Digital-twin-driven reconfigurable energy storage system real-time state monitoring method and device

Abstract

The application provides a digital twin driven reconfigurable energy storage system real-time state monitoring method and device, relates to the technical field of energy storage system monitoring, and comprises the following steps: reading a physical unit of an energy storage system, assembling an energy storage twin system based on a simplified agent unit as an energy storage twin engine; based on near field communication of the physical unit, reporting to the energy storage twin engine and reconfiguring the energy storage twin system, importing a state collapse component, performing collapse processing based on a super probability field, generating an energy storage monitoring state, and performing operation and maintenance management on the energy storage system according to the energy storage monitoring state. The technical problem that the model structure of the existing twin energy storage system is static and cannot adapt to dynamic reconfiguration of the energy storage system, resulting in low accuracy and reliability of energy storage system monitoring is solved. The technical effects of adaptive reconfiguration of the digital twin energy storage model, improvement of the real-time performance, accuracy and reliability of the energy storage system monitoring are achieved.

Description

Technical Field

[0001] This invention relates to the field of energy storage system monitoring technology, specifically to a method and apparatus for real-time status monitoring of reconfigurable energy storage systems driven by digital twins. Background Technology

[0002] With the continuous expansion of new energy power generation and the sustained growth of grid regulation needs, energy storage systems, as core equipment for smoothing fluctuations in new energy output and ensuring grid supply and demand balance, have been widely used. However, existing energy storage systems generally suffer from complex structures, variable states, and diverse fault types, leading to significant shortcomings in real-time monitoring and health assessment of their operational status. Traditional monitoring methods primarily rely on sensor data acquisition and threshold judgment, using fixed thresholds to trigger anomaly alarms. This approach fails to identify potential fault trends in advance, resulting in insufficient monitoring accuracy and delayed early warnings.

[0003] In recent years, digital twin technology has been introduced into the field of energy storage system monitoring. By constructing a digital mapping model in virtual space that interacts with the physical system in real time, it enables synchronous simulation and predictive analysis of the operating status of the energy storage system. However, existing energy storage twin technologies generally suffer from static structures, making them unable to cope with the reconstruction needs of energy storage system structures that change frequently. Furthermore, the real-time updates of the twin model largely rely on centralized data fusion, resulting in significant computational latency. Monitoring results are mostly limited to static state assessments, lacking quantifiable health trends and early warning mechanisms, which affects the accuracy of monitoring and the reliability of decision-making.

[0004] Existing twin energy storage systems suffer from technical problems such as static model structures that cannot adapt to dynamic reconfiguration of the energy storage system, resulting in low accuracy and reliability of energy storage system monitoring. Summary of the Invention

[0005] This application provides a method and apparatus for real-time status monitoring of a digital twin-driven reconfigurable energy storage system, which addresses the technical problem in the prior art where the model structure of a twin energy storage system is static and cannot adapt to the dynamic reconfiguration of the energy storage system, resulting in low accuracy and reliability of energy storage system monitoring.

[0006] In view of the above problems, this application provides a method and apparatus for real-time status monitoring of a reconfigurable energy storage system driven by digital twins.

[0007] The first aspect of this application provides a method for real-time status monitoring of a reconfigurable energy storage system driven by a digital twin. The method includes: reading the physical units of the energy storage system and constructing an energy storage twin system based on simplified agent units as an energy storage twin engine; reporting to the energy storage twin engine via near-field communication based on physical units and reconstructing the energy storage twin system as the energy storage mode switches, wherein the reconstructed energy storage twin system carries aiming point identifiers based on physical reference data and real-time data packets; importing the reconstructed energy storage twin system into a state collapse component and performing collapse processing based on a hyperprobability field to generate an energy storage monitoring status, wherein the hyperprobability field is constructed based on the multi-state structure of the energy storage system; and performing operation and maintenance management of the energy storage system according to the energy storage monitoring status.

[0008] Preferably, a physical unit is defined based on the smallest energy storage unit of the energy storage system, and N physical units are read, where N is the total number of smallest energy storage units; for the N physical units, N simplified proxy units based on simplified proxy elements are constructed; according to the physical connection relationship and energy storage interaction relationship of each physical unit in the first energy storage mode, the N simplified proxy units are networked to determine the energy storage twin system, where the first energy storage mode is a general energy storage mode; based on the energy storage twin system, the energy storage twin engine is constructed, where the energy storage twin system is a dynamic system.

[0009] Preferably, the first physical unit of the energy storage system is read, wherein the first physical unit is a battery module; for the first physical unit, a first simplified proxy element is set, wherein the first simplified proxy element is coupled with internal resistance-temperature-SOC; and a first simplified proxy unit is constructed based on the first simplified proxy element.

[0010] Preferably, as the energy storage mode switches, each physical unit determines neighboring physical units through near-field communication and generates an energy storage coding sequence, wherein each physical unit corresponds to an energy storage coding sequence, and each energy storage coding sequence consists of one physical unit and at least one neighboring physical unit; the energy storage coding sequence is reported to the energy storage twin engine to perform twin reconstruction of the energy storage twin system.

[0011] Preferably, physical reference data of the energy storage system is collected, wherein the physical reference data includes at least absolute environmental data and basic physical constants; real-time data packets of the energy storage system are collected, wherein the real-time data packets include at least electrical sensing data; using the physical reference data as a first real-world target point and the real-time data packets as a second real-world target point, the target point is marked on the reconstructed energy storage twin system.

[0012] Preferably, a hyperprobability field is constructed based on the multiple states of the energy storage system, wherein the multiple states include at least a healthy state, multiple fault states, and multiple decay paths; based on the hyperprobability field, a state collapse component is built as a central control embedded plug-in of the energy storage system.

