A fatigue damage prediction method and system for a hydro-generator set

Abstract

The present invention discloses a fatigue damage prediction method and system for a hydro-generator set, relating to the field of data processing technology. The method comprises: obtaining a top cover fixing assembly in the hydro-generator set; performing multi-dimensional operation monitoring and analysis to obtain predicted water pressure, and combining this with preload data to obtain a first fatigue damage prediction constraint for the bolts; compiling an operation log record for the hydro-generator set and analyzing it to obtain a second fatigue damage prediction constraint for multiple fixed guide vanes; and analyzing the first fatigue damage prediction constraint and the second fatigue damage prediction constraint to obtain a fatigue damage prediction result. The present invention solves the technical problem in the prior art that fatigue damage prediction of conventional hydro-generator set bolt assemblies relies on empirical formulas and expert judgment, resulting in low reliability and accuracy. This method achieves the technical effect of improving the intelligence and reliability of fatigue damage prediction for hydro-generator set bolt assemblies and accurately locating fatigue components.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a fatigue damage prediction method and system for a hydro-generator set. Background Art

[0002] With the growing demand for clean energy, the operational safety and stability of hydro-turbine generators, as crucial water-energy conversion equipment, are drawing significant attention. Fatigue damage is one of the primary challenges faced by hydro-turbine generators during operation. Long-term operation and complex operating environments can lead to fatigue accumulation and damage in turbine components. Therefore, accurately predicting fatigue damage in hydro-turbine generators is crucial for ensuring unit safety, improving operational efficiency, and reducing maintenance costs. However, traditional fatigue damage prediction methods, which rely primarily on empirical formulas and expert judgment, often suffer from low accuracy and reliability. Summary of the Invention

[0003] The present application provides a fatigue damage prediction method and system for a hydro-generator set, which is used to solve the technical problem in the prior art that fatigue damage prediction of bolt assemblies of traditional hydro-generator sets relies on empirical formulas and expert judgment, and has low reliability and accuracy.

[0004] According to a first aspect of the present application, a fatigue damage prediction method for a hydro-turbine generator set is provided, the method comprising: obtaining a top cover fixing assembly in the hydro-turbine generator set, wherein the top cover fixing assembly comprises bolts and a seat ring, and the seat ring comprises a plurality of fixed guide vanes; reading predetermined operating indicators, and performing multi-dimensional operation monitoring of the hydro-turbine generator set based on the predetermined operating indicators to obtain real-time operating information, wherein the hydro-turbine generator set has an identifier of the unit's own weight; activating a water pressure prediction model to analyze the real-time operating information to obtain a predicted water pressure, and combining the preload force data of the bolt to obtain a first fatigue damage prediction constraint for the bolt; establishing an operation log record of the hydro-turbine generator set based on the real-time operating information, the operation log record comprising a first historical record; analyzing a first historical pressure pulsation, a first historical Karman vortex frequency, a first historical fixed guide vane frequency, and the unit's own weight in the first historical record to obtain a second fatigue damage prediction constraint for the plurality of fixed guide vanes; activating the fatigue damage prediction model to analyze the first fatigue damage prediction constraint and the second fatigue damage prediction constraint to obtain a fatigue damage prediction result for the top cover fixing assembly.

