Hydropower plant dynamic response simulation method based on cloud computing

Abstract

The invention provides a hydropower station workshop dynamic response simulation method based on cloud computing. The method belongs to the field of cloud computing platform digital simulation. According to the method, firstly, digital twin of a hydropower station plant is created on a cloud computing platform, then various dynamic parameters are collected in real time through a sensor deployed in the plant, and the dynamic parameters are uploaded to the cloud computing platform in real time through the IoT technology. Next, the method combines an artificial intelligence class algorithm to enable digital twinning to self-learn, and continuously optimizes the model to predict various states and events that may occur in the future. In addition, according to the method, various virtual scene simulation can be operated on the digital twinning, and the dynamic response of the factory building under different conditions can be predicted and analyzed. Finally, the method provides a decision support system for providing data support for decision makers, and the method can quickly perform new simulation calculation by using the high processing capability of cloud computing in emergency, thereby providing an instant coping scheme.

Description

Technical Field

[0001] The present invention belongs to the field of digital simulation of cloud computing platforms, and more specifically relates to a dynamic response simulation method of a hydropower station plant based on cloud computing. Background Art

[0002] In traditional hydropower plant building management, manual monitoring and regular maintenance are usually used to ensure the normal operation of the unit equipment and prevent possible failures and whether the plant structure vibration exceeds the standard. However, manual monitoring often requires a lot of manpower and material resources, and due to the lack of comprehensive and timely monitoring, some key signs of failure may be missed, resulting in sudden failures of the unit equipment and damage to the plant structure. Although regular maintenance can prevent the occurrence of failures to a certain extent, it may not be able to quickly predict and avoid some sudden abnormalities in real time. In addition, the vibration of the plant structure often lacks an on-site real-time monitoring system, and requires local analysis and judgment after manual monitoring. It is impossible to quickly predict and prejudge the vibration safety of the plant in real time and comprehensively. It is necessary to improve and pay attention to the safety of the plant structure.

[0003] On the other hand, although some existing computer-based simulation and prediction methods can solve the above problems to a certain extent, they still have some obvious shortcomings. For example, they may not accurately predict the operating status of equipment and possible failures, because these methods often lack sufficient real-time data support and are usually unable to handle complex environmental factors. In addition, traditional simulation and prediction methods usually require specialized hardware and software to run, which increases the difficulty and cost of their implementation.

[0004] The dynamic response analysis of modern hydropower plant buildings is a very complex task, involving complex mathematical models and calculations, as well as a large amount of data processing. In order to simplify the work and improve the accuracy of the analysis, high-performance hardware equipment and excellent algorithms are often needed. However, high-performance hardware equipment is expensive, energy-intensive, and difficult to deploy and maintain; the design and optimization of algorithms also requires a lot of time and effort.

[0005] The emergence of cloud computing technology has made it possible to solve these problems. Cloud computing platforms can provide powerful computing capabilities, and can quickly and effectively process and analyze large amounts of data, thereby greatly shortening the analysis time and improving the accuracy of analysis. In addition, cloud computing platforms can perform flexible scheduling and dynamically allocate computing resources according to the needs of tasks, which can significantly reduce the purchase and maintenance costs of hardware equipment, while also reducing energy consumption.

[0006] However, how to effectively apply cloud computing technology to the dynamic response analysis of hydropower plant buildings has become an urgent problem to be solved. How to realize the construction of digital twins based on cloud computing, the collection and upload of real-time data, the autonomous learning and prediction of unit equipment status, as well as the flexible avoidance of faults and the prediction and identification of plant structure vibration characteristics. Summary of the invention

[0007] The technical problem to be solved by the present invention is mainly how to use cloud computing and artificial intelligence technology to realize real-time, accurate and efficient simulation prediction of the dynamic response of the hydropower station plant, and at the same time, provide convenient and effective decision-making support, as well as rapid response capabilities in emergency situations, further optimize the management of the hydropower station plant, and improve operational efficiency and safety.

