Digital twin increment transmission method and system for power transmission and transformation equipment

Abstract

The invention provides a digital twin increment transmission method and system for power transmission and transformation equipment, and the method comprises the steps: collecting equipment state data, screening the equipment state data with the variable quantity exceeding a preset threshold value, and compressing the equipment state data; constructing a reinforcement learning model, outputting an uploading strategy, evaluating the effect of each uploading decision through a reward function, and obtaining an optimal strategy; determining an uploading sequence according to the uploading priority, and judging whether an uploading condition is met or not according to the real-time state of the bandwidth and the priority of the data packet; only the data which is obviously changed compared with the historically uploaded data is transmitted; adjusting whether each data packet needs to be uploaded according to a plan or not or not according to the bandwidth feedback of real-time transmission and the health state of the equipment; an optimized uploading strategy is obtained; an upload execution effect evaluation index is calculated by executing feedback data, and an upload strategy is optimized for a long time based on historical upload data, a bandwidth usage mode and an equipment failure frequency.

Description

Technical Field

[0001] This invention belongs to the field of data transmission, and particularly relates to a digital twin incremental transmission method and system for power transmission and transformation equipment. Background Technology

[0002] With the rapid development of smart grids, the Internet of Things (IoT), and digital technologies, the application of digital twin systems for power transmission and transformation equipment in the power industry is gradually becoming an important trend. Digital twin technology creates virtual models of physical equipment, reflecting its operating status, health condition, and working environment in real time, enabling remote monitoring, early warning, and optimized scheduling. This technology is of great significance in improving the intelligence level of power systems, optimizing energy management, and enhancing power security. However, in existing digital twin systems for power transmission and transformation equipment, data transmission, processing, and updating still face many challenges.

[0003] Currently, digital twin systems for power transmission and transformation equipment rely on traditional data acquisition and transmission methods, typically uploading all equipment operating data to a central system for processing and analysis. While this approach was feasible in the early stages, with the expansion of power grids and the increase in the number of devices, traditional full-data transmission faces significant bottlenecks. First, the sheer volume of data becomes a critical issue. A large amount of sensor data (such as voltage, current, and temperature) needs to be uploaded in real time, leading to excessive bandwidth burden and impacting system response efficiency. Furthermore, existing transmission models neglect the timeliness and redundancy of equipment status changes, resulting in a large amount of irrelevant data being uploaded even when equipment status is relatively stable, increasing bandwidth pressure and reducing transmission efficiency. Second, existing methods have deficiencies in real-time data transmission and priority management, failing to effectively distinguish which data should be transmitted first. This can lead to delays or loss of critical data when network bandwidth is limited, affecting decision-making and response speed. Therefore, existing technologies suffer from problems such as bandwidth overload, data redundancy, and poor real-time performance. Summary of the Invention

[0004] The purpose of this invention is to design a digital twin incremental transmission method and system for power transmission and transformation equipment, which can reduce bandwidth consumption and data redundancy, and improve data transmission efficiency, especially in situations where bandwidth is limited, thereby maximizing bandwidth utilization. Secondly, this invention introduces an intelligent compression and transmission scheduling mechanism, dynamically adjusting the upload timing and priority based on changes in equipment status and network conditions, ensuring that important data is uploaded first when bandwidth is limited.

[0005] To achieve the above objectives, a method for incremental digital twin transmission of power transmission and transformation equipment is provided in a first aspect of the present invention, the method comprising: Collect device status data, filter device status data whose changes exceed a preset threshold, and compress them. The compression ratio is dynamically adjusted according to the change range of the device status data. A reinforcement learning model is constructed. The state space of the reinforcement learning model includes filtered change data, device health status, and network bandwidth usage. The action space of the reinforcement learning model is the upload strategy to be made in the current state, including the upload priority, upload timing, and upload order of data packets. The effect of each upload strategy is evaluated through a reward function to obtain the optimal strategy. The upload order is determined based on the upload priority. The upload conditions are determined by the real-time bandwidth status and the priority of the data packets. Only data that has changed significantly compared to historical upload data is transmitted. A bandwidth regularization term is added to the upload strategy of each data packet. The bandwidth regularization term represents the impact of the upload operation on the current network bandwidth. Based on real-time bandwidth feedback and device health status, adjust whether each data packet needs to be uploaded as scheduled or postponed; obtain the optimized upload strategy; output the optimized upload strategy and obtain execution feedback data; The upload performance evaluation index is calculated by executing feedback data, which represents the upload success rate, bandwidth usage, and the impact of device status. The upload strategy is optimized over the long term based on historical upload data, bandwidth usage patterns, and device failure frequency.