[0005] The second aspect of the present application provides a fatigue damage prediction system for a hydro-generator set, the system comprising: a top cover fixing assembly acquisition module, the top cover fixing assembly acquisition module being used to acquire the top cover fixing assembly in the hydro-generator set, wherein the top cover fixing assembly comprises bolts and a seat ring, and the seat ring comprises a plurality of fixed guide vanes; a multi-dimensional operation monitoring module, the multi-dimensional operation monitoring module being used to read predetermined operation indicators, and perform multi-dimensional operation monitoring of the hydro-generator set based on the predetermined operation indicators to obtain real-time operation information, wherein the hydro-generator set has an identifier of the unit's own weight; a first fatigue damage prediction constraint acquisition module, the first fatigue damage prediction constraint acquisition module being used to activate a water pressure prediction model to analyze the real-time operation information to obtain a predicted water pressure, and obtain the predicted water pressure in combination with the preload force data of the bolts. The first fatigue damage prediction constraint of the bolt; an operation log record assembly module, the operation log record assembly module is used to assemble the operation log record of the hydro-generator set based on the real-time operation information, and the operation log record includes a first historical record; a second fatigue damage prediction constraint acquisition module, the second fatigue damage prediction constraint acquisition module is used to analyze the first historical pressure pulsation, the first historical Karman vortex frequency, the first historical fixed guide vane frequency and the unit's own weight in the first historical record to obtain the second fatigue damage prediction constraint of the multiple fixed guide vanes; a fatigue damage prediction result acquisition module, the fatigue damage prediction result acquisition module is used to activate the fatigue damage prediction model to analyze the first fatigue damage prediction constraint and the second fatigue damage prediction constraint to obtain the fatigue damage prediction result of the top cover fixing assembly.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The present application provides a fatigue damage prediction method for a hydro-turbine generator set, which relates to the field of data processing technology. For a top cover fixing assembly in a hydro-turbine generator set, a predicted water pressure is obtained through multi-dimensional operation monitoring and analysis, and a first fatigue damage prediction constraint of the bolt is obtained in combination with preload data; based on the operation log records of the hydro-turbine generator set, a second fatigue damage prediction constraint of a plurality of fixed guide vanes is obtained by analysis, and the fatigue damage prediction result is obtained by combining the analysis with the first fatigue damage prediction constraint. This solves the technical problem in the prior art that fatigue damage prediction of bolt assemblies of traditional hydro-turbine generator sets relies on empirical formulas and expert judgments, and has low reliability and accuracy, thereby achieving the technical effect of improving the intelligence level and reliability of fatigue damage prediction of bolt assemblies of hydro-turbine generator sets and accurately locating fatigue components. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0009] Figure 1 A schematic flow chart of a fatigue damage prediction method for a hydro-generator set provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of a flow chart for obtaining a second fatigue damage prediction constraint in a fatigue damage prediction method for a hydro-generator set provided in an embodiment of the present application;

[0011] Figure 3 A schematic diagram of a process for obtaining fatigue damage prediction results in a fatigue damage prediction method for a hydro-generator set provided in an embodiment of the present application;

[0012] Figure 4 A schematic diagram of the structure of a fatigue damage prediction system for a hydro-generator set provided in an embodiment of the present application.

[0013] Explanation of the accompanying drawings: top cover fixing component acquisition module 11, multi-dimensional operation monitoring module 12, first fatigue damage prediction constraint acquisition module 13, operation log recording assembly module 14, second fatigue damage prediction constraint acquisition module 15, fatigue damage prediction result acquisition module 16. DETAILED DESCRIPTION

[0014] The present application provides a fatigue damage prediction method for a hydro-generator set, which is used to solve the technical problem in the prior art that fatigue damage prediction of bolt assemblies of traditional hydro-generator sets relies on empirical formulas and expert judgment, and has low reliability and accuracy.

[0015] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0016] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices.

[0017] Example 1

[0018] like Figure 1 As shown, the present application provides a fatigue damage prediction method for a hydro-generator set, the method comprising:

[0019] P10: Obtain a top cover fixing assembly in a hydro-generator set, wherein the top cover fixing assembly includes bolts and a seat ring, and the seat ring includes a plurality of fixed guide vanes;

[0020] Specifically, identify and locate the top cover fixing assembly in the hydro-turbine generator set. The top cover fixing assembly consists of two parts: bolts and seat rings. The bolts are important connecting parts of the top cover fixing assembly, which can tightly connect the top cover and the seat ring through pre-tightening force to prevent loosening or displacement during operation. The seat ring is another key part of the top cover fixing assembly. It is usually a solid annular structure used to support and fix the top cover. The seat ring also contains multiple fixed guide vanes. These guide vanes guide the water flow and control the direction and speed of the water flow to ensure that the water flow can smoothly enter the generator set and reduce the impact and vibration of the water flow on the generator set. By identifying and obtaining the top cover fixing assembly in the hydro-turbine generator set, the necessary hardware foundation is provided for subsequent operation monitoring and fatigue damage prediction.

[0021] P20: Reading predetermined operating indicators, and performing multi-dimensional operating monitoring on the hydro-generator set based on the predetermined operating indicators to obtain real-time operating information, wherein the hydro-generator set has an indicator of the unit's deadweight;

[0022] The predetermined operating indicators include speed, load, pressure, pressure pulsation, Karman vortex frequency, and fixed guide vane frequency.

[0023] It should be understood that reading predetermined operating indicators of the hydro-turbine generator set, including speed, load, pressure, pressure pulsation, Karman vortex frequency, fixed guide vane frequency, etc., can comprehensively reflect the operating status and performance of the generator set and serve as an important basis for operation monitoring. Furthermore, based on the predetermined operating indicators, the hydro-turbine generator set is subjected to multi-dimensional operation monitoring, using various sensors and monitoring equipment to measure and monitor various key parts and parameters of the generator set in real time. For example, a speed sensor is used to monitor the speed of the generator set, a pressure sensor is used to monitor pressure changes of the generator set, and a vibration sensor is used to monitor the vibration of the generator set.