[0013] Preferably, the energy storage twin system after aiming point marking is written into the state collapse component, and a first-step probabilistic collapse process is performed with the first real aiming point to determine the first collapse path; a second-step collapse process is performed with the second real aiming point to determine the second collapse path and the collapse probability field; and the energy storage monitoring state is determined based on the first collapse path, the second collapse path, and the collapse probability field.

[0014] Preferably, based on the collapse probability field, state features and probability vectors are identified to determine the energy storage state distribution; based on the first collapse path and the second collapse path, an energy storage health coefficient based on the collapse law is determined; and the energy storage state distribution and the energy storage health coefficient are used as the energy storage monitoring state.

[0015] Preferably, multiple alarm threads are set according to the energy storage risk control type and energy storage risk control level; by determining the energy storage monitoring status, targeted thread energy storage alarm and operation and maintenance management based on the multiple alarm threads are executed.

[0016] A second aspect of this application provides a digital twin-driven real-time status monitoring device for a reconfigurable energy storage system. The device includes: an energy storage twin system component module for reading the physical units of the energy storage system and constructing an energy storage twin system based on simplified agent units, serving as an energy storage twin engine; an energy storage twin system reconstruction module for reconstructing the energy storage twin system by reporting to the energy storage twin engine and performing near-field communication based on physical units as the energy storage mode switches, wherein the reconstructed energy storage twin system carries aiming point identifiers based on physical reference data and real-time data packets; an energy storage monitoring status generation module for importing the reconstructed energy storage twin system into a state collapse component, performing collapse processing based on a hyperprobability field, and generating an energy storage monitoring status, wherein the hyperprobability field is constructed based on the multi-state structure of the energy storage system; and an energy storage system management module for performing operation and maintenance management of the energy storage system according to the energy storage monitoring status.

[0017] One or more technical solutions provided in this application have at least the following technical effects or advantages: The method provided in this application reads the physical units of an energy storage system and constructs an energy storage twin system based on simplified agent units, serving as an energy storage twin engine. As the energy storage mode switches, near-field communication based on physical units is used to report to the energy storage twin engine and reconstruct the energy storage twin system. The reconstructed energy storage twin system carries aiming point identifiers based on physical reference data and real-time data packets. The reconstructed energy storage twin system is then imported into a state collapse component, where collapse processing based on a hyperprobability field is performed to generate an energy storage monitoring state. This hyperprobability field is constructed based on the multi-state structure of the energy storage system. Based on the energy storage monitoring state, the energy storage system is managed and operated. This achieves the technical effect of adaptive reconstruction of the digital twin energy storage model, improving the real-time performance, accuracy, and reliability of energy storage system monitoring. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic diagram of the real-time status monitoring method for a digital twin-driven reconfigurable energy storage system provided in this application.

[0020] Figure 2 A schematic diagram of the structure of the digital twin-driven real-time status monitoring device for reconfigurable energy storage systems provided in this application.

[0021] Explanation of reference numerals in the attached diagram: Energy storage twin system component module 11, Energy storage twin system reconfiguration module 12, Energy storage monitoring status generation module 13, Energy storage system management module 14. Detailed Implementation

[0022] This application provides a method and apparatus for real-time status monitoring of a digital twin-driven reconfigurable energy storage system. It addresses the technical problem in existing technologies where digital twin energy storage systems suffer from static model structures that cannot adapt to dynamic reconfiguration, leading to low accuracy and reliability in energy storage system monitoring. The method achieves adaptive reconfiguration of the digital twin energy storage model, thereby improving the real-time performance, accuracy, and reliability of energy storage system monitoring.

[0023] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be understood that the present invention is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. It should also be noted that, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, not all of them.

[0024] Example 1, as Figure 1 As shown, this application provides a real-time status monitoring method for a digital twin-driven reconfigurable energy storage system, the method comprising: Read the physical units of the energy storage system and build an energy storage twin system based on simplified agent units, which serves as the energy storage twin engine.

[0025] Furthermore, an energy storage twin system based on simplified agent units is constructed as an energy storage twin engine, including: defining physical units based on the smallest energy storage unit of the energy storage system, reading N physical units, where N is the total number of smallest energy storage units; constructing N simplified agent units based on simplified agent elements for the N physical units; networking the N simplified agent units according to the physical connection relationship and energy storage interaction relationship of each physical unit under the first energy storage mode to determine the energy storage twin system, where the first energy storage mode is a general energy storage mode; and constructing the energy storage twin engine based on the energy storage twin system, where the energy storage twin system is a dynamic system.

[0026] Specifically, an energy storage system refers to a collection of physical devices with energy storage and release functions, consisting of multiple battery clusters, battery modules, energy conversion devices, battery management systems, energy management systems, and control units. Energy storage systems have a multi-level structure, where various units interact through electrical and information flows, and multiple units work collaboratively to achieve the storage, conversion, and output of electrical energy. To ensure the reconfigurability and scalability of the digital twin system, the smallest energy storage unit of the energy storage system is considered the physical unit. This smallest energy storage unit refers to the indivisible basic unit in the energy storage system, possessing independent functions and characteristics, such as battery module units, PCS modules, battery management modules, and energy management modules. Through communication buses such as CAN, Modbus, and Ethernet, N independent and monitorable physical units in the energy storage system are read. Here, N represents the total number of the smallest energy storage units. When the energy storage system switches operating modes, some battery clusters may be switched on or off, and the PCS module may switch from grid-connected to off-grid mode. Modeling based on the smallest energy storage unit ensures the stability and real-time update capability of the mapping relationship in the digital twin system during topology changes. Simultaneously, basic parameters for each physical unit are collected, including rated capacity, internal resistance, temperature characteristics, and operating mode parameters.