[0006] Furthermore, the equipment status data, including voltage, current, and temperature, is collected through sensors.

[0007] Furthermore, the compression ratio is adjusted based on the magnitude of data changes. When the device status change exceeds the compression threshold, a low compression ratio is selected; when the device status change is below the compression threshold, a high compression ratio is selected.

[0008] Furthermore, the reward function is obtained by weighting the changes in device state with data upload latency and network bandwidth utilization.

[0009] Furthermore, the network architecture of the reinforcement learning model consists of an input layer, a hidden layer, and an output layer. The input layer receives state data from the device, and after calculation by the hidden layer, the output layer generates an upload strategy.

[0010] Furthermore, the bandwidth regularization term is the ratio of the current bandwidth usage to the maximum system bandwidth. By guiding the upload strategy, upload operations are reduced when bandwidth is limited, thus avoiding network congestion during the upload process.

[0011] Furthermore, in the process of obtaining the optimized upload strategy, a bandwidth occupancy control function is introduced to dynamically adjust the upload operation. When the bandwidth usage is close to the total bandwidth capacity, the bandwidth occupancy control function will increase, thereby suppressing the upload operation and prioritizing the transmission of more important data.

[0012] Furthermore, in the process of obtaining the optimized upload strategy, a health status response mechanism is implemented. When a device malfunctions or its status changes significantly, the upload priority is adjusted according to the urgency of the device's data upload. When a device malfunctions, the upload priority of the malfunctioning device is increased to ensure that its status data is uploaded first.

[0013] Furthermore, the execution feedback data includes the actual uploaded execution results, device health status, and bandwidth utilization.

[0014] A second aspect of the present invention provides a digital twin incremental transmission system for power transmission and transformation equipment, the system comprising: The status change detection unit is used to collect equipment status data, filter equipment status data whose change exceeds a preset threshold, and compress the data. The compression ratio is dynamically adjusted according to the change range of the equipment status data. A policy generation unit is used to construct a reinforcement learning model. The state space of the reinforcement learning model includes filtered change data, device health status, and network bandwidth usage. The action space of the reinforcement learning model is the upload policy made in the current state, including the upload priority, upload timing, and upload order of data packets. The effect of each upload policy is evaluated through a reward function to obtain the optimal policy. The incremental transmission unit is used to determine the upload order based on the upload priority, and to determine whether the upload conditions are met by using the real-time bandwidth status and the priority of the data packets; and only transmits data that has changed significantly compared to historical uploaded data, and adds a bandwidth regularization term to the upload strategy of each data packet, the bandwidth regularization term representing the impact of the upload operation on the current network bandwidth; The optimization unit is used to adjust whether each data packet needs to be uploaded as scheduled or postponed based on real-time transmission bandwidth feedback and device health status; obtain the optimized upload strategy; output the optimized upload strategy and obtain execution feedback data; The long-term optimization unit is used to calculate the upload performance evaluation index by executing feedback data. The upload performance evaluation index represents the upload success rate, bandwidth usage, and the impact of device status. The unit performs long-term optimization of the upload strategy based on historical upload data, bandwidth usage patterns, and device failure frequency.

[0015] The beneficial technical effects of the present invention are at least as follows: To address the aforementioned issues, this invention provides a digital twin incremental transmission method and system for power transmission and transformation equipment. One of the core innovations of this invention is incremental data transmission, which avoids the problem of uploading all data in traditional methods, transmitting only incremental data when the equipment status has significantly changed. This approach significantly reduces bandwidth consumption and data redundancy, improving data transmission efficiency, especially maximizing bandwidth utilization when bandwidth is limited. Secondly, this invention introduces intelligent compression and transmission scheduling mechanisms, dynamically adjusting upload timing and priority based on changes in equipment status and network conditions. This ensures that important data is prioritized for upload when bandwidth is limited, while non-urgent data is postponed based on real-time network load. This mechanism enhances the system's real-time performance and responsiveness, preventing delays in important data transmission during periods of high network load.

[0016] By combining incremental transmission, intelligent scheduling and other technologies, the problems of bandwidth burden, data redundancy and real-time performance of existing technologies have been solved, and the transmission efficiency and intelligence level of the digital twin system for power transmission and transformation equipment have been improved. Attached Figure Description

[0017] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0018] Figure 1 This is a flowchart of a digital twin incremental transmission method for power transmission and transformation equipment according to the present invention.

