A method, apparatus, equipment and storage medium for simulating energy consumption of building complexes.

Abstract

This invention belongs to the field of energy and discloses a method, apparatus, equipment, and storage medium for simulating the energy consumption of building complexes. The method includes: acquiring spatial topology data of the target building complex and generating a topology map based on the spatial topology data; establishing an energy consumption simulation network based on the topology map; responding to a user-issued module cultivation task, triggering the access of a new module to the energy consumption simulation network, and initiating the cultivation of the new module based on the basic cultivation information and task type included in the module cultivation task; monitoring the data flow of the new module and the associated computing modules, and performing energy regulation based on the monitoring results and dynamic control strategies until the cultivation result of the new module reaches the preset expectation. By establishing a modularly designed energy consumption simulation network, it is possible to flexibly configure the network according to the actual spatial distribution and functional requirements of the building complex, improving simulation efficiency while solving the computational resource bottleneck problem and the difficulty of hardware localization in large-scale building complex energy consumption simulation.

Description

Technical Field

[0001] This invention belongs to the field of energy, and in particular relates to a method, apparatus, equipment and storage medium for simulating the energy consumption of building complexes. Background Technology

[0002] As a cluster of buildings within a defined physical boundary, the energy consumption characteristics of a building complex have a significant impact on the stable operation of the regional power grid system. This is especially true during the transition to a new power system, where energy consumption control at the building complex level is more meaningful for assisting grid peak shaving than controlling isolated individual buildings. Therefore, accurate energy consumption simulation of building complexes is fundamental to guiding control decisions and energy optimization schemes.

[0003] In related technologies, energy consumption assessments of large-scale building clusters often face challenges such as insufficient computing resources and reduced computational efficiency, making it difficult to complete large-scale, multivariate, and high-precision simulation analyses within a reasonable timeframe, especially when energy consumption feedback is involved after energy system regulation. Regarding hardware resources, distributed computing technology provides a feasible solution for efficient simulation. Distributed computing can break down and distribute computational tasks across multiple computing nodes, thereby improving computational speed and addressing the problem of insufficient computing resources. On the software side, data-driven machine learning algorithms can extract energy consumption characteristics of individual buildings and building clusters from building energy consumption data, significantly reducing the computational resource requirements for building cluster energy consumption simulations. However, the generalizability and transferability of the models in practical engineering are the main issues restricting their application. Summary of the Invention

[0004] In view of this, the present invention discloses a method, apparatus, equipment and storage medium for simulating the energy consumption of building complexes, which can solve the shortcomings of related technologies.

[0005] To achieve the above objectives, the present invention discloses the following technical solution:

[0006] According to a first aspect of the present invention, a method for simulating the energy consumption of a building complex is proposed, comprising:

[0007] Acquire spatial topology data of the target building complex, and generate a topology map containing building locations, attributes, and connection relationships based on the spatial topology data;

[0008] An energy consumption simulation network is established based on the topology diagram. The energy consumption simulation network includes multiple computing modules. Each computing module corresponds to a single building node in the topology diagram. The computing modules are connected based on a four-way standardized data interface. Each computing module is equipped with a building thermal environment calculation model that considers the shading effect and an energy load prediction model for outputting the electricity load curve. The energy consumption simulation network performs energy regulation based on a dynamic regulation strategy of reinforcement learning.

[0009] In response to a module cultivation task issued by the user, the energy consumption simulation network is triggered to access a new module, and the cultivation of the new module is started according to the basic cultivation information and task type contained in the module cultivation task.

[0010] Monitor the data flow of the new module and the computing modules associated with the new module, and perform energy regulation based on the monitoring results and the dynamic regulation strategy until the cultivation result of the new module reaches the preset expectation.

[0011] Optionally, triggering the access of the new module to the energy consumption simulation network and initiating the cultivation of the new module according to the basic cultivation information and task type included in the module cultivation task includes:

[0012] The location information, attribute information, and neighboring computing modules of the new module are determined, and the new module is connected to the energy consumption simulation network based on the determined information;

[0013] The target thresholds for cultivation include: setting the energy consumption prediction error rate to less than 5%, increasing the energy flexibility of the regional energy system by 20%, and improving the implementation accuracy of dynamic control strategies to over 90%.