[0027] For the acquired N physical units, corresponding virtual models are established in the digital twin space, and N simplified proxy units are constructed. Each simplified proxy unit is driven by a simplified proxy element, which can simulate the main behaviors and characteristics of the physical unit with less computing resources, and is used to approximate the operating behavior of the physical unit in the virtual space. A twin network is constructed according to the first energy storage mode of the energy storage system. The first energy storage mode refers to the general energy storage mode of the energy storage system under normal operation, such as all battery clusters operating in parallel and the PCS module operating in a conventional bidirectional conversion mode. Based on the physical connection relationships and energy storage interaction relationships of each physical unit under the first energy storage mode, the N simplified proxy units are logically networked. That is, a connection relationship diagram is constructed based on the physical connection relationships, where each node represents a simplified proxy unit and the edges represent the electrical connection methods between physical units. Simultaneously, an interaction relationship diagram is constructed based on the energy storage interaction relationships, where each node represents a simplified proxy unit and the edges represent the energy interaction methods between physical units. The connection relationship diagram and the interaction relationship diagram are merged to form a virtual model network corresponding to the topology of the physical system, i.e., the energy storage twin system. The physical connection relationship between each physical unit refers to the electrical connection method between the physical units, such as series or parallel connection. This can be monitored and collected in real time based on the electrical connection diagram and design documents of the energy storage system, or by deploying sensors such as current sensors and voltage sensors between the physical units. The energy storage interaction relationship refers to the energy interaction method between the physical units, such as power transfer paths, current flow direction, and signal coupling relationships. The energy management system (EMS) in the energy storage system typically records the energy interaction between the physical units, including power transfer paths and current flow direction, and obtains the energy storage interaction relationship based on the EMS. The energy storage twin system includes the energy flow relationship, control dependency relationship, and communication link between the physical units, and can simulate the behavior and characteristics of the energy storage system in the first energy storage mode.

[0028] The energy storage twin engine is a digital twin operating platform based on the energy storage twin system as its structural framework and multi-source data interaction as its core. It manages the dynamic attributes and operating status of each simplified agent unit, receives real-time data input from the actual energy storage system, and synchronously updates the parameters of the energy storage twin system, enabling real-time management and control of the energy storage twin system. When constructing the energy storage twin engine based on the energy storage twin system, the structural parameters, state parameters, and interaction relationships of each simplified agent unit are input into the energy storage twin engine to form a unified virtual computing model. Then, the engine's data communication interface is configured to enable real-time information interaction with the sensing and control layers of the physical energy storage system, receiving operating data including voltage, current, temperature, power, and switching status. Furthermore, the energy storage twin system is a dynamic system. During operation, it adjusts the virtual topology and agent parameters in real time according to changes in the energy storage system's operating mode, such as battery cluster switching, PCS mode switching, and redundant module start-up and shutdown, through a structural reconstruction algorithm. This achieves coordinated changes between the virtual and physical systems, ensuring the adaptability and accuracy of the energy storage twin system under different operating conditions.

[0029] By reading the physical units of the energy storage system, an energy storage twin engine is constructed, ensuring that the energy storage twin system can accurately reflect the behavior and characteristics of the energy storage system in the first energy storage mode, and has dynamic reconfiguration capabilities to adapt to the switching of energy storage modes. This improves the accuracy and reliability of real-time status monitoring of the energy storage system and enhances the effectiveness of energy storage system operation and maintenance management.

[0030] Furthermore, N simplified proxy units based on simplified proxy elements are constructed, including: reading the first physical unit of the energy storage system, wherein the first physical unit is a battery module; setting a first simplified proxy element for the first physical unit, wherein the first simplified proxy element is coupled with internal resistance-temperature-SOC; and constructing a first simplified proxy unit based on the first simplified proxy element.

[0031] Specifically, based on the N physical units read, corresponding N simplified agent units are constructed. These simplified agent units are used to simulate the operating characteristics and energy behavior of the physical units in virtual space. Specifically, the first physical unit of the energy storage system is read. This first physical unit is a battery module, which serves as the smallest energy storage unit, and its performance is affected by various factors such as temperature, state of charge, and internal impedance.

[0032] In twin modeling, to avoid the computational complexity and response lag of complete electrochemical models, a full physical twin or whole energy storage system twin approach is not adopted. Instead, a simplified modeling and reconstruction based on the real-time energy conversion mechanism of energy storage components is performed. This involves forming a simplified surrogate representation through the coupling relationships of key characteristics in the energy storage mechanism of the energy storage components. For the first physical unit, a first simplified surrogate element is defined. This element couples the three key characteristics of the battery module—internal resistance, temperature, and state of charge (SOC)—forming a ternary coupling relationship of internal resistance, temperature, and SOC. This coupling relationship reflects the main behaviors and characteristics of the battery module in actual operation. For example, the internal resistance of the battery module affects the battery's charge and discharge efficiency, temperature affects the battery's performance and lifespan, and SOC represents the state of charge, reflecting the remaining battery capacity. By coupling internal resistance, temperature, and SOC, the simplified surrogate element can simulate the main behaviors of the battery module with fewer computational resources.

[0033] Based on the first simplified agent element, a simplified agent unit is constructed. The simplified agent unit is a virtual component in the digital twin model. It simulates the actual battery module operation behavior based on the simplified agent element. The simplified agent unit consists of three parts: a parameter layer, a calculation layer, and an interface layer. The parameter layer is used to store the characteristic parameters and dynamically updated parameters of the simplified agent element. The calculation layer is used to perform internal resistance-temperature-SOC coupled model calculations to realize virtual simulation of the battery module's output power, thermal state, and energy balance. The interface layer is used to provide a data interaction interface to receive real-time monitoring data from the actual energy storage system.