[0019] Figure 2 This is a framework diagram of a digital twin incremental transmission system for power transmission and transformation equipment according to the present invention. Detailed Implementation

[0020] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0021] In one or more embodiments, such as Figure 1 As shown, a digital twin incremental transmission method for power transmission and transformation equipment is disclosed, the method comprising the following: S1: Collect device status data, filter device status data whose changes exceed a preset threshold, and compress them. The compression ratio is dynamically adjusted according to the magnitude of the changes in the device status data. Specifically, the main objective in this step is to optimize bandwidth usage through device state change detection and adaptive compression, effectively reducing the transmission of redundant data while ensuring the accuracy and real-time performance of critical data. To reduce the amount of data transmitted while maintaining real-time performance and data accuracy, this solution employs a device state change detection method and adaptive compression technology. Data such as device voltage, current, and temperature are collected in real time by sensors and uploaded to the control system. By introducing a device state change detection model, this solution can intelligently filter out significantly changing device state data and process only these changed data. Simultaneously, the adaptive compression algorithm ensures that high compression is only performed when necessary, thereby minimizing bandwidth consumption.

[0022] In the equipment status change detection section, the difference between each collected data point and the data from the previous moment is first calculated using the equipment status data. If the change in equipment status Exceeding the preset threshold If this is detected, then the device is considered to have undergone a significant change. Specifically, the current sensor will transmit current data. Data is collected and transmitted to the control system in real time, read once per second. To determine whether data changes are significant, calculations are performed. ,if If a significant change is detected, the system considers that the device has changed and decides to upload that data. In this way, the system can ignore minor, insignificant data fluctuations, thus avoiding the uploading of meaningless data and significantly reducing bandwidth load.

[0023] Next, the selected changed data is adaptively compressed. The compression strategy dynamically adjusts the compression ratio based on the magnitude of the data change. If the device status changes significantly (i.e., For larger data types, a lower compression ratio is used to ensure data accuracy; for smaller data types, a higher compression ratio is used to save bandwidth. The compression function is defined here. Its function is to determine the compression ratio based on changes in equipment condition:

[0024] in, It is a factor that adjusts the compression ratio; it will be based on... The compression ratio is adjusted based on the magnitude of the current change. If the current variation is large, the system uses a lower compression ratio to ensure data accuracy; if the current variation is small, a higher compression ratio is used to reduce bandwidth consumption. Through this adaptive compression mechanism, the system can adjust the efficiency of data transmission according to the actual data requirements, maximizing bandwidth utilization while ensuring the accuracy of critical data. Compression function The implementation is based on a threshold setting for the magnitude of changes in device status, and it calculates the amount of change in device status over a continuous time period. This determines the compression strategy. Specifically, a large change in device status is defined as a change in device status data (such as current, voltage, temperature, etc.) exceeding a preset threshold. For example, when the change in current... When the change exceeds 0.5A, the system considers the change "relatively large" and therefore selects a lower compression ratio to ensure data accuracy. If the change is less than 0.5A, it is considered a "relatively small" change, and the system uses a higher compression ratio to reduce bandwidth usage. This strategy assesses the amount of change based on the difference between two consecutive device status data sets, thereby dynamically adjusting the uploaded data compression ratio. In this way, the system can minimize the amount of data when network bandwidth is limited, while ensuring high-fidelity uploading of critical device data even when significant changes occur.

[0025] After filtering and compression, the output data consists of two parts: the first part is the filtered device status change data, containing only those data that have changed significantly; the second part is the adaptively compressed data, further reducing the data volume and ensuring maximum transmission efficiency. This processed data will be used as input in subsequent steps for generating upload scheduling strategies and executing data transmission. Through this method, the system not only reduces redundant data while maintaining data accuracy but also further optimizes bandwidth usage, making the entire data transmission process more efficient and real-time.

[0026] S2: Construct a reinforcement learning model. The state space of the reinforcement learning model includes filtered change data, device health status, and network bandwidth usage. The action space of the reinforcement learning model is the upload strategy made in the current state, including the upload priority, upload timing, and upload order of data packets. The effect of each upload strategy is evaluated through a reward function to obtain the optimal strategy. Specifically, in this step, we utilize reinforcement learning algorithms to generate a data upload scheduling strategy. The aim is to dynamically optimize the timing and priority of data uploads based on device status changes and bandwidth availability, given limited network bandwidth. The input data for this step primarily comes from the output of step 1, including filtered device status change data and compressed data. Based on this data, combined with device health status and network bandwidth information, the reinforcement learning model dynamically adjusts the upload priority and timing of each data packet, ensuring that important data is uploaded first when network bandwidth is limited, thereby improving the real-time performance and bandwidth utilization of the entire system. In the design of the reinforcement learning model, we first define the system's state space. The state space includes the filtered change data from step 1 (such as changes in voltage, current, and temperature), the health status of the devices (e.g., whether the devices are malfunctioning or require urgent data upload), and the current network bandwidth usage. The device state change data is the incremental data provided in step 1, while the health status and bandwidth information are real-time system feedback. This information collectively constitutes the system state, enabling the reinforcement learning model to make optimization decisions based on the real-time status of the devices.

