Intelligent hydraulic power plant operation management system and method based on digital twinning

Abstract

According to the intelligent hydraulic power plant operation management system and method based on digital twinning, the digital twinning model of the hydraulic power plant is constructed through the digital twinning technology, the operation data of the hydraulic power plant are collected through the sensing technology, and real-time data synchronization between the digital twinning model and the hydraulic power plant is achieved through the Internet of Things technology; meanwhile, the operation data of the hydraulic power plant are intelligently analyzed, whether the working state of the turbine is abnormal or not is intelligently analyzed based on a time sequence interaction mode between the generating capacity and multiple operation parameters influencing the generating capacity, and the analysis result is synchronously displayed in a digital twinborn model of the hydraulic power plant. In this way, comprehensive sensing, real-time monitoring and intelligent analysis of the operation state of the hydraulic power plant can be effectively achieved, and therefore the overall operation efficiency and reliability of the hydraulic power plant are improved.

Description

Technical Field

[0001] The embodiments of the present disclosure belong to the technical field of hydropower plant operation management, and specifically relate to an intelligent hydropower plant operation management system and method based on digital twins. Background Art

[0002] As the proportion of renewable energy in the global energy structure continues to increase, hydropower, as an important component of clean energy, has an important role to play in ensuring energy supply and environmental protection. The efficient and stable operation of hydropower is of great significance to the protection of energy supply and environment. The operation and management of traditional hydropower plants mainly rely on manual monitoring and empirical judgment, which is not only inefficient but also prone to misjudgment due to human factors. In addition, the maintenance of hydropower plant equipment often adopts the method of planned maintenance or post-failure repair. This method makes it difficult to predict equipment failures in advance, increases the risk of unplanned downtime, and affects the economic benefits of hydropower plants and the stability of the power system.

[0003] At present, although some hydropower plants have begun to adopt automated control systems to improve operational efficiency, most of these systems operate independently and lack overall data integration and intelligent analysis capabilities. For example, the data collected by sensors is often only used to monitor the basic status of equipment, and the data is not fully utilized for deeper analysis, such as predictive maintenance and optimization of operation strategies.

[0004] In recent years, with the rise of digital twin technology, new solutions have been brought to the operation and management of hydropower plants. Therefore, we expect a smart hydropower plant operation and management solution based on digital twins to achieve a higher level of automation and intelligent management of the operation status of hydropower plants. Summary of the invention

[0005] The embodiments of the present disclosure provide an intelligent hydropower plant operation management system and method based on digital twins, aiming to solve at least one of the technical problems existing in the prior art, such as effectively realizing comprehensive perception, real-time monitoring and intelligent analysis of the operating status of the hydropower plant, thereby improving the overall operational efficiency and reliability of the hydropower plant.

[0006] On the one hand, an embodiment of the present disclosure provides an intelligent hydropower plant operation management system based on digital twins, including: a hydropower plant operation monitoring module, which is used to collect the operation data of the hydropower plant in real time through a sensor group and monitoring equipment, and the operation data includes water level value, flow value, water pressure value, turbine speed value and power generation; a digital twin module, which is used to build a digital twin model of the hydropower plant, and realize real-time data synchronization between the hydropower plant and the digital twin model through the Internet of Things technology; a turbine operation status analysis module, which is used to use the digital twin model to perform data analysis on the operation data of the hydropower plant to obtain an operation analysis result, and the operation analysis result is used to indicate whether there is an abnormality in the working state of the turbine; an operation analysis result display module, which is used to display the operation analysis result in real time using the digital twin model; wherein the turbine operation status analysis module includes: an operation data timing encoding unit, which is used to The operation data of the power plant is subjected to time series feature extraction based on parameter samples to obtain a water level time series correlation feature vector, a flow rate time series correlation feature vector, a water pressure time series correlation feature vector, a turbine speed time series correlation feature vector and a power generation amount time series correlation feature vector; a power generation dependent variable multimodal fusion unit is used to perform multimodal fusion on the water level time series correlation feature vector, the flow rate time series correlation feature vector, the water pressure time series correlation feature vector and the turbine speed time series correlation feature vector to obtain a power generation dependent variable multimodal embedding fusion representation vector; a power generation amount causal interaction analysis unit is used to perform fine-grained interaction analysis on the power generation dependent variable multimodal embedding fusion representation vector and the power generation amount time series correlation feature vector to obtain a power generation dependent variable-power generation amount decoupling full time series enhanced interaction representation vector; an operation analysis result generation unit is used to generate the operation analysis result based on the power generation dependent variable-power generation amount decoupling full time series enhanced interaction representation vector.

[0007] In one possible implementation, the operating data time series encoding unit includes: a data regularization subunit, which is used to regularize the operating data of the hydropower plant according to the parameter sample dimension to obtain a time queue of water level values, a time queue of flow values, a time queue of water pressure values, a time queue of turbine speed values, and a time queue of power generation; an operating parameter time series feature extraction subunit, which is used to input the time queue of water level values, the time queue of flow values, the time queue of water pressure values, the time queue of turbine speed values, and the time queue of power generation into a sequence encoder based on a 1D-CNN model respectively to obtain the water level time series association feature vector, the flow time series association feature vector, the water pressure time series association feature vector, the turbine speed time series association feature vector, and the power generation time series association feature vector.

[0008] In a possible implementation, the power generation dependent variable multimodal fusion unit is used to: input the water level time series associated feature vector, the flow time series associated feature vector, the water pressure time series associated feature vector and the turbine speed time series associated feature vector into the power generation dependent variable multimodal fusion module based on the embedding layer to obtain the power generation dependent variable multimodal embedded fusion representation vector.