[0008] In order to achieve the above object, the present invention is implemented by adopting the following technical scheme: the simulation method comprises:

[0009] Create digital twins: Build digital twins of hydropower plant buildings on a cloud computing platform;

[0010] Real-time data collection and upload: By deploying sensors in the plant, various dynamic parameters are collected in real time, including the operating status of the unit equipment and vibration information at typical locations of the plant structure, and uploaded to the cloud computing platform in real time through IoT;

[0011] Autonomous learning and prediction: On the cloud computing platform, combined with artificial intelligence algorithms, digital twins can self-learn based on the collected real-time data, continuously optimize the model, and predict various states and events that will occur in the future;

[0012] Virtual simulation: Run various virtual scenario simulations on the digital twin to predict and analyze the dynamic response of the plant under different conditions;

[0013] Decision support system: Provide data support to decision makers, telling them when to conduct operation scheduling, adjustment control, and equipment maintenance;

[0014] Flexible response: In emergency situations, digital twins can quickly perform new simulation calculations based on the high processing power of cloud computing and provide immediate response solutions.

[0015] In one embodiment, creating a digital twin includes:

[0016] Gather relevant resources, including building blueprints, equipment specifications and performance parameters;

[0017] 3D modeling: Use modeling software to build 3D models of plants and equipment on a cloud computing platform;

[0018] Parameter calibration: Use measured data to calibrate the model to ensure the accuracy of the model;

[0019] Synchronous operation: Once the model is built and calibrated, synchronize it with the real-world operating conditions and start the simulation.

[0020] In one embodiment, the real-time data collection and uploading includes:

[0021] Install various types of sensors in the factory to capture the dynamic parameters of equipment operation status and environmental factors inside and outside the factory;

[0022] Sensors use the MQTT communication protocol to connect and exchange data with IoT gateway devices;

[0023] Data upload: Use the API interface provided by the IoT cloud platform to upload pre-processed data to the cloud computing platform in real time;

[0024] The received data is combined with the previously created digital twin model, and is updated and calculated in real time to synchronize with the digital twin.

[0025] In one solution, the autonomous learning and prediction uses a hybrid algorithm based on quantum computing and contextual awareness, as follows:

[0026] Data preprocessing and context information extraction: First, the collected real-time data must be preprocessed, including data cleaning and normalization. At the same time, the data is analyzed and relevant context information is extracted, including the temperature and humidity environment information of the equipment, and the environment information is extracted as context information.

[0027] Quantum state preparation: Encode the preprocessed data into a quantum state, and use the Amplitude Encoding method to encode a classical data vector into a quantum state;

[0028] Quantum neural network training: Use quantum neural network to process this quantum state; define the structure of a quantum neural network and train it using a gradient-based method;

[0029] Introduction of context information: Introducing the extracted context information as control information into the quantum neural network; that is, using the context information to adjust the parameters or structure of the quantum neural network; enabling the model to perform adaptive learning and prediction according to different contexts;

[0030] Prediction and Optimization: Make predictions based on new data.

[0031] In one embodiment, the virtual simulation includes:

[0032] Establish a finite element model: In the digital twin, a finite element model is established to represent the structure of the plant; each part of the plant structure is divided into many small, simple elements;

[0033] Update boundary conditions: Update the model's boundary conditions, stress, hydraulic pulsation pressure, unit operating condition information, etc. according to real-time data;

[0034] Run a simulation: Solve global equations, calculate stress and strain for each element, then aggregate the data from all elements to get the dynamic response of the entire plant.

[0035] In one embodiment, the decision support system comprises:

[0036] Condition monitoring: After the real-time operating status of the unit equipment, the plant environment status and other data are uploaded to the cloud, the decision support system uses the model to determine the working status of the unit equipment and the vibration characteristic status of the plant, and analyze whether the real-time environment is within the safe range;

[0037] Risk warning: When the model prediction results exceed the set range, the decision support system will generate a risk warning;

[0038] Future trend forecast: predict the impact of the operating conditions of the unit equipment on the vibration safety characteristics of the plant structure in the future, and provide reference and suggestions for decision makers to plan future production arrangements.

[0039] In one embodiment, the flexible response includes:

[0040] When the decision support system detects an abnormal state of equipment or environment, the emergency response platform is activated and quickly provided to decision makers via the Web;

[0041] The emergency response platform quickly calls upon the digital twin simulation model to predict what will happen in the future and the best response plan based on the current real-time status.

[0042] Beneficial effects of the present invention:

[0043] By using digital twins and cloud computing methods, various dynamic responses of hydropower station plants can be simulated and predicted in real time and accurately, which can not only improve the accuracy of predictions, but also greatly speed up the analysis, provide a scientific basis for decision makers, and further improve the operating efficiency and safety of the plant.

