Vanadium flow battery full life cycle intelligent management platform based on digital twinning

Abstract

The application relates to a vanadium flow battery full-life-cycle intelligent management platform based on digital twinning, which comprises a sensing layer, which is installed on a measured device, is used for collecting parameters of a vanadium flow battery full-life-cycle in real time, and completes preliminary data preprocessing; a network layer, which is used for constructing a data transmission channel and realizing bidirectional data transmission between layers; a data layer, which receives data uploaded by the sensing layer through the network layer, processes original data, and transmits the processed data to a twinning layer; the twinning layer receives the data processed by the data layer, processes the data, and constructs a digital twin; an application layer, which is used for receiving simulation results of the digital twin output by the twinning layer, and converting the simulation results into intelligent management services; and a display layer, which is configured with a visual interface, a mobile terminal and a large-screen display system, and presents the results of the application layer services, the virtual model of the twinning layer and the data of the data layer in an intuitive and interactive manner.

Description

Technical Field

[0001] This invention relates to a management platform, specifically to a digital twin-based intelligent management platform for the entire lifecycle of vanadium redox flow batteries. Background Technology

[0002] As the global energy structure shifts towards clean energy, energy storage systems, as a core support for mitigating the volatility of new energy power generation and ensuring the stability of energy supply, are receiving increasing attention for their technological research and application. Vanadium redox flow batteries, with their unique advantages such as recyclable electrolyte, stable charge and discharge performance, long service life, high safety, and environmental friendliness, have broad application prospects in large-scale energy storage and distributed energy fields, and have become one of the key directions for current energy storage technology research and development and industrialization promotion.

[0003] The entire lifecycle of a vanadium redox flow battery encompasses multiple stages, including design, production, installation, operation, maintenance, and decommissioning and recycling. The operating parameters, environmental conditions, and performance degradation data at each stage are interconnected and mutually influential, directly determining the battery system's operating efficiency, safety performance, and lifespan. Therefore, building a platform capable of precise monitoring, intelligent analysis, and efficient management throughout the entire lifecycle is a key supporting technology for promoting the large-scale application of vanadium redox flow batteries.

[0004] Currently, research has been conducted both domestically and internationally on management technologies for vanadium redox flow batteries. Existing management platforms primarily focus on parameter monitoring and basic fault warnings during the operational phase. This is mainly achieved by deploying sensors on the battery itself and auxiliary equipment to collect key parameters such as voltage, current, electrolyte concentration, and ambient temperature and humidity. These data are then transmitted and displayed on a monitoring interface, with basic alarm functions implemented based on preset thresholds. Meanwhile, with the development of digital twin technology, some research is attempting to introduce it into energy storage device management. By constructing virtual models to map the physical entity's state, simulation analysis and predictive maintenance are achieved, improving the level of intelligent management to some extent.

[0005] However, in existing technologies, the entire lifecycle of vanadium redox flow batteries involves a massive number of parameters, and the importance of different parameters for battery operating status assessment and fault early warning varies significantly. Current platforms mostly adopt a full data acquisition and transmission mode without scientifically classifying the parameters. This leads to redundant data transmission and processing, increasing network transmission pressure and burdening subsequent data display and analysis. Summary of the Invention

[0006] The main objective of this invention is to provide a digital twin-based intelligent management platform for the entire lifecycle of vanadium redox flow batteries, which enables hierarchical data filtering, dynamic adaptation and transmission, and intelligent rendering scheduling.