[0024] Furthermore, the turbine generator set is equipped with an indicator indicating its deadweight. Deadweight is a crucial parameter in generator set design and operation, influencing aspects such as vibration, stress distribution, and fatigue damage. Therefore, when acquiring real-time operational information, the influence of deadweight must be fully considered to ensure the accuracy and reliability of monitoring results.

[0025] Through multi-dimensional operation monitoring, real-time operation data of the hydro-generator set can be obtained. The real-time operation data can reflect the current operating status of the generator set and provide important data support for subsequent fatigue damage prediction.

[0026] P30: Activate the water pressure prediction model to analyze the real-time operation information to obtain a predicted water pressure, and combine the preload force data of the bolt to obtain a first fatigue damage prediction constraint of the bolt;

[0027] Furthermore, in the embodiment of the present application, the process of obtaining the predicted water pressure includes:

[0028] The real-time speed, real-time load, and real-time pressure extracted from the real-time operation information are used as input information of the water pressure prediction model to obtain output information of the water pressure prediction model, wherein the output information includes the predicted water pressure. The water pressure prediction model is an intelligent model obtained by supervised learning and verification of the water pressure test data group.

[0029] Optionally, a water pressure prediction model is activated to analyze the real-time operating information, wherein the water pressure prediction model is an intelligent model obtained through supervised learning and verification, and can predict the water pressure that the generator set may encounter during operation based on the input operating information. In the process of obtaining the predicted water pressure, key parameters such as real-time speed, real-time load, and real-time pressure in the real-time operating information are first extracted and input into the water pressure prediction model as input information. Through the calculation of the model, the output information of the predicted water pressure is obtained. The predicted water pressure can more accurately reflect the water pressure conditions that the generator set may encounter during operation.

[0030] Furthermore, the predicted water pressure is combined with the bolt preload data to derive the first fatigue damage prediction constraint for the bolts. The preload is the force applied during installation to ensure the bolts securely secure the top cover and seat ring. By combining the predicted water pressure and preload data, a preliminary prediction of the stress and fatigue damage the bolts may experience during operation is made. This first fatigue damage prediction constraint is then determined, providing a reference for subsequent maintenance and management.

[0031] The water pressure prediction model is obtained by supervised learning of the water pressure test data set. The training process can be as follows: collecting the water pressure test data set, which should include the real-time speed, real-time load, real-time pressure and corresponding water pressure data of the hydro-generator set under different operating conditions, cleaning and preprocessing the collected data, and using it as training data. Furthermore, according to the nature of the data, a suitable model is selected, such as a neural network, a decision tree, a support vector machine, etc., to construct a model framework, and the preprocessed water pressure test data set is divided into a training set and a validation set. The model is trained using the training set data, and the prediction error, such as the mean square error and mean absolute error, is minimized by adjusting the model parameters. The trained model is validated using the validation set data to evaluate its prediction performance on unknown data, and the model is tuned according to the validation results until the output data of the model converges, thereby obtaining the water pressure prediction model.

[0032] P40: Build an operation log record of the hydro-generator set based on the real-time operation information, wherein the operation log record includes a first historical record;

[0033] It should be understood that by recording the real-time operating information of the hydro-turbine generator set, an operating log record of the hydro-turbine generator set is established. The operating log record includes the daily operating parameter record, equipment status record, abnormal event record, maintenance and repair record, etc. of the hydro-turbine generator set, and the operating log record contains a first historical record, that is, the operating data and information of the generator set in the past period of time (which can be one month, three months, etc., and the specific time can be adaptively adjusted according to actual needs), which can provide a reference for the performance status evaluation of the hydro-turbine generator set.

[0034] P50: Analyze the first historical pressure pulsation, the first historical Karman vortex frequency, the first historical fixed guide vane frequency, and the deadweight of the turbine unit in the first historical record to obtain a second fatigue damage prediction constraint for the plurality of fixed guide vanes;

[0035] Further, such as Figure 2 As shown, step P50 in the embodiment of the present application also includes:

[0036] P51: Determine whether the first historical Karman vortex frequency is coupled with the first historical fixed guide vane frequency;

[0037] P52: If it is coupled, generate a first unit historical coupling period, and add the first unit historical coupling period to obtain the total historical coupling period length;

[0038] P53: Comparing the first historical pressure pulsation with a predetermined pressure pulsation to obtain a first pressure pulsation comparison deviation;

[0039] P54: Add the first pressure pulsation comparison deviation to obtain the total historical pressure pulsation deviation;

[0040] P55: The deadweight of the unit, the total historical coupling period, and the total historical pressure pulsation deviation are used as the second fatigue damage prediction constraints.

