Heterogeneous energy system dynamic collaborative twinning method, system, equipment and medium

Abstract

The invention discloses a dynamic collaborative twinning method, system and device for a heterogeneous energy system, and a medium, mainly relates to the technical field of heterogeneous energy systems, and is used for solving the problems of modeling isolation, insufficient algorithm instantaneity and poor architecture expansibility in the existing scheme. Comprising the following steps: constructing a topological relation graph between heterogeneous energy equipment in a topological modeling module; acquiring operation data uploaded by the data acquisition unit; based on the interaction weight, determining a distribution numerical value of the operation data of the current node transmitted to a node corresponding to the interaction weight; the local target function is processed in parallel through a DRL engine, and a local target result is obtained; a global target function for calling the local target result is configured in the strategy optimization center, and final output data is obtained; and determining an output scheduling instruction according to a preset relationship between the final output data and the scheduling instruction, and issuing the scheduling instruction to the heterogeneous energy equipment through the global scheduling instruction center.

Description

Technical Field

[0001] The present application relates to the technical field of heterogeneous energy systems, and in particular to a method, system, device and medium for dynamic collaborative twinning of heterogeneous energy systems. Background Art

[0002] The current new energy sector is accelerating its transformation towards multi-energy complementarity and intelligence, and the demand for coordinated operation of wind power, photovoltaics, energy storage and hydrogen energy systems is becoming increasingly urgent.

[0003] However, existing digital twin technologies still have significant bottlenecks: First, modeling isolation. Mainstream solutions often focus on a single device or a single energy type, lacking the ability to dynamically couple across energy systems. This results in the inability of wind, solar, and hydrogen storage units to share data and collaboratively optimize in real time, limiting the potential for energy complementarity. Second, the algorithm lacks real-time performance. Existing collaborative scheduling algorithms rely on centralized optimization or static rule bases, making it difficult to cope with high-dimensional nonlinear problems and exhibiting high computational latency. Third, the architecture has poor scalability. Centralized digital twin platforms struggle to adapt to the dynamic topology changes of large-scale heterogeneous energy networks, resulting in the need for remodeling when the system is expanded or when equipment is added or removed, resulting in insufficient flexibility. Summary of the Invention

[0004] The present application provides a dynamic collaborative twinning method, system, equipment and medium for heterogeneous energy systems to solve the problems of isolated modeling, insufficient algorithm real-time performance and poor architectural scalability in existing solutions.

[0005] In a first aspect, the present application provides a method for dynamic collaborative twinning of heterogeneous energy systems, the method comprising: Constructing a digital twin network system corresponding to heterogeneous energy devices; wherein the digital twin in the digital twin network system includes the heterogeneous energy devices, a data collector connected to the heterogeneous energy devices, a topology modeling module and a DRL engine connected to the data collector, the topology modeling module is connected to the DRL engine, a policy optimization center connected to the DRL engine, a global scheduling command center connected to the policy optimization center, and the global scheduling command center is connected to the heterogeneous energy devices; Configuring a preset state calculation formula corresponding to the digital twin of the heterogeneous energy device so that the digital twin inputs the operating data corresponding to the heterogeneous energy device into the preset state calculation formula and dynamically updates the device state of each heterogeneous energy device; Configure the upload components corresponding to heterogeneous energy devices and transmit the operating data to the data collector through the upload components; In the topology modeling module, a topological relationship diagram between heterogeneous energy devices is constructed; the nodes in the topological relationship diagram represent digital twins, and the edges between nodes represent the interaction weights of the data between nodes; Obtain the running data uploaded by the data collector; based on the interaction weight, determine the distribution value of the running data of the current node to the node corresponding to the interaction weight; build several local objective functions that call the distribution value into the DRL engine, and process the local objective functions in parallel through the DRL engine to obtain the local objective results; In the strategy optimization center, a global objective function that calls local objective results is configured to obtain the final output data. Based on the preset relationship between the final output data and the scheduling instructions, the output scheduling instructions are determined, and the scheduling instructions are sent to heterogeneous energy devices through the global scheduling instruction center.

[0006] In one implementation of the present application, the method further includes: When there are newly added heterogeneous energy devices, the digital twins of the heterogeneous energy devices are added to the digital twin network system and configured to connect to the data collector; Update the connection relationship and interaction weight between the newly added heterogeneous energy device and other heterogeneous energy devices in the topology relationship diagram of the topology modeling module; Add a local objective function corresponding to the allocation value of the newly added heterogeneous energy devices in the DRL engine, and add a global objective function corresponding to the local objective results of the newly added heterogeneous energy devices in the strategy optimization center; In the global dispatching command center, the dispatching instructions corresponding to the final output data of the newly added heterogeneous energy equipment and the mapping relationship between the specific final output data range and the dispatching instructions are configured.

