Water turbine safety management method and system based on digital twinning

Abstract

A water turbine safety management method and system based on digital twinning. The method comprises: obtaining a three-dimensional static scanning video of a current pumped-storage hydropower station, and a three-dimensional dynamic scanning video and a water turbine historical data set of a water turbine system (S101); on the basis of the three-dimensional static scanning video of the current pumped-storage hydropower station and the three-dimensional dynamic scanning video of the water turbine system, constructing a water turbine system safety management three-dimensional dynamic model (S102); on the basis of the water turbine historical data set, matching a corresponding fault diagnosis algorithm, and on the basis of the matched fault diagnosis algorithm, inputting the water turbine historical data set for optimization training, and constructing a water turbine system fault diagnosis model (S103); obtaining real-time water turbine data, and inputting the real-time water turbine data of the water turbine system into the water turbine system fault diagnosis model to obtain water turbine fault information (S104); and performing safety management on a water turbine on the basis of the water turbine fault information (S105).

Description

A hydraulic turbine safety management method and system based on digital twin Technical Field

[0001] The present application relates to the field of hydraulic turbine safety management, and specifically, to a hydraulic turbine safety management method and system based on digital twins. Background Art

[0002] Petri nets are a widely used technology for discrete event modeling and analysis. They primarily include process-oriented Petri nets (POPN) and resource-oriented Petri nets (ROPN). ROPN models are smaller than POPN models and can more effectively analyze issues such as deadlock and liveness in discrete event systems. They are relatively closer to practical applications. Electricity is a key energy source for industrial production and daily life. With the dwindling availability of non-renewable resources, the advantages of hydropower are becoming increasingly apparent. Pumped-storage hydropower stations, with their rapid startup and shutdown capabilities, also play a role in peak load regulation, frequency regulation, phase regulation, and emergency backup within power systems. Pumped-storage hydropower stations utilize turbines to convert the impact of water flow into power for generators, providing an environmentally friendly and clean way to generate electricity. As a key component of pumped-storage hydropower stations, turbines play a critical role in their normal operation. Therefore, safe management of turbine systems and real-time monitoring of sudden failures or operating status are crucial to the normal and stable operation of pumped-storage hydropower stations.

[0003] In the existing safety management of the turbine system in the pumped storage hydropower station, the turbine system is monitored by analyzing real-time data. This management method is not intuitive enough and lacks overall control of the turbine system. In addition, the existing safety management method cannot provide reference guidance for the control of the turbine system, has weak functionality, and cannot adapt to existing needs. In response to the above problems, effective technical solutions are urgently needed.

[0004] Summary of the Invention

[0005] The purpose of the embodiments of the present application is to provide a turbine safety management method and system based on digital twins. By establishing a three-dimensional dynamic model for turbine system safety management, it is convenient for staff to understand the overall layout of the turbine system of a pumped storage hydropower station, thereby improving the overall control of the turbine system. A visual interface is provided to display the actual situation of the turbine system of the current pumped storage hydropower station in real time, greatly improving the intuitiveness of observation. Real-time fault diagnosis is performed through the turbine system fault diagnosis model, thereby improving the safety of the turbine system and the functionality of the method.

[0006] The present application also provides a hydraulic turbine safety management method based on digital twins, including:

[0007] Obtain the current 3D static scanning video of the pumped storage hydropower station, the 3D dynamic scanning video of the turbine system, and the turbine history dataset;

[0008] Based on the current 3D static scanning video of the pumped storage hydropower station and the 3D dynamic scanning video of the turbine system, a 3D dynamic model for turbine system safety management is constructed;

[0009] According to the turbine historical data set, the corresponding fault diagnosis algorithm is matched, and the turbine historical data set is input into the matched fault diagnosis algorithm for optimization training to build a turbine system fault diagnosis model;

[0010] Acquire real-time data of the turbine, and input the real-time data of the turbine into the fault diagnosis model of the turbine system to obtain turbine fault information;

[0011] Carry out safety management of turbines based on turbine fault information.

[0012] Optionally, in the digital twin-based turbine safety management method described in the embodiment of the present application, a three-dimensional dynamic model for turbine system safety management is constructed based on the current three-dimensional static scanning video of the pumped storage hydropower station and the three-dimensional dynamic scanning video of the turbine system, including the following steps:

[0013] Constructing a 3D static model of a pumped storage hydropower station based on the current 3D static scanning video of the pumped storage hydropower station;

[0014] Based on the corresponding turbine system installation drawings and 3D dynamic scanning videos, use digital twin technology to build a 3D dynamic model of the turbine system;

[0015] The three-dimensional dynamic model of the turbine system is added to the three-dimensional static model of the pumped storage hydropower station to obtain the three-dimensional dynamic model of the turbine system safety management.

[0016] Optionally, in the digital twin-based turbine safety management method described in the embodiment of the present application, the fault diagnosis algorithm includes one of an artificial neural network algorithm, a deep learning algorithm, a fuzzy reasoning algorithm, and a fault tree algorithm.

[0017] Optionally, in the digital twin-based hydraulic turbine safety management method described in an embodiment of the present application, matching a corresponding fault diagnosis algorithm based on a hydraulic turbine historical data set, and inputting the hydraulic turbine historical data set for optimization training based on the matched fault diagnosis algorithm to construct a hydraulic turbine system fault diagnosis model includes the following steps:

[0018] Match the corresponding fault diagnosis algorithm based on the historical data set of the turbine system;

[0019] Using the matching fault diagnosis algorithm, the original fault diagnosis model of the turbine system is constructed;

[0020] Performing data preprocessing on a turbine history data set of a turbine system to obtain a preprocessed turbine fault diagnosis sample set, and dividing the turbine fault diagnosis sample set into a turbine fault diagnosis training sample set and a turbine fault diagnosis test sample set;

[0021] Inputting the hydraulic turbine fault diagnosis training sample set into the original hydraulic turbine system fault diagnosis model for optimization training, and obtaining the optimized hydraulic turbine system fault diagnosis model;

[0022] The turbine fault diagnosis test sample set is input into the optimized turbine system fault diagnosis model for testing, and the test accuracy is obtained;

[0023] If the test accuracy is greater than the preset threshold, the final turbine system fault diagnosis model is output;

[0024] If the test accuracy is less than the preset threshold, optimization training will be performed again.