[0044] By deploying sensors in the factory to collect data in real time and uploading it to the cloud computing platform in real time using Internet of Things technology, comprehensive, accurate and real-time factory operation data can be obtained, laying a good foundation for analysis.

[0045] By introducing artificial intelligence algorithms, autonomous learning and prediction can be carried out, which can not only predict the operating status of equipment based on historical data, but also predict future environmental changes, providing decision makers with strong decision support.

[0046] By utilizing the powerful computing power of cloud computing, various complex calculations can be performed quickly, providing real-time solutions for dealing with emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a flow chart of the method of the present invention;

[0048] Figure 2 Creating a digital twin flow chart for the present invention;

[0049] Figure 3 It is the autonomous learning and prediction flow chart of the present invention;

[0050] Figure 4 It is a virtual simulation flow chart of the present invention. DETAILED DESCRIPTION

[0051] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Typical embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described in the present invention. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0052] Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0053] like Figure 1 As shown, the simulation method comprises:

[0054] Create a digital twin: On the cloud computing platform, build a digital twin of the hydropower plant building to fully simulate every detail of the real plant, including equipment and structure.

[0055] like Figure 2 As shown, the following steps are included:

[0056] S101. Data Collection: Collect relevant resources, including detailed building blueprints, equipment specifications and performance parameters, etc. These data will be used to create and calibrate the digital twin model.

[0057] S102, 3D modeling: Use modeling software to build 3D models of factories and equipment on the cloud platform. Every detail, including every device and structure, is simulated in detail.

[0058] S103, Physical property simulation: Add physical functions to the model to simulate the operating principle of each part. This can be achieved by setting various mathematical formulas and equations, such as Newton's second law to simulate the force and movement of objects, fluid mechanics equations to simulate water flow, structural motion equations to simulate plant structure vibration, etc.

[0059] S104, parameter calibration: Use measured data to calibrate the model to ensure the accuracy of the model. The calibration process can use an optimization algorithm such as a genetic algorithm (GA) to find the model parameters that best match the measured data.

[0060] S105. Synchronous operation: After the model is established and calibrated, it can be synchronized with the real-world operating conditions and start simulation.

[0061] Real-time data collection and upload: By deploying sensors in the factory, various dynamic parameters are collected in real time, including equipment operation status and typical structural vibration information of the factory structure, and uploaded to the cloud computing platform in real time through IoT. The detailed process is as follows:

[0062] Various types of sensors are installed in the plant, such as unit operating condition signals, unit swing sensors, hydraulic pulsation pressure sensors, structural vibration sensors, etc., to capture various dynamic parameters such as equipment operating status and environmental factors inside and outside the plant.

[0063] Use appropriate communication protocols, such as MQTT (Message Queuing Telemetry Transport) and CoAP (Constrained Application Protocol), to connect these sensors and exchange data with IoT gateway devices.

[0064] IoT gateway devices receive data from various sensors in real time, and may need to perform certain data preprocessing, such as data cleaning and format conversion, to prepare for uploading the data to the cloud platform.

[0065] Use IoT cloud platforms, such as AWS IoT, Google Cloud IoT Core, etc., and upload the pre-processed data to the cloud in real time through the API interfaces they provide.

[0066] In the cloud platform, the received data is combined with the previously created digital twin model for real-time update and calculation. Finally, the data uploaded to the cloud is synchronized to the corresponding digital twin model to update the model's status.

[0067] like Figure 3As shown, autonomous learning and prediction: On the cloud computing platform, combined with artificial intelligence algorithms, digital twins can self-learn based on the collected real-time data, continuously optimize the model, and predict various states and events that will occur in the future. A hybrid algorithm based on quantum computing and context-awareness is used.

[0068] Quantum Machine Learning: Quantum machine learning refers to the use of the properties of quantum computing when designing and implementing machine learning algorithms in order to achieve higher computing speeds and higher model performance. Quantum computing can handle parallel computing tasks and has greater potential than traditional computing. With the help of quantum computing, massive amounts of data can be analyzed and predicted more efficiently.

[0069] Context-Aware: Context-aware machine learning means that when a machine is learning or making decisions, it can perceive and understand the environment and context of the current task. This contextual information can be used to guide the machine to adaptively adjust its learning or decision-making strategies to achieve better performance.