[0034] Repeat the above process for all N physical units to construct N simplified proxy units based on simplified proxy elements. Combine the physical connections and energy storage interactions of each physical unit under the first energy storage mode to determine the energy storage twin system. By constructing the energy storage twin system based on simplified proxy elements, not only is the response speed of the digital twin system to real-time data changes improved, but it also ensures that the digital twin system can accurately reflect the actual operating status of the energy storage system.

[0035] As the energy storage mode switches, the energy storage twin engine is reported via near-field communication based on physical units, and the energy storage twin system is reconstructed. The reconstructed energy storage twin system has aiming point identifiers based on physical reference data and real-time data packets.

[0036] Furthermore, as the energy storage mode switches, near-field communication based on physical units is used to report to the energy storage twin engine and reconstruct the energy storage twin system. This includes: as the energy storage mode switches, each physical unit determines neighboring physical units through near-field communication and generates an energy storage coding sequence, wherein each physical unit corresponds to one energy storage coding sequence, and each energy storage coding sequence consists of one physical unit and at least one neighboring physical unit; the energy storage coding sequence is reported to the energy storage twin engine to perform twin reconstruction of the energy storage twin system.

[0037] Specifically, during the operation of an energy storage system, the energy storage model will switch according to different operational needs and strategies. Examples include battery cluster switching, PCS operating mode switching from grid-connected to off-grid, and backup module start-up and shutdown. To ensure that the energy storage twin system can accurately reflect the actual state of the energy storage system, the energy storage twin system is reconstructed during energy storage mode switching. Specifically, when the energy storage mode switches, each physical unit in the energy storage system, such as battery modules and PCS modules, actively identifies its domain through built-in near-field communication technologies, such as Bluetooth, Zigbee, or other point-to-point communication technologies. That is, after detecting a change in operating status, each physical unit broadcasts its own unit identifier and operating status information within the near-field range, while simultaneously receiving signals from surrounding units, thereby determining the physical units in the domain with which it has a direct electrical or logical connection. The near-field communication refers to the point-to-point interaction between physical units in the energy storage system within a limited space using short-range wireless communication technology. After identifying and determining the domain physical units, each physical unit generates an energy storage coding sequence based on its correspondence with the domain units. This energy storage coding sequence reflects the local connection topology of the physical unit under the current energy storage mode, including information such as the unit's unique identifier, the domain unit identifier, the connection type, and the current energy interaction direction. Each physical unit corresponds to one energy storage coding sequence, and each energy storage change sequence is composed of one physical unit and at least one domain physical unit.

[0038] After generating the energy storage coding sequence, the physical unit reports the sequence to the energy storage twin engine via a communication bus or wireless synchronization channel. Upon receiving all the energy storage coding sequences, the energy storage twin engine no longer analyzes and processes them based on real-time conditions at the twin center. Instead, it determines the domain relationships through the physical unit's own near-field communication and directly reports to the twin engine for synchronization, achieving a twin reconstruction of the energy storage twin system. The entire process is simple and efficient, enabling rapid response to energy storage mode switching. Furthermore, the reconstructed energy storage twin system carries aiming point markers based on physical baseline data and real-time data packets, ensuring that the energy storage twin system maintains consistency with the actual energy storage system's state, improving the reliability and effectiveness of real-time status monitoring of the energy storage system.

[0039] Furthermore, after performing twin reconstruction on the energy storage twin system, the process includes: collecting physical reference data of the energy storage system, wherein the physical reference data includes at least absolute environmental data and basic physical constants; collecting real-time data packets of the energy storage system, wherein the real-time data packets include at least electrical sensing data; and marking the reconstructed energy storage twin system with the physical reference data as a first real-world target point and the real-time data packets as a second real-world target point.

[0040] Specifically, after reconstructing the energy storage twin system using near-field communication based on physical units, physical reference data of the energy storage system is collected. This physical reference data reflects the stable external environment and fundamental physical properties of the energy storage system, providing a stable reference benchmark for the energy storage twin system and ensuring that the energy storage twin system can accurately reflect the state of the energy storage system. The physical reference data includes at least absolute environmental data and fundamental physical constants. The absolute environmental data reflects the external environmental conditions of the energy storage system, including temperature, humidity, and air pressure, and is acquired through temperature sensors, humidity sensors, and air pressure sensors. Fundamental physical constants are extracted from the energy storage system configuration file. These fundamental physical constants refer to the basic parameters of the energy storage system, including the rated capacity of the battery and the rated power of the PCS module. Simultaneously, real-time data packets of the energy storage system are collected. These real-time data packets reflect the actual electrical state of each physical unit in the energy storage system at the current moment, including at least electrical sensing data, which includes at least voltage, current, and power, acquired through current sensors and voltage sensors.

[0041] After acquiring physical reference data and real-time data packets, the reconstructed energy storage twin system is marked with target points. These target points are used to calibrate the state of the energy storage twin system with that of the actual energy storage system, ensuring that the reconstructed energy storage twin system accurately reflects the operating state of the actual energy storage system. Specifically, the physical reference data is used as the first real-world target point. In the reconstructed energy storage twin system, the static characteristic layer of the energy storage twin system is calibrated to ensure that the energy storage twin system is consistent with the actual energy storage system in terms of environmental parameters and fundamental physical constants. The real-time data packets are used as the second real-world target point. In the reconstructed energy storage twin system, real-time data, including current, voltage, and power, are input to the dynamic parameter interface of each agent unit to drive the dynamic state update of the energy storage twin system, ensuring that the energy storage twin system is consistent with the operating state of the actual energy storage system.