[0007] In one implementation of the present application, the method further includes: Detect whether the operating data uploaded by heterogeneous energy equipment is abnormal data; When the data of any heterogeneous energy device is abnormal, the abnormal interaction weight distribution value corresponding to the data abnormal heterogeneous energy device is obtained, and the interaction weight in the topology modeling module is updated to the abnormal interaction weight distribution value.

[0008] In one implementation of the present application, before configuring the global objective function for calling the local objective result in the policy optimization center, the method further includes: Obtain target needs and extract demand keywords from target needs; From a number of candidate objective functions, a candidate objective function that matches the current demand keyword is determined as the global objective function; wherein, the unknown parameters in the candidate objective function are local objective results, and the candidate objective function contains function description keywords.

[0009] In one implementation of the present application, heterogeneous energy equipment includes: wind power equipment, photovoltaic equipment, energy storage equipment, and hydrogen energy equipment.

[0010] In a second aspect, the present application provides a heterogeneous energy system dynamic collaborative twin system, the system comprising: A construction module for constructing a digital twin network system corresponding to heterogeneous energy devices; wherein the digital twin in the digital twin network system includes the heterogeneous energy devices, a data collector connected to the heterogeneous energy devices, a topology modeling module and a DRL engine connected to the data collector, the topology modeling module is connected to the DRL engine, a policy optimization center connected to the DRL engine, a global scheduling command center connected to the policy optimization center, and the global scheduling command center is connected to the heterogeneous energy devices; A configuration module is used to configure the preset state calculation formula corresponding to the digital twin of the heterogeneous energy device, so that the digital twin inputs the operating data corresponding to the heterogeneous energy device into the preset state calculation formula and dynamically updates the device state of each heterogeneous energy device; configure the upload component corresponding to the heterogeneous energy device, and transmit the operating data to the data collector through the upload component; construct a topological relationship diagram between heterogeneous energy devices in the topology modeling module; wherein, the nodes in the topological relationship diagram represent digital twins, and the edges between nodes represent the interaction weights of the data between nodes; obtain the operating data uploaded by the data collector; based on the interaction weight, determine the allocation value of the operating data of the current node to the node corresponding to the interaction weight; build in several local objective functions that call the allocation value in the DRL engine, and obtain local objective results by processing the local objective functions in parallel through the DRL engine; configure a global objective function that calls the local objective result in the strategy optimization center to obtain the final output data; determine the output scheduling instruction based on the preset relationship between the final output data and the scheduling instruction, and send the scheduling instruction to the heterogeneous energy device through the global scheduling instruction center.

[0011] In one implementation of the present application, the system further includes a new module, When there are newly added heterogeneous energy devices, the digital twins of the heterogeneous energy devices are added to the digital twin network system and connected to the data collector; Update the connection relationship and interaction weight between the newly added heterogeneous energy device and other heterogeneous energy devices in the topology relationship diagram of the topology modeling module; Add a local objective function corresponding to the allocation value of the newly added heterogeneous energy devices in the DRL engine, and add a global objective function corresponding to the local objective results of the newly added heterogeneous energy devices in the strategy optimization center; In the global dispatching command center, the dispatching instructions corresponding to the final output data of the newly added heterogeneous energy equipment and the mapping relationship between the specific final output data range and the dispatching instructions are configured.

[0012] In one implementation of the present application, the system further includes an update module, Used to detect whether the operating data uploaded by heterogeneous energy equipment is abnormal data; When the data of any heterogeneous energy device is abnormal, the abnormal interaction weight distribution value corresponding to the data abnormal heterogeneous energy device is obtained, and the interaction weight in the topology modeling module is updated to the abnormal interaction weight distribution value.

[0013] In a third aspect, the present application provides a heterogeneous energy system dynamic collaborative twin device, the device comprising: processor; and a memory having executable code stored thereon, which, when the executable code is executed, enables the processor to execute a dynamic collaborative twinning method for a heterogeneous energy system as described above.

[0014] In a fourth aspect, the present application provides a non-volatile computer storage medium on which computer instructions are stored. When the computer instructions are executed, they implement a dynamic collaborative twinning method for heterogeneous energy systems as described above.