[0070] The combination of these two technologies can introduce the real-time data of the plant (such as the status of the unit equipment, vibration information of typical structural positions, etc.) as contextual information into the quantum machine learning model, so that the model can autonomously learn and predict the possible future states and events based on the specific real-time context. At the same time, the parallel processing capability of quantum computing also enables the model to perform large-scale data processing and model optimization more efficiently.

[0071] This hybrid algorithm based on quantum computing and context awareness is implemented through the following steps.

[0072] S301, data preprocessing and context information extraction: First, the collected real-time data must be preprocessed, including data cleaning, normalization, etc. At the same time, the data can be analyzed and relevant context information can be extracted. For example, if the data contains environmental information such as the operating status of the equipment and the vibration of the plant structure, the environmental information can be extracted as context information.

[0073] S302, Quantum state preparation: The first step of quantum machine learning is the preparation of quantum state. We need to encode the preprocessed data into a quantum state. Specifically, we can use the Amplitude Encoding method to encode a classical data vector into a quantum state. The mathematical formula is:

[0074] |x>=Σi xi|i>

[0075] Among them, x is the classical data vector and i represents the basis vector of the quantum state.

[0076] S303, quantum neural network training: Then use the quantum neural network to process the quantum state. Specifically, define the structure of a quantum neural network and use a gradient-based method to train it.

[0077] |Ψ(θ)>=U(θU|x>

[0078] Among them, U(θ) represents the quantum neural network, θ represents the parameters of the network, and |Ψ(θ)> represents the quantum state after processing by the neural network.

[0079] S304, context information introduction: The extracted context information is introduced into our quantum neural network as control information. That is, the context information can be used to adjust the parameters or structure of the quantum neural network. In this way, the model can perform adaptive learning and prediction according to different contexts.

[0080] This is achieved by "parameter adjustment". A vector of environmental parameters E = [e1, e2, ..., en] is set up. This vector reflects various factors of the current environment, such as temperature, humidity, etc. During the training process of the quantum neural network, this vector of environmental parameters is used as an additional input to the control part of the network. In this way, the various parameters of the network can change adaptively due to environmental changes, and ultimately achieve adaptive learning and prediction for different situations.

[0081] S305, prediction and optimization: Finally, the new data is predicted. After the prediction is completed, the prediction result needs to be decoded from the quantum state into classical data.

[0082] like Figure 4 As shown, virtual simulation: various virtual scenario simulations are run on the digital twin to predict and analyze the dynamic response of the plant under different conditions. Finite element analysis (FEA) method is used for analysis.

[0083] S401. Establish a finite element model: In the digital twin, a finite element model is established to represent the structure of the plant. In this model, each part of the plant structure is divided into many small, simple elements. The behavior of each element can be expressed by a mathematical formula, and then the formulas of all elements are combined to form a global equation that describes the behavior of the entire system.

[0084] In finite element analysis, the local stiffness matrix (k) is used, which can be used to calculate the stress and strain of each element in the structure. The basic form is as follows:

[0085] [k]=∫[B]T[D][B]dV

[0086] Where [k] is the element stiffness matrix, [B] is the strain-displacement matrix, [D] is the matrix of the material's elastic constants (containing material properties such as Poisson's ratio and elastic modulus), and V is the integrated volume.

[0087] S402. Update boundary conditions: Update the boundary conditions of the model according to real-time data, such as the unit operation status, hydraulic pulsation pressure, local vibration acceleration of the plant, etc.

[0088] S403 performs simulation: solves global equations, calculates stress, strain and vibration characteristics of each element, and then aggregates the data of all elements to obtain the dynamic response of the entire plant.

[0089] Decision support system: Provides data support for decision makers, telling them when to schedule operations, adjust controls, and perform equipment maintenance. Mainly using the models trained in the previous steps, combined with current real-time data, it provides decision makers with reports on the status of unit equipment, plant environment, and possible risk warnings. It uses the fast processing capabilities of big data and cloud computing to analyze and provide real-time status and possible future situations.

[0090] The specific process is as follows:

[0091] Decision support system: A decision support system can be a web application deployed through a cloud platform, which obtains real-time data from a cloud database and uses trained models to make predictions, analyze, and generate reports.

