A power equipment modular real-time updating system

Abstract

The application discloses a kind of electric power equipment modular real-time updating system, belongs to electric power equipment automation technical field, including equipment modularization module, electric power equipment is divided to obtain electric power equipment submodule;Data acquisition module obtains the update data of electric power equipment submodule, data preprocessing module, update data is preprocessed, obtain the update data after pre-processing, prediction analysis module, based on dependency, cost and time establish multi-objective function, determine the electric power equipment submodule needing updating, decision update module determines the update order priority of electric power equipment submodule;Real-time updating module, according to the directed graph of dependency, update request is forwarded to source node according to priority order, ensure layer-by-layer response and execute update, to improve equipment reliability and update efficiency.The problems that the update strategy is not reasonable due to the unclear update order priority and the update process system crashes due to the unclear dependency in the prior art are solved.

Description

Technical Field

[0001] This invention relates to the field of power equipment automation technology, and in particular to a modular real-time update system for power equipment. Background Technology

[0002] In recent years, modular upgrade technology for power equipment has become an important direction for the development of modern power systems. This technology divides power equipment into multiple independent, functionally defined modules, each with specific functions and standardized interfaces, facilitating replacement and upgrades. Modular design significantly improves the flexibility, scalability, and maintenance efficiency of power systems. Existing modular upgrade technologies for power equipment include distributed design, interface standardization, and intelligent monitoring and maintenance systems. Distributed design separates different functional modules, avoiding mutual interference between modules and improving system stability and reliability. Interface standardization enables seamless connection of equipment from different manufacturers, reducing replacement and upgrade costs. Intelligent monitoring and maintenance systems can monitor the operating status of equipment in real time, detect potential faults early, and provide strong support for module updates and maintenance.

[0003] While modular upgrade technology for power equipment offers numerous advantages, it also faces several challenges in practical applications. Two key issues are determining dependencies and update priorities. First, complex dependencies often exist between power equipment modules; updating one module can affect the normal operation of others. These dependencies make module updates complex and unpredictable, requiring consideration of multiple factors such as equipment functionality, interfaces, and compatibility. Second, given the large number of modules in a power system, determining which modules require priority updates is challenging. Different modules have varying importance within the system, and their update priorities should also differ. However, existing technologies often struggle to accurately assess the importance and update priority of each module, leading to a lack of scientific basis for update strategies and potentially causing system instability or performance degradation. Therefore, resolving the issues of dependency determination and update priority is crucial for the development of modular upgrade technology for power equipment. Summary of the Invention

[0004] The technical problem solved by this invention is to address the issues of unreasonable update strategies due to unclear update priorities and system crashes during the update process caused by unclear dependencies in the modular update of existing power equipment.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] This invention provides a modular real-time update system for power equipment, comprising:

[0007] The modular equipment module is used to divide power equipment into sensor sub-modules, control sub-modules, and execution sub-modules to obtain power equipment sub-modules.

[0008] The data acquisition module is used to acquire updated data from the power equipment sub-module, including current, voltage, temperature, humidity, vibration, fault records, and maintenance data.

[0009] The data preprocessing module is used to preprocess the updated data to obtain preprocessed updated data;

[0010] The predictive analysis module is used to extract features from the pre-trained digital twin model module to obtain feature vectors, classify the feature vectors using logistic regression, map the feature vectors to the probability distribution of fault types, and obtain fault type classification prediction results; perform linear transformation on the feature vectors to obtain the remaining life prediction value of the power equipment; and obtain the prediction result of the digital twin model module using a weighted average method based on the fault type classification prediction result and the remaining life prediction value of the power equipment, wherein the prediction result of the digital twin model module includes the power equipment sub-modules that need to be updated.

[0011] The update decision module is used to establish a multi-objective function based on the dependencies, update costs, and update times of the updated power equipment sub-modules, obtain the optimization results of the multi-objective function using a multi-objective evolutionary algorithm, and output an update decision based on the optimization results of the multi-objective function. The update decision includes the update order priority of the power equipment sub-modules that need to be updated.

[0012] The real-time update module is used to form a directed graph structure including source nodes and multiple layers of intermediate nodes based on dependencies and the power equipment sub-modules that need to be updated. According to the update order priority, the power equipment sub-modules that need to be updated are received as receiving nodes in sequence to receive update requests. If the receiving node is not the source node, the update request is forwarded to the preceding node of the receiving node until the update request is forwarded to the source node to respond to the update request. Among them, all nodes that receive the update request respond to the update request. If the receiving node is the source node, the receiving node responds to the update request.