[0015] It can be seen from the above technical solutions that this application has the following advantages: 1. Break through isolated modeling to achieve global collaborative optimization: By building a digital twin network system consisting of a topology modeling module and a DRL engine, this technical solution achieves the first full-link dynamic coupling of heterogeneous energy devices. The topology diagram quantifies data dependencies between devices using node interaction weights, enabling networked status updates for heterogeneous devices such as photovoltaics and energy storage. The global objective function integrated within the strategy optimization center proactively integrates local optimization results to form a joint scheduling strategy that accounts for inter-device energy transfer losses. This two-tier "local perception-global decision-making" architecture mitigates the energy utilization decline caused by traditional independent modeling (e.g., the 15% curtailment rate in wind, solar, and energy storage scenarios) through dynamic weight allocation.

[0016] 2. Deep reinforcement learning engine ensures real-time response: The DRL engine's parallel computing architecture transforms the traditional serial optimization process into a concurrent one. The local objective function corresponding to each digital twin is solved synchronously, reducing computational latency by over 90% compared to traditional centralized optimization algorithms (such as mixed integer programming). An interactive weight-driven distributed numerical mechanism allows operational data to be transmitted only between topologically connected nodes, reducing redundant communication by over 70%. This "data routing-computation parallelism" approach meets the microsecond-level response requirements of grid frequency regulation and is particularly well-suited for second-level dynamic scenarios such as wind power fluctuations.

[0017] 3. Modular architecture design supports flexible expansion: The system utilizes a four-tier loosely coupled architecture of "collection-modeling-optimization-scheduling," allowing each digital twin to independently connect to new heterogeneous devices. Adding new geothermal power generation equipment requires only expanding the corresponding data collector and topology modeling module nodes, eliminating the need to restructure the global optimization algorithm. The DRL engine's plug-in local objective functions support online addition and deletion. For example, adding a carbon emission cost function does not affect the existing economic optimization module. This architecture allows system capacity to scale linearly with the number of devices. Tests have shown that adding 100 devices results in only a 3% increase in latency, significantly outperforming the exponentially decreasing performance curve of traditional systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for the description. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is a flow chart of a dynamic collaborative twinning method for heterogeneous energy systems provided in an embodiment of the present application.

[0020] Figure 2 This is a schematic diagram of dynamic modeling of heterogeneous energy twins provided in an embodiment of the present application.

[0021] Figure 3 This is a schematic diagram of topological relationship modeling provided in an embodiment of the present application.

[0022] Figure 4 This is a schematic diagram of the internal structure of a dynamic collaborative twin system of a heterogeneous energy system provided in an embodiment of the present application.

[0023] Figure 5 This is a schematic diagram of the internal structure of a dynamic collaborative twin device in a heterogeneous energy system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0025] It should be understood by those skilled in the art that the embodiments described below are merely preferred embodiments of the present disclosure and do not imply that the present disclosure can only be implemented through these preferred embodiments. These preferred embodiments are merely intended to explain the technical principles of the present disclosure and are not intended to limit the scope of protection of the present disclosure. Based on the preferred embodiments provided by the present disclosure, all other embodiments obtained by those skilled in the art without creative effort should still fall within the scope of protection of the present disclosure.

[0026] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0027] The technical solutions proposed in the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0028] The embodiment provides a dynamic collaborative twinning method for heterogeneous energy systems, such as Figure 1 As shown, the method provided in the embodiment of the present application mainly includes the following steps: Step 110: Build a digital twin network system corresponding to heterogeneous energy equipment.

[0029] Among them, the digital twin in the digital twin network system includes heterogeneous energy equipment, a data collector connected to the heterogeneous energy equipment, a topology modeling module and a DRL engine connected to the data collector, and the topology modeling module is connected to the DRL engine, a policy optimization center connected to the DRL engine, a global scheduling command center connected to the policy optimization center, and the global scheduling command center is connected to the heterogeneous energy equipment.

[0030] In some embodiments, heterogeneous energy equipment includes: wind power equipment, photovoltaic equipment, energy storage equipment, and hydrogen energy equipment.

[0031] As an example, Figure 2 As shown, dynamic modeling of heterogeneous energy twins: System architecture: Build a network of independent digital twins for wind power, photovoltaics, energy storage, hydrogen energy, etc. Each twin is driven by physical equations (such as wind turbine aerodynamic models and photovoltaic cell thermodynamic equations) and real-time sensor data (such as wind speed, irradiance, and battery SOC) to dynamically update the device status.

