimize the rotor temperature model.

[0033] Rotor temperature calculation formula: Tt=A×I f 2 +B×U f +C×P+D×Q+E×T0+F

[0034] Among them: A and B are parameters in the static model;

[0035] C, D, E, and F are constants for a certain generator;

[0036] I f is the excitation current, U f is the excitation voltage, P is the active power, Q is the reactive power, and T0 is the inlet hydrogen temperature;

[0037] 2.3) Heat conduction principle modeling

[0038] Based on the heat conduction principle, a dynamic model is created according to the coupling relationship of the cooling media in various parts of the steam turbine generator. The fitting degree of different algorithms is tested, and the appropriate combination algorithm is selected to optimize the rotor temperature model.

[0039] The input parameters are set as the average temperature of cold hydrogen, the average temperature of hot hydrogen, and the average temperature of iron core, and the output parameter is the rotor temperature. The algorithm models try various linear regression and nonlinear regression classic algorithms including AlAgent, Linear, Ridge, GBDT, and MLP.

[0040] 2.4) Combine electromagnetic principles and heat conduction principle models, use historical data, and train an appropriate dynamic temperature model.

[0041] Furthermore, the overheating degree of the turbine generator rotor is predicted based on the predicted temperature. This includes introducing dynamic classification criteria such as the deviation between the dynamic model output value and the electromagnetic mechanism calculation value and the rotor temperature rise rate on the basis of the static rotor temperature threshold classification to predict the rotor overheating degree. The details are as follows:

[0042] 1) Rotor temperature static threshold

[0043] Taking a 630MW unit as an example, the rotor temperature static threshold classification criteria are:

[0044] Working conditions: Level I: 100℃> calculated rotor average temperature ≥ 90℃; Level II: 105℃> calculated rotor average temperature ≥ 100℃; Level III: calculated rotor average temperature ≥ 110℃;

[0045] Late phase condition: When P≥600MW: Level I: 100℃> calculated rotor average temperature ≥90℃; Level II: 105℃> calculated rotor average temperature ≥100℃; Level III: calculated rotor average temperature ≥110℃;

[0046] Late phase condition: 600MW>P≥300MW: Level I: 95℃>calculated rotor average temperature ≥90℃; Level II: 100℃>calculated rotor average temperature ≥95℃; Level III:calculated rotor average temperature ≥100℃;

[0047] Late phase operating conditions: When P<300MW: Level I: 90℃> calculated rotor average temperature ≥85℃; Level II: 95℃> calculated rotor average temperature ≥90℃; Level III: calculated rotor average temperature ≥95℃;

[0048] 2) Deviation between the dynamic model output value and the electromagnetic mechanism calculation value

[0049] The deviation between the output value of each rotor dynamic model and the calculated value of the electromagnetic mechanism is calculated according to different operating conditions, and the deviation is verified by comparing the current operating conditions. Taking the 630MW unit as an example, the rotor temperature deviation classification criteria are:

[0050] Working conditions: Level I: 25K> rotor temperature deviation ≥ 10K; Level II: 40K> rotor temperature deviation ≥ 25K; Level III: 55K> rotor temperature deviation ≥ 40K;

[0051] Late phase condition: When P≥600MW: Level I: 25K> rotor temperature deviation ≥10K; Level II: 40K> rotor temperature deviation ≥25K; Level III: 55K> rotor temperature deviation ≥40K;

[0052] Late phase condition: 600MW>P≥300MW: Level I: 20K> rotor temperature deviation ≥10K; Level II: 30K> rotor temperature deviation ≥20K; Level III: 40K> rotor temperature deviation ≥30K;

[0053] Late phase condition: When P<300MW: Level I: 15K> rotor temperature deviation ≥10K; Level II: 20K> rotor temperature deviation ≥15K; Level III: 25K> rotor temperature deviation ≥20K;

[0054] 3) Rotor temperature rise rate

[0055] Automatically compare the temperature rise rate of the turbine generator rotor model value with the calculated value according to different operating conditions, automatically select the operating conditions with higher temperature rise rate during the period, and perform deviation verification by comparing with the current operating conditions. Taking a 630MW unit as an example, the rotor temperature rise rate classification criteria are:

[0056] Working conditions: Level I: 2.5K / min> rotor temperature rise rate ≥ 1K / min; Level II: 4K / min> rotor temperature rise rate ≥ 2.5K / min; Level III: 5.5K / min> rotor temperature rise rate ≥ 4K / min;