[0013] Furthermore, the training method for the digital twin model module includes:

[0014] The online learning module receives preprocessed updated data in real time and updates the parameters of the digital twin model module. An adaptive learning rate adjustment strategy is adopted to dynamically adjust the learning rate according to the gradient of the loss function, so as to obtain the trained digital twin model module.

[0015] Further, the updated data is preprocessed to obtain preprocessed updated data, including:

[0016] The missing values ​​in the updated data are filled with the mode to obtain the updated data after filling the missing values.

[0017] Use 3 The principle is to identify and delete outliers in the updated data after filling in missing values, thus obtaining preprocessed updated data.

[0018] Furthermore, logistic regression is used to classify the feature vectors, mapping them to the probability distribution of fault types to obtain fault type classification prediction results, wherein the fault type classification prediction results are expressed as:

[0019] ;

[0020] In the formula, In the eigenvector Under these conditions, fault type For category indexing The probability, Represents category index The transpose of the weights, Represents category index The bias, Indicates the total number of category indexes. Indicates the first The transpose of the weights of each category index. Indicates the first Bias of each category index This represents an exponential function.

[0021] Further, a linear transformation is performed on the feature vector to obtain the predicted remaining life of the power equipment, wherein the predicted remaining life of the power equipment is expressed as:

[0022] ;

[0023] In the formula, RUL represents the predicted remaining life of the electrical equipment. Representing the regression coefficients, these are factors influencing the remaining lifespan of electrical equipment. This represents the intercept, which is the initial lifespan of the power equipment.

[0024] Furthermore, a multi-objective function is established based on the dependencies, update costs, and update times of the updated power equipment sub-modules, and the multi-objective function is expressed as:

[0025] ;

[0026] In the formula, Represents the combination of decision variables The minimum multi-objective function value, where the combination of decision variables is... For all power equipment sub-modules that need updating; Indicates the update cost function The weighting coefficients, Indicates the function to update time The weighting coefficients, Dependency strength function The weighting coefficients.

[0027] Furthermore, the update cost function Represented as:

[0028] ;

[0029] In the formula, This indicates the total number of updated power equipment sub-modules. Indicates the first Decision variables of each power equipment submodule The corresponding update cost, Indicates the relationship with the first The power equipment submodule has dependencies on the first Decision variables for each power equipment submodule This indicates that the first one is updated simultaneously. The power equipment submodule and the first The power equipment submodule has dependencies on the first The update cost corresponding to each power equipment sub-module This represents the fixed cost, which is a decision variable independent of the power equipment submodule.

[0030] Furthermore, the update time function Represented as:

[0031] ;

[0032] In the formula, Indicates the update of the first The maximum bottleneck time in the process of each power equipment sub-module. This indicates the total number of updated power equipment sub-modules. Indicates the first Decision variables of each power equipment submodule The corresponding update time, Indicates the first The power equipment submodule belongs to the first A set of power equipment submodules with dependencies. , This indicates the additional update time due to dependencies, meaning the first update must be performed first. Only the power equipment submodule can be updated. The update time corresponding to each power equipment sub-module Indicates the relationship with the first The power equipment submodules have dependencies on the first Decision variables for each power equipment sub-module.

[0033] Furthermore, the dependency strength function Represented as:

[0034] ;

[0035] In the formula, Indicates the first The power equipment submodule and the first Physical connection dependencies of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Data exchange dependencies of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Functional dependencies of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Energy supply dependence of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Spatial location dependency of each power equipment submodule Intensity control factor.