[0014] Optionally, the dynamic control strategy of reinforcement learning includes a cluster-level strategy, a single-entity strategy, and a real-time update mechanism; wherein, the cluster-level strategy allocates the total energy consumption limit of the building group based on a multi-agent deep deterministic policy gradient algorithm, the single-entity strategy applies the Q-learning algorithm to optimize the air conditioning setpoint temperature, and the real-time update mechanism updates the strategy parameters through transfer learning every preset time interval.

[0015] Optionally, the execution of the cluster-level policy includes:

[0016] The energy allocation between new modules and associated modules is coordinated through a multi-agent deep deterministic policy gradient algorithm.

[0017] By combining energy allocation strategies, the energy system of the building complex is dynamically regulated through nonlinear optimization algorithms;

[0018] Compare the output of the new module with the preset standard deviation; if the standard deviation is not met, incremental training is triggered.

[0019] Optionally, the building thermal environment calculation model may consider shading effects in the following ways:

[0020] Calculate the solar shading coefficient and heat island effect weight based on the building height and spacing data sent by the adjacent calculation module;

[0021] The indoor temperature is solved using a building thermal process model, and the temperature change results are corrected according to weights.

[0022] Optionally, the method further includes:

[0023] Once the new module has been developed, it will be migrated to the actual building automation system.

[0024] Optionally, the four-way standardized data interface includes: a power interface, a data communication interface, a clock synchronization interface, and an analog signal interface; wherein, the power interface supports 24V DC power supply and PoE dual-mode input, the data communication interface adopts RS-485 / CAN bus dual protocol stack, the clock synchronization interface achieves microsecond-level synchronization through the IEEE 1588 protocol, and the analog signal interface collects cultivation task information data and module information data.

[0025] According to a second aspect of the present invention, a building complex energy consumption simulation device is provided, the device comprising:

[0026] Acquisition Unit: Acquires spatial topology data of the target building complex and generates a topology map containing building locations, attributes, and connection relationships based on the spatial topology data;

[0027] Construction Unit: An energy consumption simulation network is established based on the topology diagram. The energy consumption simulation network includes multiple computing modules. Each computing module corresponds to a single building node in the topology diagram. The computing modules are connected based on a four-way standardized data interface. Each computing module is equipped with a building thermal environment calculation model that considers the shading effect and an energy load prediction model for outputting the electricity load curve. The energy consumption simulation network performs energy regulation based on a dynamic regulation strategy of reinforcement learning.

[0028] Triggering unit: In response to the module cultivation task issued by the user, triggers the access of the new module to the energy consumption simulation network, and starts the cultivation of the new module according to the basic cultivation information and task type contained in the module cultivation task;

[0029] Control unit: Monitors the data flow of the new module and the computing modules associated with the new module, and performs energy regulation based on the monitoring results and the dynamic control strategy until the cultivation result of the new module reaches the preset expectation.

[0030] According to a third aspect of the present invention, an electronic device is provided, comprising:

[0031] processor;

[0032] Memory used to store processor-executable instructions;

[0033] The processor implements the steps of the method as described in the first aspect by running the executable instructions.

[0034] According to a fourth aspect of the invention, a computer-readable storage medium is provided having computer instructions stored thereon that, when executed by a processor, implement the steps of the method as described in the first aspect.