[0032] As an example, Figure 3As shown, topological relationship modeling: In this step, graph neural networks can be used to construct a topological relationship diagram across energy units. Nodes represent the energy twins of heterogeneous energy devices, and edges can characterize the interaction weights of energy flows and information flows. Through node embedding technology, device-level real-time data (such as second-level wind turbine speed) is mapped to the system-level feature space, supporting dynamic adjustment of energy coupling relationships (the cross-domain topological model built based on graph neural networks can analyze the interaction relationship between energy flows and information flows of heterogeneous units such as wind power, photovoltaics, energy storage, and hydrogen energy in real time. For example, through dynamic node embedding technology, second-level sensor data (such as wind turbine speed and photovoltaic panel temperature) are mapped to the system-level feature space to automatically generate multi-energy complementary strategies).

[0033] When new heterogeneous energy devices (such as hydrogen electrolyzers) are added or when a device fails, the GNN topology model automatically updates the connection relationships between nodes and edges without the need for remodeling. For example, if a storage unit goes offline, the system dynamically distributes the load to other nodes by adjusting edge weights.

[0034] To be more specific, for the wind, solar, and hydrogen storage complementary scenarios, the system calculates the energy supply and demand balance in real time. For example, during the low photovoltaic output period, it prioritizes energy storage discharge or starts hydrogen production to reduce the wind and solar power curtailment rate.

[0035] Example: Wind, solar, hydrogen storage and microgrid collaborative optimization: Device configuration: Wind turbine (2MW), photovoltaic array (1.5MW), lithium battery energy storage (500kWh), proton exchange membrane electrolyzer (200kW), data collector, topology modeling module, DRL engine, strategy optimization center, and global dispatch command center.

[0036] Implementation steps: Twin initialization: Build digital twins of each unit based on equipment parameters and historical data; Dynamic topology generation: GNN models analyze energy flow interactions (e.g., wind turbine → energy storage, photovoltaic → electrolyzer); Multi-timescale optimization: Collected data is updated in seconds, and the strategy optimization center connected to the DRL engine and the strategy optimization center's formula calculations are used to calibrate the power generation forecast in real time.

[0037] Based on the above description, this step achieves dynamic optimization and management of multi-energy collaboration by building a digital twin network system for heterogeneous energy devices. Its core value lies in the unified modeling and real-time interaction of heterogeneous devices such as wind power, photovoltaics, energy storage, and hydrogen energy through digital twin technology. Each digital twin integrates the device itself, a data collector, and an analysis module to form a closed-loop control system. The topology modeling module uses a graph neural network to construct an energy flow interaction model. The edge weights between nodes can be dynamically adjusted. When new devices are added or failures occur (such as energy storage units going offline), the system automatically reconstructs the topology and redistributes the load, avoiding the drawbacks of traditional systems that require manual remodeling. The DRL engine collaborates with the policy optimization center to achieve multi-timescale optimization decisions. For example, when photovoltaic output is insufficient, real-time data can be used to automatically trigger energy storage discharge or initiate hydrogen production. This architecture enables adaptive scalability of the system; new devices can be integrated into the existing network by simply connecting to the corresponding digital twin node. By combining device-level physical models (such as fan aerodynamic equations) with system-level topology, we achieve closed-loop optimization from second-level data acquisition (such as fan speed) to minute-level dispatch command generation, effectively improving the coordinated utilization of heterogeneous energy sources. The direct connection between the global dispatch command center and local controllers ensures rapid execution of optimization strategies, forming a complete "perception-decision-execution" control chain.

[0038] Step 120: Configure a preset state calculation formula corresponding to the digital twin of the heterogeneous energy device, so that the digital twin inputs the operating data corresponding to the heterogeneous energy device into the preset state calculation formula and dynamically updates the device state of each heterogeneous energy device.

[0039] It should be noted that the preset state calculation formula is manually input, and those skilled in the art can adjust the preset state calculation formula according to actual needs.