[0027] Action space of reinforcement learning models This refers to the upload strategy adopted by the model in the current state, specifically including the upload priority, timing, and order of data packets. The model's goal is to select an optimal action that ensures the priority transmission of critical data under different network loads and device states, while conserving bandwidth as much as possible. To achieve this goal, we designed a reward function. It is used to evaluate the effectiveness of each upload strategy:

[0028] in: These are changes in equipment status, reflecting the importance of the data; This refers to the data upload latency; the longer the latency, the lower the reward value. This indicates the current network bandwidth utilization. When bandwidth is limited, the system will avoid uploading low-priority data. These are weighting coefficients that control the impact of device changes, upload latency, and bandwidth utilization on the reward function.

[0029] The goal of reinforcement learning is to learn an optimal policy by maximizing a reward function, prioritizing the upload of critical data when bandwidth is limited, thus avoiding bandwidth waste and latency. During training, the model adjusts its upload strategy based on environmental feedback, ensuring optimal upload order and timing for each data packet. Through the reward function, the system balances bandwidth usage and device status responsiveness, avoiding the limitations of traditional methods that rely solely on static thresholds. The network architecture of the reinforcement learning model consists of multiple neural network layers, including an input layer, hidden layers, and an output layer. The input layer receives state data from the device (such as device changes, device health status, and bandwidth conditions). After computation by multiple hidden layers, the output layer generates the final upload strategy. Each layer of the neural network is trained using a backpropagation algorithm, continuously adjusting network weights to enable the model to achieve optimal upload scheduling in dynamic environments. In this way, the reinforcement learning model can dynamically adjust upload priorities based on changes in device status and network load. When network bandwidth is sufficient, the system prioritizes uploading data with significant device status changes; while under high network load, the system delays uploading low-priority data, ensuring timely transmission of important data. The innovation of reinforcement learning models lies in their ability to make adaptive decisions based on real-time network conditions and device status, rather than relying solely on preset upload order or priority.

[0030] Finally, this step outputs an upload scheduling strategy, providing clear guidance for subsequent upload execution. This strategy includes the upload priority, timing, and order of each data packet, ensuring real-time data transmission and optimizing bandwidth usage. Through reinforcement learning algorithms, we implemented an adaptive and flexible upload scheduling strategy that can optimize data transmission under different network environments and device conditions, improving the efficiency and stability of the entire system. This solution not only avoids the upload of redundant data but also ensures that critical data is transmitted first when bandwidth is limited, thus solving the bandwidth burden and real-time performance issues in existing technologies.

[0031] S3: Determine the upload order based on the upload priority, and determine whether the upload conditions are met by using the real-time bandwidth status and the priority of the data packets; and only transmit data that has changed significantly compared to historical upload data, and add a bandwidth regularization term to the upload strategy of each data packet, the bandwidth regularization term representing the impact of the upload operation on the current network bandwidth; Specifically, in this step, our goal is to execute data uploads and incremental transmissions based on the upload scheduling strategy generated in Step 2. The upload scheduling strategy, based on a reinforcement learning model, allows the system to dynamically determine the upload timing and priority of each data packet according to device status changes and network bandwidth conditions. Through this innovative upload mechanism, the system ensures that critical data is transmitted first when bandwidth is limited, avoiding data loss or delayed responses due to bandwidth overload or unnecessary latency. The upload priority is output in Step 2. and upload timing This information will serve as the core input for this step, and the system will use it to determine how to perform incremental data uploads.