[0007] To achieve the above objectives, this invention provides a digital twin-based intelligent management platform for the entire lifecycle of vanadium redox flow batteries, comprising a sensing layer, a network layer, a data layer, a digital twin layer, an application layer, and a presentation layer; wherein: The sensing layer, installed on the device under test, includes various sensors, smart meters, and industrial-grade edge computing devices. It is used to collect parameters of the vanadium redox flow battery throughout its entire life cycle in real time and to perform preliminary data preprocessing. The network layer is used to build data transmission channels and enable bidirectional data transmission between layers. The network layer has a dynamic transmission optimization module that adapts the transmission strategy by combining the filtering parameters output by the data layer, and coordinates the data layer and presentation layer to transmit data. The data layer receives data uploaded from the perception layer through the network layer, processes the raw data, and transmits the processed data to the twin layer. The data layer has a quantization and grading filtering unit, which uses parameter importance weight quantization calculation logic and data filtering threshold quantization calculation logic to realize parameter priority division and differentiated data filtering, reducing the amount of data transmitted to the presentation layer. The twin layer receives data processed by the data layer, performs data processing, and constructs a digital twin. The application layer is used to receive the simulation results of the digital twin output by the twin layer and transform the simulation results into intelligent management services. The presentation layer is equipped with a visual interface, mobile terminal and large screen display system to present the results of application layer services, virtual models of twin layer and data of data layer in an intuitive interactive way; the presentation layer has an intelligent rendering scheduling module that dynamically adjusts the rendering strategy in combination with the priority parameters output by data layer.

[0008] Preferably, the formula for calculating the importance weights of the data layer parameters is: , in, For the first The importance weight of each parameter The degree of influence of the parameter on the core performance of the battery. For the real-time requirement coefficient of the parameter, The correlation between parameters and fault warnings, The weighting coefficients are satisfied. .

[0009] Further optimized parameter importance weights The value range of is [0,1], and according to Prioritization: It is the highest priority. Medium priority It has low priority.

[0010] Preferably, the data filtering threshold formula for the data layer is: , in, For the first Data filtering threshold for each parameter This represents the maximum data transfer volume corresponding to the currently available network bandwidth. The original data volume of the j-th parameter. This represents the total number of parameters involved in data transmission.

[0011] Further preferred, the formulaic data hierarchical filtering unit is based on To achieve differentiated processing: Highest priority parameter To retain all data, medium priority parameters are set according to... Perform sampling dimensionality reduction, with low-priority parameters sorted by Perform aggregate statistics.

[0012] Preferably, the dynamic transmission optimization module at the network layer includes a dynamic bandwidth adaptation unit and a data compression unit. The dynamic bandwidth adaptation unit uses the data output from the data layer... Based on the allocation of transmission priorities, high-speed transmissions are given priority. The transmission bandwidth of the parameters.

[0013] More preferably, the data compression unit compresses data according to the output of the data layer. Adjust the compression intensity: use light compression for high-priority parameters and deep compression for low-priority parameters.

[0014] Preferably, the intelligent rendering scheduling module of the presentation layer includes an intelligent rendering scheduling unit, a local caching unit, and a multi-threaded rendering unit, wherein the intelligent rendering scheduling unit is configured to... As the criterion for determining rendering priority, high-quality rendering is given priority. parameter.

[0015] More preferably, the local cache unit prioritizes caching. and The core parameters, multi-threaded rendering units are based on Thread priority is determined.

[0016] More preferably, the weighting coefficient It can be dynamically adjusted according to the application scenario; the default value is... .

[0017] The advantages of this invention are: This invention sets up a quantitative and hierarchical filtering unit in the data layer, and combines the quantitative calculation logic of parameter importance weight with the quantitative calculation logic of data screening threshold to achieve scientific parameter classification and differentiated filtering. By classifying parameter priorities and processing them in a targeted manner, the amount of data transmitted to subsequent layers is greatly reduced while ensuring that the core parameters are fully retained. This effectively reduces network transmission pressure and data processing burden, thereby avoiding low management efficiency caused by data redundancy.

[0018] The dynamic transmission optimization module of the network layer of this invention combines the filtering parameters output by the data layer (i.e.) , The transmission strategy is dynamically adjusted by prioritizing the transmission bandwidth of high-priority parameters through a dynamic bandwidth adaptation unit, while the data compression unit adjusts the compression intensity according to parameter priority. This design allows the transmission strategy to flexibly adapt to network bandwidth fluctuations, reducing latency and packet loss in high-priority data transmission and improving overall data transmission efficiency.