[0041] Optionally, after obtaining the real-time operation information and operation log records of the hydro-turbine generator set, the key parameters in the first historical record are further analyzed to predict fatigue damage of the fixed guide vanes, the key parameters including the first historical pressure pulsation, the first historical Karman vortex frequency, the first historical fixed guide vane frequency and the deadweight of the unit in the first historical record.

[0042] Among them, the first historical pressure pulsation reflects the change in pressure during the operation of the hydro-generator set. By analyzing the first historical pressure pulsation, the pressure distribution and change pattern of the set under different operating conditions can be obtained. The Karman vortex is a vortex phenomenon generated when water flows through the fixed guide vanes. Its frequency is closely related to factors such as the water flow velocity and the shape and size of the fixed guide vanes. By analyzing the first historical Karman vortex frequency and the first historical fixed guide vane frequency, it is determined whether there is a coupling relationship between the two, that is, whether the two affect each other and act together on the operation of the set.

[0043] Specifically, first determine whether the first historical Karman vortex frequency and the first historical fixed guide vane frequency are coupled. If they are coupled, it means that the change trends of the two are correlated and have a common impact on the fatigue damage of the unit. It is necessary to further analyze the duration of this coupling phenomenon and generate a first unit historical coupling period, that is, multiple time periods in which the two remain coupled within a period of time. By adding the first unit historical coupling periods, the total historical coupling period length is obtained, that is, the total duration of the coupling phenomenon in the entire historical record.

[0044] Furthermore, the first historical pressure pulsation in the actual record is compared with the pressure pulsation within a predetermined ideal or safe range to obtain a pressure pulsation comparison deviation, and the first pressure pulsation comparison deviation is added to obtain a total historical pressure pulsation deviation, that is, the total difference in pressure pulsation compared with the ideal or safe state throughout the entire operating history.

[0045] Finally, the unit's deadweight, the historical total coupling period, and the historical total deviation of pressure pulsation are combined to serve as the second fatigue damage prediction constraint, i.e., the second constraint for predicting fatigue damage to the fixed guide vanes. The deadweight of the unit is a fixed physical parameter that affects the overall structural characteristics and stress distribution of the unit. The historical total coupling period and the historical total deviation of pressure pulsation reflect the coupling between the Karman vortex and the fixed guide vane frequency and the changes in pressure pulsation during unit operation, respectively. Together, they constitute the constraints for predicting fatigue damage to the fixed guide vanes.

[0046] P60: Activate the fatigue damage prediction model to analyze the first fatigue damage prediction constraint and the second fatigue damage prediction constraint to obtain a fatigue damage prediction result of the top cover fixing assembly.

[0047] Further, such as Figure 3 As shown, step P60 in this embodiment of the application also includes:

[0048] Step a: The fatigue damage prediction model includes a bolt fatigue damage prediction channel and a fixed guide vane fatigue damage prediction channel;

[0049] Step b: analyzing the first fatigue damage prediction constraint through the bolt fatigue damage prediction channel to obtain a bolt fatigue damage prediction index;

[0050] Step c: analyzing the second fatigue damage prediction constraint through the fixed guide vane fatigue damage prediction channel to obtain a fixed guide vane fatigue damage prediction index;

[0051] Step d1: If the bolt fatigue damage prediction index is greater than the fixed guide vane fatigue damage prediction index, the bolt fatigue damage prediction index is used as the fatigue damage prediction result;

[0052] Step d2: If the bolt fatigue damage prediction index is less than the fixed guide vane fatigue damage prediction index, the fixed guide vane fatigue damage prediction index is used as the fatigue damage prediction result.

[0053] Furthermore, in the embodiment of the present application, the process of constructing the fixed guide vane fatigue damage prediction channel includes:

[0054] P61a: A fixed guide vane fatigue test data set is established based on the big data, and a first data set is extracted from the fixed guide vane fatigue test data set, wherein the first data set includes a first unit deadweight, a first total coupling period, a first total pressure pulsation deviation, and first fatigue damage detection data;

[0055] P62a: Calculating the normalized first fatigue damage detection data using the coefficient of variation principle to obtain a first fixed guide vane fatigue damage index;

[0056] P63a: Supervised learning and verification are performed on the deadweight of the first unit, the first total coupling period, the first total pressure pulsation deviation, and the first fixed guide vane fatigue damage index to obtain the fixed guide vane fatigue damage prediction channel.