[0032] The input to this step includes the upload scheduling policy output from step 2, i.e., the upload priority of each data packet. and upload timing In addition, the system needs to further optimize the upload strategy based on information such as the current health status of the devices (e.g., whether a malfunction has occurred) and network bandwidth usage. This data comes from real-time system feedback and device status change data provided in step 1. By combining this information, the system can determine which data packets should be uploaded first and which can be delayed. When performing data uploads, the upload order and timing of each data packet are first determined based on the upload priority generated in step 2. Specifically, the higher the upload priority of a data packet, the more likely it is to be uploaded when network bandwidth is sufficient; if the data packet has a lower priority, it will only be uploaded if network bandwidth allows, and other low-priority data packets will be delayed. The system dynamically evaluates the priority of each data packet and the current bandwidth availability to decide whether to execute the upload task. When network bandwidth is sufficient and the data packet has a high priority, the system will immediately execute the upload task for that data packet. The system determines whether the upload conditions are met based on the real-time bandwidth status and the data packet priority. If bandwidth is insufficient or the data packet has a low priority, the system will delay the upload of that data packet until bandwidth becomes available or the priority is increased. Through this mechanism, the system can ensure that important data is uploaded first when network load is high, guaranteeing the timely transmission of critical data and avoiding unnecessary bandwidth consumption. This priority and bandwidth adaptive upload strategy effectively improves the real-time performance and bandwidth utilization of data uploads.

[0033] To further optimize upload efficiency, especially under bandwidth constraints, this step introduces an incremental transmission mechanism. In traditional full data transmission, all data is transmitted, while this patented solution uses incremental transmission to ensure that the system only uploads device data that has undergone significant changes. Step 1 has already filtered out and compressed the data showing status changes; this incremental data is then transmitted after being prioritized using an upload scheduling strategy. We introduced an incremental data upload mechanism, transmitting only data that has changed significantly since the last upload, thereby saving bandwidth and improving transmission efficiency. Specifically, we designed a bandwidth control regular expression. The purpose of this regular expression is to introduce bandwidth usage constraints into the upload strategy, avoiding excessive bandwidth consumption during the upload process, which could lead to network congestion or delays in important data. We added a bandwidth regular expression to the upload strategy for each data packet. This item indicates the impact of the upload operation on the current network bandwidth:

[0034] in: This represents the bandwidth usage at the current moment. This is the system's maximum bandwidth capacity; It is an adjustment factor used to control the degree of influence of the bandwidth regularization term.

[0035] This regular expression guides upload strategies to reduce upload operations when bandwidth is limited, thus avoiding network congestion during the upload process. For example, when the system's bandwidth is already close to its maximum capacity, the regular expression... This mechanism increases and suppresses upload operations, prioritizing the transmission of important data. It ensures system stability under bandwidth constraints, taking into account current bandwidth usage limitations in each upload strategy, thus avoiding excessive or unreasonable bandwidth consumption.

[0036] The output of this step is the actual transmitted data packets, including sorted incremental data and a priority order adjusted according to the upload policy. Based on the output of the upload scheduling policy, the system performs incremental uploads of data packets and optimizes bandwidth usage using bandwidth regularization. Ultimately, the system maximizes bandwidth utilization while ensuring the timely upload of critical data.

[0037] S4: Based on real-time bandwidth feedback and device health status, adjust whether each data packet needs to be uploaded as scheduled or postponed; obtain the optimized upload strategy; output the optimized upload strategy and obtain execution feedback data; Specifically, in this step, our goal is to further optimize the system execution process based on the upload scheduling strategy implemented in step 3, and to respond to real-time feedback, thereby ensuring the efficient completion of data upload tasks. In this process, we don't just rely on a pre-generated upload scheduling strategy, but also adjust the upload operation based on real-time feedback information from the system. Especially in situations with limited bandwidth, sudden changes in device status, or large fluctuations in network load, we can flexibly adjust the upload timing, order, and priority to ensure that critical data is transmitted first, avoiding data delays or loss due to insufficient bandwidth or system changes.

[0038] The input for this step comes from the upload scheduling strategy in step 3, which provides the upload priority for each data packet. Upload timing And upload strategy functions under bandwidth-limited conditions. Based on these inputs, the system determines whether each data packet needs to be uploaded as scheduled or postponed, according to real-time bandwidth feedback and device health status. Real-time feedback information includes current network bandwidth usage, device health status (e.g., whether a device has malfunctioned or requires urgent data upload), and upload latency. This information is used in conjunction with upload scheduling strategies to ensure that the upload of each data packet can be adjusted according to real-time conditions.

[0039] During data uploads, we introduced a dynamic bandwidth adjustment mechanism. When the system detects network bandwidth constraints, we adjust the upload strategy, fine-tuning the upload priority to ensure that the most important data packets are transmitted first. In cases of bandwidth overload, low-priority data packets will be delayed in upload, and the system will adjust the upload strategy to ensure that network resources are prioritized for critical data. Specifically, we use a bandwidth occupancy control function. To dynamically adjust the upload operation, the formula is as follows:

[0040] in, This is the bandwidth usage at the current moment. It is the total bandwidth capacity of the network. This represents the allowed bandwidth usage tolerance. Through this control function, the system can dynamically adjust its upload strategy based on bandwidth usage. Specifically, when bandwidth usage approaches the total bandwidth capacity, This will increase the size, thereby suppressing upload operations and prioritizing the transmission of more important data.