[0025] Optionally, in the hydraulic turbine safety management method based on digital twins described in the embodiment of the present application, after obtaining the three-dimensional static scanning video of the current pumped storage hydropower station, the three-dimensional dynamic scanning video of the hydraulic turbine system, and the hydraulic turbine history data set, the method further includes:

[0026] Construct an original 3D static model of the pumped storage hydropower station based on the current 3D static scanning video of the pumped storage hydropower station;

[0027] Obtaining a first perspective center of the original 3D static model of the pumped storage hydropower station, and performing perspective alignment on at least three frames of 3D static scan images in a 3D static scan video of the pumped storage hydropower station based on the first perspective center to obtain at least three frames of perspective-aligned 3D static scan images;

[0028] The original three-dimensional static model of the pumped storage hydropower station is corrected according to at least three frames of three-dimensional static scanned images after the view angles are aligned, so as to obtain a final three-dimensional static model of the pumped storage hydropower station.

[0029] Optionally, in the hydraulic turbine safety management method based on digital twins described in an embodiment of the present application, real-time hydraulic turbine data is obtained, and the real-time hydraulic turbine data of the hydraulic turbine system is input into a hydraulic turbine system fault diagnosis model to obtain hydraulic turbine fault information; and safety management of the hydraulic turbine is performed based on the hydraulic turbine fault information, including the following steps:

[0030] Acquire the real-time data of the turbine of the current pumped storage hydropower station's turbine system and store the real-time data in the turbine system safety management database;

[0031] Based on the real-time data of the turbine, the turbine system fault diagnosis model is used to diagnose the turbine system fault and obtain the turbine system fault diagnosis result;

[0032] Obtaining a control scheme for the turbine system based on the turbine system fault diagnosis results;

[0033] Adjust turbine operating data according to the turbine system control plan and conduct safety management of the turbine system of the pumped storage hydropower station.

[0034] In a second aspect, an embodiment of the present application provides a hydraulic turbine safety management system based on digital twins, the system comprising: a memory and a processor, the memory comprising a program for a hydraulic turbine safety management method based on digital twins, the program for the hydraulic turbine safety management method based on digital twins, when executed by the processor, implementing the following steps:

[0035] Obtain the current 3D static scanning video of the pumped storage hydropower station, the 3D dynamic scanning video of the turbine system, and the turbine history dataset;

[0036] Based on the current 3D static scanning video of the pumped storage hydropower station and the 3D dynamic scanning video of the turbine system, a 3D dynamic model for turbine system safety management is constructed;

[0037] According to the turbine historical data set, the corresponding fault diagnosis algorithm is matched, and the turbine historical data set is input into the matched fault diagnosis algorithm for optimization training to build a turbine system fault diagnosis model;

[0038] Acquire real-time data of the turbine, and input the real-time data of the turbine into the fault diagnosis model of the turbine system to obtain turbine fault information;

[0039] Carry out safety management of turbines based on turbine fault information.

[0040] Optionally, in the digital twin-based turbine safety management system described in the embodiment of the present application, a three-dimensional dynamic model for turbine system safety management is constructed based on the current three-dimensional static scanning video of the pumped storage hydropower station and the three-dimensional dynamic scanning video of the turbine system, including the following steps:

[0041] Constructing a 3D static model of a pumped storage hydropower station based on the current 3D static scanning video of the pumped storage hydropower station;

[0042] Based on the corresponding turbine system installation drawings and 3D dynamic scanning videos, use digital twin technology to build a 3D dynamic model of the turbine system;

[0043] The three-dimensional dynamic model of the turbine system is added to the three-dimensional static model of the pumped storage hydropower station to obtain the three-dimensional dynamic model of the turbine system safety management.

[0044] Optionally, in the digital twin-based hydraulic turbine safety management system described in the embodiment of the present application, matching a corresponding fault diagnosis algorithm based on a hydraulic turbine historical data set, and inputting the hydraulic turbine historical data set for optimization training based on the matched fault diagnosis algorithm to construct a hydraulic turbine system fault diagnosis model includes the following steps:

[0045] Match the corresponding fault diagnosis algorithm based on the historical data set of the turbine system;

[0046] Using the matching fault diagnosis algorithm, the original fault diagnosis model of the turbine system is constructed;

[0047] Performing data preprocessing on a turbine history data set of a turbine system to obtain a preprocessed turbine fault diagnosis sample set, and dividing the turbine fault diagnosis sample set into a turbine fault diagnosis training sample set and a turbine fault diagnosis test sample set;

[0048] Inputting the hydraulic turbine fault diagnosis training sample set into the original hydraulic turbine system fault diagnosis model for optimization training, and obtaining the optimized hydraulic turbine system fault diagnosis model;

[0049] The turbine fault diagnosis test sample set is input into the optimized turbine system fault diagnosis model for testing, and the test accuracy is obtained;

[0050] If the test accuracy is greater than the preset threshold, the final turbine system fault diagnosis model is output;

[0051] If the test accuracy is less than the preset threshold, optimization training will be performed again.