[0019] The intelligent rendering scheduling module of the presentation layer of this invention uses parameter importance weights. Based on this core principle, the system utilizes an intelligent rendering scheduling unit to prioritize rendering, a local caching unit to prioritize core parameters, and a multi-threaded rendering unit to prioritize threads. This optimizes the front-end display logic from multiple dimensions, including rendering strategy, data caching, and thread allocation. It effectively reduces stuttering and slow response issues caused by synchronous rendering across the entire interface, ensuring that administrators can monitor the core status of the vanadium redox flow battery throughout its entire lifecycle in real time and smoothly, thus improving decision-making efficiency.

[0020] The weight coefficients of the parameter importance weight calculation logic in this invention It can be dynamically adjusted according to different application scenarios such as large-scale energy storage power stations and distributed energy, so that the parameter classification standard can match the management needs under different scenarios, effectively improving the platform's versatility and adaptability, and expanding the scope of technology application. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Many specific details are set forth in the following description to provide a thorough understanding of the present invention; however, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] The vanadium redox flow battery full lifecycle intelligent management platform based on digital twins described in this embodiment includes a sensing layer, a network layer, a data layer, a twin layer, an application layer, and a presentation layer. Each layer module works collaboratively according to preset logic. The specific implementation process is as follows: I. Data Acquisition and Preliminary Preprocessing Implementation in the Perception Layer The sensing layer is deployed at key locations in the vanadium redox flow battery-related testing equipment. It consists of various sensors, smart instruments, and industrial-grade edge computing devices adapted to the full lifecycle monitoring needs of vanadium redox flow batteries. Among them, the sensors and smart instruments are used to collect parameters at each stage of the vanadium redox flow battery's lifecycle, covering core battery operating parameters (such as electrolyte concentration, battery terminal voltage, charge and discharge current density, etc.), auxiliary monitoring parameters (such as equipment operating environment temperature and humidity, electrolyte circulation flow rate, etc.), and full lifecycle process parameters (such as component accuracy parameters in the production stage, maintenance record parameters in the maintenance stage, and performance degradation parameters in the decommissioning stage, etc.).

[0023] After receiving the raw data collected by various sensors and smart meters, the industrial-grade edge computing device completes the preliminary preprocessing work, which includes data format standardization conversion (converting heterogeneous data output by different devices into a preset digital format), preliminary removal of abnormal data (removing data that is obviously out of reasonable range due to sensor failure), and data time sequence alignment (synchronizing and normalizing data from different sources according to a unified timestamp). The preprocessed data is then transmitted to the data layer through the network layer.

[0024] II. Implementation of Quantization and Hierarchical Filtering in the Data Layer After receiving the preliminary preprocessed data uploaded by the perception layer through the network layer, the data layer first performs further standardization on the raw data to ensure data consistency and usability. Then, it uses the built-in quantization and grading filtering unit to prioritize parameters and filter differentiated data. The specific implementation process is as follows: Parameter Importance Weight Calculation: The quantization and grading filtering unit calls the parameter importance weight quantization calculation logic to obtain the influence degree corresponding to each preset parameter. Real-time demand coefficient and correlation with fault warning (These parameters can be preset using a vanadium redox flow battery performance simulation model combined with empirical data, and can be adjusted as needed later.) Simultaneously, preset weighting coefficients can be obtained. (Default value) It can be dynamically adjusted according to the needs of actual application scenarios, and Substitute into the formula The importance weights of each parameter are calculated. ( The value range is [0,1].