[0041] In this step, we designed a health status response mechanism based on real-time changes in device health status. When a device malfunctions or its status changes significantly, the system automatically adjusts the upload priority based on the urgency of the data upload. For example, if a device malfunctions, the system immediately increases its data upload priority to ensure that fault information is transmitted in the shortest possible time. The system monitors the device's health status in real time and dynamically adjusts the upload priority based on changes in device status. Specifically, when a device malfunctions, the system sets the upload priority of the malfunctioning device to a higher level, ensuring that its status data is uploaded first and avoiding upload delays caused by bandwidth limitations or high network load.

[0042] If the device is operating normally, the system will maintain its data upload priority at a normal level, waiting for suitable bandwidth and network conditions before uploading. When the system detects a change in device status, it will respond immediately, adjusting the device's upload priority to ensure that critical fault data is processed promptly. This mechanism enhances the system's emergency response capability, ensuring that critical device information can be quickly uploaded when problems occur, guaranteeing timely handling and response to device malfunctions. Finally, this step outputs an optimized upload scheduling execution strategy, which combines dynamic network bandwidth adjustment, changes in device health status, and adjustments to upload timing. This strategy ensures that the system can flexibly adjust its upload strategy when bandwidth is limited, device status changes, or network conditions fluctuate, prioritizing the transmission of critical data. Through real-time response and optimization, this step further improves the real-time performance of data uploads and system stability, resolving upload latency issues caused by excessive network load, bandwidth waste, and device failure.

[0043] S5: Calculate the upload performance evaluation index by executing feedback data. The upload performance evaluation index represents the upload success rate, bandwidth usage, and the impact of device status. Optimize the upload strategy over the long term based on historical upload data, bandwidth usage patterns, and device failure frequency.

[0044] Specifically, in this step, our goal is to perform periodic feedback and long-term system optimization based on the upload execution feedback from step 4. The core task of this step is to dynamically adjust the upload scheduling strategy based on feedback information such as actual upload execution results, device health status, and bandwidth utilization. This periodic feedback mechanism can adaptively optimize the upload strategy based on the data upload performance over long-term system operation, thereby improving the system's stability and efficiency in complex environments. Through this optimization process, the system can continuously improve its upload scheduling strategy based on real-time feedback information, enabling the system to adapt to factors such as bandwidth changes, device failures, and network fluctuations during long-term operation, ensuring the efficient execution of data upload tasks.

[0045] The input for this step comes from the output of step 4, namely the upload scheduling strategy and execution feedback data, including the upload priority of each data packet. Upload timing and upload execution feedback data This feedback data provides the basis for the actual execution results of this step, helping the system to evaluate key information such as the success rate of upload tasks, bandwidth utilization efficiency, and device health status. Based on these inputs, the system will periodically evaluate the upload execution and optimize and adjust the upload scheduling strategy according to the analysis results, ensuring that the system can adapt to changing bandwidth, device status, and network conditions.

[0046] During the optimization process, the system first uses execution feedback data to calculate an evaluation metric for upload performance. This metric integrates the impact of upload success rate, bandwidth usage, and device status to quantify the actual effectiveness of the upload strategy. Specifically, the system evaluates the success rate of upload tasks to measure whether data uploads are completed smoothly. Simultaneously, the system considers bandwidth usage, especially how to allocate bandwidth resources rationally when bandwidth is overloaded. To evaluate the effectiveness of the upload strategy, the system also considers the impact of device status, ensuring that critical data is prioritized for upload when device status changes. Through a comprehensive evaluation of these factors, the system can quantify the effectiveness of the upload strategy. Based on the evaluation results, the system will adjust according to the priority of upload tasks. If the system finds a low upload success rate or excessive bandwidth usage, it will appropriately adjust the upload strategy, prioritizing the transmission of high-priority data packets and delaying the upload of low-priority data packets to ensure reasonable allocation of bandwidth resources and maintain the efficiency of upload tasks. This feedback mechanism enables the system to dynamically optimize the upload strategy based on real-time data, ensuring the timely transmission of important data even when network and device loads change.