[0035] As can be seen from the above technical solutions, the building cluster energy consumption simulation method disclosed in this invention achieves the following technical effects: On the one hand, an energy consumption simulation network is established. Through modular design, it can be flexibly configured according to the actual spatial distribution and functional requirements of the building cluster. Each module represents a single building and completes functional verification and development in a semi-realistic physical environment. The developed modules can be directly transferred to actual engineering applications, providing localized load calculation and energy optimization and control methods to ensure consistency with actual applications. On the other hand, when calculating energy consumption, the network splits the calculation task into multiple calculation modules, improving simulation efficiency through parallel and iterative calculations, and solving the bottleneck problem of computing resources in large-scale building cluster energy consumption simulation. In addition, unlike traditional simulation methods, this invention considers factors such as the shading effect and heat island effect between buildings, making the energy consumption simulation more accurate and better reflecting the actual energy consumption characteristics of the building cluster. Attached Figure Description

[0036] Figure 1 This is an exemplary embodiment of an architecture diagram of a building complex energy consumption simulation system.

[0037] Figure 2 This is a flowchart of an exemplary embodiment of a method for simulating the energy consumption of a building complex;

[0038] Figure 3 This is a schematic diagram of an energy consumption simulation network provided in an exemplary embodiment;

[0039] Figure 4 This is a schematic diagram of a three-level structure topology provided in an exemplary embodiment;

[0040] Figure 5 This is a schematic structural diagram of a device provided in an exemplary embodiment;

[0041] Figure 6 This is a block diagram of a building complex energy consumption simulation device provided in an exemplary embodiment. Detailed Implementation

[0042] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of one or more embodiments of the present invention as detailed in the appended claims.

[0043] It should be noted that the steps of the corresponding methods in other embodiments are not necessarily performed in the order shown and described in this invention. In some other embodiments, the methods may include more or fewer steps than those described in this invention. Furthermore, a single step described in this invention may be broken down into multiple steps in other embodiments; and multiple steps described in this invention may be combined into a single step in other embodiments.

[0044] To further illustrate the present invention, the following embodiments are provided:

[0045] As a cluster of buildings within a defined physical boundary, the energy consumption characteristics of a building complex have a significant impact on the stable operation of the regional power grid system. This is especially true during the transition to a new power system, where energy consumption control at the building complex level is more meaningful for assisting grid peak shaving than controlling isolated individual buildings. Therefore, accurate energy consumption simulation of building complexes is fundamental to guiding control decisions and energy optimization schemes.

[0046] In related technologies, energy consumption assessments of large-scale building clusters often face challenges such as insufficient computing resources and reduced computational efficiency, making it difficult to complete large-scale, multivariate, and high-precision simulation analyses within a reasonable timeframe, especially when energy consumption feedback is involved after energy system regulation. Regarding hardware resources, distributed computing technology provides a feasible solution for efficient simulation. Distributed computing can break down and distribute computational tasks across multiple computing nodes, thereby improving computational speed and addressing the problem of insufficient computing resources. On the software side, data-driven machine learning algorithms can extract energy consumption characteristics of individual buildings and building clusters from building energy consumption data, significantly reducing the computational resource requirements for building cluster energy consumption simulations. However, the generalizability and transferability of the models in practical engineering are the main issues restricting their application.

[0047] Therefore, there is an urgent need for a building cluster energy consumption simulation technology that combines the advantages of efficient computing algorithms and distributed computing hardware to meet the practical application needs of energy consumption planning and optimization for building clusters of different sizes. Compared to single-building simulation, a significant difference in building cluster energy consumption simulation is the interference between buildings, including shading and heat island effects. These inter-building connections can be iteratively calculated through the exchange of limited information. Therefore, a building cluster computing network can be constructed using distributed computing modules. While completing calculations for individual buildings, communication between modules can achieve clustered calculations of the overall energy consumption of the building cluster. This hardware physical platform composed of distributed computing modules can be regarded as a miniature sandbox of the building cluster, used for cultivating distributed computing modules. After individual modules are trained and validated as computing nodes in the physical platform, they can be directly migrated to energy consumption calculations and system control in actual projects. This hardware technology, which maps buildings to hardware modules and cultivates modules through a distributed computing network, fills a gap in existing technologies.

[0048] To address the shortcomings of related technologies, this invention proposes a method, apparatus, equipment, and storage medium for simulating energy consumption in building complexes.