[0042] By using aiming point marking, the reconstructed energy storage twin system has clear real-world parameter benchmarks, ensuring that the twin system can accurately reflect the real-time operating status of the energy storage system, thereby improving the reliability and accuracy of energy storage system monitoring and diagnosis.

[0043] The reconstructed energy storage twin system is imported into the state collapse component, and a collapse process based on a hyperprobability field is performed to generate the energy storage monitoring state, wherein the hyperprobability field is constructed based on the multivariate state of the energy storage system.

[0044] Furthermore, before importing the reconstructed energy storage twin system into the state collapse component, the construction of the state collapse component includes: constructing a hyperprobability field based on the multiple states of the energy storage system, wherein the multiple states include at least a healthy state, multiple fault states, and multiple decay paths; and building a state collapse component based on the hyperprobability field as a central control embedded plug-in of the energy storage system.

[0045] Specifically, by analyzing historical data of the energy storage system, a multi-dimensional state of the system is obtained. This multi-dimensional state reflects various possible operating conditions of the energy storage system, including at least a healthy state, multiple fault states, and multiple degradation paths. A healthy state refers to the energy storage system operating normally under its design conditions, with all physical units operating within the expected range. Multiple fault states refer to various faults that the energy storage system may experience, including but not limited to single-cell failure, battery overheating, short circuits, and over-discharge. These fault types can be obtained through historical fault records and / or experimental data. Multiple degradation paths refer to the possible paths through which an energy storage system gradually degrades with usage time, charge-discharge cycles, or environmental changes. These include, but are not limited to, battery capacity degradation and energy efficiency reduction. Modeling can be achieved using long-term operational data and cycle life models. Mathematical functions or state transition matrices can be used to describe the evolution of the physical units of the energy storage system over time. For example, key parameter data of the energy storage system during long-term operation, including battery capacity, charge-discharge efficiency, and internal resistance, can be collected. The collected data is then cleaned, filtered, and normalized to remove outliers. Statistical analysis methods, such as multinomial regression, are used to analyze data trends and determine the degradation pattern. For instance, the degradation of battery capacity over time can be described using an exponential function. Cycle life models are used to describe the performance changes of a battery after multiple charge-discharge cycles. Key parameters of the battery during multiple charge-discharge cycles, such as capacity and internal resistance, are obtained. Data fitting methods, such as least squares, are used to fit the key parameters and establish a cycle life model.

[0046] Based on the multivariate states of the energy storage system, a hyperprobability field is constructed using probability theory and multivariate random field theory. The construction steps include: defining a state space, mapping the multivariate states to a multidimensional state space, where each dimension represents a possible state of the energy storage system, and each state is represented by its probability of occurrence. In the state space, for each state, its probability value is estimated based on historical data, experimental results, and real-time monitoring data. For example, historical operating data, experimental results, and real-time monitoring data of the energy storage system are collected, and the frequency of each state in the historical operating data is calculated using frequency statistics as the initial probability value for that state. Bayes' theorem is then used in conjunction with experimental results and real-time monitoring data to update the initial probability value of the state, obtaining the probability value corresponding to each state. The probability value is assigned to the corresponding state dimension, reflecting the likelihood of the energy storage system being in that state. For situations where each state may occur simultaneously, a joint probability distribution or conditional probability method is used to describe the correlation between states. The state dimensions and their corresponding probability values ​​are integrated to form a multi-state hyperprobability field, which is a multidimensional probability distribution space encompassing the probability distributions of all possible states.

[0047] Furthermore, based on the hyperprobability field, a state collapse component is constructed. This component includes an input interface, a probability operation unit, a path analysis unit, and an output interface. The input interface receives real-world target point data (physical baseline data and real-time data packets) marked by the twin system. The probability operation unit performs multi-step probability deduction on the input data using the hyperprobability field. The path analysis unit identifies the most probable state evolution path based on the probability deduction results. The output interface outputs the collapsed state information. The state of each simplified agent unit in the twin system is mapped to a corresponding state node in the hyperprobability field. The state collapse component performs conditional constraint calculations on the hyperprobability field based on the real-world target point data to achieve probability collapse. The state collapse component is embedded as a hollow insert into the energy storage twin engine, supporting real-time invocation and data updates. The reconstructed energy storage twin system is imported into the state collapse component, and hyperprobability field-based collapse processing is performed to generate energy storage monitoring status. This enables real-time dynamic monitoring and health assessment of the energy storage system's status, improving the stability and safety of the energy storage system's operation.

[0048] Furthermore, a collapse process based on a hyperprobability field is performed to generate an energy storage monitoring status, including: writing the energy storage twin system after aiming point marking into the state collapse component; performing a one-step probabilistic collapse process with the first real aiming point to determine a first collapse path; performing a two-step collapse process with the second real aiming point to determine a second collapse path and a collapse probability field; and determining the energy storage monitoring status based on the first collapse path, the second collapse path, and the collapse probability field.

[0049] Specifically, based on the constructed hyperprobability field and state collapse component, a collapse process based on the hyperprobability field is performed. The collapse process refers to the process in the energy storage twin system where, by introducing real data target markers into the hyperprobability field, the originally uncertain multivariate state probability distribution is gradually collapsed into the current state probability distribution under the action of real constraints. This is a dynamic convergence process of projecting from the virtual possibility space to the real state space. The collapse process result reflects the current real operating state and probability characteristics of the energy storage system.