[0057] Late phase condition: When P≥600MW: Level I: 2.5K / min> rotor temperature rise rate ≥1K / min; Level II: 4K / min> rotor temperature rise rate ≥2.5K / min; Level III: 5.5K / min> rotor temperature rise rate ≥4K / min;

[0058] Late phase condition: 600MW>P≥300MW: Level I: 2K / min>rotor temperature rise rate ≥1K / min; Level II: 3K / min>rotor temperature rise rate ≥2K / min; Level III: 4K / min>rotor temperature rise rate ≥3K / min;

[0059] Late phase operating conditions: When P<300MW: Level I: 1.5K / min > rotor temperature rise rate ≥ 1K / min; Level II: 2K / min > rotor temperature rise rate ≥ 1.5K / min; Level III: 2.5K / min > rotor temperature rise rate ≥ 2K / min;

[0060] 5) Prediction of rotor overheating

[0061] According to the analysis of historical data, the predicted value of rotor overheating degree A TZGR =0.4*A ZJ +0.4*A ZP +0.2*A ZS

[0062] Among them: A TZGR is the predicted value of rotor overheating degree, A ZJ Assign a static threshold value for the rotor temperature, A ZP Assign a value to the deviation between the dynamic model output value and the electromagnetic mechanism calculation value, A ZS Assign a value to the rotor temperature rise rate;

[0063] A ZJ No warning value: 1.0; Level I value: 0.9; Level II value: 0.8; Level III value: 0.7;

[0064] A ZP No warning value: 1.0; Level I value: 0.9; Level II value: 0.8; Level III value: 0.7;

[0065] A ZS No warning value: 1.0; Level I value: 0.9; Level II value: 0.8; Level III value: 0.7;

[0066] A TZGR :Level I warning is: 0.9; Level II warning is: 0.8; Level III warning is: 0.

[0067] Furthermore, based on the overheating condition of the turbine generator rotor, a diagnosis and early warning of the overheating condition of the turbine generator rotor is performed, including:

[0068] The system conducts self-learning on historical normal operation data and fault operation data, continuously improves the dynamic model, automatically completes the configuration of quantitative warning levels, generates a turbine generator rotor overheating degree warning model, and diagnoses the rotor overheating status according to different overheating degree levels;

[0069] According to the overheating warning scores of the three types of models, the appropriate rotor overheating state judgment criteria are obtained.

[0070] In another aspect, the present invention further discloses a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the above method.

[0071] On the other hand, the present invention further discloses a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above method.

[0072] As can be seen from the above technical solution, the present invention provides a method for predicting rotor overheating in a water-hydrogen-hydrogen steam turbine generator. Based on the prediction of the turbine generator rotor's overheating trend and degree, it enables online, real-time diagnosis of turbine generator rotor overheating conditions. This method primarily addresses the issues of offline test diagnosis being limited by the unit's maintenance schedule and the low sensitivity and accuracy of online diagnosis. The implementation process is simple and easy to operate, and it incorporates offline maintenance and test conditions for comprehensive judgment.

[0073] In general, the present invention aims to solve the current problem of generator rotors having no temperature measurement points to monitor operating temperature. It effectively combines the physical process of rotor heat exchange and the temperature variation law during load changes, and fully utilizes the advantages of big data for physical modeling and machine learning, in order to achieve accurate modeling and dynamic prediction of rotor overheating.

[0074] The beneficial effects of the present invention are as follows:

[0075] 1) Improve safety. By real-time monitoring of the rotor's temperature field changes, abnormal rotor temperature increases can be detected in a timely manner, preventing mechanical failure or damage caused by overheating, thereby ensuring the safe and stable operation of the unit.

[0076] 2) Economic optimization. Online temperature monitoring provides operators with real-time temperature information on the rotor and other key components, helping them optimize operational strategies such as ramp rate, load change rate, and steam temperature change rate. This helps extend component life, reduce unnecessary maintenance costs, and improve operational economy.

[0077] 3.) Condition-based maintenance support. The online rotor overheat analysis system enables on-demand maintenance, meaning maintenance or inspection is performed only when components require it. This reduces unnecessary maintenance activities, improves maintenance efficiency, and ensures optimal unit operation.

[0078] 4) Operation optimization guidance. The online temperature calculation results can be used directly to guide the operations of on-site operators, such as adjusting operating parameters to avoid or reduce excessive thermal stress, further extending the service life of key components such as the rotor.