[0049] Figure 1 This is an exemplary embodiment of an architecture diagram of a building complex energy consumption simulation system. (See diagram for example.) Figure 1 As shown, the system includes: a distributed computing module cluster, a server, and a PC (personal computer). The PC and the server are connected via an uplink local area network, while the server is connected to the module cluster via a downlink local area network.

[0050] Modules connected to the server are defined as access point modules. Users can issue tasks via PC. The system configuration, module information, and calculation tasks issued by the PC are all sent to the access point modules by the server, and then transmitted to other computing modules through the inter-module data channel. To ensure data transmission efficiency, multiple access point modules can be defined. The building cluster energy consumption simulation system incorporates spatial factors such as the physical layout, thermal environment, and shading effects of the building cluster into the simulation process. Each distributed computing module represents a single building, possesses independent computing capabilities, and can perform tasks such as energy efficiency calculations, thermal environment simulation, and energy system optimization. It also simulates the energy consumption impact between buildings and the interaction of energy systems through a semi-realistic physical environment. The physical information exchange between modules is achieved through data transmission channels. This interaction ensures a high degree of consistency between the system simulation and the actual building cluster environment. Cross-sectional monitoring can be performed at the data transmission channel of any module, and the system can interface with the actual building automation system to verify the simulation results in the real environment.

[0051] Figure 2 This is a flowchart illustrating a method for simulating the energy consumption of a building complex, as provided in an exemplary embodiment. The method may include the following steps:

[0052] Step 201: Obtain the spatial topology data of the target building complex, and generate a topology map containing building locations, attributes, and connection relationships based on the spatial topology data.

[0053] Before establishing a simulation network, it is necessary to perform spatial topology topology on the building complex. The simulation method involved in this invention performs real-world simulation of the building complex and generates a topology containing nodes, connecting lines and node partitions based on the acquired spatial topology data, which guides the arrangement of distributed computing modules and network design.

[0054] Spatial topology data can be acquired in two ways: First, by parsing vector data obtained through GIS technology. This data should include the building location information, attribute information, and connection relationships between buildings within a building complex. Analyzing building attribute information clarifies building type, height, area, age, and latitude and longitude (building outline coordinates) for subsequent initialization and definition of the calculation module. Location information clarifies the physical boundaries of the building complex, the positional relationships between buildings, and building distribution characteristics to guide module network design, thereby generating spatial topology of the building complex and visualizing it on a local PC connected to the hardware platform. Second, the above information can be entered into the local PC to complete the spatial topology and visualization of the building complex.

[0055] like Figure 3 As shown, the topology map generated in this embodiment has a three-level structure of boundary-partition-node. Nodes represent the smallest unit of a building complex, and are represented semi-physically using distributed computing modules. Partitions are additional partitioning identifiers added to building nodes within a region to address specific functional clusters or differences in business types within the building complex. For example, buildings in a certain region may have distributed photovoltaic power generation capabilities. The physical boundary represents the physical boundary and energy flow cross-section of the building complex. The building complex topology map guides the module networking. Each node deploys one module, and data cables are connected between modules via tie lines. Tie lines represent interference between buildings that affects energy consumption calculation and energy system optimization, including shading effects calculated from building coordinates and height, and shared energy systems. Partition and physical boundary information are written during the module initialization phase, thus completing the module networking process and collectively forming a semi-physical simulation hardware platform system capable of mapping actual building complexes.

[0056] Step 202: Establish an energy consumption simulation network based on the topology diagram. The energy consumption simulation network includes multiple computing modules. Each computing module corresponds to a single building node in the topology diagram. The computing modules are connected based on a four-way standardized data interface. Each computing module is equipped with a building thermal environment calculation model that considers the shading effect and an energy load prediction model for outputting the electricity load curve. The energy consumption simulation network performs energy regulation based on a dynamic regulation strategy of reinforcement learning.

[0057] In the network, modules represent individual buildings. The module network strictly follows the actual building distribution, with each module providing four data interfaces to connect to other modules corresponding to adjacent buildings. Based on the building spatial topology map in step 201, a corresponding number of modules are deployed for each node, and communication lines between modules are established according to connecting lines. For node partitioning, modules representing the two closest buildings between regions are selected and connected based on their positional relationships, enabling information exchange between regions. This ultimately forms a three-tiered energy consumption simulation network system with a cluster-region-node structure. Furthermore, energy stations involved at the building cluster scale are also integrated into the module network in the form of modules.