[0009] In a possible implementation, the power generation causal interaction analysis unit is used to: input the power generation dependent variable multimodal embedding fusion representation vector and the power generation time series associated feature vector into a feature decoupled local interaction fine-grained aggregation analysis network to obtain the power generation dependent variable-power generation decoupled full time series enhanced interaction representation vector.

[0010] In a possible implementation, the power generation causal interaction analysis unit includes: a feature decoupling subunit, which is used to perform feature decoupling on the power generation dependent variable multimodal embedding fusion representation vector and the power generation time series association feature vector respectively to obtain a sequence of power generation dependent variable multimodal fusion feature local representation vectors and a sequence of power generation time series association feature local representation vectors; a local semantic association interaction subunit, which is used to calculate the local semantic association interaction vector between each group of corresponding power generation dependent variable multimodal fusion feature local representation vectors and power generation time series association feature local representation vectors in the sequence of power generation dependent variable multimodal fusion feature local representation vectors and the sequence of power generation time series association feature local representation vectors to obtain the power generation dependent variable-power generation. A sequence of local semantic association interaction vectors; an attention weight calculation subunit, used to calculate the fine-grained attention weights of each power generation dependent variable-power generation local semantic association interaction vector in the sequence of power generation dependent variable-power generation local semantic association interaction vectors to obtain a sequence of fine-grained attention weights; a random deactivation subunit, used to input the sequence of fine-grained attention weights into a random deactivation module to obtain a sequence of fine-grained attention weights after deactivation processing; a local interaction information aggregation subunit, used to use the sequence of fine-grained attention weights after deactivation processing as a weight distribution, calculate the weighted sum of the sequence of power generation dependent variable-power generation local semantic association interaction vectors to obtain the power generation dependent variable-power generation decoupled full-time series reinforcement interaction representation vector.

[0011] In a possible implementation, the local semantic association interaction subunit is used to: calculate the point-by-point addition, point-by-point subtraction and point-by-point multiplication between the local representation vector of the multimodal fusion feature of the power generation dependent variable and the corresponding local representation vector of the time series association feature of the power generation amount to obtain a first fusion result, a second fusion result and a third fusion result; after cascading the first fusion result, the second fusion result and the third fusion result into a power generation dependent variable-power generation multi-scale fusion feature vector, perform one-dimensional convolution processing and maximum pooling processing on it to obtain the power generation dependent variable-power generation local semantic association interaction vector.

[0012] In one possible implementation, the attention weight calculation subunit is used to: multiply the power generation dependent variable-power generation local semantic association interaction vector by a weight coefficient matrix, add the result to the bias vector and pass it through a tanh function to obtain a power generation dependent variable-power generation local semantic association interaction encoding feature vector; use an attention score reference vector to multiply the power generation dependent variable-power generation local semantic association interaction encoding feature vector to obtain the fine-grained attention weight.

[0013] In a possible implementation, the operation analysis result generating unit is used to: input the power generation dependent variable-power generation decoupled full-time reinforcement interaction representation vector into a classifier-based operation analysis result generator to obtain the operation analysis result.

[0014] On the other hand, an embodiment of the present disclosure provides an intelligent hydropower plant operation management method based on digital twins, including: collecting operating data of the hydropower plant in real time through a sensor group and monitoring equipment, the operating data including water level value, flow value, water pressure value, turbine speed value and power generation; constructing a digital twin model of the hydropower plant, and realizing real-time data synchronization between the hydropower plant and the digital twin model through Internet of Things technology; using the digital twin model to perform data analysis on the operating data of the hydropower plant to obtain an operation analysis result, and the operation analysis result is used to indicate whether there is any abnormality in the working state of the turbine; using the digital twin model to display the operation analysis result in real time; wherein, using the digital twin model to perform data analysis on the operating data of the hydropower plant to obtain an operation analysis result, and the operation analysis result is used to indicate whether there is any abnormality in the working state of the turbine Whether there is an abnormality in the state, including: extracting the time series features based on parameter samples on the operation data of the hydropower plant to obtain a water level time series correlation feature vector, a flow rate time series correlation feature vector, a water pressure time series correlation feature vector, a turbine speed time series correlation feature vector and a power generation amount time series correlation feature vector; performing multimodal fusion on the water level time series correlation feature vector, the flow rate time series correlation feature vector, the water pressure time series correlation feature vector and the turbine speed time series correlation feature vector to obtain a power generation dependent variable multimodal embedding fusion representation vector; performing fine-grained interactive analysis on the power generation dependent variable multimodal embedding fusion representation vector and the power generation amount time series correlation feature vector to obtain a power generation dependent variable-power generation amount decoupling full time series enhanced interactive representation vector; generating the operation analysis result based on the power generation dependent variable-power generation amount decoupling full time series enhanced interactive representation vector.