[0036] Furthermore, the optimization results of the multi-objective function are obtained using a multi-objective evolutionary algorithm, including:

[0037] (1) There are n power equipment sub-modules that need to be updated, of which the nth sub-module is the first one. The update cost for each power equipment submodule is: and update time is , No. Decision variables of each power equipment submodule Decision variables A value of 1 indicates an update; the decision variable... A value of 0 indicates no update;

[0038] Based on logical constraints and priority constraints, constraints are established according to the dependencies between power equipment sub-modules. Specifically, during the update of the... Before updating the power equipment submodule, the first one must be updated. Other submodules that a power equipment submodule depends on;

[0039] (2) The multi-objective function is used as the optimization objective;

[0040] (3) Randomly generate a set of decision variable combinations that satisfy dependency constraints as the initial population;

[0041] (4) Based on the multi-objective function, perform fitness evaluation on each set of decision variable combinations and calculate the multi-objective function value for each set of decision variable combinations;

[0042] (5) Take the combination of decision variables whose multi-objective function values ​​exceed the threshold as the parent generation, perform crossover and mutation operations to generate new offspring individuals, and remove offspring individuals that do not meet the constraints to obtain all combinations of decision variables;

[0043] (6) Repeat steps (4)-(5) for all combinations of decision variables until the predetermined number of iterations or convergence condition is reached, and obtain all combinations of decision variables after iteration;

[0044] (7) Select a set of Pareto optimal solutions from all combinations of decision variables after iteration as the optimization result of the multi-objective function.

[0045] The beneficial effects of this invention are:

[0046] 1. Compared with the prior art, the present invention, based on the fault type classification prediction results and the remaining life prediction value of the power equipment, uses a weighted average method to more reasonably determine the power equipment sub-modules that need to be updated, thereby avoiding unnecessary update operations, reducing update costs, and improving the stability and operating efficiency of the power system.

[0047] 2. This invention innovatively considers the dependencies between power equipment sub-modules. Based on the dependencies of the updated power equipment sub-modules, update costs, and update times, a multi-objective function is established to determine the update priority of the power equipment sub-modules that need to be updated, thus optimizing the update strategy and solving the problem of unreasonable update strategies caused by unclear update priorities in existing modular updates of power equipment.

[0048] 3. This invention uses a directed graph structure to clearly represent dependencies and designs an update mechanism that forwards update requests layer by layer, ensuring the sequentiality and coordination of the update process and effectively avoiding system crashes caused by unclear dependencies. Compared with existing technologies, this invention can more robustly handle dependencies during the update process, ensuring stable system operation and improving the reliability and security of the update process. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the modular structure of a real-time update system for power equipment according to an embodiment of the present invention;

[0050] Figure 2 This is a schematic diagram illustrating the basic process of obtaining the optimization result of a multi-objective function using a multi-objective evolutionary algorithm, as provided in one embodiment of the present invention.

[0051] Figure 3 This is a schematic diagram of an update mechanism for forwarding update requests layer by layer, provided in one embodiment of the present invention. Detailed Implementation

[0052] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0053] Example 1

[0054] like Figure 1 As shown in the figure, this embodiment introduces a modular real-time update system for power equipment, including:

[0055] The equipment modularization module is used to divide power equipment into sensor sub-modules, control sub-modules, and execution sub-modules to obtain power equipment sub-modules.

[0056] Based on the functional and structural characteristics of power equipment, dividing power equipment into sensor sub-modules, control sub-modules, and execution sub-modules helps to achieve modular management and updates of the equipment.

[0057] This invention decomposes power equipment into multiple independent sub-modules based on modular equipment modules, which makes it easier to monitor, maintain and update each sub-module, improving the maintainability and scalability of the equipment. When power equipment fails or needs to be upgraded, the problematic power equipment sub-module can be located and replaced more quickly, reducing the downtime of the entire system.

[0058] The data acquisition module is used to acquire updated data from the power equipment sub-module, including current, voltage, temperature, humidity, vibration, fault records, and maintenance data.

[0059] This invention, based on modular equipment modules, uses sensors and other data acquisition devices to acquire real-time updated data from sub-modules of power equipment, including current, voltage, temperature, humidity, vibration, fault records, and maintenance data. This allows for real-time monitoring of the operating status of power equipment, timely detection of potential problems, and implementation of corresponding measures. It ensures the accuracy and timeliness of update decisions, helps reduce equipment failure rates, and improves the stability and security of the power system.

[0060] The data preprocessing module is used to preprocess the updated data to obtain preprocessed updated data.

[0061] This invention uses a data preprocessing module to clean, transform, and format the acquired updated data to eliminate noise and outliers, thereby improving data quality. The data preprocessing module provides high-quality data input to the predictive analysis module, improving the efficiency and accuracy of predictive analysis and reducing the risk of misjudgment due to data quality issues.