[0050] The state collapse component is a centrally controlled embedded plugin used to process and analyze the collapse process of the probability field. It completely writes the energy storage twin system, after aiming point marking, into the state collapse component. The aiming point marking includes physical reference data and real-time data packets. The state collapse component uses the first real aiming point, i.e., the physical reference data, as the initial aiming point and inputs it into the hyperprobability field. The physical reference data includes absolute environmental stability and fundamental physical constants, providing a stable reference point for the hyperprobability field. This allows the hyperprobability field to undergo a one-step probability collapse process around the first real aiming point in the initial stage. During this one-step probability collapse process, the originally unconstrained state cloud in the hyperprobability field begins to converge, forming a primary state distribution that conforms to the physical boundary conditions, and generating the first collapse path. The first collapse path refers to the convergence process of the hyperprobability field under the constraints of the physical reference data.

[0051] Then, the real-time data packets are used as the second real-world target point for a two-step collapse process. The state collapse component maps multiple real-time data to each state dimension of the hyperprobability field, causing the probability field that forms the initial state distribution that meets the physical boundary conditions to collapse further under the drive of dynamic input data, converging to a state that matches the second real-world target point, obtaining the final collapse result, and forming a collapse probability field. The collapse probability field is the state probability analysis of the current energy storage system, and a second collapse path is determined. The second collapse path refers to the path from the initial collapse state to the final collapse group state under the action of the second real-world target point, reflecting the further convergence process of the hyperprobability field under the constraints of real-time data.

[0052] Then, based on the analysis of the first collapse path, the second collapse path, and the collapse probability field, the energy storage monitoring status is comprehensively determined. Specifically, when the collapse path is smooth, the probability cloud converges rapidly, and the single-peak distribution is obvious, it indicates that the energy storage system is operating stably and in good health. When the collapse path has multiple branches, and the probability distribution oscillates or exhibits a multi-peak distribution during the collapse process, it indicates that there are potential anomalies or faults within the energy storage system. Therefore, based on the evolution speed and convergence difficulty of the first and second collapse paths, combined with the collapse probability field, the energy storage monitoring status is assessed and determined. Specifically, the first and second collapse paths of the hyperprobability field are observed and analyzed, and the collapse velocity, collapse stability, and collapse bifurcation degree are extracted. The collapse velocity is the convergence rate of the standard deviation of the probability distribution, the collapse stability is the degree of fluctuation of the collapse path, and the collapse bifurcation degree is the number and intensity of multiple path distributions. Based on the collapse processing of the hyperprobability field, accurate and reliable energy storage monitoring status is generated, enabling real-time monitoring and early warning of the health status of the energy storage system, thereby improving the stability and safety of the energy storage system operation.

[0053] Furthermore, the energy storage monitoring status is determined based on the first collapse path, the second collapse path, and the collapse probability field, including: identifying state features and probability vectors based on the collapse probability field to determine the energy storage status distribution; determining an energy storage health coefficient based on the collapse law based on the first collapse path and the second collapse path; and using the energy storage status distribution and the energy storage health coefficient as the energy storage monitoring status.

[0054] Specifically, the collapse probability field, through the application of real-world constraints to a multivariate state space, forms a dynamic probability distribution after considering two types of target data. This distribution reflects the probability of each possible state of the energy storage system at the current moment. Multidimensional analysis of the collapse probability field is performed to extract state features, which represent the various possible states the energy storage system can currently be in. Simultaneously, the probability distributions corresponding to these state features are obtained, forming probability vectors. Based on the state features and probability vectors, the energy storage state distribution is determined. This energy storage state distribution reflects the probability distribution of various operating states within the energy storage system.

[0055] The first and second collapse paths are analyzed to extract collapse features, including collapse velocity, collapse stability, and path bifurcation. These features are normalized to form collapse velocity factor, collapse stability factor, and path bifurcation factor. Based on these collapse features, an energy storage health coefficient is determined according to the collapse pattern. A faster collapse velocity, a smoother collapse path, and fewer path bifurcations indicate a healthier energy storage system. The collapse velocity, collapse stability, and path bifurcation are weighted and summed to obtain the energy storage health coefficient. The energy storage health coefficient H = w1 × collapse velocity factor + w2 × collapse stability factor + w3 × path bifurcation factor, where w1, w2, and w3 are weighting coefficients obtained through historical operating data and manual calibration, and adjusted according to actual needs. The closer the energy storage health coefficient is to 1, the higher the health level of the energy storage system. Furthermore, the energy storage probability distribution and energy storage health coefficient are integrated to serve as the energy storage monitoring status, enabling real-time and accurate monitoring and evaluation of the health status of the energy storage system.

[0056] By analyzing the state characteristics and probability vectors in the collapse probability field, we can accurately obtain the various possible states of the energy storage system and their probability distributions, thereby gaining a comprehensive understanding of the system's operational status. Combining the first and second collapse paths, we can further analyze the evolution patterns and trends of the energy storage system's state, obtaining the energy storage monitoring status. This allows for an accurate and intuitive reflection of the system's operational status, enabling early detection of potential faults, optimization of operation and maintenance strategies, improvement of the system's reliability and safety, and ensuring its efficient and stable operation.

[0057] Based on the energy storage monitoring status, the energy storage system is operated and maintained.

[0058] Furthermore, the operation and maintenance management of the energy storage system includes: setting multiple alarm threads based on energy storage risk control type and energy storage risk control level; and executing targeted thread energy storage alarm and operation and maintenance management based on the multiple alarm threads by determining the energy storage monitoring status.