[0015] The embodiment of the present disclosure is a digital twin-based intelligent hydropower plant operation management system and method, which constructs a digital twin model of the hydropower plant through digital twin technology, collects the operation data of the hydropower plant through sensor technology, and uses the Internet of Things technology to achieve real-time data synchronization between the digital twin model and the hydropower plant, and at the same time, intelligently analyzes the operation data of the hydropower plant, and intelligently analyzes whether there is an abnormality in the working state of the turbine based on the time-series interaction mode between the power generation and the various operating parameters that affect the power generation, and synchronously displays the analysis results in the digital twin model of the hydropower plant. In this way, comprehensive perception, real-time monitoring and intelligent analysis of the operating state of the hydropower plant can be effectively achieved, thereby improving the overall operational efficiency and reliability of the hydropower plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic block diagram of an intelligent hydropower plant operation and management system based on digital twins according to an embodiment of the present disclosure;

[0017] Figure 2A schematic block diagram of a turbine operation status analysis module in a digital twin-based intelligent hydropower plant operation management system according to another embodiment of the present disclosure;

[0018] Figure 3 This is a data flow diagram of a turbine operation status analysis module in a digital twin-based intelligent hydropower plant operation management system according to another embodiment of the present disclosure;

[0019] Figure 4 A schematic block diagram of an operation data timing encoding unit in a digital twin-based intelligent hydropower plant operation management system according to another embodiment of the present disclosure;

[0020] Figure 5 A schematic block diagram of a power generation causal interaction analysis unit in a digital twin-based intelligent hydropower plant operation management system according to another embodiment of the present disclosure;

[0021] Figure 6 is a schematic flow chart of an intelligent hydropower plant operation management method based on digital twins according to another embodiment of the present disclosure;

[0022] Figure 7 In another embodiment of the present disclosure, in the intelligent hydropower plant operation management method based on digital twin, a schematic flowchart of using the digital twin model to perform data analysis on the operation data of the hydropower plant to obtain an operation analysis result. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the technical solution of the present disclosure is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0024] Although many hydropower plants today are equipped with automated control systems to improve their operational efficiency, these control systems usually operate in isolation and cannot achieve unified data management and high-level intelligent analysis. This means that although sensors can continuously collect information about the status of equipment, this information is mainly used for simple status monitoring and is not used for more complex analysis tasks, such as predicting potential equipment failures or developing more efficient operating strategies. This situation limits the possibility of hydropower plants gaining greater benefits from existing data resources.

[0025] In response to the above technical problems, the technical concept of the disclosed embodiment is: construct a digital twin model of a hydropower plant through digital twin technology, collect the operating data of the hydropower plant through sensor technology, and use the Internet of Things technology to achieve real-time data synchronization between the digital twin model and the hydropower plant, and at the same time, intelligently analyze the operating data of the hydropower plant, based on the time-series interaction mode between the power generation and the various operating parameters that affect the power generation, intelligently analyze whether there is an abnormality in the working state of the turbine, and synchronously display the analysis results in the digital twin model of the hydropower plant. In this way, comprehensive perception, real-time monitoring and intelligent analysis of the operating state of the hydropower plant can be effectively achieved, thereby improving the overall operational efficiency and reliability of the hydropower plant.

[0026] Based on this, the embodiment of the present disclosure proposes an intelligent hydropower plant operation management system based on digital twins. Figure 1 FIG. 1 is a schematic block diagram of an intelligent hydropower plant operation and management system based on digital twins according to an embodiment of the present disclosure. Figure 1 As shown, the intelligent hydropower plant operation management system 100 based on digital twins includes: a hydropower plant operation monitoring module 110, which is used to collect the operation data of the hydropower plant in real time through a sensor group and monitoring equipment, and the operation data includes water level value, flow value, water pressure value, turbine speed value and power generation. A digital twin module 120 is used to build a digital twin model of the hydropower plant, and to achieve real-time data synchronization between the hydropower plant and the digital twin model through the Internet of Things technology. A turbine operation status analysis module 130 is used to use the digital twin model to perform data analysis on the operation data of the hydropower plant to obtain an operation analysis result, and the operation analysis result is used to indicate whether there is an abnormality in the working state of the turbine. An operation analysis result display module 140 is used to display the operation analysis result in real time using the digital twin model.

[0027] In the above-mentioned intelligent hydropower plant operation management system based on digital twin, the hydropower plant operation monitoring module 110 is used to collect the operation data of the hydropower plant in real time through the sensor group and monitoring equipment, and the operation data includes water level value, flow value, water pressure value, turbine speed value and power generation. It should be understood that the real-time collection of the operation data of the hydropower plant through the sensor group and monitoring equipment is to ensure that the hydropower plant can effectively monitor the status of its key equipment, especially the working condition of the turbine. These data include water level value, flow value, water pressure value, turbine speed value and power generation, which are all key parameters affecting power generation efficiency. By continuously monitoring these data, abnormal conditions in the operation of the equipment, such as reduced efficiency or equipment failure, can be discovered in time, thereby reducing the risk of unplanned downtime and improving the overall operation efficiency and reliability of the hydropower plant. Optionally, in one embodiment of the present disclosure, a series of sensors and monitoring equipment are deployed in the hydropower plant. These sensors will be installed in key locations, such as near the turbine, in the water flow channel and around the generator set to measure the water level, water flow velocity, water pressure, turbine rotation speed and the final amount of electricity generated. Sensors convert these physical quantities into electrical or digital signals and transmit them to the central monitoring system through the network. The monitoring equipment is responsible for receiving and processing these signals and converting them into meaningful data, such as charts or alarm signals. In this way, operators can view these data in real time, understand the health status of the equipment, and take necessary preventive measures to avoid failures.

[0028] In the above-mentioned intelligent hydropower plant operation management system based on digital twin, the digital twin module 120 is used to construct a digital twin model of the hydropower plant, and to achieve real-time data synchronization between the hydropower plant and the digital twin model through the Internet of Things technology. It should be understood that the digital twin is a virtual model that can reflect the state of the physical equipment, the working performance and the changes in the external environment in real time. In the application scenario of the hydropower plant, the digital twin technology can simulate the actual operation of the hydropower plant, and provide support for operational decisions by collecting, processing and analyzing the data of the physical world. Through the Internet of Things technology, the hydropower plant can connect various sensors and monitoring equipment distributed in different locations. These devices can collect the operation data of the hydropower plant in real time, including but not limited to water level value, flow value, water pressure value, turbine speed value and power generation and other key parameters. These data are transmitted to the digital twin model in real time, so that the model can accurately reflect the actual operation status of the hydropower plant. With this real-time data synchronization, managers can intuitively see the current status of the hydropower plant on the digital twin interface and discover any potential problems or abnormalities in time.