[0057] Optionally, a fatigue damage prediction model is activated to further analyze the first and second fatigue damage prediction constraints. The fatigue damage prediction model includes a bolt fatigue damage prediction channel and a fixed guide vane fatigue damage prediction channel, which can predict fatigue damage to specific components based on given constraints. Furthermore, the bolt fatigue damage prediction channel and the fixed guide vane fatigue damage prediction channel correspond to two key components of the top cover fixing assembly, the bolts and the fixed guide vanes, respectively. Because these two components may have different materials, structures, and stress conditions, separate predictions are required.

[0058] Among them, the construction process of the fixed guide vane fatigue damage prediction channel can be: first, a large amount of fixed guide vane fatigue test data is collected using big data technology, including fixed guide vane fatigue test data from different hydro-turbine generator sets, different operating conditions and different time periods, and a fixed guide vane fatigue test data group is formed, and a first data group is randomly extracted therefrom. The first data group includes four key parameters: the first unit dead weight, the first coupling total period, the first pressure pulsation total deviation and the first fatigue damage detection data, all of which are important factors affecting the fatigue damage of the fixed guide vane.

[0059] Furthermore, the first fatigue damage detection data is normalized to eliminate the effects of different dimensions and units, and the normalized data is calculated using the coefficient of variation principle. The coefficient of variation is an indicator that measures the degree of data dispersion and can reflect the relative degree of change between data. By calculating the coefficient of variation, the first fixed guide vane fatigue damage index is obtained. Furthermore, a supervised learning method is used to train the first unit's deadweight, the first total coupling period, the first total pressure pulsation deviation, and the first fixed guide vane fatigue damage index. During the training process, the model parameters and structure are continuously adjusted to ensure that the model's predicted results are as close as possible to the actual results until convergence is achieved, thereby obtaining the fixed guide vane fatigue damage prediction channel.

[0060] Similarly, the bolt fatigue damage prediction channel is trained using the above method. Furthermore, the first fatigue damage prediction constraint is analyzed using the bolt fatigue damage prediction channel to obtain a bolt fatigue damage prediction index. This bolt fatigue damage prediction index reflects the degree of fatigue damage that may occur to the bolt under given constraints. The second fatigue damage prediction constraint is analyzed using the fixed guide vane fatigue damage prediction channel to obtain a fixed guide vane fatigue damage prediction index. This fixed guide vane fatigue damage prediction index reflects the fatigue damage of the fixed guide vane under specific constraints.

[0061] Furthermore, if the bolt fatigue damage prediction index is greater than the fixed guide vane fatigue damage prediction index, this means that under the current operating conditions, the bolt portion is more likely to suffer fatigue damage than the fixed guide vane, and the bolt fatigue damage prediction index is used as the final fatigue damage prediction result. If the bolt fatigue damage prediction index is less than the fixed guide vane fatigue damage prediction index, it indicates that the fixed guide vane is more likely to suffer fatigue damage under the current conditions, and the fixed guide vane fatigue damage prediction index is used as the final prediction result. Through the above analysis and comparison, the parts of the top cover fixing assembly that are more prone to fatigue damage can be more accurately determined, thereby providing targeted recommendations for subsequent maintenance and management.

[0062] Furthermore, the embodiment of the present application further includes step P70, which further includes:

[0063] Calling a predetermined loss analysis function to perform loss analysis on the fatigue damage prediction result to obtain a loss analysis result. The expression of the predetermined loss analysis function is as follows:

[0064]

[0065] Wherein, L(x, x′) refers to the predetermined loss analysis function between the fatigue damage prediction result x and the loss analysis result x′, x i Refers to the fatigue damage prediction result of the top cover fixing component at the i-th moment in the n-moment prediction of the fatigue damage prediction result x, x i ′ refers to the fatigue damage prediction loss analysis result at the i-th moment in the loss analysis result x′, α refers to the ambient temperature feedback adjustment coefficient, β refers to the ambient humidity feedback adjustment coefficient, and γ refers to the working condition feedback adjustment coefficient.