[0079] 5) Preventive maintenance: Real-time analysis of rotor overheating can help establish a more effective preventive maintenance plan. Predictive analysis can identify potential maintenance needs in advance, avoiding equipment failures and downtime.

[0080] 6) Improve operational flexibility. The online rotor overheat analysis system can provide real-time temperature data, helping operators make quick adjustments when necessary to respond to changes in grid demand or other operational requirements, thereby improving the unit's operational flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 This is a flow chart of the water-hydrogen-hydrogen turbine generator rotor overheating early warning method according to an embodiment of the invention;

[0082] Figure 2 The simulation data of the 630MW unit in delayed phase operation is simulated in the calculation model of the embodiment of the present invention. DETAILED DESCRIPTION

[0083] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0084] like Figure 1 As shown, the water-hydrogen-hydrogen steam turbine generator rotor overheating early warning method described in the embodiment of the present invention includes three parts: steam turbine generator rotor temperature prediction, steam turbine generator rotor overheating degree prediction, and steam turbine generator rotor overheating condition diagnosis. The method includes first predicting the steam turbine generator rotor temperature, then predicting the steam turbine generator rotor overheating degree based on the predicted temperature, and finally diagnosing and warning the steam turbine generator rotor overheating condition based on the steam turbine generator rotor overheating condition; the following are detailed descriptions:

[0085] 1. The contents of turbine generator rotor temperature prediction are as follows:

[0086] 1. Calculate rotor temperature using electromagnetic mechanism

[0087] The average rotor temperature under different operating conditions is calculated based on the excitation current, excitation voltage, rotor DC resistance, and inlet hydrogen temperature.

[0088] The calculation formula of rotor average temperature is:

[0089] θ x =KR x -235, of which

[0090] Where: R0 is the resistance value when the rotor winding temperature is θ0.

[0091] R X ——Calculate the resistance value of the rotor at any time, its value is R X =U F / I F (Where: U F is the excitation voltage, I F is the excitation current).

[0092] 2. "AI + physical mechanism" algorithm calculates rotor temperature

[0093] Based on the correlation calculation, the parameters affecting the rotor temperature are found, and the rotor static temperature model and dynamic temperature model are constructed.

[0094] 1) Rotor static temperature model

[0095] According to the correlation degree of the physical parameters on the turbine generator rotor temperature, a temperature calculation formula was established. Using the no-load test data of the unit, a preliminary static temperature model was built:

[0096] Rotor temperature calculation formula: T t =A×I f 2 +B×U f +T0

[0097] Among them: A and B are constants for a certain generator;

[0098] I f is the excitation current, U f is the excitation voltage, T0 is the inlet hydrogen temperature;

[0099] 2) Rotor dynamic temperature model

[0100] Based on the principles of electromagnetics and heat conduction, combined with the characteristics of different operating conditions and the coupling relationship between the turbine generator and the cooling of the parts, the dynamic temperature model is built and trained:

[0101] 2.1) Working condition model

[0102] According to the different active power and reactive power values ​​of the generator, the representative working conditions are named, such as:

[0103] "Generator phase leading" represents the operating condition when the reactive power is less than 0;

[0104] "Generator heavy load operation" refers to the operating condition when the active power is greater than 300MW;

[0105] “Generator normal operating condition 210” represents the operating condition when the active power is less than 210MW;

[0106] "Generator normal operating condition 210-240" represents the operating condition when the active power is within the range of 210 to 240 MW;

[0107] And so on.

[0108] 2.2) Electromagnetic principle modeling

[0109] Based on electromagnetic principles, a dynamic model is created according to different operating conditions of the generator, and explicit function algorithms are tried to select appropriate coefficients for different operating conditions and optimize the rotor temperature model.

[0110] Rotor temperature calculation formula: Tt=A×I f 2 +B×U f +C×P+D×Q+E×T0+F

[0111] Among them: A and B are parameters in the static model;

[0112] C, D, E, and F are constants for a certain generator;

[0113] I f is the excitation current, U f is the excitation voltage, P is the active power, Q is the reactive power, and T0 is the inlet hydrogen temperature.

[0114] 2.3) Heat conduction principle modeling

[0115] Based on the principle of heat conduction and the coupling relationship of cooling media in various parts of the steam turbine generator, a dynamic model is created. The fitting degree of different algorithms is tried, and the appropriate combination algorithm is selected to optimize the rotor temperature model.