[0029] In the above-mentioned intelligent hydropower plant operation management system based on digital twin, the turbine operation status analysis module 130 is used to use the digital twin model to perform data analysis on the operation data of the hydropower plant to obtain an operation analysis result, and the operation analysis result is used to indicate whether there is an abnormality in the working state of the turbine. It should be understood that the turbine is one of the core equipment of the hydropower station, which mainly converts the kinetic energy of the water flow into mechanical energy, thereby driving the generator to generate electrical energy. Specifically, the hydropower station guides the water flow to the turbine through a water diversion channel or a pressure pipe. The water flow accumulated in a reservoir with a high water level usually has a high flow rate and pressure. When the water flows through the turbine, it carries energy to impact the blades of the turbine, driving the turbine blades to rotate. The turbine shaft is connected to the generator, and the rotational motion of its blades is transmitted to the generator through the shaft, thereby realizing energy conversion. In this process, the operating state of the turbine is directly related to the power generation efficiency and stability of the hydropower station. Since the water level value, flow value, water pressure value and turbine speed value are key operating parameters that affect power generation, in the technical solution disclosed herein, power generation is interactively analyzed with water level value, flow value, water pressure value and turbine speed value to assess whether the working state of the turbine is abnormal. For example, if the power generation is not proportional to the increase in flow value and water pressure value, it may indicate that the turbine has efficiency problems or is blocked.

[0030] Figure 2 It is a schematic block diagram of a turbine operation status analysis module in a digital twin-based intelligent hydropower plant operation management system according to an embodiment of the present disclosure. Figure 3 Schematic diagram of data flow of turbine operation status analysis module in the intelligent hydropower plant operation management system based on digital twin according to the embodiment of the present disclosure. Figure 2 and Figure 3As shown, the turbine operation status analysis module 130 includes: an operation data time series encoding unit 131, which is used to extract the time series features of the operation data of the hydropower plant based on parameter samples to obtain a water level time series correlation feature vector, a flow rate time series correlation feature vector, a water pressure time series correlation feature vector, a turbine speed time series correlation feature vector and a power generation time series correlation feature vector. A power generation dependent variable multimodal fusion unit 132, which is used to perform multimodal fusion on the water level time series correlation feature vector, the flow rate time series correlation feature vector, the water pressure time series correlation feature vector and the turbine speed time series correlation feature vector to obtain a power generation dependent variable multimodal embedding fusion representation vector. A power generation causal interaction analysis unit 133, which is used to perform fine-grained interaction analysis on the power generation dependent variable multimodal embedding fusion representation vector and the power generation time series correlation feature vector to obtain a power generation dependent variable-power generation decoupling full-time series enhanced interaction representation vector. The operation analysis result generating unit 134 is used to generate the operation analysis result based on the power generation dependent variable-power generation decoupled full-time enhanced interaction representation vector.

[0031] In the above-mentioned intelligent hydropower plant operation and management system based on digital twin, the operation data timing encoding unit 131 is used to extract the timing features of the operation data of the hydropower plant based on parameter samples to obtain water level timing correlation feature vectors, flow timing correlation feature vectors, water pressure timing correlation feature vectors, turbine speed timing correlation feature vectors and power generation timing correlation feature vectors. It should be understood that, considering that the water level value, flow value, water pressure value, turbine speed value and power generation are parameters that change dynamically over time, therefore, in order to effectively capture the time series change characteristics of the operating status of the hydropower plant, the present disclosure adopts a sequence encoder based on the 1D-CNN model to perform time series encoding on the time queue of the water level value, the time queue of the flow value, the time queue of the water pressure value, the time queue of the turbine speed value and the time queue of the power generation, respectively, and performs a one-dimensional sliding convolution operation on each parameter in the time dimension to extract the dynamic change characteristics of each parameter over time, capture the time series change trend of each parameter, and thus obtain the water level time series correlation feature vector, flow time series correlation feature vector, water pressure time series correlation feature vector, turbine speed time series correlation feature vector and power generation time series correlation feature vector, so as to provide effective time series data support for the subsequent turbine operating status analysis.

[0032] Figure 4It is a schematic block diagram of an operation data time series encoding unit in a digital twin-based intelligent hydropower plant operation management system according to an embodiment of the present disclosure. Optionally, in one embodiment of the present disclosure, the operation data time series encoding unit 131 includes: a data regularization subunit 1311, which is used to regularize the operation data of the hydropower plant according to the parameter sample dimension to obtain a time queue of water level values, a time queue of flow values, a time queue of water pressure values, a time queue of turbine speed values, and a time queue of power generation. An operation parameter time series feature extraction subunit 1312 is used to input the time queue of water level values, the time queue of flow values, the time queue of water pressure values, the time queue of turbine speed values, and the time queue of power generation into a sequence encoder based on a 1D-CNN model to obtain the water level time series association feature vector, the flow time series association feature vector, the water pressure time series association feature vector, the turbine speed time series association feature vector, and the power generation time series association feature vector.