[0062] The predictive analysis module is used to extract features from the pre-trained digital twin model module to obtain feature vectors, classify the feature vectors using logistic regression, map the feature vectors to the probability distribution of fault types, and obtain fault type classification prediction results; perform linear transformation on the feature vectors to obtain the remaining life prediction value of the power equipment; and obtain the prediction result of the digital twin model module using a weighted average method based on the fault type classification prediction result and the remaining life prediction value of the power equipment. The prediction result of the digital twin model module includes the power equipment sub-modules that need to be updated.

[0063] This invention utilizes a pre-trained digital twin model module to extract features and classify updated data based on a predictive analysis module. The digital twin model is a highly realistic virtual model capable of simulating the actual operating state of power equipment. Through in-depth mining and analysis of real-time power equipment data, the predictive analysis module accurately predicts the fault types and remaining lifespan of the power equipment, providing a scientific basis for update decisions. This helps to detect equipment faults early and take corresponding maintenance measures, extending the equipment's service life and reducing maintenance costs.

[0064] The update decision module is used to establish a multi-objective function based on the dependencies, update costs, and update times of the updated power equipment sub-modules, obtain the optimization results of the multi-objective function using a multi-objective evolutionary algorithm, and output an update decision based on the optimization results of the multi-objective function. The update decision includes the update order priority of the power equipment sub-modules that need to be updated.

[0065] This invention establishes a multi-objective function based on the dependencies, update costs, and update time of the updated power equipment sub-modules in the update decision module, and uses a multi-objective evolutionary algorithm for optimization. By comprehensively considering multiple factors, the update decision module derives the optimal update strategy, including the equipment sub-modules that need to be updated and their priorities, ensuring the scientific and rational nature of the update process, reducing the risk of system collapse due to improper updates, and improving the stability and reliability of the power system.

[0066] The real-time update module is used to form a directed graph structure including source nodes and multiple layers of intermediate nodes based on dependencies and the power equipment sub-modules that need to be updated. According to the update order priority, the power equipment sub-modules that need to be updated are received as receiving nodes in sequence to receive update requests. If the receiving node is not the source node, the update request is forwarded to the preceding node of the receiving node until the update request is forwarded to the source node to respond to the update request. Among them, all nodes that receive the update request respond to the update request. If the receiving node is the source node, the receiving node responds to the update request.

[0067] This invention, based on a real-time update module, forms a directed graph structure including source nodes and multiple layers of intermediate nodes according to dependencies and the power equipment sub-modules that need to be updated. According to the update order priority, update requests are forwarded layer by layer to the dependent power equipment sub-modules, realizing fast and orderly updates of power equipment sub-modules, ensuring the smooth progress of the update process, improving the efficiency and accuracy of the update process, reducing system downtime caused by improper updates, and improving the operating efficiency and service quality of the power system.

[0068] Example 2

[0069] Similar to the inventive concept of Embodiment 1, this embodiment introduces a modular real-time update system for power equipment, the specific implementation process of which includes:

[0070] I. Modular Equipment Modules

[0071] The equipment is modularized into sub-modules, which divide the power equipment into sensor sub-modules, control sub-modules, and execution sub-modules.

[0072] In this embodiment, firstly, a comprehensive functional analysis of the power equipment is performed to clarify the functional characteristics of each component and the interrelationships between them. Based on the results of the functional analysis, the part of the power equipment responsible for data acquisition and monitoring is divided into a sensor submodule, the module responsible for data processing and decision-making is divided into a control submodule, and the part responsible for executing control commands and driving the equipment is divided into an execution submodule, thus obtaining the power equipment submodules.

[0073] Secondly, standardized interfaces are designed for each power equipment submodule to ensure seamless connection and communication between them. Furthermore, the interface design considers future scalability and compatibility, allowing for easy addition of new power equipment submodules or updates to existing ones when needed.

[0074] Finally, the divided power equipment sub-modules are integrated to form a complete power equipment. During the integration process, it is necessary to ensure the collaborative work between the various power equipment sub-modules, as well as the stability and reliability of the entire power equipment and the system containing the power equipment.

[0075] II. Data Acquisition Module

[0076] The data acquisition module is used to acquire updated data from the power equipment sub-module, including current, voltage, temperature, humidity, vibration, fault records, and maintenance data.