[0052] Optionally, in the digital twin-based hydraulic turbine safety management system described in the embodiment of the present application, after obtaining the three-dimensional static scanning video of the current pumped storage hydropower station, the three-dimensional dynamic scanning video of the hydraulic turbine system, and the hydraulic turbine history data set, the system further includes:

[0053] Construct an original 3D static model of the pumped storage hydropower station based on the current 3D static scanning video of the pumped storage hydropower station;

[0054] Obtaining a first perspective center of the original 3D static model of the pumped storage hydropower station, and performing perspective alignment on at least three frames of 3D static scan images in a 3D static scan video of the pumped storage hydropower station based on the first perspective center to obtain at least three frames of perspective-aligned 3D static scan images;

[0055] The original three-dimensional static model of the pumped storage hydropower station is corrected according to at least three frames of three-dimensional static scanned images after the view angles are aligned, so as to obtain a final three-dimensional static model of the pumped storage hydropower station.

[0056] From the above, it can be seen that the embodiment of the present application provides a hydraulic turbine safety management method and system based on digital twin, which obtains the current 3D static scanning video of the pumped storage hydropower station, the 3D dynamic scanning video of the hydraulic turbine system and the hydraulic turbine historical data set; constructs a 3D dynamic model for hydraulic turbine system safety management based on the current 3D static scanning video of the hydraulic storage hydropower station and the 3D dynamic scanning video of the hydraulic turbine system; matches the corresponding fault diagnosis algorithm based on the hydraulic turbine historical data set, and inputs the hydraulic turbine historical data set for optimization training based on the matched fault diagnosis algorithm to construct a hydraulic turbine system fault diagnosis model; obtains real-time data of the hydraulic turbine, and converts The real-time data of the turbine of the turbine system is input into the fault diagnosis model of the turbine system to obtain the turbine fault information; the turbine is safely managed according to the turbine fault information; by establishing a three-dimensional dynamic model of the safety management of the turbine system, it is convenient for the staff to understand the overall layout of the turbine system of the pumped storage hydropower station, and improve the overall control of the turbine system. In addition, a visual interface is provided to display the actual situation of the turbine system of the current pumped storage hydropower station in real time, which greatly improves the intuitiveness of observation. Real-time fault diagnosis is performed through the turbine system fault diagnosis model, which improves the safety of the turbine system and the functionality of the method.

[0057] Other features and advantages of the present application will be described in the following description, and the advantages of the present application can be inferred from the description or understood by practicing the embodiments of the present application. The objectives and other advantages of the present application can be achieved and obtained through the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0059] FIG1 is a flow chart of a hydraulic turbine safety management method based on digital twins provided in an embodiment of the present application;

[0060] FIG2 is a flowchart of constructing a three-dimensional dynamic model for water turbine system safety management according to a water turbine safety management method based on digital twins provided in an embodiment of the present application;

[0061] FIG3 is a flowchart of a fault diagnosis model optimization method for a hydraulic turbine safety management method based on digital twins according to an embodiment of the present application;

[0062] Figure 4 is a structural diagram of a digital twin-based turbine safety management system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work fall within the scope of protection of the present application.

[0064] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0065] Please refer to Figure 1, which is a flow chart of a hydraulic turbine safety management method based on digital twins in some embodiments of the present application. The hydraulic turbine safety management method based on digital twins is used in a terminal device and includes the following steps:

[0066] S101, obtaining a 3D static scanning video of the current pumped storage hydropower station, a 3D dynamic scanning video of the turbine system, and a turbine history dataset;

[0067] S102, constructing a three-dimensional dynamic model for safety management of the turbine system based on the current three-dimensional static scanning video of the pumped storage hydropower station and the three-dimensional dynamic scanning video of the turbine system;

[0068] S103, matching a corresponding fault diagnosis algorithm based on the turbine historical data set, and inputting the turbine historical data set into the matched fault diagnosis algorithm for optimization training to construct a turbine system fault diagnosis model;

[0069] S104, acquiring real-time data of the turbine, inputting the real-time data of the turbine into a fault diagnosis model of the turbine system to obtain turbine fault information;

[0070] S105: Perform safety management on the turbine according to the turbine fault information.

[0071] It should be noted that the current pumped-storage hydropower station's architectural drawings and 3D static scan videos, as well as the corresponding turbine system's installation drawings, 3D dynamic scan videos, and turbine historical datasets, are obtained. Based on these drawings and 3D static scan videos, as well as the corresponding turbine system's installation drawings and 3D dynamic scan videos, a 3D dynamic model for turbine system safety management is constructed using digital twin technology.

[0072] Furthermore, a hydraulic turbine system safety management database is constructed based on the hydraulic turbine historical data set of the hydraulic turbine system, and several fault diagnosis algorithms are set in the hydraulic turbine system safety management database;

[0073] Furthermore, based on the turbine system safety management database, a corresponding fault diagnosis algorithm is matched according to the turbine historical data set of the turbine system, and the turbine historical data set is input into the matched fault diagnosis algorithm for optimization training to construct a turbine system fault diagnosis model;

[0074] In addition, based on the turbine system safety management database and the turbine history data set of the turbine system, the ROPN modeling method is used to construct the turbine system control ROPN model;

[0075] Specifically, based on the turbine system safety management database and the turbine history data set of the turbine system, the ROPN modeling method is used to construct a turbine system control ROPN model, which includes the following steps:

[0076] Based on the turbine system safety management database and the control objectives of the turbine system, the original ROPN model is constructed;

[0077] According to the data type of any hydraulic turbine historical data in the hydraulic turbine historical data set of the hydraulic turbine system, ROPN element information is obtained;

[0078] According to the ROPN element information, the original ROPN model is optimized and trained to obtain the turbine system control ROPN model.