[0116] The input parameters are set as the average temperature of cold hydrogen, the average temperature of hot hydrogen, and the average temperature of iron core, and the output parameter is the rotor temperature. The algorithm models try various linear regression and nonlinear regression classic algorithms such as AlAgent, Linear, Ridge, GBDT, and MLP.

[0117] 2.4) Combine electromagnetic and heat conduction principle models and utilize historical data to train an appropriate dynamic temperature model. Specifically, the model training method provided by the machine learning algorithm can be to continuously characterize the polynomial coefficients using historical data, or to compare typical machine learning algorithms after clarifying the input and output parameters to find an algorithm or algorithm combination with high fit. The selection of historical data should be based on the generator maintenance and operating conditions. The historical data should be obtained when the generator rotor has not undergone major maintenance or replacement of major components, and preferably covers various operating conditions.

[0118] 2. Overheating prediction

[0119] On the basis of the static threshold classification of rotor temperature, dynamic classification criteria such as the deviation between the dynamic model output value and the electromagnetic mechanism calculation value and the rotor temperature rise rate are introduced to predict the rotor overheating degree, such as Figure 2 As shown;

[0120] 1) Rotor temperature static threshold

[0121] Taking a 630MW unit as an example, the rotor temperature static threshold classification criteria are:

[0122] Working conditions: Level I: 100℃> calculated rotor average temperature ≥ 90℃; Level II: 105℃> calculated rotor average temperature ≥ 100℃; Level III: calculated rotor average temperature ≥ 110℃

[0123] Late phase condition: When P≥600MW: Level I: 100℃> calculated rotor average temperature value ≥90℃; Level II: 105℃> calculated rotor average temperature value ≥100℃; Level III: calculated rotor average temperature value ≥110℃

[0124] Late phase condition: 600MW>P≥300MW: Level I: 95℃>calculated rotor average temperature ≥90℃; Level II: 100℃>calculated rotor average temperature ≥95℃; Level III:calculated rotor average temperature ≥100℃

[0125] Late phase condition: When P<300MW: Level I: 90℃> calculated rotor average temperature value ≥85℃; Level II: 95℃> calculated rotor average temperature value ≥90℃; Level III: calculated rotor average temperature value ≥95℃

[0126] 2) Deviation between the dynamic model output value and the electromagnetic mechanism calculation value

[0127] The deviation between the output value of each rotor dynamic model and the calculated value of the electromagnetic mechanism is calculated according to different operating conditions, and the deviation is checked by comparing the current operating conditions. Taking the 630MW unit as an example, the rotor temperature deviation classification criteria are:

[0128] Working conditions: Level I: 25K> rotor temperature deviation ≥ 10K; Level II: 40K> rotor temperature deviation ≥ 25K; Level III: 55K> rotor temperature deviation ≥ 40K

[0129] Delayed phase condition: When P≥600MW: Level I: 25K> rotor temperature deviation ≥10K; Level II: 40K> rotor temperature deviation ≥25K; Level III: 55K> rotor temperature deviation ≥40K

[0130] Late phase condition: 600MW>P≥300MW: Level I: 20K> rotor temperature deviation ≥10K; Level II: 30K> rotor temperature deviation ≥20K; Level III: 40K> rotor temperature deviation ≥30K

[0131] Late phase condition: When P<300MW: Level I: 15K> rotor temperature deviation ≥10K; Level II: 20K> rotor temperature deviation ≥15K; Level III: 25K> rotor temperature deviation ≥20K

[0132] 3) Rotor temperature rise rate

[0133] Automatically compare the temperature rise rate of the turbine generator rotor model value with the calculated value according to different operating conditions, automatically select the operating conditions with higher temperature rise rate, and perform deviation verification by comparing with the current operating conditions. Taking a 630MW unit as an example, the rotor temperature rise rate classification criteria are:

[0134] Working conditions: Level I: 2.5K / min > rotor temperature rise rate ≥ 1K / min; Level II: 4K / min > rotor temperature rise rate ≥ 2.5K / min; Level III: 5.5K / min > rotor temperature rise rate ≥ 4K / min

[0135] Late phase condition: When P≥600MW: Level I: 2.5K / min > rotor temperature rise rate ≥1K / min; Level II: 4K / min > rotor temperature rise rate ≥2.5K / min; Level III: 5.5K / min > rotor temperature rise rate ≥4K / min