[0047] Meanwhile, we introduced a long-term optimization mechanism. By analyzing long-term operational data, the system can perform long-term adjustments and optimizations of its strategies based on historical upload data, bandwidth usage patterns, and device failure frequency. This long-term optimization mechanism predicts potential bandwidth pressure and device health issues by analyzing historical data, thereby adjusting upload strategies to avoid bandwidth waste and device status loss. We utilize a long-term optimization function... To optimize the upload scheduling strategy:

[0048] in, and These represent the average historical upload success rate and the long-term trend of bandwidth usage, respectively. and These are weighting coefficients optimized over the long term. Through the long-term optimization function, the system can finely adjust the upload strategy over the long term, ensuring that the system can maintain excellent performance under constantly changing environmental conditions.

[0049] Finally, the output of this step is the optimized upload scheduling strategy. This strategy has been adjusted based on periodic feedback and long-term optimization mechanisms to ensure that the system can adaptively adjust upload timing and priority. The optimized strategy can adapt to different situations such as changes in device status and fluctuations in network load, ensuring real-time data uploads and efficient system operation. Through this innovative solution, the system can maintain long-term stability in a large-scale device environment, avoiding upload delays or data loss caused by excessive bandwidth usage or device failures, and ensuring the efficient execution of upload tasks.

[0050] In one or more embodiments, such as Figure 2 As shown, a digital twin incremental transmission system for power transmission and transformation equipment is disclosed, the system comprising: The status change detection unit is used to collect equipment status data, filter equipment status data whose change exceeds a preset threshold, and compress the data. The compression ratio is dynamically adjusted according to the change range of the equipment status data. A policy generation unit is used to construct a reinforcement learning model. The state space of the reinforcement learning model includes filtered change data, device health status, and network bandwidth usage. The action space of the reinforcement learning model is the upload policy made in the current state, including the upload priority, upload timing, and upload order of data packets. The effect of each upload policy is evaluated through a reward function to obtain the optimal policy. The incremental transmission unit is used to determine the upload order based on the upload priority, and to determine whether the upload conditions are met by using the real-time bandwidth status and the priority of the data packets; and only transmits data that has changed significantly compared to historical uploaded data, and adds a bandwidth regularization term to the upload strategy of each data packet, the bandwidth regularization term representing the impact of the upload operation on the current network bandwidth; The optimization unit is used to adjust whether each data packet needs to be uploaded as scheduled or postponed based on real-time transmission bandwidth feedback and device health status; obtain the optimized upload strategy; output the optimized upload strategy and obtain execution feedback data; The long-term optimization unit is used to calculate the upload performance evaluation index by executing feedback data. The upload performance evaluation index represents the upload success rate, bandwidth usage, and the impact of device status. The unit performs long-term optimization of the upload strategy based on historical upload data, bandwidth usage patterns, and device failure frequency.

[0051] It is worth noting that the specific workflow of the incremental transmission system for digital twins of power transmission and transformation equipment provided in this embodiment of the invention is the same as that of the incremental transmission method for digital twins of power transmission and transformation equipment described in the above embodiment, and will not be repeated here.

[0052] This invention also provides a digital twin incremental transmission device for power transmission and transformation equipment, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps described in the above embodiment of the digital twin incremental transmission method for power transmission and transformation equipment, for example... Figure 1 The steps S1 to S6 described above; or, when the processor executes the computer program, it implements the functions of each module in the above system embodiments.

[0053] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the digital twin incremental transmission device of the power transmission and transformation equipment.

[0054] The aforementioned digital twin incremental transmission device for power transmission and transformation equipment can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. This device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the device may also include input / output devices, network access devices, buses, etc.

[0055] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. This processor is the control center of the digital twin incremental transmission equipment for power transmission and transformation, connecting all parts of the equipment via various interfaces and lines.

[0056] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the digital twin incremental transmission device for power transmission and transformation equipment by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc.; the data storage area may store data created based on the operation of the air conditioning controller, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart memory card (SMC), secure digital card (SD), flash memory card, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0057] If the integrated module of the digital twin incremental transmission device for power transmission and transformation equipment is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0058] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0059] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A power transmission and transformation equipment digital twin incremental transmission method, characterized in that, The method comprises: Collecting device state data, screening device state data with a change amount exceeding a preset threshold, and compressing the same, wherein a compression ratio of the compression is dynamically adjusted according to a change range of the device state data; A reinforcement learning model is constructed, wherein a state space of the reinforcement learning model comprises the screened change data, a health state of the device, and network bandwidth usage, an action space of the reinforcement learning model is an uploading strategy made in a current state, comprising an uploading priority, an uploading timing, and an uploading sequence of a data packet; an effect of each uploading strategy is evaluated by a reward function, and an optimal strategy is obtained; An uploading sequence is determined according to the uploading priority, whether the uploading condition is met is judged through a real-time state of the bandwidth and the priority of the data packet, and only data with a significant change compared with historical uploading data is transmitted; a bandwidth regularization term is added to the uploading strategy of each data packet, wherein the bandwidth regularization term represents an influence of the uploading operation on the current network bandwidth; Whether each data packet needs to be uploaded according to a plan or delayed is adjusted according to real-time transmission bandwidth feedback and the health state of the device; an optimized uploading strategy is obtained; and the optimized uploading strategy is outputted to obtain execution feedback data; An uploading execution effect evaluation index is calculated through the execution feedback data, wherein the uploading execution effect evaluation index represents an uploading success rate, a bandwidth occupation condition, and an influence on the device state; and a long-term optimization is performed on the uploading strategy based on historical uploading data, a bandwidth usage mode, and a device failure frequency.