[0059] Specifically, multiple alarm threads are set up based on the energy storage risk control type and level. The energy storage risk control type refers to various fault types categorized according to the nature of the fault in the energy storage system, such as thermal management risk, electrical risk, and equipment performance risk. Thermal management risks include abnormal temperature rise and thermal runaway; electrical risks include overvoltage, short circuit, and insulation abnormalities; and equipment performance risks include PCS failure or battery cluster degradation. The energy storage risk control level reflects the severity of the risk and is divided into four levels: normal, minor fault, moderate fault, and severe fault. The level classification is determined based on the threshold range of the energy storage health system and the probability weight of each abnormal state in the state distribution. After determining the energy storage risk control type and level, multiple alarm threads are set up according to different energy storage risk control types and levels, including high-priority threads, medium-priority threads, and normal threads. Each alarm thread corresponds to an independent risk channel and response strategy to achieve multi-dimensional parallel operation and maintenance control.

[0060] Based on state matching and threshold ranges, the energy storage monitoring status is determined to belong to the energy storage risk control type and level, enabling targeted thread-based energy storage alarm and operation and maintenance management of the multi-alarm threads. When the monitoring status of a certain energy storage unit in the energy storage system enters a specific risk range, the alarm thread corresponding to the current area is automatically retrieved, and corresponding operation and maintenance actions are executed. For example, when frequent mode switching of the PCS module and high latency in the communication thread are detected, the communication and electrical dual threads are triggered to execute real-time diagnosis and local isolation strategies. If multiple threads are triggered simultaneously, they are prioritized according to the risk control level, with the highest risk level as the scheduling core, coordinating the execution order and response resource allocation of each thread.

[0061] Through a multi-threaded intelligent operation and maintenance mechanism driven by energy storage monitoring status, real-time, hierarchical, and targeted management of the energy storage system is achieved, which improves the operational safety and management efficiency of the energy storage system and ensures its safe and stable operation.

[0062] Example 2, based on the same inventive concept as the digital twin-driven real-time status monitoring method for reconfigurable energy storage systems in the foregoing examples, such as... Figure 2 As shown, this application provides a real-time status monitoring device for a digital twin-driven reconfigurable energy storage system, wherein the device includes: The energy storage twin system component module 11 is used to read the physical units of the energy storage system and build an energy storage twin system based on simplified agent units, serving as the energy storage twin engine. The energy storage twin system reconstruction module 12 is used to report to the energy storage twin engine and reconstruct the energy storage twin system according to the switching of energy storage modes, using near-field communication based on physical units. The reconstructed energy storage twin system has aiming point identifiers based on physical reference data and real-time data packets. The energy storage monitoring status generation module 13 is used to import the reconstructed energy storage twin system into the state collapse component, perform collapse processing based on a hyperprobability field, and generate the energy storage monitoring status. The hyperprobability field is constructed based on the multi-state of the energy storage system. The energy storage system management module 14 is used to perform operation and maintenance management of the energy storage system according to the energy storage monitoring status.

[0063] Furthermore, the energy storage twin system component module 11 is also used for: defining physical units based on the smallest energy storage unit of the energy storage system, reading N physical units, where N is the total number of smallest energy storage units; constructing N simplified proxy units based on simplified proxy elements for the N physical units; networking the N simplified proxy units according to the physical connection relationship and energy storage interaction relationship of each physical unit under the first energy storage mode to determine the energy storage twin system, where the first energy storage mode is a general energy storage mode; and constructing the energy storage twin engine according to the energy storage twin system, where the energy storage twin system is a dynamic system.

[0064] Furthermore, the energy storage twin system component module 11 is also used to: read the first physical unit of the energy storage system, wherein the first physical unit is a battery module; set a first simplified proxy element for the first physical unit, wherein the first simplified proxy element is coupled with internal resistance-temperature-SOC; and construct a first simplified proxy unit based on the first simplified proxy element.

[0065] Furthermore, the energy storage twin system reconstruction module 12 is also used for: as the energy storage mode switches, each physical unit determines the neighboring physical units through near-field communication and generates an energy storage coding sequence, wherein each physical unit corresponds to an energy storage coding sequence, and each energy storage coding sequence consists of one physical unit and at least one neighboring physical unit; and reporting the energy storage coding sequence to the energy storage twin engine to perform twin reconstruction of the energy storage twin system.

[0066] Furthermore, the energy storage twin system reconstruction module 12 is also used to: collect physical reference data of the energy storage system, wherein the physical reference data includes at least absolute environmental data and basic physical constants; collect real-time data packets of the energy storage system, wherein the real-time data packets include at least electrical sensing data; and mark the energy storage twin system after twin reconstruction by using the physical reference data as a first real-world aiming point and the real-time data packets as a second real-world aiming point.

[0067] Furthermore, the energy storage monitoring status generation module 13 is also used to: construct a hyperprobability field based on the multi-state of the energy storage system, wherein the multi-state includes at least a healthy state, multiple fault states, and multiple decay paths; and build a state collapse component based on the hyperprobability field as a central control embedded plug-in of the energy storage system.

[0068] Furthermore, the energy storage monitoring status generation module 13 is also used to: write the energy storage twin system after the aiming point is marked into the status collapse component; perform a one-step probabilistic collapse process with the first real aiming point to determine the first collapse path; perform a two-step collapse process with the second real aiming point to determine the second collapse path and the collapse probability field; and determine the energy storage monitoring status based on the first collapse path, the second collapse path, and the collapse probability field.

[0069] Furthermore, the energy storage monitoring status generation module 13 is also used to: identify state features and probability vectors based on the collapse probability field, and determine the energy storage status distribution; determine the energy storage health coefficient based on the collapse law based on the first collapse path and the second collapse path; and use the energy storage status distribution and the energy storage health coefficient as the energy storage monitoring status.

[0070] Furthermore, the energy storage system management module 14 is also used to: set multiple alarm threads according to the energy storage risk control type and energy storage risk control level; and execute targeted thread energy storage alarm and operation and maintenance management based on the multiple alarm threads by determining the energy storage monitoring status.