[0033] In the above-mentioned intelligent hydropower plant operation and management system based on digital twin, the power generation dependent variable multimodal fusion unit 132 is used to perform multimodal fusion on the water level time series associated feature vector, the flow rate time series associated feature vector, the water pressure time series associated feature vector and the turbine speed time series associated feature vector to obtain a multimodal embedded fusion representation vector of the power generation dependent variable. It should be understood that in order to comprehensively consider the influence of water level, flow rate, water pressure and turbine speed on power generation, the water level time series associated feature vector, the flow rate time series associated feature vector, the water pressure time series associated feature vector and the turbine speed time series associated feature vector are further fused to comprehensively utilize the complementary information between multimodal data and improve the accuracy and comprehensiveness of the analysis. In the technical solution disclosed in the present invention, a multimodal fusion module of power generation dependent variables based on an embedding layer is used to implement this fusion process. Specifically, the embedding layer maps the time-series correlation feature vectors of various parameters into a common feature space, so that the features of different modal data can effectively interact and fuse in the shared feature space, thereby effectively capturing the intrinsic connections and interactions between various parameters and generating a multimodal embedding fusion representation vector of the power generation dependent variable.

[0034] Optionally, in one embodiment of the present disclosure, the power generation dependent variable multimodal fusion unit is used to: input the water level time series associated feature vector, the flow time series associated feature vector, the water pressure time series associated feature vector and the turbine speed time series associated feature vector into the power generation dependent variable multimodal fusion module based on the embedding layer to obtain the power generation dependent variable multimodal embedded fusion representation vector.

[0035] In the above-mentioned intelligent hydropower plant operation and management system based on digital twin, the power generation causal interaction analysis unit 133 is used to perform fine-grained interaction analysis on the multimodal embedding fusion representation vector of the power generation dependent variable and the power generation time series correlation feature vector to obtain the power generation dependent variable-power generation decoupling full time series enhanced interaction representation vector. It should be understood that in order to reveal the complex correlation between power generation and water level, flow, water pressure and turbine speed, the embodiment of the present disclosure further introduces a feature decoupling local interaction fine-grained aggregation analysis network to perform correlation interaction analysis on the multimodal embedding fusion representation vector of the power generation dependent variable and the power generation time series correlation feature vector. Specifically, the network first decouples the multimodal embedding fusion representation vector of the power generation dependent variable and the power generation time series correlation feature vector respectively, and decomposes the two into a sequence of two groups of local feature vectors to enhance the sensitivity to local detail features. Next, the local semantic association modeling is performed on the sequence of the two groups of local feature vectors after decoupling to generate a local semantic association interaction vector sequence, and on this basis, the feature interaction response strength is further quantified to generate an attention weight sequence, so that the model can focus on the most important feature interactions, thereby improving the quality of feature representation and prediction accuracy. At the same time, in order to prevent overfitting, the generated attention weight sequence is further subjected to random inactivation processing to enhance the generalization ability of the model. Finally, the attention weight sequence after random inactivation processing is used as the weight distribution, and the local semantic association interaction vector sequence is weighted and summed to synthesize the local interaction information of the multimodal embedding fusion representation vector of the power generation dependent variable and the power generation time series association feature vector, and obtain the global power generation dependent variable-power generation decoupled full time series enhanced interaction representation vector, thereby achieving a comprehensive understanding of the turbine operation status.

[0036] Optionally, in one embodiment of the present disclosure, the power generation causal interaction analysis unit is used to: input the power generation dependent variable multimodal embedding fusion representation vector and the power generation time series correlation feature vector into a feature decoupled local interaction fine-grained aggregation analysis network to obtain the power generation dependent variable-power generation decoupled full time series reinforcement interaction representation vector.

[0037] Figure 5It is a schematic block diagram of the power generation causal interaction analysis unit in the digital twin-based intelligent hydropower plant operation and management system of the embodiment of the present disclosure. Optionally, in one embodiment of the present disclosure, the power generation causal interaction analysis unit 133 includes: a feature decoupling subunit 1331, which is used to perform feature decoupling on the power generation dependent variable multimodal embedding fusion representation vector and the power generation time series association feature vector to obtain a sequence of power generation dependent variable multimodal fusion feature local representation vectors and a sequence of power generation time series association feature local representation vectors. A local semantic association interaction subunit 1332 is used to calculate the local semantic association interaction vector between each group of corresponding power generation dependent variable multimodal fusion feature local representation vectors and power generation time series association feature local representation vectors in the sequence of power generation dependent variable multimodal fusion feature local representation vectors and the sequence of power generation time series association feature local representation vectors to obtain a sequence of power generation dependent variable-power generation local semantic association interaction vectors. The attention weight calculation subunit 1333 is used to calculate the fine-grained attention weights of each power generation dependent variable-power generation local semantic association interaction vector in the sequence of power generation dependent variable-power generation local semantic association interaction vectors to obtain a sequence of fine-grained attention weights. The random deactivation subunit 1334 is used to input the sequence of fine-grained attention weights into the random deactivation module to obtain a sequence of fine-grained attention weights after deactivation processing. The local interaction information aggregation subunit 1335 is used to use the sequence of fine-grained attention weights after deactivation processing as the weight distribution, and calculate the weighted sum of the sequence of power generation dependent variable-power generation local semantic association interaction vectors to obtain the power generation dependent variable-power generation decoupled full-time series reinforcement interaction representation vector.

[0038] Optionally, in one embodiment of the present disclosure, the local semantic association interaction subunit is used to: calculate the position point addition, position point subtraction and position point multiplication between the local representation vector of the multimodal fusion feature of the power generation dependent variable and the local representation vector of the time series association feature of the corresponding power generation amount to obtain the first fusion result, the second fusion result and the third fusion result. After the first fusion result, the second fusion result and the third fusion result are cascaded into a multi-scale fusion feature vector of the power generation dependent variable-power generation amount, one-dimensional convolution processing and maximum pooling processing are performed on them to obtain the local semantic association interaction vector of the power generation dependent variable-power generation amount.