[0025] Parameter priority classification: based on calculations The values ​​are used to prioritize parameters, where The parameters are determined to be of the highest priority (such as electrolyte concentration, battery terminal voltage and other core operating parameters). The parameters are classified as medium priority (such as auxiliary monitoring parameters such as ambient temperature and humidity, electrolyte circulation flow rate, etc.). The parameters are classified as low priority (such as historical parameters of component accuracy during the production stage, and statistical parameters of performance degradation during the retirement stage).

[0026] Data filtering threshold calculation and differential processing: The quantization and grading filtering unit calls the data filtering threshold quantization calculation logic to collect the maximum data transmission volume corresponding to the current available network bandwidth in real time. Calculate the amount of raw data for each parameter. and the total number of parameters involved in data transmission Substitute into the formula , The data filtering thresholds for each parameter were calculated. Subsequently, according to Differentiated processing for parameters of different priorities: Setting the highest priority parameter Retain all data; medium priority parameters are set according to... Sampling and dimensionality reduction are performed on the corresponding proportions (e.g.) (At that time, retain one valid data point every other data point in the time series); low priority parameters are... The corresponding aggregation granularity is subjected to aggregation statistical processing (e.g. At that time, the raw data for each hour is aggregated into an hourly average. After the differentiation processing is completed, the data layer transmits the processed data to the twin layer, and simultaneously sets the parameters... and It is fed back to the network layer as a filtering parameter.

[0027] III. Implementation of Dynamic Transmission Optimization at the Network Layer The network layer enables bidirectional data transmission between the perception layer and the data layer, the data layer and the twin layer / presentation layer, and the twin layer and the application layer by constructing a bidirectional data transmission channel that integrates wired and wireless technologies. Simultaneously, the network layer's built-in dynamic transmission optimization module combines filtering parameters fed back from the data layer (i.e.,...) and The transmission strategy is dynamically adjusted, and the specific implementation process is as follows: Dynamic bandwidth adaptation: The dynamic bandwidth adaptation unit in the dynamic transmission optimization module monitors the bandwidth occupancy of the transmission link in real time, and outputs the data layer bandwidth as needed. Based on the division of transmission priority, high The highest priority parameter corresponds to the highest transmission priority. The medium priority parameter corresponds to medium transmission priority, low priority. The lowest priority parameter corresponds to the lowest transmission priority. When insufficient transmission link bandwidth is detected, the dynamic bandwidth adaptation unit prioritizes the transmission bandwidth of the highest transmission priority data. For medium and low transmission priority data, a combination of off-peak transmission (avoiding peak periods for high priority data transmission) and fragmented transmission (splitting data into multiple small data packets for transmission in different time periods) is used to avoid data transmission congestion.

[0028] Differential data compression: The data compression unit in the dynamic transmission optimization module compresses data based on the output of the data layer. Adjust the data compression strength, for The highest priority parameters employ a light compression algorithm (such as LZ4, with a compression ratio controlled between 1.2 and 1.5) to maximize data accuracy while ensuring data transmission efficiency; medium and low priority parameters are handled according to... Adjusting the size to control compression strength The smaller the value, the higher the compression intensity (e.g., using the LZMA algorithm for low-priority parameters, the compression ratio can be controlled between 3 and 5). Differentiated compression further reduces the data transmission volume and improves transmission efficiency.

[0029] IV. Construction and Implementation of Digital Twins in Twin Layers The twin layer receives differentiated data transmitted from the data layer. First, it performs further anomaly detection and interference mitigation on the data (using a Kalman filter algorithm to remove noise introduced during transmission). Then, based on the processed data, it constructs a digital twin of the vanadium redox flow battery's entire lifecycle. Specifically, using the physical entity of the vanadium redox flow battery as a prototype, a virtual mapping model is established, encompassing all stages: design, production, installation, operation, maintenance, and decommissioning / recycling. Parameters from each stage transmitted from the data layer are mapped in real-time to the corresponding modules in the virtual model, achieving real-time synchronization between the virtual model and the physical entity. Simultaneously, the twin layer performs full lifecycle simulation analysis based on the virtual model, including operational status assessment, performance degradation prediction, and fault tracing. The simulation results are then transmitted to the application layer.