[0092] Condition monitoring: After the real-time operating status of the equipment, typical structural vibration of the factory building and other data are uploaded to the cloud, the decision support system will use the model to predict whether the equipment is in a normal state and whether the real-time environment is within a safe range.

[0093] Risk warning: When the model's predicted results exceed the set range, such as possible equipment failure or abnormal plant vibration, the decision support system will generate a risk warning to remind decision makers to take preventive measures.

[0094] Future trend prediction: The model can also predict the operating status of the unit equipment and the changes in the vibration characteristics of the plant structure in the future, providing reference and suggestions for decision makers in planning future production arrangements.

[0095] Flexible response: In emergencies, digital twins can quickly perform new simulation calculations based on the high processing power of cloud computing and provide immediate response plans. This mainly involves emergency decision-making and rapid response. When certain emergencies occur, the decision support system needs to quickly run new simulation models and provide emergency response plans. This step mainly depends on the fast processing power of cloud computing and the digital twin simulation model prepared in the previous step.

[0096] The specific process can be as follows:

[0097] (1) Emergency response platform: When the decision support system detects an abnormal state of equipment or environment, the emergency response platform can be activated. This platform can be quickly provided to decision makers through the Web and other means.

[0098] (2) Rapid simulation: The emergency response platform can quickly call the digital twin simulation model to predict the conditions that may occur in the future and the best possible response plan based on the current real-time status.

[0099] (3) Providing response plans: Based on the results of rapid simulation, the emergency response platform combines existing experience and data to provide emergency plans.

[0100] Embodiment 2:

[0101] Taking a pumped-storage hydropower station plant as an example, assuming that the plant includes different types of unit equipment from several manufacturers, the operating status of these unit equipment may show complex dynamic changes due to factors such as working conditions, workload and the physical characteristics of the equipment itself.

[0102] First, we create a digital twin model of the plant on the cloud computing platform. The specific process is as follows:

[0103] (1) Collect relevant resources, including the plant's architectural blueprint, equipment models and specifications, and historical data on equipment operating status;

[0104] (2) Build 3D models of plant and equipment on the cloud computing platform; for example, build a pumped storage underground plant, including four generators;

[0105] (3) Use measured data to calibrate the model, for example, adjust the position of the unit equipment and the size of the volute in the model to ensure the accuracy of the model.

[0106] Secondly, we deployed sensors for unit status, hydraulic pulsation pressure, structural vibration, etc. in the plant, and collected data from these sensors in real time through the Internet of Things technology and uploaded them to the cloud computing platform. For example, under a certain operating condition on a certain working day, we collected a 1-minute time-course signal dynamically monitored by the plant type sensor and uploaded this data in real time.

[0107] Then, we use a hybrid algorithm based on quantum computing and contextual awareness for autonomous learning and prediction, which includes steps such as preparation of quantum states, training of quantum neural networks, and introduction of contextual information.

[0108] We use the finite element model to carry out numerical simulation of the digital twin. On a certain working day, we update the boundary conditions, such as a certain operating condition of the unit. Through simulation calculations, we find that the vibration characteristics of a certain plant structure location exceed the safety standards.

[0109] In the decision support system, by real-time monitoring of equipment status and generating risk warnings, when the equipment operating status is abnormal or the vibration characteristics of the plant exceed the safe range, the system can immediately notify the staff to intervene. We predict that continuing this operating condition will be detrimental to the structural safety of the plant and the operating condition needs to be adjusted.

[0110] When the current operating conditions of the unit are not conducive to structural safety, we initiate an emergency response and provide decision makers with quick solutions through flexible responses. For example, we adjust the operating parameters of the unit equipment or propose a plant structure optimization plan to ensure the safe operation of the plant.

[0111] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0112] It should be understood that the detailed description of the technical solutions of the present invention by means of the preferred embodiments is illustrative rather than restrictive. A person skilled in the art may modify the technical solutions described in the embodiments, or replace some of the technical features by equivalents, based on reading the specification of the present invention; and these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for simulating the dynamic response of a hydropower plant building based on cloud computing, characterized in that: The simulation method comprises: Create digital twins: Build digital twins of hydropower plant buildings on a cloud computing platform; Real-time data collection and upload: By deploying sensors in the plant, various dynamic parameters are collected in real time, including the operating status of the unit equipment and environmental factors inside and outside the plant, and uploaded to the cloud computing platform in real time through IoT; Autonomous learning and prediction: On the cloud computing platform, combined with artificial intelligence algorithms, digital twins can self-learn based on the collected real-time data, continuously optimize the model, and predict various states and events that will occur in the future; Virtual simulation: Run various virtual scenario simulations on the digital twin to predict and analyze the dynamic response of the plant under different unit operating conditions; Decision support system: Provides data support to decision makers, telling them when to conduct operation scheduling, adjustment control, and equipment maintenance; In case of emergencies, digital twins perform new simulation calculations based on cloud computing and provide immediate response solutions.