[0079] Specifically, the ROPN element information includes a first place for representing each data type, a second place for generating a control scheme for the hydro-turbine system, a transition, and a token.

[0080] Based on the three-dimensional dynamic model of the hydraulic turbine system safety management, a visualization interface is set, and a link is set between the visualization interface and a hydraulic turbine system safety management database, thereby obtaining the three-dimensional dynamic model of the hydraulic turbine system safety management provided with the visualization interface;

[0081] In this way, by obtaining the real-time data of the turbine of the turbine system of the current pumped storage hydropower station, safety management can be performed using a three-dimensional dynamic model of turbine system safety management provided with a visual interface based on the real-time data of the turbine of the turbine system.

[0082] Furthermore, the turbine historical data includes but is not limited to turbine control data, turbine speed regulation data, turbine power supply data, number of turbines, turbine model, turbine location, turbine installed capacity, turbine power generation, turbine power, utilization rate, operation plan, maintenance plan, total power generation, control plan and fault data of the turbine system of the current pumped storage hydropower station.

[0083] Please refer to Figure 2, which is a flowchart of constructing a three-dimensional dynamic model for turbine system safety management in a digital twin-based turbine safety management method according to some embodiments of the present application. According to an embodiment of the present invention, constructing a three-dimensional dynamic model for turbine system safety management based on a three-dimensional static scan video of the current pumped storage hydropower station and a three-dimensional dynamic scan video of the turbine system includes the following steps:

[0084] S201, constructing a 3D static model of the pumped storage hydropower station based on the current 3D static scanning video of the pumped storage hydropower station;

[0085] S202, constructing a three-dimensional dynamic model of the turbine system using digital twin technology based on the corresponding turbine system installation drawings and three-dimensional dynamic scanning video;

[0086] S203: Add the three-dimensional dynamic model of the turbine system to the three-dimensional static model of the pumped storage hydropower station to obtain a three-dimensional dynamic model of the turbine system safety management.

[0087] According to an embodiment of the present invention, the fault diagnosis algorithm includes one of an artificial neural network algorithm, a deep learning algorithm, a fuzzy reasoning algorithm, and a fault tree algorithm. The above algorithms are commonly used algorithms that can be used by those skilled in the art, and the present invention will not elaborate on them one by one.

[0088] Please refer to Figure 3, which is a flowchart of a fault diagnosis model optimization process for a hydraulic turbine safety management method based on digital twins in some embodiments of the present application. According to an embodiment of the present invention, a corresponding fault diagnosis algorithm is matched based on a hydraulic turbine historical data set, and the hydraulic turbine historical data set is input for optimization training based on the matched fault diagnosis algorithm to construct a hydraulic turbine system fault diagnosis model, including the following steps:

[0089] S301, matching a corresponding fault diagnosis algorithm based on a historical data set of a hydraulic turbine of the hydraulic turbine system;

[0090] S302, using a matching fault diagnosis algorithm to construct an original hydraulic turbine system fault diagnosis model;

[0091] S303, performing data preprocessing on a historical data set of a hydraulic turbine of the hydraulic turbine system to obtain a preprocessed hydraulic turbine fault diagnosis sample set, and dividing the hydraulic turbine fault diagnosis sample set into a hydraulic turbine fault diagnosis training sample set and a hydraulic turbine fault diagnosis test sample set;

[0092] S304, inputting the turbine fault diagnosis training sample set into the original turbine system fault diagnosis model for optimization training to obtain an optimized turbine system fault diagnosis model;

[0093] S305, input the turbine fault diagnosis test sample set into the optimized turbine system fault diagnosis model for testing, and obtain the test accuracy; if the test accuracy is greater than the preset threshold, output the final turbine system fault diagnosis model; if the test accuracy is less than the preset threshold, re-optimization training is performed.

[0094] According to an embodiment of the present invention, after obtaining the 3D static scanning video of the current pumped storage hydropower station, the 3D dynamic scanning video of the turbine system, and the turbine history data set, the method further includes:

[0095] Construct an original 3D static model of the pumped storage hydropower station based on the current 3D static scanning video of the pumped storage hydropower station;

[0096] Obtaining a first perspective center of the original 3D static model of the pumped storage hydropower station, and performing perspective alignment on at least three frames of 3D static scan images in a 3D static scan video of the pumped storage hydropower station based on the first perspective center to obtain at least three frames of perspective-aligned 3D static scan images;

[0097] The original three-dimensional static model of the pumped storage hydropower station is corrected according to at least three frames of three-dimensional static scanned images after the view angles are aligned, so as to obtain a final three-dimensional static model of the pumped storage hydropower station.

[0098] According to an embodiment of the present invention, real-time turbine data is acquired, and the real-time turbine data of the turbine system is input into a turbine system fault diagnosis model to obtain turbine fault information; and safety management of the turbine is performed based on the turbine fault information, including the following steps:

[0099] Acquire the real-time data of the turbine of the current pumped storage hydropower station's turbine system and store the real-time data in the turbine system safety management database;

[0100] Based on the real-time data of the turbine, the turbine system fault diagnosis model is used to diagnose the turbine system fault and obtain the turbine system fault diagnosis result;

[0101] Obtaining a control scheme for the turbine system based on the turbine system fault diagnosis results;

[0102] Adjust turbine operating data according to the turbine system control plan and conduct safety management of the turbine system of the pumped storage hydropower station.

[0103] It should be noted that the real-time data of the turbine includes but is not limited to the turbine control data, turbine speed regulation data, turbine power supply data, number of turbines, turbine model, turbine location, turbine installed capacity, turbine power generation, turbine power and utilization rate of the turbine system of the current pumped storage hydropower station.