[0136] Late phase condition: 600MW>P≥300MW: Level I: 2K / min>rotor temperature rise rate ≥1K / min; Level II: 3K / min>rotor temperature rise rate ≥2K / min; Level III: 4K / min>rotor temperature rise rate ≥3K / min

[0137] Late phase condition: When P<300MW: Level I: 1.5K / min > rotor temperature rise rate ≥ 1K / min; Level II: 2K / min > rotor temperature rise rate ≥ 1.5K / min; Level III: 2.5K / min > rotor temperature rise rate ≥ 2K / min

[0138] 4) Prediction of rotor overheating

[0139] According to the analysis of historical data, the predicted value of rotor overheating degree A TZGR =0.4*A ZJ +0.4*A ZP +0.2*A ZS

[0140] Among them: A TZGR is the predicted value of rotor overheating degree, AZJ Assign a static threshold value for the rotor temperature, A ZP Assign a value to the deviation between the dynamic model output value and the electromagnetic mechanism calculation value, A ZS Assign a value to the rotor temperature rise rate.

[0141] A ZJ No warning value: 1.0; Level I value: 0.9; Level II value: 0.8; Level III value: 0.7

[0142] A ZP No warning value: 1.0; Level I value: 0.9; Level II value: 0.8; Level III value: 0.7

[0143] A ZS No warning value: 1.0; Level I value: 0.9; Level II value: 0.8; Level III value: 0.7

[0144] A TZGR :Level I warning is: 0.9; Level II warning is: 0.8; Level III warning is: 0.7.

[0145] 3. Overheating diagnosis

[0146] It conducts self-learning on historical normal operation data and fault operation data, continuously improves the dynamic model, automatically completes the configuration of quantitative warning levels, generates a turbine generator rotor overheating warning model, and diagnoses the rotor overheating status according to different overheating levels.

[0147] According to the overheating warning scores of the three types of models, the appropriate rotor overheating state judgment criteria are obtained.

[0148] Individual score ≥80 points and comprehensive score ≥90 points, the rotor is healthy;

[0149] 80 points > single item score ≥ 70 points and 90 points > comprehensive score ≥ 80 points, rotor sub-health;

[0150] 70 points > single item score ≥ 60 points or 80 points > comprehensive score ≥ 70 points, there is a risk of rotor overheating;

[0151] 60 points > single item score or 70 points > comprehensive score ≥ 60 points, the rotor is overheated.

[0152] In summary, the embodiments of the present invention address the current problem of generator rotors having no temperature measurement points to monitor operating temperature. They effectively combine the physical process of rotor heat exchange with the temperature variation patterns during load changes, and fully utilize the advantages of big data for physical modeling and machine learning, in order to achieve accurate modeling and dynamic prediction of rotor overheating.

[0153] like Figure 2As shown in the figure, the calculation model simulates the simulation data of the 630MW unit in late phase operation, with an active power of 387.877MW. Its excitation current increases from 2537.136A to 2818.518A, the corresponding excitation voltage increases from 230.413V to 280.361V, and the calculated rotor average temperature increases from 65.692℃ to 94.349℃, meeting the late phase operating conditions: 600MW>P≥300MW: Level I: 95℃>rotor average temperature calculated value ≥90℃ alarm conditions, and the alarm has been triggered.

[0154] In another aspect, the present invention further discloses a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the above method.

[0155] On the other hand, the present invention further discloses a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above method.

[0156] In another embodiment provided in the present application, a computer program product containing instructions is also provided, which, when run on a computer, enables the computer to execute any of the water-hydrogen-hydrogen steam turbine generator rotor overheating warning methods in the above embodiments.

[0157] It is understandable that the system, device and storage medium provided in the embodiments of the present invention correspond to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant contents can refer to the corresponding parts of the above methods.

[0158] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0159] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0160] Each embodiment in this specification is described in a related manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiment is generally similar to the method embodiment, so the description is relatively simple. For related parts, refer to the description of the method embodiment.