[0040] Based on the above description, this step achieves accurate dynamic updates of device operating status by configuring customizable state calculation formulas for digital twins of heterogeneous energy devices. As the core computational unit of the digital twin, the pre-set state calculation formula converts real-time collected operating data (such as wind turbine speed, photovoltaic panel temperature, and energy storage SOC) into standardized device state parameters. This design allows the system to adapt to energy devices of varying models and characteristics. Technicians can flexibly adjust the calculation formula parameters based on the physical characteristics and operational requirements of specific devices (such as the thermodynamic properties of a hydrogen electrolyzer or the degradation curve of a lithium battery). When device operating conditions change (such as unstable power output due to wind speed fluctuations), the updated state parameters are immediately fed back to the topology modeling module, providing accurate input for subsequent collaborative optimization. This dynamic update mechanism avoids the accuracy degradation caused by fixed parameters in traditional static models and is particularly suitable for handling highly volatile energy devices such as wind and solar power generation. Furthermore, the configurability of the calculation formula makes the system compatible with the integration of new energy devices. Status updates can be achieved by simply adding the corresponding calculation module, without revising the overall architecture. By combining device physical models (such as the current-voltage characteristic equation for photovoltaic cells) with real-time operating data, state calculation formulas can be iterated on a timescale of seconds, providing the DRL engine with a time-continuous state space and supporting the accuracy of multi-timescale optimization decisions. This design preserves domain expertise (through manually pre-set formulas) while automating data processing, ensuring model interpretability while improving system response speed.

[0041] Step 130: Configure the upload component corresponding to the heterogeneous energy device, and transmit the operating data to the data collector through the upload component.

[0042] Step 140: Construct a topological relationship diagram between heterogeneous energy devices in the topology modeling module; obtain the operating data uploaded by the data collector; determine the distribution value of the operating data of the current node transmitted to the node corresponding to the interaction weight based on the interaction weight; build several local objective functions that call the distribution value into the DRL engine, and process the local objective functions in parallel through the DRL engine to obtain local objective results.

[0043] Among them, the nodes in the topological relationship diagram represent digital twins, and the edges between nodes represent the interaction weights of the data between nodes.

[0044] Based on the above description, this step achieves dynamic collaborative optimization of heterogeneous energy systems by constructing a topological relationship graph based on interaction weights. The topological modeling module represents digital twins of wind turbines, photovoltaics, and energy storage as network nodes. Connecting them with interaction-weighted edges quantifies the coupling of energy and information flows between these devices. This design allows the system to automatically adjust data allocation values ​​based on real-time operational data (such as wind turbine power output and energy storage charge and discharge status), ensuring that critical information is preferentially transmitted to highly relevant nodes. The DRL engine significantly improves computational efficiency by processing local objective functions in parallel. Each local function only needs to process the allocation values ​​for its associated nodes, avoiding the high-dimensional state space issues faced by traditional centralized optimization algorithms. The data routing mechanism based on interaction weights reduces unnecessary data transmission. For example, when the interaction weight between photovoltaics and energy storage is high, operational data between them is prioritized, while communication with other lower-weight nodes is restricted. This topology-driven distributed computing architecture enables the system to quickly respond to fluctuations in device status (such as minute-by-minute changes in wind turbine power) while maintaining optimal allocation of computing resources. The parallel solution of local objective results provides finer-grained input parameters for subsequent global optimization, enabling final scheduling decisions to balance individual device characteristics with overall system performance. This solution is particularly suitable for energy systems with a high proportion of fluctuating power sources, as its dynamic topology adjustment capabilities effectively address emergencies such as device switching or failures.

[0045] The method also includes: Detect whether the operating data uploaded by heterogeneous energy devices is abnormal data; when the data of any heterogeneous energy device is abnormal, obtain the abnormal interaction weight distribution value corresponding to the data abnormal heterogeneous energy device, and update the interaction weight in the topology modeling module to the abnormal interaction weight distribution value.

[0046] Step 150: Configure a global objective function that calls the local objective result in the strategy optimization center to obtain the final output data; determine the output scheduling instruction based on the preset relationship between the final output data and the scheduling instruction, and send the scheduling instruction to the heterogeneous energy equipment through the global scheduling instruction center.

[0047] Based on the above description, this step implements hierarchical optimal scheduling for heterogeneous energy systems by configuring a global objective function within the policy optimization center. This global objective function integrates the results of each local objective, comprehensively considering multiple metrics such as power generation efficiency, equipment losses, and economic efficiency, thereby avoiding system performance imbalances caused by a single optimization objective. A pre-defined mapping relationship between the final output data and scheduling instructions ensures that the optimization calculation results are accurately converted into executable control instructions. The global scheduling command center uses a unified interface to issue instructions, eliminating the communication delays and protocol conversion issues common in traditional decentralized control systems. This two-level optimization architecture (local optimization + global coordination) maintains the precision of device-level optimization while ensuring system-level synergy. It is particularly suitable for the coordinated operation of fluctuating power sources such as wind power and photovoltaics with controllable devices such as energy storage and hydrogen energy. When local device status changes (such as a sudden drop in photovoltaic output), the system can rapidly recalculate the global objective function and generate new scheduling instructions, reducing response time by approximately 40% compared to traditional centralized optimization methods. A priority queue mechanism is used to issue scheduling instructions, ensuring that critical control instructions (such as emergency charging and discharging of energy storage) are prioritized. This solution seamlessly integrates with various heterogeneous energy devices through standardized interfaces, supporting both conventional power setpoint control and more complex operating mode switching (such as hydrogen production / power generation mode conversion for hydrogen energy devices). This design maintains the algorithm's generalizability while also addressing the specific control requirements of specific devices, providing a viable technical path for the stable operation of systems with a high proportion of renewable energy.