[0104] According to an embodiment of the present invention, the method further includes constructing a three-dimensional dynamic model of the turbine system using digital twin technology based on the corresponding installation drawings and three-dimensional dynamic scanning video of the turbine system, including the following steps:

[0105] Construct a three-dimensional static model of the turbine system according to the corresponding installation drawings of the turbine system;

[0106] Obtaining a second perspective center of the three-dimensional static model of the turbine system, and performing perspective alignment on the second perspective center according to the first perspective center of the original three-dimensional static model of the pumped storage hydropower station to obtain the aligned second perspective center;

[0107] Performing perspective alignment on at least three frames of three-dimensional dynamic scanning images in the three-dimensional dynamic scanning video of the turbine system according to the second perspective center after perspective alignment, to obtain at least three frames of three-dimensional dynamic scanning images after perspective alignment;

[0108] Correcting the three-dimensional static model of the hydraulic turbine system based on at least three frames of three-dimensional dynamic scanning images after view angle alignment to obtain a three-dimensional dynamic model of the hydraulic turbine system;

[0109] The 3D dynamic model of the turbine system is added to the 3D static model of the pumped storage hydropower station to obtain the 3D dynamic model of the turbine system safety management, including the following steps:

[0110] According to the current architectural drawings of the pumped storage hydropower station and the corresponding installation drawings of the turbine system, the installation position of the turbine system in the pumped storage hydropower station is obtained;

[0111] According to the installation position of the turbine system in the pumped storage hydropower station, the 3D dynamic model of the turbine system is added to the corresponding position of the final 3D static model of the pumped storage hydropower station to obtain the original 3D dynamic model of the turbine system safety management;

[0112] According to the first perspective center of the original 3D static model of the pumped storage hydropower station, the perspective alignment of the original 3D dynamic model of the turbine system safety management is performed to obtain the final 3D dynamic model of the turbine system safety management.

[0113] It should be noted that a turbine system safety management database is generated based on the turbine system safety management three-dimensional dynamic model, and turbine system fault diagnosis is performed based on turbine real-time data using the turbine system fault diagnosis model to obtain turbine system fault diagnosis results.

[0114] Based on the turbine system safety management database and the real-time turbine data, the turbine system control ROPN model is used to predict the turbine system control and obtain the turbine system control plan;

[0115] Based on the link between the visualization interface and the turbine system safety management database, the turbine system fault diagnosis results, turbine system control plan and turbine real-time data are displayed on the visualization interface;

[0116] Based on a three-dimensional dynamic model of turbine system safety management with a visual interface, the turbine system of the current pumped storage hydropower station is safely managed according to the turbine system fault diagnosis results, turbine system control scheme and real-time turbine data displayed on the visual interface.

[0117] Please refer to Figure 4, which is a schematic diagram of the structure of a hydraulic turbine safety management system based on digital twins in some embodiments of the present application. In a second aspect, the embodiments of the present application provide a hydraulic turbine safety management system 4 based on digital twins, which includes: a memory 41 and a processor 42. The memory 41 includes a program for a hydraulic turbine safety management method based on digital twins. When the program for the hydraulic turbine safety management method based on digital twins is executed by the processor, the following steps are implemented:

[0118] Obtain the current 3D static scanning video of the pumped storage hydropower station, the 3D dynamic scanning video of the turbine system, and the turbine history dataset;

[0119] Based on the current 3D static scanning video of the pumped storage hydropower station and the 3D dynamic scanning video of the turbine system, a 3D dynamic model for turbine system safety management is constructed;

[0120] According to the turbine historical data set, the corresponding fault diagnosis algorithm is matched, and the turbine historical data set is input into the matched fault diagnosis algorithm for optimization training to build a turbine system fault diagnosis model;

[0121] Acquire real-time data of the turbine, and input the real-time data of the turbine into the fault diagnosis model of the turbine system to obtain turbine fault information;

[0122] Carry out safety management of turbines based on turbine fault information.

[0123] It should be noted that the current pumped-storage hydropower station's architectural drawings and 3D static scan videos, as well as the corresponding turbine system's installation drawings, 3D dynamic scan videos, and turbine historical datasets, are obtained. Based on these drawings and 3D static scan videos, as well as the corresponding turbine system's installation drawings and 3D dynamic scan videos, a 3D dynamic model for turbine system safety management is constructed using digital twin technology.

[0124] Furthermore, a hydraulic turbine system safety management database is constructed based on the hydraulic turbine historical data set of the hydraulic turbine system, and several fault diagnosis algorithms are set in the hydraulic turbine system safety management database;

[0125] Furthermore, based on the turbine system safety management database, a corresponding fault diagnosis algorithm is matched according to the turbine historical data set of the turbine system, and the turbine historical data set is input into the matched fault diagnosis algorithm for optimization training to construct a turbine system fault diagnosis model;

[0126] In addition, based on the turbine system safety management database and the turbine history data set of the turbine system, the ROPN modeling method is used to construct the turbine system control ROPN model;

[0127] Specifically, based on the turbine system safety management database and the turbine history data set of the turbine system, the ROPN modeling method is used to construct a turbine system control ROPN model, which includes the following steps:

[0128] Based on the turbine system safety management database and the control objectives of the turbine system, the original ROPN model is constructed;

[0129] Obtaining ROPN element information according to the data type of any turbine history data in the turbine history data set of the turbine system;

[0130] According to the ROPN element information, the original ROPN model is optimized and trained to obtain the turbine system control ROPN model.

[0131] Specifically, the ROPN element information includes a first place for representing each data type, a second place for generating a control scheme for the hydro-turbine system, a transition, and a token.