[0030] V. Implementation of Intelligent Management Services at the Application Layer After receiving the simulation results output by the twin layer, the application layer transforms the simulation results into targeted intelligent management services through the built-in service conversion module. These services include generating operation optimization suggestions (generating charging and discharging parameter adjustment suggestions based on operation status assessment results), predictive maintenance reminders (generating maintenance time and maintenance content reminders based on performance degradation prediction results), fault early warning and handling plan push (generating fault type judgment and corresponding handling measures based on fault tracing results), and full lifecycle data traceability services (providing historical data query and traceability functions based on mapping data at each stage). All types of intelligent management service data are transmitted to the presentation layer.

[0031] VI. Implementation of Intelligent Rendering Scheduling in the Presentation Layer The presentation layer is configured with a visual interface, mobile terminals, and a large-screen display system. After receiving intelligent management service data transmitted from the application layer, virtual model data transmitted from the twin layer, and differentiated processing data transmitted from the data layer, it achieves efficient and intuitive data presentation through a built-in intelligent rendering scheduling module. The specific implementation process is as follows: Rendering priority scheduling: The intelligent rendering scheduling unit in the intelligent rendering scheduling module monitors the hardware performance parameters (including CPU utilization, video memory utilization, and memory utilization) of the display terminals (visual interface, mobile terminal, large screen) in real time, and simultaneously obtains various parameters fed back from the data layer. ,by As the rendering priority criterion, high priority is given to rendering high-quality rendering. The highest priority parameters (such as core operating parameters and fault warning data) are rendered, while medium and low priority parameters are rendered according to... Rendering is performed in order of size; when the terminal hardware performance is detected to be close to the threshold (e.g., CPU utilization ≥ 80%), real-time rendering of low-priority parameters is automatically paused, and only static data is displayed. Real-time rendering is restarted after the hardware performance recovers.

[0032] Core data local caching: The local caching unit in the intelligent rendering scheduling module is designed for... and For the highest priority core parameters, a local cache pool is established to store this type of frequently accessed data locally on the terminal, avoiding repeated data requests from the data layer / application layer for each display, thus shortening data loading time. Simultaneously, the local cache unit periodically synchronizes and updates data with the data layer to ensure the accuracy of the cached data.

[0033] Multi-threaded rendering allocation: Multi-threaded rendering units in the intelligent rendering scheduling module are allocated according to parameters. Rendering threads are prioritized, with the highest priority parameters assigned to the core rendering thread, medium priority parameters to regular rendering threads, and low priority parameters to auxiliary rendering threads. This multi-threaded parallel rendering avoids single-threaded congestion and improves overall rendering efficiency. Simultaneously, the thread allocation unit monitors the running status of each thread in real time. When congestion occurs in a thread, low-priority rendering tasks within that thread are automatically migrated to idle auxiliary threads, ensuring a smooth rendering process.

[0034] Through the collaborative work of the above modules, this platform realizes hierarchical filtering, dynamic adaptation and transmission, and intelligent rendering scheduling of vanadium redox flow battery data throughout its entire life cycle. This effectively solves the problems of data redundancy, inefficient transmission, and display lag in existing technologies, ensuring the efficiency and reliability of the management process.