[0066] In a possible embodiment of the present application, in order to understand the potential risk of the top cover fixing assembly, a predetermined loss analysis function is called to perform loss analysis on the fatigue damage prediction result. The expression of the predetermined loss analysis function is:

[0067] Wherein, L(x, x′) refers to the predetermined loss analysis function between the fatigue damage prediction result x and the loss analysis result x′, x i Refers to the fatigue damage prediction result of the top cover fixing component at the i-th moment in the n-moment prediction of the fatigue damage prediction result x, x i ′ refers to the fatigue damage prediction loss analysis result at the i-th moment in the loss analysis result x′, α refers to the ambient temperature feedback adjustment coefficient, β refers to the ambient humidity feedback adjustment coefficient, and γ refers to the working condition feedback adjustment coefficient.

[0068] Through the predetermined loss analysis function, combined with the fatigue damage prediction results of the top cover fixing assembly, fatigue loss analysis is performed to obtain loss analysis results, including economic losses, production losses, safety risks, etc., to provide strong support for the operation management, maintenance decision-making and risk control of the hydro-generator set.

[0069] In summary, the embodiments of the present application have at least the following technical effects:

[0070] This application targets the top cover fixing assembly in a hydro-turbine generator set, obtains predicted water pressure through multi-dimensional operation monitoring and analysis, and obtains the first fatigue damage prediction constraint of the bolt in combination with the preload data; based on the operation log records of the hydro-turbine generator set, the second fatigue damage prediction constraint of multiple fixed guide vanes is analyzed and obtained, and combined with the first fatigue damage prediction constraint for analysis to obtain the fatigue damage prediction result.

[0071] The technical effect of improving the intelligence level and reliability of fatigue damage prediction of bolt assemblies of hydro-turbine generator sets and accurately locating fatigue components has been achieved.

[0072] Example 2

[0073] Based on the same inventive concept as the fatigue damage prediction method of a hydro-generator set in the above embodiment, Figure 4 As shown, the present application provides a fatigue damage prediction system for a hydro-generator set. The system and method embodiments in the present application are based on the same inventive concept. The system includes:

[0074] A top cover fixing assembly acquisition module 11 is used to acquire a top cover fixing assembly in a hydro-generator set, wherein the top cover fixing assembly includes bolts and a seat ring, and the seat ring includes a plurality of fixed guide vanes;

[0075] A multi-dimensional operation monitoring module 12 is used to read predetermined operation indicators and perform multi-dimensional operation monitoring of the hydro-generator set based on the predetermined operation indicators to obtain real-time operation information, wherein the hydro-generator set has an identifier of the unit's deadweight;

[0076] a first fatigue damage prediction constraint acquisition module 13, configured to activate a water pressure prediction model to analyze the real-time operation information to obtain a predicted water pressure, and to obtain a first fatigue damage prediction constraint for the bolt in combination with the preload force data of the bolt;

[0077] An operation log record forming module 14, wherein the operation log record forming module 14 is configured to form an operation log record of the hydro-generator set based on the real-time operation information, wherein the operation log record includes a first historical record;

[0078] a second fatigue damage prediction constraint acquisition module 15, configured to analyze a first historical pressure pulsation, a first historical Karman vortex frequency, a first historical fixed guide vane frequency, and the deadweight of the turbine in the first historical records to obtain second fatigue damage prediction constraints for the plurality of fixed guide vanes;

[0079] The fatigue damage prediction result acquisition module 16 is used to activate the fatigue damage prediction model to analyze the first fatigue damage prediction constraint and the second fatigue damage prediction constraint to obtain the fatigue damage prediction result of the top cover fixing assembly.

[0080] Furthermore, the first fatigue damage prediction constraint acquisition module 13 is further configured to perform the following steps:

[0081] The real-time speed, real-time load, and real-time pressure extracted from the real-time operation information are used as input information of the water pressure prediction model to obtain output information of the water pressure prediction model, wherein the output information includes the predicted water pressure. The water pressure prediction model is an intelligent model obtained by supervised learning and verification of the water pressure test data group.

[0082] Furthermore, the second fatigue damage prediction constraint acquisition module 15 is further configured to perform the following steps:

[0083] determining whether the first historical Karman vortex frequency is coupled to the first historical fixed guide vane frequency;

[0084] If it is coupled, generate a first unit historical coupling period, and add the first unit historical coupling period to obtain the total historical coupling period length;

[0085] comparing the first historical pressure pulsation with a predetermined pressure pulsation to obtain a first pressure pulsation comparison deviation;

[0086] Adding the first pressure pulsation comparison deviation to obtain a total historical pressure pulsation deviation;

[0087] The deadweight of the unit, the total historical coupling period, and the total historical pressure pulsation deviation are used as the second fatigue damage prediction constraints.