[0039] Optionally, in one embodiment of the present disclosure, the attention weight calculation subunit is used to: multiply the power generation dependent variable-power generation local semantic association interaction vector by a weight coefficient matrix, add it to the bias vector and pass it through a tanh function to obtain a power generation dependent variable-power generation local semantic association interaction encoding feature vector. Use an attention score reference vector to multiply the power generation dependent variable-power generation local semantic association interaction encoding feature vector to obtain the fine-grained attention weight.

[0040] In summary, in the above embodiment, the power generation causal interaction analysis unit 133 is used to process the battery temperature time series propagation aggregation representation vector and the battery real-time power time series propagation aggregation representation vector using the following interaction response analysis formula to obtain the power generation dependent variable-power generation decoupling full time series reinforcement interaction representation vector, wherein the interaction response analysis formula is:

[0041] Decouple(x1)={x 11 , x 12 , ..., x 1i ..., x 1n}

[0042] Decouple(x2)={x 21 , x 22 , ..., x 2i , ..., x 2n}

[0043]

[0044]

[0045] W i = dropout(E i )

[0046]

[0047] Wherein, x1 represents the battery temperature time series propagation aggregation representation vector, x2 represents the battery real-time power time series propagation aggregation representation vector, Decouple(·) represents decoupling processing, x 11 、x 12 、x 1i and x 1n They respectively represent the first, second, i-th and n-th local representation vectors of multimodal fusion features of the power generation dependent variable in the sequence of the local representation vectors of multimodal fusion features of the power generation dependent variable, and the value of n is the number of feature vectors in the sequence of the local representation vectors of multimodal fusion features of the power generation dependent variable, x 21 、x 22 、x 1i and x 2n denote the first, second, i-th and n-th local representation vectors of the time series correlation feature of the power generation respectively, [·, ·] denotes a cascade operation, MaxPool(·) denotes a maximum pooling operation, conv1D(·) denotes a one-dimensional convolution operation, ⊙ denotes a dot product, Indicates point addition, Indicates point minus, fi (x 1i , x 2i ) represents the i-th power generation dependent variable-power generation amount local semantic association interaction vector in the sequence of the power generation dependent variable-power generation amount local semantic association interaction vector, W r is the weight coefficient matrix, b r is the bias vector, tanh is the hyperbolic tangent function, is the attention score reference vector, E i represents the i-th fine-grained attention weight, dropout(·) represents random inactivation processing, and W i represents the fine-grained attention weight after the i-th inactivation process, and v represents the power generation dependent variable-power generation decoupled full-time reinforcement interaction representation vector.

[0048] In the above-mentioned intelligent hydropower plant operation and management system based on digital twin, the operation analysis result generating unit 134 is used to generate the operation analysis result based on the power generation dependent variable-power generation decoupling full time series enhanced interaction representation vector. It should be understood that generating the operation analysis result based on the power generation dependent variable-power generation decoupling full time series enhanced interaction representation vector can improve the depth of understanding of the operation status of the hydropower plant by comprehensively analyzing multi-source heterogeneous data, thereby providing technical support for intelligent monitoring and preventive maintenance, and ultimately improving the overall operation efficiency and reliability of the hydropower plant.

[0049] Optionally, in one embodiment of the present disclosure, the operation analysis result generating unit is used to: input the power generation dependent variable-power generation decoupled full-time reinforcement interaction representation vector into a classifier-based operation analysis result generator to obtain the operation analysis result.

[0050] In the above-mentioned intelligent hydropower plant operation management system based on digital twin, the operation analysis result display module 140 is used to display the operation analysis results in real time using the digital twin model. It should be understood that the digital twin model can reflect the state of the hydropower plant in the physical world in real time. By displaying the operation analysis results in the digital twin model, managers can quickly understand whether the working status of the turbine is normal, discover potential problems in time and take measures to avoid downtime or other adverse consequences caused by equipment failure. This real-time monitoring and visualization method improves the operating efficiency of the hydropower plant, reduces the losses caused by fault detection delays, and enhances the reliability of the system.

[0051] In summary, according to the embodiment of the present disclosure, the intelligent hydropower plant operation management system based on digital twin is explained, which constructs a digital twin model of the hydropower plant through digital twin technology, collects the operation data of the hydropower plant through sensing technology, and uses the Internet of Things technology to achieve real-time data synchronization between the digital twin model and the hydropower plant. At the same time, the operation data of the hydropower plant is intelligently analyzed, and based on the time-series interaction mode between the power generation and the various operating parameters that affect the power generation, the working state of the turbine is intelligently analyzed to see if there is any abnormality, and the analysis results are synchronously displayed in the digital twin model of the hydropower plant. In this way, comprehensive perception, real-time monitoring and intelligent analysis of the operating state of the hydropower plant can be effectively achieved, thereby improving the overall operational efficiency and reliability of the hydropower plant.

[0052] Figure 6 FIG. 1 is a schematic flow chart of an intelligent hydropower plant operation management method based on digital twins according to an embodiment of the present disclosure. Figure 6 As shown, the operation management method of an intelligent hydropower plant based on digital twins includes: S1, collecting the operation data of the hydropower plant in real time through a sensor group and monitoring equipment, and the operation data includes water level value, flow value, water pressure value, turbine speed value and power generation. S2, constructing a digital twin model of the hydropower plant, and realizing real-time synchronization of data between the hydropower plant and the digital twin model through Internet of Things technology. S3, using the digital twin model to analyze the operation data of the hydropower plant to obtain an operation analysis result, and the operation analysis result is used to indicate whether there is an abnormality in the working state of the turbine. S4, using the digital twin model to display the operation analysis result in real time.