[0077] In this embodiment, various types of sensors, including current sensors, voltage sensors, temperature sensors, humidity sensors, and vibration sensors, are deployed at key locations in the power equipment submodule. The sensors collect updated data of the power equipment submodule in real time and transmit it to the data acquisition module for processing. At the same time, the collected data is stored in the database.

[0078] In this embodiment, the data acquisition module also has a data monitoring function, which can monitor the operating status of the power equipment sub-module in real time for training the digital twin model.

[0079] III. Data Preprocessing Module

[0080] The data preprocessing module is used to preprocess the updated data to obtain preprocessed updated data.

[0081] In this embodiment, the updated data is preprocessed to obtain preprocessed updated data, including:

[0082] In the updated data, there are some fields with empty or missing values. In most cases, the missing values ​​are most likely to be close to the mode of the updated data. In order to maintain the integrity of the data, this embodiment uses the mode, that is, the value that appears most frequently, to fill the missing values ​​in the data preprocessing module, and obtains the updated data after filling the missing values.

[0083] Outliers are data points that differ significantly from the majority of data points. These outliers may be caused by factors such as power equipment malfunctions, sensor errors, or data acquisition errors. In this embodiment, 3 is used in the data preprocessing module. The principle is to identify and delete outliers in the updated data after filling in missing values, thus obtaining preprocessed updated data. Among them, 3 The rule for identifying outliers is: if a data point's value exceeds the range of the average of its field plus or minus three standard deviations, then that data point is considered an outlier.

[0084] IV. Predictive Analysis Module

[0085] The predictive analysis module is used to extract features from the pre-trained digital twin model module to obtain feature vectors, classify the feature vectors using logistic regression, map the feature vectors to the probability distribution of fault types, and obtain fault type classification prediction results; perform linear transformation on the feature vectors to obtain the remaining life prediction value of the power equipment; and obtain the prediction result of the digital twin model module using a weighted average method based on the fault type classification prediction result and the remaining life prediction value of the power equipment. The prediction result of the digital twin model module includes the power equipment sub-modules that need to be updated.

[0086] In this embodiment, the training method for the digital twin model module includes:

[0087] The online learning module receives preprocessed updated data in real time and updates the parameters of the digital twin model module. An adaptive learning rate adjustment strategy is adopted to dynamically adjust the learning rate according to the gradient of the loss function, so as to obtain the trained digital twin model module.

[0088] In this embodiment, a trained digital twin model module is used to extract features from the preprocessed updated data to obtain feature vectors that reflect the operating status of power equipment. The feature vectors include current, voltage, temperature, humidity, and vibration data of the power equipment, which are crucial for fault type prediction and remaining life assessment of power equipment.

[0089] In this embodiment, a logistic regression algorithm is used to classify the feature vectors, mapping them to a probability distribution of fault types. Through logistic regression, the probability of a feature vector belonging to a particular fault type can be determined, thus obtaining a fault type classification prediction result. The fault types include generator faults, transformer faults, switchgear faults, short-circuit faults, open-circuit faults, overvoltage faults, undervoltage faults, frequency deviation faults, poor contact faults, and equipment aging faults.

[0090] This embodiment, while obtaining the feature vector, also performs a linear transformation on the feature vector to estimate the remaining lifespan of the power equipment. The linear transformation, based on the training data of the digital twin model module, obtains an accurate predicted value of the remaining lifespan of the power equipment by fitting the relationship between the power equipment's state and its remaining lifespan.

[0091] Based on the above fault types and the predicted remaining life of the power equipment, the final prediction result of the digital twin model module is obtained by using the weighted average method. This allows us to identify components with high fault risk or short remaining life and to determine the power equipment sub-modules that need to be updated.

[0092] In this embodiment, logistic regression is used to classify the feature vectors, mapping the feature vectors to the probability distribution of fault types to obtain fault type classification prediction results. These fault type classification prediction results are expressed as follows:

[0093] ;

[0094] In the formula, In the eigenvector Under these conditions, fault type For category indexing The probability, Represents category index The transpose of the weights, Represents category index The bias, Indicates the total number of category indexes. Indicates the first The transpose of the weights of each category index. Indicates the first Bias of each category index This represents an exponential function.

[0095] In this embodiment, a linear transformation is performed on the feature vector to obtain the predicted remaining life of the power equipment, wherein the predicted remaining life of the power equipment is expressed as:

[0096] ;

[0097] In the formula, RUL represents the predicted remaining life of the electrical equipment. Representing the regression coefficients, these are factors influencing the remaining lifespan of electrical equipment. This represents the intercept, which is the initial lifespan of the power equipment.