[0088] Furthermore, the fatigue damage prediction result acquisition module 16 is further configured to perform the following steps:

[0089] Step a: The fatigue damage prediction model includes a bolt fatigue damage prediction channel and a fixed guide vane fatigue damage prediction channel;

[0090] Step b: analyzing the first fatigue damage prediction constraint through the bolt fatigue damage prediction channel to obtain a bolt fatigue damage prediction index;

[0091] Step c: analyzing the second fatigue damage prediction constraint through the fixed guide vane fatigue damage prediction channel to obtain a fixed guide vane fatigue damage prediction index;

[0092] Step d1: If the bolt fatigue damage prediction index is greater than the fixed guide vane fatigue damage prediction index, the bolt fatigue damage prediction index is used as the fatigue damage prediction result;

[0093] Step d2: If the bolt fatigue damage prediction index is less than the fixed guide vane fatigue damage prediction index, the fixed guide vane fatigue damage prediction index is used as the fatigue damage prediction result.

[0094] Furthermore, the fatigue damage prediction result acquisition module 16 is further configured to perform the following steps:

[0095] A fixed guide vane fatigue test data group is established based on the big data, and a first data group is extracted from the fixed guide vane fatigue test data group, wherein the first data group includes a first unit deadweight, a first total coupling period, a first total pressure pulsation deviation, and a first fatigue damage detection data;

[0096] Calculating the normalized first fatigue damage detection data using the coefficient of variation principle to obtain a first fixed guide vane fatigue damage index;

[0097] The fixed guide vane fatigue damage prediction channel is obtained by performing supervised learning and verification on the deadweight of the first unit, the first total coupling period, the first total pressure pulsation deviation, and the first fixed guide vane fatigue damage index.

[0098] Furthermore, the system further comprises:

[0099] A loss analysis module is used to call a predetermined loss analysis function to perform loss analysis on the fatigue damage prediction result to obtain a loss analysis result. The expression of the predetermined loss analysis function is as follows:

[0100]

[0101] Wherein, L(x, x′) refers to the predetermined loss analysis function between the fatigue damage prediction result x and the loss analysis result x′, x i Refers to the fatigue damage prediction result of the top cover fixing component at the i-th moment in the n-moment prediction of the fatigue damage prediction result x, x i ′ refers to the fatigue damage prediction loss analysis result at the i-th moment in the loss analysis result x′, α refers to the ambient temperature feedback adjustment coefficient, β refers to the ambient humidity feedback adjustment coefficient, and γ refers to the working condition feedback adjustment coefficient.

[0102] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0103] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

[0104] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A fatigue damage prediction method for a hydro-generator set, characterized in that: include: Obtain a top cover fixing assembly in a hydro-generator set, wherein the top cover fixing assembly includes bolts and a seat ring, and the seat ring includes a plurality of fixed guide vanes; Reading predetermined operating indicators, and performing multi-dimensional operating monitoring on the hydro-generator set based on the predetermined operating indicators to obtain real-time operating information, wherein the hydro-generator set has an identifier of the unit's deadweight; activating a water pressure prediction model to analyze the real-time operation information to obtain a predicted water pressure, and combining the preload force data of the bolt to obtain a first fatigue damage prediction constraint of the bolt; Establishing an operation log record of the hydro-generator set based on the real-time operation information, wherein the operation log record includes a first historical record; Analyzing a first historical pressure pulsation, a first historical Karman vortex frequency, a first historical fixed guide vane frequency, and the deadweight of the unit in the first historical records to obtain a second fatigue damage prediction constraint for the plurality of fixed guide vanes; activating a fatigue damage prediction model to analyze the first fatigue damage prediction constraint and the second fatigue damage prediction constraint to obtain a fatigue damage prediction result of the top cover fixing assembly; The process of obtaining the predicted water pressure includes: using the real-time speed, real-time load, and real-time pressure extracted from the real-time operation information as input information of the water pressure prediction model, obtaining output information of the water pressure prediction model, wherein the output information includes the predicted water pressure, and the water pressure prediction model is an intelligent model obtained by supervised learning and verification of a water pressure test data set; The process of obtaining the second fatigue damage prediction constraint includes: determining whether the first historical Karman vortex frequency is coupled to the first historical fixed guide vane frequency; If it is coupled, generate a first unit historical coupling period, and add the first unit historical coupling period to obtain the total historical coupling period length; comparing the first historical pressure pulsation with a predetermined pressure pulsation to obtain a first pressure pulsation comparison deviation; Adding the first pressure pulsation comparison deviation to obtain a total historical pressure pulsation deviation; The unit's own weight, the total historical coupling period, and the total historical pressure pulsation deviation are used as the second fatigue damage prediction constraint; The process of obtaining fatigue damage prediction results includes: Step a: The fatigue damage prediction model includes a bolt fatigue damage prediction channel and a fixed guide vane fatigue damage prediction channel; Step b: analyzing the first fatigue damage prediction constraint through the bolt fatigue damage prediction channel to obtain a bolt fatigue damage prediction index; Step c: analyzing the second fatigue damage prediction constraint through the fixed guide vane fatigue damage prediction channel to obtain a fixed guide vane fatigue damage prediction index; Step d1: If the bolt fatigue damage prediction index is greater than the fixed guide vane fatigue damage prediction index, the bolt fatigue damage prediction index is used as the fatigue damage prediction result; Step d2: If the bolt fatigue damage prediction index is less than the fixed guide vane fatigue damage prediction index, the fixed guide vane fatigue damage prediction index is used as the fatigue damage prediction result.