[0053] Figure 7 The present invention is a schematic flow chart of an intelligent hydropower plant operation management method based on digital twins in an embodiment of the present invention, in which the digital twin model is used to analyze the operation data of the hydropower plant to obtain an operation analysis result. Figure 7As shown, the digital twin model is used to perform data analysis on the operation data of the hydropower plant to obtain an operation analysis result, and the operation analysis result is used to indicate whether there is an abnormality in the working state of the turbine, including: S31, extracting the time series features of the operation data of the hydropower plant based on parameter samples to obtain a water level time series correlation feature vector, a flow time series correlation feature vector, a water pressure time series correlation feature vector, a turbine speed time series correlation feature vector and a power generation time series correlation feature vector. S32, multi-modal fusion is performed on the water level time series correlation feature vector, the flow time series correlation feature vector, the water pressure time series correlation feature vector and the turbine speed time series correlation feature vector to obtain a power generation dependent variable multi-modal embedding fusion representation vector. S33, fine-grained interaction analysis is performed on the power generation dependent variable multi-modal embedding fusion representation vector and the power generation time series correlation feature vector to obtain a power generation dependent variable-power generation decoupling full-time series enhanced interaction representation vector. S34, based on the power generation dependent variable-power generation decoupling full-time series enhanced interaction representation vector, the operation analysis result is generated.

[0054] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned intelligent hydropower plant operation management method based on digital twins have been referred to above. Figures 1 to 5 It has been introduced in detail in the description of the digital twin-based intelligent hydropower plant operation and management system, and therefore, its repeated description will be omitted.

[0055] The basic principles of the embodiments of the present disclosure are described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the embodiments of the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details of the above embodiments are only for the purpose of illustration and ease of understanding, rather than limitation, and the above details do not limit the embodiments of the present disclosure to the necessity of adopting the above specific details to be implemented.

[0056] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided in the present disclosure, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the module division is only a logical function division, and there may be other division methods in actual implementation. The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0057] In addition, each functional module in each embodiment of the present disclosure may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0058] It is obvious to those skilled in the art that the embodiments of the present disclosure are not limited to the details of the exemplary embodiments described above, and that the embodiments of the present disclosure can be implemented in other specific forms without departing from the spirit or basic features of the embodiments of the present disclosure. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the embodiments of the present disclosure is defined by the attached claims rather than the above description, and it is intended that all changes that fall within the meaning and scope of the equivalent elements of the claims are included in the embodiments of the present disclosure. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0059] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units stated in the device claim can also be implemented by one unit through software or hardware.

[0060] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the embodiments of the present disclosure, but the embodiments of the present disclosure are not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the embodiments of the present disclosure, and these modifications and improvements are also considered to be within the protection scope of the embodiments of the present disclosure.

Claims

1. An intelligent hydropower plant operation management system based on digital twins, characterized in that: include: A hydropower plant operation monitoring module, which is used to collect the operation data of the hydropower plant in real time through a sensor group and monitoring equipment, wherein the operation data includes water level value, flow value, water pressure value, turbine speed value and power generation; A digital twin module, used to construct a digital twin model of the hydropower plant and to achieve real-time data synchronization between the hydropower plant and the digital twin model through Internet of Things technology; a turbine operation status analysis module, used to perform data analysis on the operation data of the hydropower plant using the digital twin model to obtain an operation analysis result, wherein the operation analysis result is used to indicate whether there is an abnormality in the working status of the turbine; An operation analysis result display module, used to display the operation analysis result in real time using the digital twin model; Wherein, the turbine operation status analysis module comprises: An operation data time series encoding unit, used for extracting time series features of the operation data of the hydropower plant based on parameter samples to obtain a water level time series correlation feature vector, a flow rate time series correlation feature vector, a water pressure time series correlation feature vector, a turbine speed time series correlation feature vector and a power generation time series correlation feature vector; A power generation dependent variable multimodal fusion unit, used for performing multimodal fusion on the water level time series associated feature vector, the flow rate time series associated feature vector, the water pressure time series associated feature vector and the turbine speed time series associated feature vector to obtain a power generation dependent variable multimodal embedded fusion representation vector; A power generation causal interaction analysis unit, used for performing fine-grained interaction analysis on the power generation dependent variable multimodal embedding fusion representation vector and the power generation time series correlation feature vector to obtain a power generation dependent variable-power generation decoupling full time series enhanced interaction representation vector; An operation analysis result generating unit is used to generate the operation analysis result based on the power generation dependent variable-power generation decoupled full-time enhanced interaction representation vector.

2. The intelligent hydropower plant operation and management system based on digital twin according to claim 1 is characterized in that: The operation data timing encoding unit comprises: A data regularization subunit, used for regularizing the operation data of the hydropower plant according to the parameter sample dimension to obtain a time queue of water level values, a time queue of flow values, a time queue of water pressure values, a time queue of turbine speed values ​​and a time queue of power generation; The operating parameter timing feature extraction subunit is used to input the time queue of the water level value, the time queue of the flow value, the time queue of the water pressure value, the time queue of the turbine speed value and the time queue of the power generation into a sequence encoder based on the 1D-CNN model to obtain the water level timing association feature vector, the flow timing association feature vector, the water pressure timing association feature vector, the turbine speed timing association feature vector and the power generation timing association feature vector.