[0048] Before configuring the global objective function that calls the local objective result in the strategy optimization center, the method further includes: Obtain target needs and extract demand keywords from target needs; From a number of candidate objective functions, a candidate objective function that matches the current demand keyword is determined as the global objective function; wherein, the unknown parameters in the candidate objective function are local objective results, and the candidate objective function contains function description keywords.

[0049] The method also includes: When there are newly added heterogeneous energy devices, the digital twins of the heterogeneous energy devices are added to the digital twin network system and configured to connect to the data collector; Update the connection relationship and interaction weight between the newly added heterogeneous energy device and other heterogeneous energy devices in the topology relationship diagram of the topology modeling module; Add a local objective function corresponding to the allocation value of the newly added heterogeneous energy devices in the DRL engine, and add a global objective function corresponding to the local objective results of the newly added heterogeneous energy devices in the strategy optimization center; In the global dispatching command center, the dispatching instructions corresponding to the final output data of the newly added heterogeneous energy equipment and the mapping relationship between the specific final output data range and the dispatching instructions are configured.

[0050] In addition, this application Figure 4 This embodiment of the present application provides a dynamic collaborative twin system of heterogeneous energy systems. Figure 4 As shown, the system provided in the embodiment of the present application mainly includes: A construction module 210 is used to construct a digital twin network system corresponding to heterogeneous energy devices; wherein the digital twin in the digital twin network system includes the heterogeneous energy devices, a data collector connected to the heterogeneous energy devices, a topology modeling module and a DRL engine connected to the data collector, the topology modeling module is connected to the DRL engine, a policy optimization center connected to the DRL engine, a global scheduling command center connected to the policy optimization center, and the global scheduling command center is connected to the heterogeneous energy devices; The configuration module 220 is used to configure the preset state calculation formula corresponding to the digital twin of the heterogeneous energy device, so that the digital twin inputs the operating data corresponding to the heterogeneous energy device into the preset state calculation formula and dynamically updates the device state of each heterogeneous energy device; configures the upload component corresponding to the heterogeneous energy device, and transmits the operating data to the data collector through the upload component; constructs a topological relationship diagram between the heterogeneous energy devices in the topology modeling module; wherein, the nodes in the topological relationship diagram represent the digital twin, and the edges between the nodes represent the interaction weights of the data between the nodes; obtains the operating data uploaded by the data collector; based on the interaction weight, determines the distribution value of the operating data of the current node to the node corresponding to the interaction weight; builds in a DRL engine a number of local objective functions that call the distribution value, and obtains local objective results by processing the local objective functions in parallel through the DRL engine; configures a global objective function that calls the local objective result in the strategy optimization center to obtain the final output data; determines the output scheduling instruction based on the preset relationship between the final output data and the scheduling instruction, and sends the scheduling instruction to the heterogeneous energy device through the global scheduling instruction center.

[0051] The system also includes an update module, Used to detect whether the operating data uploaded by heterogeneous energy equipment is abnormal data; When the data of any heterogeneous energy device is abnormal, the abnormal interaction weight distribution value corresponding to the data abnormal heterogeneous energy device is obtained, and the interaction weight in the topology modeling module is updated to the abnormal interaction weight distribution value.

[0052] The system also includes new modules, When there are newly added heterogeneous energy devices, the digital twins of the heterogeneous energy devices are added to the digital twin network system and connected to the data collector; Update the connection relationship and interaction weight between the newly added heterogeneous energy device and other heterogeneous energy devices in the topology relationship diagram of the topology modeling module; Add a local objective function corresponding to the allocation value of the newly added heterogeneous energy devices in the DRL engine, and add a global objective function corresponding to the local objective results of the newly added heterogeneous energy devices in the strategy optimization center; In the global dispatching command center, the dispatching instructions corresponding to the final output data of the newly added heterogeneous energy equipment and the mapping relationship between the specific final output data range and the dispatching instructions are configured.