[0132] Based on the three-dimensional dynamic model of the hydraulic turbine system safety management, a visualization interface is set, and a link is set between the visualization interface and a hydraulic turbine system safety management database, thereby obtaining the three-dimensional dynamic model of the hydraulic turbine system safety management provided with the visualization interface;

[0133] In this way, by obtaining the real-time data of the turbine of the turbine system of the current pumped storage hydropower station, safety management can be performed using a three-dimensional dynamic model of turbine system safety management provided with a visual interface based on the real-time data of the turbine of the turbine system.

[0134] Furthermore, the turbine historical data includes but is not limited to turbine control data, turbine speed regulation data, turbine power supply data, number of turbines, turbine model, turbine location, turbine installed capacity, turbine power generation, turbine power, utilization rate, operation plan, maintenance plan, total power generation, control plan and fault data of the turbine system of the current pumped storage hydropower station.

[0135] According to an embodiment of the present invention, a 3D dynamic model for safety management of a turbine system is constructed based on a 3D static scanning video of a current pumped storage hydropower station and a 3D dynamic scanning video of a turbine system, including the following steps:

[0136] Constructing a 3D static model of a pumped storage hydropower station based on the current 3D static scanning video of the pumped storage hydropower station;

[0137] Based on the corresponding turbine system installation drawings and 3D dynamic scanning videos, use digital twin technology to build a 3D dynamic model of the turbine system;

[0138] The three-dimensional dynamic model of the turbine system is added to the three-dimensional static model of the pumped storage hydropower station to obtain the three-dimensional dynamic model of the turbine system safety management.

[0139] According to an embodiment of the present invention, a corresponding fault diagnosis algorithm is matched based on a historical data set of a hydraulic turbine, and the matched fault diagnosis algorithm is input into the historical data set of the hydraulic turbine for optimization training to construct a hydraulic turbine system fault diagnosis model, including the following steps:

[0140] Match the corresponding fault diagnosis algorithm based on the historical data set of the turbine system;

[0141] Using the matching fault diagnosis algorithm, the original fault diagnosis model of the turbine system is constructed;

[0142] Performing data preprocessing on a turbine history data set of a turbine system to obtain a preprocessed turbine fault diagnosis sample set, and dividing the turbine fault diagnosis sample set into a turbine fault diagnosis training sample set and a turbine fault diagnosis test sample set;

[0143] Inputting the hydraulic turbine fault diagnosis training sample set into the original hydraulic turbine system fault diagnosis model for optimization training, and obtaining the optimized hydraulic turbine system fault diagnosis model;

[0144] The turbine fault diagnosis test sample set is input into the optimized turbine system fault diagnosis model for testing, and the test accuracy is obtained;

[0145] If the test accuracy is greater than the preset threshold, the final turbine system fault diagnosis model is output;

[0146] If the test accuracy is less than the preset threshold, optimization training will be performed again.

[0147] According to an embodiment of the present invention, after obtaining the 3D static scanning video of the current pumped storage hydropower station, the 3D dynamic scanning video of the turbine system, and the turbine history data set, the method further includes:

[0148] Construct an original 3D static model of the pumped storage hydropower station based on the current 3D static scanning video of the pumped storage hydropower station;

[0149] Obtaining a first perspective center of the original 3D static model of the pumped storage hydropower station, and performing perspective alignment on at least three frames of 3D static scan images in a 3D static scan video of the pumped storage hydropower station based on the first perspective center to obtain at least three frames of perspective-aligned 3D static scan images;

[0150] The original three-dimensional static model of the pumped storage hydropower station is corrected according to at least three frames of three-dimensional static scanned images after the view angles are aligned, so as to obtain a final three-dimensional static model of the pumped storage hydropower station.

[0151] According to an embodiment of the present invention, real-time turbine data is acquired, and the real-time turbine data of the turbine system is input into a turbine system fault diagnosis model to obtain turbine fault information; and safety management of the turbine is performed based on the turbine fault information, including the following steps:

[0152] Acquire the real-time data of the turbine of the current pumped storage hydropower station's turbine system and store the real-time data in the turbine system safety management database;

[0153] Based on the real-time data of the turbine, the turbine system fault diagnosis model is used to diagnose the turbine system fault and obtain the turbine system fault diagnosis result;

[0154] Obtaining a control scheme for the turbine system based on the turbine system fault diagnosis results;

[0155] Adjust turbine operating data according to the turbine system control plan and conduct safety management of the turbine system of the pumped storage hydropower station.

[0156] It should be noted that the real-time data of the turbine includes but is not limited to the turbine control data, turbine speed regulation data, turbine power supply data, number of turbines, turbine model, turbine location, turbine installed capacity, turbine power generation, turbine power and utilization rate of the turbine system of the current pumped storage hydropower station.

[0157] According to an embodiment of the present invention, the method further includes constructing a three-dimensional dynamic model of the turbine system using digital twin technology based on the corresponding installation drawings and three-dimensional dynamic scanning video of the turbine system, including the following steps:

[0158] Construct a three-dimensional static model of the turbine system according to the corresponding installation drawings of the turbine system;

[0159] Obtaining a second perspective center of the three-dimensional static model of the turbine system, and performing perspective alignment on the second perspective center according to the first perspective center of the original three-dimensional static model of the pumped storage hydropower station to obtain the aligned second perspective center;

[0160] Performing perspective alignment on at least three frames of three-dimensional dynamic scanning images in the three-dimensional dynamic scanning video of the turbine system according to the second perspective center after perspective alignment, to obtain at least three frames of three-dimensional dynamic scanning images after perspective alignment;