[0071] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0072] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for real-time status monitoring of a reconfigurable energy storage system driven by digital twins, characterized in that, The method includes: Read the physical units of the energy storage system and build an energy storage twin system based on simplified agent units, which serves as the energy storage twin engine; As the energy storage mode switches, the energy storage twin engine is reported to the energy storage twin engine via near-field communication based on physical units, and the energy storage twin system is reconstructed. The reconstructed energy storage twin system has aiming point identifiers based on physical reference data and real-time data packets. The reconstructed energy storage twin system is imported into the state collapse component, and a collapse process based on a hyperprobability field is performed to generate the energy storage monitoring state, wherein the hyperprobability field is constructed based on the multivariate state of the energy storage system. Based on the energy storage monitoring status, the energy storage system is operated and maintained. Before introducing the reconstructed energy storage twin system into the state collapse component, the construction of the state collapse component includes: Based on the multi-state of the energy storage system, a hyperprobability field is constructed, wherein the multi-state includes at least a healthy state, multiple fault states, and multiple decay paths; Based on the hyperprobability field, a state collapse component is constructed as a central control embedded plug-in for the energy storage system. Perform collapse processing based on a hyperprobability field to generate energy storage monitoring status, including: The energy storage twin system marked with the aiming point is written into the state collapse component, and a one-step probabilistic collapse process is performed with the first real aiming point to determine the first collapse path. A two-step collapse process is performed using the second real aiming point to determine the second collapse path and the collapse probability field. The energy storage monitoring status is determined based on the first collapse path, the second collapse path, and the collapse probability field.

2. The real-time status monitoring method for a digital twin-driven reconfigurable energy storage system as described in claim 1, characterized in that, To construct an energy storage twin system based on simplified agent units, serving as the energy storage twin engine, including: Define physical units based on the smallest energy storage unit of the energy storage system, and read N physical units, where N is the total number of smallest energy storage units; For the N physical units, construct N simplified proxy units based on simplified proxy elements; Based on the physical connection relationship and energy storage interaction relationship of each physical unit in the first energy storage mode, the N simplified agent units are networked to determine the energy storage twin system. The first energy storage mode is a general energy storage mode. Based on the energy storage twin system, the energy storage twin engine is constructed, wherein the energy storage twin system is a dynamic system.

3. The real-time status monitoring method for a digital twin-driven reconfigurable energy storage system as described in claim 2, characterized in that, Construct N simplified proxy units based on simplified proxy elements, including: Read the first physical unit of the energy storage system, wherein the first physical unit is a battery module; For the first physical unit, a first simplified proxy element is set, wherein the first simplified proxy element is coupled with internal resistance-temperature-SOC. Based on the first simplified proxy element, construct the first simplified proxy unit.

4. The real-time status monitoring method for a digital twin-driven reconfigurable energy storage system as described in claim 1, characterized in that, As the energy storage mode switches, near-field communication based on physical units is used to report to the energy storage twin engine and reconstruct the energy storage twin system, including: As the energy storage mode switches, each physical unit determines its neighboring physical units through near-field communication and generates an energy storage coding sequence. Each physical unit corresponds to an energy storage coding sequence, and each energy storage coding sequence consists of one physical unit and at least one neighboring physical unit. The energy storage coding sequence is reported to the energy storage twin engine to perform twin reconstruction of the energy storage twin system.

5. The real-time status monitoring method for a digital twin-driven reconfigurable energy storage system as described in claim 4, characterized in that, After performing a twin reconfiguration on the energy storage twin system, the following are included: Collect physical reference data of the energy storage system, wherein the physical reference data includes at least absolute environmental data and basic physical constants; Collect real-time data packets from the energy storage system, wherein the real-time data packets contain at least electrical sensing data; Using the physical reference data as the first real-world aiming point and the real-time data packet as the second real-world aiming point, aiming point marking is performed on the reconstructed energy storage twin system.

6. The real-time status monitoring method for a digital twin-driven reconfigurable energy storage system as described in claim 1, characterized in that, Based on the first collapse path, the second collapse path, and the collapse probability field, the energy storage monitoring status is determined, including: Based on the collapse probability field, identify state characteristics and probability vectors to determine the energy storage state distribution; Based on the first collapse path and the second collapse path, determine the energy storage health coefficient based on the collapse law; The energy storage state distribution and the energy storage health coefficient are used as the energy storage monitoring state.

7. The real-time status monitoring method for a digital twin-driven reconfigurable energy storage system as described in claim 1, characterized in that, The operation and maintenance management of the energy storage system includes: Multiple alarm threads are set according to the type and level of energy storage risk control; By determining the energy storage monitoring status, targeted thread energy storage alarm and operation and maintenance management based on the multi-alarm thread is executed.

8. A real-time status monitoring device for a reconfigurable energy storage system driven by a digital twin, characterized in that, The steps for implementing the real-time status monitoring method for a digital twin-driven reconfigurable energy storage system according to any one of claims 1 to 7 include: The energy storage twin system component module is used to read the physical units of the energy storage system and build an energy storage twin system based on simplified agent units, serving as the energy storage twin engine; The energy storage twin system reconfiguration module is used to report to the energy storage twin engine and reconfigure the energy storage twin system via near-field communication based on physical units as the energy storage mode switches. The reconfigured energy storage twin system has aiming point identifiers based on physical reference data and real-time data packets. The energy storage monitoring state generation module is used to import the reconstructed energy storage twin system into the state collapse component, perform collapse processing based on the hyperprobability field, and generate the energy storage monitoring state, wherein the hyperprobability field is constructed based on the multivariate state of the energy storage system. The energy storage system management module is used to perform operation and maintenance management of the energy storage system based on the energy storage monitoring status.

Images