2. The power transmission equipment digital twin incremental transmission method according to claim 1, characterized in that, The device state data comprises voltage, current, and temperature, and is collected by a sensor.

3. The power transmission equipment digital twin incremental transmission method of claim 1, wherein, The compression adjusts a compression ratio through a change range of the data; when the device state changes exceeds a compression threshold, a low compression ratio is selected; and when the device state changes is lower than the compression threshold, a high compression ratio is selected.

4. The power transmission equipment digital twin incremental transmission method of claim 1, wherein, The reward function is obtained by calculating a weighted sum of a device state change amount, a data uploading time delay, and network bandwidth utilization.

5. The power transmission equipment digital twin delta transmission method of claim 1, wherein, A network architecture of the reinforcement learning model comprises an input layer, a hidden layer, and an output layer; the input layer receives state data from the device; after calculation by the hidden layer, the output layer generates the uploading strategy.

6. The power transmission equipment digital twin delta transmission method of claim 1, wherein, The bandwidth regularization term is a ratio of a current bandwidth usage amount to a maximum system bandwidth; the uploading strategy is guided to reduce the uploading operation when the bandwidth is limited, so as to avoid network congestion in the uploading process.

7. The power transmission equipment digital twin delta transmission method of claim 1, wherein, A bandwidth occupation control function is introduced in the process of obtaining the optimized uploading strategy to dynamically adjust the uploading operation; when the bandwidth usage amount is close to the total bandwidth capacity, the bandwidth occupation control function is increased, thereby inhibiting the uploading operation and preferentially transmitting more important data.

8. The power transmission equipment digital twin delta transmission method of claim 1, wherein, A health state response mechanism is introduced in the process of obtaining the optimized uploading strategy; when a device fails or a device state changes greatly, an uploading priority is adjusted according to an emergency degree of data uploading of the device; when the device fails, the uploading priority of the failed device is increased to ensure that the state data of the device is uploaded preferentially.

9. The power transmission equipment digital twin delta transmission method of claim 1, wherein, The execution feedback data comprises an actual uploading execution result, a device health state, and a bandwidth utilization rate.

10. A power transmission and transformation equipment digital twin incremental transmission system, characterized in that, The system comprises: The state change detection unit is configured to collect device state data, filter device state data with a change amount exceeding a preset threshold, and compress the filtered device state data, wherein a compression ratio of the compression is dynamically adjusted according to a change range of the device state data. The policy generation unit is configured to construct a reinforcement learning model, wherein a state space of the reinforcement learning model includes filtered change data, a health state of the device, and network bandwidth usage, and an action space of the reinforcement learning model is an upload policy made in a current state, including an upload priority, an upload timing, and an upload sequence of a data packet; an effect of each upload policy is evaluated by a reward function, and an optimal policy is obtained. The incremental transmission unit is configured to determine an upload sequence according to the upload priority, determine whether an upload condition is met by using a real-time state of the bandwidth and the priority of the data packet, and only transmit data with a significant change compared with historical upload data, wherein a bandwidth regularization term is added to the upload policy of each data packet, and the bandwidth regularization term represents an influence of an upload operation on a current network bandwidth. The execution optimization unit is configured to adjust whether each data packet needs to be uploaded according to a plan or delayed according to a real-time transmission bandwidth feedback and a device health state, obtain an optimized upload policy, output the optimized upload policy, and obtain an execution feedback data. The long-term optimization unit is configured to calculate an upload execution effect evaluation index by using the execution feedback data, wherein the upload execution effect evaluation index represents an upload success rate, a bandwidth occupation condition, and an influence on the device state, and perform long-term optimization on the upload policy based on historical upload data, a bandwidth usage mode, and a device failure frequency.

Images