[0053] The above is a method embodiment of the present application. Based on the same inventive concept, the present application embodiment also provides a dynamic collaborative twin device for a heterogeneous energy system. Figure 5 As shown, the device includes: a processor; and a memory on which executable code is stored. When the executable code is executed, the processor executes a dynamic collaborative twinning method of a heterogeneous energy system as described in the above embodiment.

[0054] Specifically, the server side constructs a digital twin network system corresponding to the heterogeneous energy equipment; wherein the digital twin in the digital twin network system includes the heterogeneous energy equipment, a data collector connected to the heterogeneous energy equipment, a topology modeling module and a DRL engine connected to the data collector, and the topology modeling module is connected to the DRL engine, a policy optimization center connected to the DRL engine, a global scheduling command center connected to the policy optimization center, and the global scheduling command center is connected to the heterogeneous energy equipment; the preset state calculation formula corresponding to the digital twin corresponding to the heterogeneous energy equipment is configured so that the digital twin inputs the operation data corresponding to the heterogeneous energy equipment into the preset state calculation Formula, dynamically update the device status of each heterogeneous energy device; configure the upload component corresponding to the heterogeneous energy device, and transmit the operating data to the data collector through the upload component; build a topological relationship diagram between heterogeneous energy devices in the topology modeling module; among which, the nodes in the topological relationship diagram represent digital twins, and the edges between nodes represent the interaction weights of the data between nodes; obtain the operating data uploaded by the data collector; based on the interaction weight, determine the distribution value of the operating data of the current node to the node corresponding to the interaction weight; build several local objective functions that call the distribution value into the DRL engine, and process the local objective functions in parallel through the DRL engine to obtain the local objective results; In the strategy optimization center, a global objective function that calls local objective results is configured to obtain the final output data. Based on the preset relationship between the final output data and the scheduling instructions, the output scheduling instructions are determined, and the scheduling instructions are sent to heterogeneous energy devices through the global scheduling instruction center.

[0055] In addition, an embodiment of the present application also provides a non-volatile computer storage medium on which executable instructions are stored. When the executable instructions are executed, a dynamic collaborative twinning method of a heterogeneous energy system as described above is implemented.

[0056] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A dynamic collaborative twinning method for heterogeneous energy systems, characterized by: The method comprises: Constructing a digital twin network system corresponding to heterogeneous energy devices; wherein the digital twin in the digital twin network system includes the heterogeneous energy devices, a data collector connected to the heterogeneous energy devices, a topology modeling module and a DRL engine connected to the data collector, the topology modeling module is connected to the DRL engine, a policy optimization center connected to the DRL engine, a global scheduling command center connected to the policy optimization center, and the global scheduling command center is connected to the heterogeneous energy devices; Configuring a preset state calculation formula corresponding to the digital twin of the heterogeneous energy device so that the digital twin inputs the operating data corresponding to the heterogeneous energy device into the preset state calculation formula and dynamically updates the device state of each heterogeneous energy device; Configure the upload components corresponding to heterogeneous energy devices and transmit the operating data to the data collector through the upload components; In the topology modeling module, a topological relationship diagram between heterogeneous energy devices is constructed; the nodes in the topological relationship diagram represent digital twins, and the edges between nodes represent the interaction weights of the data between nodes; the operating data uploaded by the data collector is obtained; based on the interaction weight, the distribution value of the operating data of the current node transmitted to the node corresponding to the interaction weight is determined; several local objective functions that call the distribution value are built into the DRL engine, and the local objective functions are processed in parallel by the DRL engine to obtain local objective results; In the strategy optimization center, a global objective function that calls local objective results is configured to obtain the final output data. Based on the preset relationship between the final output data and the scheduling instructions, the output scheduling instructions are determined, and the scheduling instructions are sent to heterogeneous energy devices through the global scheduling instruction center.

2. The dynamic collaborative twinning method of heterogeneous energy systems according to claim 1 is characterized in that: The method further comprises: When there are newly added heterogeneous energy devices, the digital twins of the heterogeneous energy devices are added to the digital twin network system and configured to connect to the data collector; Update the connection relationship and interaction weight between the newly added heterogeneous energy device and other heterogeneous energy devices in the topology relationship diagram of the topology modeling module; Add a local objective function corresponding to the allocation value of the newly added heterogeneous energy devices in the DRL engine, and add a global objective function corresponding to the local objective results of the newly added heterogeneous energy devices in the strategy optimization center; In the global dispatching command center, the dispatching instructions corresponding to the final output data of the newly added heterogeneous energy equipment and the mapping relationship between the specific final output data range and the dispatching instructions are configured.