[0098] V. Update the decision-making module

[0099] The update decision module is used to establish a multi-objective function based on the dependencies, update costs, and update times of the updated power equipment sub-modules, obtain the optimization results of the multi-objective function using a multi-objective evolutionary algorithm, and output an update decision based on the optimization results of the multi-objective function. The update decision includes the update order priority of the power equipment sub-modules that need to be updated.

[0100] In this embodiment, a multi-objective function is established based on the dependencies of the updated power equipment sub-modules, the update cost, and the update time. This multi-objective function is expressed as:

[0101] ;

[0102] In the formula, Represents the combination of decision variables The minimum multi-objective function value, where the combination of decision variables is... For all power equipment sub-modules that need updating; Indicates the update cost function The weighting coefficients, Indicates the function to update time The weighting coefficients, Dependency strength function The weighting coefficients.

[0103] In this embodiment, the update cost function is... Represented as:

[0104] ;

[0105] In the formula, This indicates the total number of updated power equipment sub-modules. Indicates the first Decision variables of each power equipment submodule The corresponding update cost, Indicates the relationship with the first The power equipment submodules have dependencies on the first Decision variables for each power equipment submodule This indicates that the first one is updated simultaneously. The power equipment submodule and the first The power equipment submodules have dependencies on the first The update cost corresponding to each power equipment sub-module This represents the fixed cost, which is a decision variable independent of the power equipment submodule.

[0106] In this embodiment, the update time function is... Represented as:

[0107] ;

[0108] In the formula, Indicates the update of the first The maximum bottleneck time in the process of each power equipment sub-module. This indicates the total number of updated power equipment sub-modules. Indicates the first Decision variables of each power equipment submodule The corresponding update time, Indicates the first The power equipment submodule belongs to the first A set of power equipment submodules with dependencies. , This indicates the additional update time due to dependencies, meaning the first update must be performed first. Only the power equipment submodule can be updated. The update time corresponding to each power equipment sub-module Indicates the relationship with the first The power equipment submodules have dependencies on the first Decision variables for each power equipment sub-module.

[0109] In this embodiment, the dependency strength function is... Represented as:

[0110] ;

[0111] In the formula, Indicates the first The power equipment submodule and the first Physical connection dependencies of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Data exchange dependencies of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Functional dependencies of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Energy supply dependence of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Spatial location dependency of each power equipment submodule Intensity control factor.

[0112] In this embodiment, a multi-objective evolutionary algorithm is used to obtain the optimization result of the multi-objective function, such as... Figure 2 As shown, it includes:

[0113] (1) There are n power equipment sub-modules that need to be updated, of which the nth sub-module is the first one. The update cost for each power equipment submodule is: and update time is , No. Decision variables of each power equipment submodule Decision variables A value of 1 indicates an update; the decision variable... A value of 0 indicates no update;

[0114] Based on logical constraints and priority constraints, constraints are established according to the dependencies between power equipment sub-modules. Specifically, during the update of the... Before updating the power equipment submodule, the first one must be updated. Other submodules that a power equipment submodule depends on;

[0115] (2) The multi-objective function is used as the optimization objective;

[0116] (3) Randomly generate a set of decision variable combinations that satisfy dependency constraints as the initial population;

[0117] (4) Based on the multi-objective function, perform fitness evaluation on each set of decision variable combinations and calculate the multi-objective function value for each set of decision variable combinations;

[0118] (5) Take the combination of decision variables whose multi-objective function values ​​exceed the threshold as the parent generation, perform crossover and mutation operations to generate new offspring individuals, and remove offspring individuals that do not meet the constraints to obtain all combinations of decision variables;

[0119] (6) Repeat steps (4)-(5) for all combinations of decision variables until the predetermined number of iterations or convergence condition is reached, and obtain all combinations of decision variables after iteration;

[0120] (7) Select a set of Pareto optimal solutions from all combinations of decision variables after iteration as the optimization result of the multi-objective function.