[0161] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A water-hydrogen-hydrogen turbine generator rotor overheating early warning method, characterized in that: The system first predicts the temperature of the turbine generator rotor, then predicts the degree of overheating of the turbine generator rotor based on the predicted temperature, and finally diagnoses and warns of overheating of the turbine generator rotor based on the overheating of the turbine generator rotor. The turbine generator rotor temperature prediction specifically includes: Calculate the average rotor temperature under different operating conditions based on the excitation current, excitation voltage, rotor DC resistance, and inlet hydrogen temperature; According to the correlation calculation, the rotor temperature influencing parameters are found, and the rotor static temperature model and dynamic temperature model are built, including: 1) Rotor static temperature model According to the correlation degree of the physical parameters on the turbine generator rotor temperature, a temperature calculation formula is established. Using the unit no-load test data, a preliminary rotor static temperature model is built: T t =A×I f 2 +B×U f +T0 Where: T t is the rotor temperature; A and B are constants for a certain generator; I f is the excitation current, U f is the excitation voltage, T0 is the inlet hydrogen temperature; 2) Rotor dynamic temperature model According to the electromagnetic principle and heat conduction principle, combined with the characteristics of different working conditions and the coupling relationship between the turbine generator and the cooling of the parts, the rotor dynamic temperature model is built and trained. The rotor dynamic temperature model is: T t =A×I f 2 +B×U f +C×P+D×Q+E×T0+F; Electromagnetic principle modeling includes, Based on electromagnetic principles, a dynamic model is created according to different operating conditions of the generator. The explicit function algorithm is tried to select appropriate coefficients for different operating conditions and optimize the rotor dynamic temperature model: Where: T t is the rotor temperature; A and B are parameters in the static model; C, D, E, and F are constants for a certain generator; I f is the excitation current, U f is the excitation voltage, P is the active power, Q is the reactive power, and T0 is the inlet hydrogen temperature.

2. The water-hydrogen-hydrogen turbine generator rotor overheating early warning method according to claim 1 is characterized in that: The construction of the rotor dynamic temperature model also includes: Working condition model construction: According to the different active power and reactive power values ​​of the generator operation, representative working conditions are named, including: "Generator phase leading" represents the operating condition when the reactive power is less than 0; "Generator high load operation" refers to the operating condition when the active power is greater than 300MW; "Generator normal operating condition 210" represents the operating condition when the active power is less than 210MW; "Generator normal operating condition 210-240" represents the operating condition when the active power is within the range of 210~240MW; Heat conduction principle modeling: Based on the heat conduction principle, a dynamic model is created according to the coupling relationship of the cooling media in various parts of the steam turbine generator. The fit of different algorithms is tested, and the appropriate combination algorithm is selected to optimize the rotor temperature model. The input parameters are set as the average temperature of cold hydrogen, the average temperature of hot hydrogen, and the average temperature of iron core, and the output parameter is the rotor temperature. The algorithm models try various linear regression and nonlinear regression classic algorithms including AlAgent, Linear, Ridge, GBDT, and MLP. Combining electromagnetic principles and heat conduction principle models, and using historical data, an appropriate dynamic temperature model is trained.

3. The water-hydrogen-hydrogen turbine generator rotor overheating early warning method according to claim 1 is characterized in that: Calculate the average rotor temperature under different working conditions based on the excitation current, excitation voltage, rotor DC resistance, and inlet hydrogen temperature; specifically, The calculation formula of rotor average temperature is: Where: The rotor winding temperature is The resistance value when R X Indicates the calculated resistance value of the rotor at any time, and its value is R X = U F / I F ,in: U F is the excitation voltage, I F is the excitation current.