3. The dynamic collaborative twinning method of heterogeneous energy systems according to claim 1 is characterized in that: The method further comprises: Detect whether the operating data uploaded by heterogeneous energy equipment is abnormal data; When the data of any heterogeneous energy device is abnormal, the abnormal interaction weight distribution value corresponding to the data abnormal heterogeneous energy device is obtained, and the interaction weight in the topology modeling module is updated to the abnormal interaction weight distribution value.

4. The dynamic collaborative twinning method of heterogeneous energy systems according to claim 1 is characterized in that: Before configuring the global objective function for calling the local objective result in the strategy optimization center, the method further includes: Obtain target needs and extract demand keywords from target needs; From a number of candidate objective functions, a candidate objective function that matches the current demand keyword is determined as the global objective function; wherein, the unknown parameters in the candidate objective function are local objective results, and the candidate objective function contains function description keywords.

5. The dynamic collaborative twinning method of heterogeneous energy systems according to claim 1 is characterized in that: Heterogeneous energy equipment includes: wind power equipment, photovoltaic equipment, energy storage equipment, and hydrogen energy equipment.

6. A dynamic collaborative twin system of heterogeneous energy systems, characterized by: The system comprises: A construction module for constructing a digital twin network system corresponding to heterogeneous energy devices; wherein the digital twin in the digital twin network system includes the heterogeneous energy devices, a data collector connected to the heterogeneous energy devices, a topology modeling module and a DRL engine connected to the data collector, the topology modeling module is connected to the DRL engine, a policy optimization center connected to the DRL engine, a global scheduling command center connected to the policy optimization center, and the global scheduling command center is connected to the heterogeneous energy devices; A configuration module is used to configure the preset state calculation formula corresponding to the digital twin of the heterogeneous energy device, so that the digital twin inputs the operating data corresponding to the heterogeneous energy device into the preset state calculation formula and dynamically updates the device state of each heterogeneous energy device; configure the upload component corresponding to the heterogeneous energy device, and transmit the operating data to the data collector through the upload component; construct a topological relationship diagram between heterogeneous energy devices in the topology modeling module; wherein, the nodes in the topological relationship diagram represent digital twins, and the edges between nodes represent the interaction weights of the data between nodes; obtain the operating data uploaded by the data collector; based on the interaction weight, determine the allocation value of the operating data of the current node to the node corresponding to the interaction weight; build in several local objective functions that call the allocation value in the DRL engine, and obtain local objective results by processing the local objective functions in parallel through the DRL engine; configure a global objective function that calls the local objective result in the strategy optimization center to obtain the final output data; determine the output scheduling instruction based on the preset relationship between the final output data and the scheduling instruction, and send the scheduling instruction to the heterogeneous energy device through the global scheduling instruction center.

7. The heterogeneous energy system dynamic collaborative twin system according to claim 6 is characterized in that: The system also includes a new module, When there are newly added heterogeneous energy devices, the digital twins of the heterogeneous energy devices are added to the digital twin network system and connected to the data collector; Update the connection relationship and interaction weight between the newly added heterogeneous energy device and other heterogeneous energy devices in the topology relationship diagram of the topology modeling module; Add a local objective function corresponding to the allocation value of the newly added heterogeneous energy devices in the DRL engine, and add a global objective function corresponding to the local objective results of the newly added heterogeneous energy devices in the strategy optimization center; In the global dispatching command center, the dispatching instructions corresponding to the final output data of the newly added heterogeneous energy equipment and the mapping relationship between the specific final output data range and the dispatching instructions are configured.

8. The heterogeneous energy system dynamic collaborative twin system according to claim 6 is characterized in that: The system further includes an update module, Used to detect whether the operating data uploaded by heterogeneous energy equipment is abnormal data; When the data of any heterogeneous energy device is abnormal, the abnormal interaction weight distribution value corresponding to the data abnormal heterogeneous energy device is obtained, and the interaction weight in the topology modeling module is updated to the abnormal interaction weight distribution value.

9. A dynamic collaborative twin device for heterogeneous energy systems, characterized in that: The device comprises: processor; and a memory having executable code stored thereon, which, when the executable code is executed, enables the processor to execute a dynamic collaborative twinning method for heterogeneous energy systems as described in any one of claims 1-5.

10. A non-volatile computer storage medium, characterized in that Computer instructions are stored thereon, and when the computer instructions are executed, they implement a dynamic collaborative twinning method for heterogeneous energy systems as described in any one of claims 1 to 5.

Images