[0121] VI. Real-time Update Module

[0122] The real-time update module is used to form a directed graph structure including source nodes and multiple layers of intermediate nodes based on dependencies and the power equipment sub-modules that need to be updated. According to the update order priority, the power equipment sub-modules that need to be updated are received as receiving nodes in sequence to receive update requests. If the receiving node is not the source node, the update request is forwarded to the preceding node of the receiving node until the update request is forwarded to the source node to respond to the update request. Among them, all nodes that receive the update request respond to the update request. If the receiving node is the source node, the receiving node responds to the update request.

[0123] like Figure 3As shown, this embodiment constructs a directed graph structure based on the dependencies between power equipment sub-modules, including a source node A, multiple layers of intermediate nodes B, C, D, E, F, G, H, I, J, K, and a sink node L. Each node in the directed graph represents a power equipment sub-module, and arrows indicate the dependencies between modules. The source node is usually the core module and starting module of the entire power equipment, while other power equipment sub-modules are arranged as intermediate nodes or receiving nodes according to their dependencies.

[0124] In this embodiment, the power equipment sub-modules that need to be updated are selected as receiving nodes in sequence according to the update order priority, and update requests are generated accordingly.

[0125] If the receiving node is not the source node, the update request is forwarded to the receiving node's forward node, i.e., the power equipment submodule that depends on the receiving node. This process is repeated layer by layer until the update request is forwarded to the source node. Figure 3 In this process, intermediate node H is the highest priority power equipment submodule that needs to be updated. At this time, intermediate node H acts as the receiving node, not the source node. Therefore, the update request is forwarded to the preceding nodes of intermediate node H, namely intermediate nodes B and C. Intermediate nodes B and C then forward the update request to the preceding node, namely the source node A. During the forwarding process, each node that receives the update request will respond to the request. Among them, the source node A, intermediate nodes B, intermediate nodes C, and receiving node H will all respond to the request, indicating that the power equipment submodule represented by the node is ready to be updated or has completed the preparatory work before the update.

[0126] If the receiving node is the source node A, the source node A will directly respond to the update request and start the update process, while other nodes will not perform updates.

[0127] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0128] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A modular real-time update system for power equipment, characterized in that, include: The modular equipment module is used to divide power equipment into sensor sub-modules, control sub-modules, and execution sub-modules to obtain power equipment sub-modules. The data acquisition module is used to acquire updated data from the power equipment sub-module, including current, voltage, temperature, humidity, vibration, fault records, and maintenance data. The data preprocessing module is used to preprocess the updated data to obtain preprocessed updated data; The predictive analysis module is used to extract features from the pre-trained digital twin model module to obtain feature vectors, classify the feature vectors using logistic regression, map the feature vectors to the probability distribution of fault types, and obtain fault type classification prediction results; perform linear transformation on the feature vectors to obtain the remaining life prediction value of the power equipment; and obtain the prediction result of the digital twin model module using a weighted average method based on the fault type classification prediction result and the remaining life prediction value of the power equipment, wherein the prediction result of the digital twin model module includes the power equipment sub-modules that need to be updated. The update decision module is used to establish a multi-objective function based on the dependencies, update costs, and update times of the updated power equipment sub-modules, obtain the optimization results of the multi-objective function using a multi-objective evolutionary algorithm, and output an update decision based on the optimization results of the multi-objective function. The update decision includes the update order priority of the power equipment sub-modules that need to be updated. The real-time update module is used to form a directed graph structure including source nodes and multiple layers of intermediate nodes based on dependencies and the power equipment sub-modules that need to be updated. According to the update order priority, it sequentially receives update requests from the power equipment sub-modules that need to be updated as receiving nodes. If the receiving node is not the source node, the update request is forwarded to the node preceding the receiving node, and so on, until the update request is forwarded layer by layer to the source node to respond to the update request. All nodes receiving the update request respond to the update request. If the receiving node is the source node, the receiving node responds to the update request. A multi-objective function is established based on the dependencies, update costs, and update times of the updated power equipment sub-modules. This multi-objective function is expressed as: ； In the formula, Represents the combination of decision variables The minimum multi-objective function value, where the combination of decision variables is... For all power equipment sub-modules that need updating; Indicates the update cost function The weighting coefficients, Indicates the function to update time The weighting coefficients, Dependency strength function Weighting coefficients; The update cost function Represented as: ； In the formula, This indicates the total number of updated power equipment sub-modules. Indicates the first Decision variables of each power equipment submodule The corresponding update cost, Indicates the relationship with the first The power equipment submodule has dependencies on the first Decision variables for each power equipment submodule This indicates that the first one is updated simultaneously. The power equipment submodule and the first The power equipment submodule has dependencies on the first The update cost corresponding to each power equipment sub-module The fixed cost represents the decision variable independent of the power equipment submodule; The update time function Represented as: ； In the formula, Indicates the update of the first The maximum bottleneck time in the process of each power equipment sub-module. This indicates the total number of updated power equipment sub-modules. Indicates the first Decision variables of each power equipment submodule The corresponding update time, Indicates the first The power equipment submodule belongs to the first A set of power equipment submodules with dependencies. , This indicates the additional update time due to dependencies, meaning the first update must be performed first. Only the power equipment submodule can be updated. The update time corresponding to each power equipment sub-module Indicates the relationship with the first The power equipment submodule has dependencies on the first Decision variables for each power equipment submodule; The dependency strength function Represented as: ； In the formula, Indicates the first The power equipment submodule and the first Physical connection dependencies of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Data exchange dependencies of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Functional dependencies of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Energy supply dependence of each power equipment submodule Intensity control factor, Indicates the first The power equipment submodule and the first Spatial location dependency of each power equipment submodule Intensity control factor.