2. According to claim 1, a method for simulating the dynamic response of a hydropower plant building based on cloud computing is characterized by: The creation of digital twins includes: Collect relevant resources, including architectural blueprints, unit equipment specifications and performance parameters; 3D modeling: Use modeling software to build 3D models of plant buildings and equipment on the cloud computing platform; Parameter calibration: Use measured data to calibrate the model to ensure the accuracy of the model; Synchronous operation: Once the model is built and calibrated, synchronize it with the real-world operating conditions and start the simulation.

3. The method for simulating dynamic response of a hydropower plant building based on cloud computing according to claim 1 is characterized in that: The real-time data collection and uploading includes: Install various types of sensors in the plant to capture the operating status of the unit equipment and the dynamic parameters of the plant vibration; Sensors use the MQTT communication protocol to connect and exchange data with IoT gateway devices; Data upload: Use the API interface provided by the IoT cloud platform to upload pre-processed data to the cloud computing platform in real time; The received data is combined with the previously created digital twin model, and is updated and calculated in real time to synchronize with the digital twin.

4. The method for simulating dynamic response of a hydropower plant building based on cloud computing according to claim 1 is characterized by: The autonomous learning and prediction described above adopts a hybrid algorithm based on quantum computing and situational awareness, as follows: Data preprocessing and context information extraction: First, the collected real-time monitoring data is preprocessed, including data cleaning and normalization; at the same time, the data is analyzed and relevant context information is extracted; including the working status information of the unit equipment under different operating conditions and the plant structure vibration information, and the working status information is extracted as context information; Quantum state preparation: Encode the preprocessed data into a quantum state, and use the Amplitude Encoding method to encode a classical data vector into a quantum state; Quantum neural network training: Use quantum neural network to process this quantum state; define the structure of a quantum neural network and train it using a gradient-based method; Introduction of context information: Introducing the extracted context information as control information into the quantum neural network; that is, using the context information to adjust the parameters or structure of the quantum neural network; enabling the model to perform adaptive learning and prediction according to different contexts; Prediction and Optimization: Make predictions based on new data.

5. The method for simulating dynamic response of a hydropower plant building based on cloud computing according to claim 1 is characterized by: The virtual simulation includes: Establish a finite element model: In the digital twin, a finite element model is established to represent the structure of the plant; each part of the plant structure is divided into many small, simple elements; Update boundary conditions: Update the boundary conditions of the model according to real-time data, including unit operating information, unit swing, and hydraulic pulsation pressure; Perform simulation: Solve global equations, calculate stress and strain as well as vibration acceleration, velocity, and displacement for each element, and then summarize the data of all elements to obtain the dynamic response of the entire plant.

6. The method for simulating dynamic response of a hydropower plant building based on cloud computing according to claim 1 is characterized by: The decision support system comprises: Condition monitoring: After the real-time operating status of the unit equipment and the typical structural vibration status data of the plant are uploaded to the cloud, the decision support system uses the model to predict whether the vibration characteristics of the plant structure are in a normal state and within a safe range; Risk warning: When the model prediction results exceed the set range, the decision support system will generate a risk warning; Future trend forecast: predict the operating conditions of the unit equipment and the vibration characteristics of the plant structure in the future, and provide reference and suggestions for decision makers to plan future production arrangements.

7. The method for simulating dynamic response of a hydropower plant building based on cloud computing according to claim 1 is characterized by: The digital twin performs new simulation calculations based on cloud computing and provides immediate response solutions including: When the decision support system detects abnormal vibration of the unit equipment or plant structure, the emergency response platform is activated and quickly provided to decision makers via the Web; The emergency response platform quickly calls upon the digital twin simulation model to predict what will happen in the future and the best response plan based on the current real-time status.

Images