2. A fatigue damage prediction method for a hydro-generator set according to claim 1, characterized in that: The predetermined operating indicators include speed, load, pressure, pressure pulsation, Karman vortex frequency, and fixed guide vane frequency.

3. The fatigue damage prediction method for a hydro-generator set according to claim 1, characterized in that: The construction process of the fixed guide vane fatigue damage prediction channel includes: A fixed guide vane fatigue test data group is established based on the big data, and a first data group is extracted from the fixed guide vane fatigue test data group, wherein the first data group includes a first unit deadweight, a first total coupling period, a first total pressure pulsation deviation, and a first fatigue damage detection data; Calculating the normalized first fatigue damage detection data using the coefficient of variation principle to obtain a first fixed guide vane fatigue damage index; The fixed guide vane fatigue damage prediction channel is obtained by performing supervised learning and verification on the deadweight of the first unit, the first total coupling period, the first total pressure pulsation deviation, and the first fixed guide vane fatigue damage index.

4. A fatigue damage prediction method for a hydro-generator set according to claim 3, characterized in that: The first fatigue damage detection data includes a first fixed guide vane stress and a first seat annular variable.

5. The fatigue damage prediction method for a hydro-generator set according to claim 1, characterized in that: Also includes: Calling a predetermined loss analysis function to perform loss analysis on the fatigue damage prediction result to obtain a loss analysis result. The expression of the predetermined loss analysis function is as follows: ； in, Refers to the fatigue damage prediction results The loss analysis results The predetermined loss analysis function between Refers to the fatigue damage prediction results in The first moment in the forecast The fatigue damage prediction results of the top cover fixing assembly at the moment, Refers to the loss analysis results The The fatigue damage prediction loss analysis results at each moment are: Refers to the ambient temperature feedback adjustment coefficient, Refers to the ambient humidity feedback adjustment coefficient, Refers to the operating condition feedback adjustment coefficient.

6. A fatigue damage prediction system for a hydro-generator set, characterized in that: The system is used to perform the method according to any one of claims 1 to 5, and the system includes: A top cover fixing assembly acquisition module, wherein the top cover fixing assembly acquisition module is used to acquire a top cover fixing assembly in a hydro-generator set, wherein the top cover fixing assembly includes bolts and a seat ring, and the seat ring includes a plurality of fixed guide vanes; a multi-dimensional operation monitoring module, the multi-dimensional operation monitoring module being used to read predetermined operation indicators and perform multi-dimensional operation monitoring of the hydro-generator set based on the predetermined operation indicators to obtain real-time operation information, wherein the hydro-generator set has an identifier of the unit's deadweight; a first fatigue damage prediction constraint acquisition module, configured to activate a water pressure prediction model to analyze the real-time operation information to obtain a predicted water pressure, and to obtain a first fatigue damage prediction constraint for the bolt in combination with the preload force data of the bolt; An operation log record forming module, wherein the operation log record forming module is used to form an operation log record of the hydro-generator set based on the real-time operation information, wherein the operation log record includes a first historical record; a second fatigue damage prediction constraint acquisition module, configured to analyze a first historical pressure pulsation, a first historical Karman vortex frequency, a first historical fixed guide vane frequency, and the deadweight of the turbine in the first historical records to obtain second fatigue damage prediction constraints for the plurality of fixed guide vanes; A fatigue damage prediction result acquisition module is used to activate the fatigue damage prediction model to analyze the first fatigue damage prediction constraint and the second fatigue damage prediction constraint to obtain a fatigue damage prediction result of the top cover fixing assembly.

Images