2. The modular real-time update system for power equipment as described in claim 1, characterized in that, The training method for the digital twin model module includes: The online learning module receives preprocessed updated data in real time and updates the parameters of the digital twin model module. An adaptive learning rate adjustment strategy is adopted to dynamically adjust the learning rate according to the gradient of the loss function, so as to obtain the trained digital twin model module.

3. The modular real-time update system for power equipment as described in claim 2, characterized in that, The updated data is preprocessed to obtain preprocessed updated data, including: The missing values ​​in the updated data are filled with the mode to obtain the updated data after filling the missing values. Use 3 The principle is to identify and delete outliers in the updated data after filling in missing values, thus obtaining preprocessed updated data.

4. The modular real-time update system for power equipment as described in claim 3, characterized in that, Using logistic regression to classify the feature vector maps the feature vector to the probability distribution of fault types, obtaining a fault type classification prediction result, wherein the fault type classification prediction result is expressed as: ; In the formula, In the eigenvector Under these conditions, fault type For category indexing The probability, Represents category index The transpose of the weights, Represents category index The bias, Indicates the total number of category indexes. Indicates the first The transpose of the weights of each category index. Indicates the first Bias of each category index This represents an exponential function.

5. The modular real-time update system for power equipment as described in claim 4, characterized in that, A linear transformation is performed on the feature vector to obtain the predicted remaining life of the power equipment, wherein the predicted remaining life of the power equipment is expressed as: ； In the formula, RUL represents the predicted remaining life of the electrical equipment. Representing the regression coefficients, these are factors influencing the remaining lifespan of electrical equipment. This represents the intercept, which is the initial lifespan of the power equipment.

6. The modular real-time update system for power equipment as described in claim 1, characterized in that, The optimization results of the multi-objective function are obtained using a multi-objective evolutionary algorithm, including: (1) There are n power equipment sub-modules that need to be updated, of which the nth sub-module is the first one. The update cost for each power equipment submodule is: and update time is , No. Decision variables of each power equipment submodule Decision variables A value of 1 indicates an update; the decision variable... A value of 0 indicates no update; Based on logical constraints and priority constraints, constraints are established according to the dependencies between power equipment sub-modules. Specifically, during the update of the... Before updating the power equipment submodule, the first one must be updated. Other submodules that a power equipment submodule depends on; (2) The multi-objective function is used as the optimization objective; (3) Randomly generate a set of decision variable combinations that satisfy dependency constraints as the initial population; (4) Based on the multi-objective function, perform fitness evaluation on each set of decision variable combinations and calculate the multi-objective function value for each set of decision variable combinations; (5) Take the combination of decision variables whose multi-objective function values ​​exceed the threshold as the parent generation, perform crossover and mutation operations to generate new offspring individuals, and remove offspring individuals that do not meet the constraints to obtain all combinations of decision variables; (6) Repeat steps (4)-(5) for all combinations of decision variables until the predetermined number of iterations or convergence condition is reached, and obtain all combinations of decision variables after iteration; (7) Select a set of Pareto optimal solutions from all combinations of decision variables after iteration as the optimization result of the multi-objective function.

Images