3. The intelligent hydropower plant operation management system based on digital twin according to claim 2 is characterized in that: The power generation dependent variable multi-modal fusion unit is used to: The water level time series associated feature vector, the flow rate time series associated feature vector, the water pressure time series associated feature vector and the turbine speed time series associated feature vector are input into a power generation dependent variable multimodal fusion module based on an embedding layer to obtain the power generation dependent variable multimodal embedding fusion representation vector.

4. The digital twin-based intelligent hydropower plant operation and management system according to claim 3 is characterized in that: The power generation causal interaction analysis unit is used to: The multimodal embedding fusion representation vector of the power generation dependent variable and the power generation amount time series associated feature vector are input into a feature decoupled local interaction fine-grained aggregation analysis network to obtain the power generation dependent variable-power generation amount decoupled full time series enhanced interaction representation vector.

5. The digital twin-based intelligent hydropower plant operation and management system according to claim 4 is characterized in that: The power generation causal interaction analysis unit comprises: A feature decoupling subunit, used to perform feature decoupling on the power generation dependent variable multimodal embedding fusion representation vector and the power generation amount time series associated feature vector to obtain a sequence of power generation dependent variable multimodal fusion feature local representation vectors and a sequence of power generation amount time series associated feature local representation vectors; A local semantic association interaction subunit is used to calculate the local semantic association interaction vector between each group of corresponding local representation vectors of the multimodal fusion features of the power generation dependent variable and the local representation vectors of the time series association features of the power generation amount in the sequence of the local representation vectors of the multimodal fusion features of the power generation dependent variable and the sequence of the local representation vectors of the time series association features of the power generation amount, so as to obtain a sequence of local semantic association interaction vectors of the power generation dependent variable-power generation amount; An attention weight calculation subunit, used to calculate the fine-grained attention weight of each power generation dependent variable-power generation amount local semantic association interaction vector in the sequence of power generation dependent variable-power generation amount local semantic association interaction vectors to obtain a sequence of fine-grained attention weights; A random dropout subunit, configured to input the sequence of fine-grained attention weights into a random dropout module to obtain a sequence of fine-grained attention weights after dropout processing; The local interaction information aggregation subunit is used to calculate the weighted sum of the sequence of the power generation dependent variable-power generation local semantic association interaction vectors using the sequence of fine-grained attention weights after the deactivation process as the weight distribution to obtain the power generation dependent variable-power generation decoupled full-time series reinforcement interaction representation vector.

6. The digital twin-based intelligent hydropower plant operation and management system according to claim 5 is characterized in that: The local semantic association interaction subunit is used to: Calculate the position point addition, position point subtraction and position point multiplication between the local representation vector of the multimodal fusion feature of the power generation dependent variable and the local representation vector of the power generation time series associated feature corresponding thereto to obtain a first fusion result, a second fusion result and a third fusion result; After the first fusion result, the second fusion result and the third fusion result are cascaded into a power generation dependent variable-power generation multi-scale fusion feature vector, one-dimensional convolution processing and maximum pooling processing are performed on them to obtain the power generation dependent variable-power generation local semantic association interaction vector.

7. The digital twin-based intelligent hydropower plant operation and management system according to claim 6 is characterized in that: The attention weight calculation subunit is used to: The power generation dependent variable-power generation amount local semantic association interaction vector is multiplied by a weight coefficient matrix, and then added to the bias vector and passed through a tanh function to obtain a power generation dependent variable-power generation amount local semantic association interaction encoding feature vector; The attention score reference vector is multiplied by the power generation dependent variable-power generation local semantic association interaction encoding feature vector to obtain the fine-grained attention weight.

8. The intelligent hydropower plant operation and management system based on digital twin according to claim 7 is characterized in that: The operation analysis result generating unit is used to: The power generation dependent variable-power generation decoupled full-time reinforcement interaction representation vector is input into the classifier-based operation analysis result generator to obtain the operation analysis result.

9. A method for operation and management of an intelligent hydropower plant based on digital twins, characterized in that: include: Collecting the operation data of the hydropower plant in real time through the sensor group and monitoring equipment, the operation data including water level value, flow value, water pressure value, turbine speed value and power generation; Constructing a digital twin model of the hydropower plant, and achieving real-time data synchronization between the hydropower plant and the digital twin model through Internet of Things technology; Using the digital twin model to perform data analysis on the operating data of the hydropower plant to obtain an operating analysis result, wherein the operating analysis result is used to indicate whether there is an abnormality in the working state of the turbine; Using the digital twin model to display the operation analysis results in real time; The digital twin model is used to perform data analysis on the operation data of the hydropower plant to obtain an operation analysis result, and the operation analysis result is used to indicate whether there is an abnormality in the working state of the turbine, including: Extracting time series features based on parameter samples from the operation data of the hydropower plant to obtain a water level time series correlation feature vector, a flow rate time series correlation feature vector, a water pressure time series correlation feature vector, a turbine speed time series correlation feature vector, and a power generation time series correlation feature vector; Performing multimodal fusion on the water level time series associated feature vector, the flow rate time series associated feature vector, the water pressure time series associated feature vector, and the turbine speed time series associated feature vector to obtain a multimodal embedded fusion representation vector of a power generation dependent variable; Performing fine-grained interactive analysis on the multimodal embedding fusion representation vector of the power generation dependent variable and the power generation amount time series associated feature vector to obtain a power generation dependent variable-power generation amount decoupling full time series enhanced interactive representation vector; The operation analysis result is generated based on the power generation dependent variable-power generation decoupled full-time enhanced interaction representation vector.

Images