[0161] Correcting the three-dimensional static model of the hydraulic turbine system based on at least three frames of three-dimensional dynamic scanning images after view angle alignment to obtain a three-dimensional dynamic model of the hydraulic turbine system;

[0162] The 3D dynamic model of the turbine system is added to the 3D static model of the pumped storage hydropower station to obtain the 3D dynamic model of the turbine system safety management, including the following steps:

[0163] According to the current architectural drawings of the pumped storage hydropower station and the corresponding installation drawings of the turbine system, the installation position of the turbine system in the pumped storage hydropower station is obtained;

[0164] According to the installation position of the turbine system in the pumped storage hydropower station, the 3D dynamic model of the turbine system is added to the corresponding position of the final 3D static model of the pumped storage hydropower station to obtain the original 3D dynamic model of the turbine system safety management;

[0165] According to the first perspective center of the original 3D static model of the pumped storage hydropower station, the perspective alignment of the original 3D dynamic model of the turbine system safety management is performed to obtain the final 3D dynamic model of the turbine system safety management.

[0166] It should be noted that a turbine system safety management database is generated based on the turbine system safety management three-dimensional dynamic model, and turbine system fault diagnosis is performed based on turbine real-time data using the turbine system fault diagnosis model to obtain turbine system fault diagnosis results.

[0167] Based on the turbine system safety management database and the real-time turbine data, the turbine system control ROPN model is used to predict the turbine system control and obtain the turbine system control plan;

[0168] Based on the link between the visualization interface and the turbine system safety management database, the turbine system fault diagnosis results, turbine system control plan and turbine real-time data are displayed on the visualization interface;

[0169] Based on a three-dimensional dynamic model of turbine system safety management with a visual interface, the turbine system of the current pumped storage hydropower station is safely managed according to the turbine system fault diagnosis results, turbine system control scheme and real-time turbine data displayed on the visual interface.

[0170] The present invention discloses a hydraulic turbine safety management method and system based on digital twin, which obtains the current 3D static scanning video of the pumped storage hydropower station, the 3D dynamic scanning video of the hydraulic turbine system and the hydraulic turbine historical data set; constructs a 3D dynamic model of hydraulic turbine system safety management based on the current 3D static scanning video of the hydraulic storage hydropower station and the 3D dynamic scanning video of the hydraulic turbine system; matches the corresponding fault diagnosis algorithm based on the hydraulic turbine historical data set, and inputs the hydraulic turbine historical data set for optimization training based on the matched fault diagnosis algorithm to construct a hydraulic turbine system fault diagnosis model; obtains real-time data of the hydraulic turbine, and converts the hydraulic turbine system into a 3D dynamic model of hydraulic turbine system safety management; The real-time data of the turbine of the system is input into the turbine system fault diagnosis model to obtain the turbine fault information; the turbine is safely managed according to the turbine fault information; by establishing a three-dimensional dynamic model of turbine system safety management, it is convenient for staff to understand the overall layout of the turbine system of the pumped storage hydropower station, and improve the overall control of the turbine system. In addition, a visual interface is provided to display the actual situation of the turbine system of the current pumped storage hydropower station in real time, which greatly improves the intuitiveness of observation. Real-time fault diagnosis is performed through the turbine system fault diagnosis model, which improves the safety of the turbine system and the functionality of the method.

[0171] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0172] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0173] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0174] Those skilled in the art will understand that all or part of the steps of the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc. Various media that can store program codes.

[0175] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

Claims

1. A method for turbine safety management based on digital twin, characterized in that, it includes: Obtain the three-dimensional static scan video of the current pumped-storage power station, the three-dimensional dynamic scan video of the turbine system, and the turbine historical data set; Construct a three-dimensional dynamic model for turbine system safety management according to the three-dimensional static scan video of the current pumped-storage power station and the three-dimensional dynamic scan video of the turbine system; Match the corresponding fault diagnosis algorithm according to the turbine historical data set, and input the turbine historical data set according to the matched fault diagnosis algorithm for optimization training to construct a turbine system fault diagnosis model; Obtain the real-time turbine data, input the real-time turbine data of the turbine system into the turbine system fault diagnosis model to obtain turbine fault information; Conduct safety management on the turbine according to the turbine fault information.

2. The method for turbine safety management based on digital twin according to claim 1, characterized in that, Constructing a three-dimensional dynamic model for turbine system safety management according to the three-dimensional static scan video of the current pumped-storage power station and the three-dimensional dynamic scan video of the turbine system includes the following steps: Construct a three-dimensional static model of the pumped-storage power station according to the three-dimensional static scan video of the current pumped-storage power station; Construct a three-dimensional dynamic model of the turbine system using digital twin technology according to the three-dimensional dynamic scan video of the turbine system; Add the three-dimensional dynamic model of the turbine system to the three-dimensional static model of the pumped-storage power station to obtain a three-dimensional dynamic model for turbine system safety management.

3. The method for turbine safety management based on digital twin according to claim 2, characterized in that, The fault diagnosis algorithm includes one of artificial neural network algorithm, deep learning algorithm, fuzzy inference algorithm, and fault tree algorithm.

4. The method for turbine safety management based on digital twin according to claim 3, characterized in that ， The step of matching the corresponding fault diagnosis algorithm according to the turbine historical data set, and inputting the turbine historical data set according to the matched fault diagnosis algorithm for optimization training to construct a turbine system fault diagnosis model includes the following steps: Match the corresponding fault diagnosis algorithm according to the turbine historical data set of the turbine system; Use the matched fault diagnosis algorithm to construct an original turbine system fault diagnosis model; Perform data preprocessing on the turbine historical data set of the turbine system to obtain a preprocessed turbine fault diagnosis sample set, and divide the turbine fault diagnosis sample set into a turbine fault diagnosis training sample set and a turbine fault diagnosis test sample set; Input the turbine fault diagnosis training sample set into the original turbine system fault diagnosis model for optimization training to obtain an optimized turbine system fault diagnosis model; Input the turbine fault diagnosis test sample set into the optimized turbine system fault diagnosis model for testing, and obtain the test accuracy rate; If the test accuracy rate is greater than the preset threshold, output the final turbine system fault diagnosis model; If the test accuracy rate is less than the preset threshold, re-perform optimization training.