4. The water-hydrogen-hydrogen turbine generator rotor overheating early warning method according to claim 1 is characterized in that: The overheating degree of the turbine generator rotor is predicted based on the predicted temperature. This includes introducing dynamic classification criteria such as the deviation between the dynamic model output value and the electromagnetic mechanism calculation value and the rotor temperature rise rate on the basis of the static rotor temperature threshold classification to predict the rotor overheating degree. The details are as follows: 1) Rotor temperature static threshold For 630MW units, the rotor temperature static threshold classification criteria include: Phase-advancing condition: Level I: 100°C > calculated rotor average temperature ≥ 90°C; Level II: 105°C > calculated rotor average temperature ≥ 100°C; Level III: calculated rotor average temperature ≥ 110°C; Late phase condition: When P≥600MW: Level I: 100℃> calculated rotor average temperature ≥90℃; Level II: 105℃> calculated rotor average temperature ≥100℃; Level III: calculated rotor average temperature ≥110℃; Late phase condition: 600MW>P≥300MW: Level I: 95℃>calculated rotor average temperature ≥90℃; Level II: 100℃>calculated rotor average temperature ≥95℃; Level III:calculated rotor average temperature ≥100℃; Late phase operating conditions: When P<300MW: Level I: 90℃> calculated rotor average temperature ≥85℃; Level II: 95℃> calculated rotor average temperature ≥90℃; Level III: calculated rotor average temperature ≥95℃; 2) Deviation between the dynamic model output value and the electromagnetic mechanism calculation value The deviation between the output value of each rotor dynamic model and the calculated value of the electromagnetic mechanism is calculated according to different operating conditions, and the deviation is verified by comparing the current operating conditions. For the 630MW unit, the rotor temperature deviation classification criteria are: Phase-advancing conditions: Level I: 25K> rotor temperature deviation ≥ 10K; Level II: 40K> rotor temperature deviation ≥ 25K; Level III: 55K> rotor temperature deviation ≥ 40K; Late phase condition: When P≥600MW: Level I: 25K> rotor temperature deviation ≥10K; Level II: 40K> rotor temperature deviation ≥25K; Level III: 55K> rotor temperature deviation ≥40K; Late phase condition: 600MW>P≥300MW: Level I: 20K> rotor temperature deviation ≥10K; Level II: 30K> rotor temperature deviation ≥20K; Level III: 40K> rotor temperature deviation ≥30K; Late phase condition: When P<300MW: Level I: 15K> rotor temperature deviation ≥10K; Level II: 20K> rotor temperature deviation ≥15K; Level III: 25K> rotor temperature deviation ≥20K; 3) Rotor temperature rise rate Automatically compare the temperature rise rate of the turbine generator rotor model value with the calculated value according to different operating conditions, automatically select the operating conditions with higher temperature rise rate during the period, and perform deviation verification by comparing with the current operating conditions; for 630MW units, the rotor temperature rise rate classification criteria are: Phase-advancing conditions: Level I: 2.5K / min > rotor temperature rise rate ≥ 1K / min; Level II: 4K / min > rotor temperature rise rate ≥ 2.5K / min; Level III: 5.5K / min > rotor temperature rise rate ≥ 4K / min; Late phase condition: When P≥600MW: Level I: 2.5K / min> rotor temperature rise rate ≥1K / min; Level II: 4K / min> rotor temperature rise rate ≥2.5K / min; Level III: 5.5K / min> rotor temperature rise rate ≥4K / min; Late phase condition: 600MW>P≥300MW: Level I: 2K / min>rotor temperature rise rate ≥1K / min; Level II: 3K / min>rotor temperature rise rate ≥2K / min; Level III: 4K / min>rotor temperature rise rate ≥3K / min; Late phase operating conditions: When P<300MW: Level I: 1.5K / min > rotor temperature rise rate ≥ 1K / min; Level II: 2K / min > rotor temperature rise rate ≥ 1.5K / min; Level III: 2.5K / min > rotor temperature rise rate ≥ 2K / min; 4) Prediction of rotor overheating According to the analysis of historical data, the predicted value of rotor overheating degree A TZGR =0.4*A ZJ +0.4*A ZP +0.2*A ZS Among them: A TZGR is the predicted value of rotor overheating degree, A ZJ Assign a static threshold value for the rotor temperature, A ZP Assign a value to the deviation between the dynamic model output value and the electromagnetic mechanism calculation value, A ZS Assign a value to the rotor temperature rise rate; A ZJ No warning value: 1.0; Level I value: 0.9; Level II value: 0.8; Level III value: 0.7; A ZP No warning value: 1.0; Level I value: 0.9; Level II value: 0.8; Level III value: 0.7; A ZS No warning value: 1.0; Level I value: 0.9; Level II value: 0.8; Level III value: 0.7; A TZGR : Level I warning is: 0.9; Level II warning is: 0.8; Level III warning is:

0.

5. The water-hydrogen-hydrogen turbine generator rotor overheating early warning method according to claim 1 is characterized in that: Based on the overheating of the turbine generator rotor, the turbine generator rotor overheating condition diagnosis and warning are carried out, including: The system conducts self-learning on historical normal operation data and fault operation data, continuously improves the dynamic model, automatically completes the configuration of quantitative warning levels, generates a turbine generator rotor overheating degree warning model, and diagnoses the rotor overheating status according to different overheating degree levels; According to the overheating warning scores of the three types of models, the appropriate rotor overheating state judgment criteria are obtained.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 5.

Images