[0035] Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

Claims

1. A vanadium redox flow battery full lifecycle intelligent management platform based on digital twins, characterized in that, It includes the perception layer, network layer, data layer, twin layer, application layer, and presentation layer; among which: The sensing layer, installed on the device under test, includes various sensors, smart meters, and industrial-grade edge computing devices. It is used to collect parameters of the vanadium redox flow battery throughout its entire life cycle in real time and to perform preliminary data preprocessing. The network layer is used to build data transmission channels and enable bidirectional data transmission between layers. The network layer has a dynamic transmission optimization module that adapts the transmission strategy by combining the filtering parameters output by the data layer, and coordinates the data layer and presentation layer to transmit data. The data layer receives data uploaded from the perception layer through the network layer, processes the raw data, and transmits the processed data to the twin layer. The data layer has a quantization and grading filtering unit, which uses parameter importance weight quantization calculation logic and data filtering threshold quantization calculation logic to realize parameter priority division and differentiated data filtering, reducing the amount of data transmitted to the presentation layer. The twin layer receives data processed by the data layer, performs data processing, and constructs a digital twin. The application layer is used to receive the simulation results of the digital twin output by the twin layer and transform the simulation results into intelligent management services. The presentation layer is equipped with a visual interface, mobile terminal and large screen display system to present the results of application layer services, virtual models of twin layer and data of data layer in an intuitive interactive way; the presentation layer has an intelligent rendering scheduling module that dynamically adjusts the rendering strategy in combination with the priority parameters output by data layer.

2. The intelligent management platform for the entire life cycle of vanadium redox flow batteries based on digital twins as described in claim 1, characterized in that, The formula for calculating the importance weights of parameters in the data layer is: ， in, For the first The importance weight of each parameter The degree of influence of the parameter on the core performance of the battery. For the real-time requirement coefficient of the parameter, The correlation between parameters and fault warnings, The weighting coefficients are and satisfy the following conditions: .

3. The intelligent management platform for the entire life cycle of vanadium redox flow batteries based on digital twins as described in claim 2, characterized in that, Parameter importance weight The value range of is [0,1], and according to Prioritization: As the highest priority, Medium priority It has low priority.

4. The intelligent management platform for the entire life cycle of vanadium redox flow batteries based on digital twins as described in claim 1, characterized in that, The formula for the data filtering threshold in the data layer is: ， in, For the first Data filtering threshold for each parameter This represents the maximum data transfer volume corresponding to the currently available network bandwidth. The original data volume of the j-th parameter. This represents the total number of parameters involved in data transmission.

5. The intelligent management platform for the entire life cycle of vanadium redox flow batteries based on digital twins as described in claim 4, characterized in that, Formulated data hierarchical filtering unit based on Achieve differentiated processing: Highest priority parameter To retain all data, medium priority parameters are set according to... Perform sampling dimensionality reduction, with low-priority parameters sorted by Perform aggregate statistics.

6. The intelligent management platform for the entire life cycle of vanadium redox flow batteries based on digital twins as described in claim 1, characterized in that, The network layer's dynamic transmission optimization module includes a dynamic bandwidth adaptation unit and a data compression unit. The dynamic bandwidth adaptation unit uses the data layer output... Based on the allocation of transmission priorities, high-speed transmissions are given priority. The transmission bandwidth of the parameters.

7. The intelligent management platform for the entire life cycle of vanadium redox flow batteries based on digital twins as described in claim 6, characterized in that, The data compression unit compresses data based on the output of the data layer. Adjust the compression intensity: use light compression for high-priority parameters and deep compression for low-priority parameters.

8. The intelligent management platform for the entire life cycle of vanadium redox flow batteries based on digital twins according to claim 1, characterized in that, The intelligent rendering scheduling module of the presentation layer includes an intelligent rendering scheduling unit, a local caching unit, and a multi-threaded rendering unit. The intelligent rendering scheduling unit uses... As the criterion for determining rendering priority, high-quality rendering is given priority. parameter.

9. The intelligent management platform for the entire life cycle of vanadium redox flow batteries based on digital twins as described in claim 8, characterized in that, The local cache unit prioritizes caching. and The core parameters, multi-threaded rendering units are based on Thread priority is determined.

10. The intelligent management platform for the entire life cycle of vanadium redox flow batteries based on digital twins according to claim 2, characterized in that, The weighting coefficient It can be dynamically adjusted according to the application scenario; the default value is... .

Images