5. The method for turbine safety management based on digital twin according to claim 4, characterized in that, After obtaining the three-dimensional static scan video of the current pumped storage power station, the three-dimensional dynamic scan video of the turbine system, and the turbine historical data set, it further includes: Construct an original three-dimensional static model of the pumped storage power station based on the three-dimensional static scan video of the current pumped storage power station; Obtain the first perspective center of the original three-dimensional static model of the pumped storage power station. According to the first perspective center, perform perspective alignment on at least three three-dimensional static scan images in the three-dimensional static scan video of the pumped storage power station to obtain at least three perspective-aligned three-dimensional static scan images; Modify the original three-dimensional static model of the pumped storage power station according to at least three perspective-aligned three-dimensional static scan images to obtain the final three-dimensional static model of the pumped storage power station.

6. The method for turbine safety management based on digital twin according to claim 5, characterized in that When obtaining the real-time turbine data, input the real-time turbine data of the turbine system into the turbine system fault diagnosis model to obtain turbine fault information; perform safety management on the turbine according to the turbine fault information, including the following steps: Obtain the real-time turbine data of the turbine system of the current pumped storage power station and store the real-time turbine data in the turbine system safety management database; Perform turbine system fault diagnosis using the turbine system fault diagnosis model according to the real-time turbine data to obtain the turbine system fault diagnosis result; Obtain the turbine system control scheme according to the turbine system fault diagnosis result; Adjust the turbine operation data according to the turbine system control scheme to perform safety management on the turbine system of the pumped storage power station.

7. A turbine safety management system based on digital twin, characterized in that The system includes: a memory and a processor. The memory includes a program for the method for turbine safety management based on digital twin. When the program for the method for turbine safety management based on digital twin is executed by the processor, the following steps are implemented: Obtain the three-dimensional static scan video of the current pumped storage power station, the three-dimensional dynamic scan video of the turbine system, and the turbine historical data set; Construct a three-dimensional dynamic model for turbine system safety management based on the three-dimensional static scan video of the current pumped storage power station and the three-dimensional dynamic scan video of the turbine system; Match the corresponding fault diagnosis algorithm according to the turbine historical data set, and input the turbine historical data set for optimization training according to the matched fault diagnosis algorithm to construct a turbine system fault diagnosis model; Obtain the real-time turbine data, and input the real-time turbine data of the turbine system into the turbine system fault diagnosis model to obtain turbine fault information; Perform safety management on the turbine according to the turbine fault information.

8. The turbine safety management system based on digital twin according to claim 7, characterized in that The step of constructing a three-dimensional dynamic model for turbine system safety management based on the three-dimensional static scan video of the current pumped storage power station and the three-dimensional dynamic scan video of the turbine system includes the following steps: Construct a three-dimensional static model of the pumped storage power station based on the three-dimensional static scan video of the current pumped storage power station; According to the installation drawings and 3D dynamic scanning videos of the corresponding hydroturbine system, use digital twin technology to construct a 3D dynamic model of the hydroturbine system; Add the 3D dynamic model of the hydroturbine system to the 3D static model of the pumped storage power station to obtain a 3D dynamic model for the safety management of the hydroturbine system.

9. The digital twin-based hydroturbine safety management system according to claim 8, characterized in that, for the hydroturbine historical data set, match the corresponding fault diagnosis algorithm, and according to the matched fault diagnosis algorithm, input the hydroturbine historical data set for optimization training to construct a fault diagnosis model for the hydroturbine system, including the following steps: Match the corresponding fault diagnosis algorithm according to the hydroturbine historical data set of the hydroturbine system; Use the matched fault diagnosis algorithm to construct an original fault diagnosis model for the hydroturbine system; Perform data preprocessing on the hydroturbine historical data set of the hydroturbine system to obtain a preprocessed hydroturbine fault diagnosis sample set, and divide the hydroturbine fault diagnosis sample set into a hydroturbine fault diagnosis training sample set and a hydroturbine fault diagnosis test sample set; Input the hydroturbine fault diagnosis training sample set into the original fault diagnosis model of the hydroturbine system for optimization training to obtain an optimized fault diagnosis model of the hydroturbine system; Input the hydroturbine fault diagnosis test sample set into the optimized fault diagnosis model of the hydroturbine system for testing, and obtain the test accuracy rate; If the test accuracy rate is greater than the preset threshold, output the final fault diagnosis model of the hydroturbine system; If the test accuracy rate is less than the preset threshold, re-perform optimization training.

10. The digital twin-based hydroturbine safety management system according to claim 9, characterized in that, after obtaining the 3D static scanning video of the current pumped storage power station, the 3D dynamic scanning video of the hydroturbine system and the hydroturbine historical data set, further includes: Construct an original 3D static model of the pumped storage power station according to the 3D static scanning video of the current pumped storage power station; Obtain the first perspective center of the original 3D static model of the pumped storage power station, and according to the first perspective center, align the perspectives of at least three 3D static scanning images in the 3D static scanning video of the pumped storage power station to obtain at least three perspective-aligned 3D static scanning images; Modify the original 3D static model of the pumped storage power station according to at least three perspective-aligned 3D static scanning images to obtain the final 3D static model of the pumped storage power station.

Images