[0058] The advantage of distributed computing lies in its ability to decompose tasks and execute large-scale calculations through parallel computing. In building cluster energy consumption simulation, individual buildings are not calculated independently; rather, the shading effects between buildings must be considered. Simultaneously, in the optimization and control of the building cluster's energy system, the heating and cooling supply of multiple buildings needs to be optimized, and energy consumption feedback from these buildings needs to be obtained. Therefore, during the calculation process, modules need to interact regarding the information required for shading and energy optimization and control calculations, including the orientation and height information of adjacent buildings, and sharing energy consumption information of buildings with the same energy system, thereby completing iterative calculations. In this step, a name is defined for the module using a local PC connected to the hardware platform. Based on the building attributes and spatial information obtained in step 201, basic module information is defined, interaction information variables are set, and initial values ​​are assigned. Subsequently, commands are sent to the module to read the names, building attributes, spatial information, and initial values ​​of interaction variables from adjacent modules, and the corresponding information is returned to the PC, completing the module information interaction test.

[0059] After completing module networking and interaction testing, a building thermal environment calculation model, an energy consumption calculation model, and a control strategy library need to be deployed for each module. The building thermal environment calculation model can calculate the indoor environment based on the previously defined module basic information and the outdoor environment specified by the subsequent tasks. Based on the definition of the basic information of each module, an energy consumption calculation model corresponding to the building type is deployed. This model can output the building's heating and cooling loads and sub-item electricity loads according to specific tasks and input parameters. In addition, a control strategy library is deployed for each module, including operation strategies, cluster optimization strategies, and optimization control strategies (including rule-based control, heuristic optimization control, and reinforcement learning optimization control strategies) for lighting, air conditioning systems, energy stations, and integrated energy systems. The corresponding strategies in the control strategy library are invoked according to the module's task, and the corresponding parameters for the controlled object are output. The building thermal environment calculation model, energy consumption calculation model, and control strategy library enable positive data flow and feedback between models, forming an iterative calculation closed loop for each individual module.

[0060] Step 203: In response to the module cultivation task issued by the user, trigger the access of the new module to the energy consumption simulation network, and start the cultivation of the new module according to the basic cultivation information and task type contained in the module cultivation task.

[0061] This step involves the simulation network performing functional verification and training on the module's achievable calculations and controls according to a specific task. Modules that have undergone this step can be directly transferred to practical engineering applications. Standardized training information is written to the PC, including: the location of the building complex, building thermal environment control standards, and building energy system configuration (including energy station configuration, building HVAC configuration, photovoltaics, energy storage, etc., corresponding one-to-one with the actual configuration of the target building complex). A functional training plan is selected based on the module's actual application intentions, including: building energy consumption calculation (prediction), building energy consumption optimization configuration, and building energy system optimization control. Through the node network constructed by the platform, the module can complete functional training in a semi-realistic physical environment (considering building spatial distribution and system physical constraints).

[0062] Specifically, such as Figure 4As shown, the key links in the process include module and system configuration, communication configuration and data interaction, task definition and library call, module training and evaluation, and module migration and deployment. In the module and system configuration link, through the PC and via the server, the module firmware driver and data interface driver are downloaded and upgraded; according to the building complex space topology, the modules are networked, and the access point module is specified, and then the access point module is connected to the server; after the server setup is completed, the LAN IP is configured, and the basic application configuration file for the module is set, including the debugging query application, node name and type setting application, topology data call application, etc.; finally, in this link, the format of the module interaction data is defined, which determines the parsing and visualization methods of the output signal in the subsequent steps. After the basic configuration of the module is completed, it is necessary to complete the communication configuration and data interaction test between modules and between the module and the server; based on the basic application configuration file deployed in the previous link, the basic information of the module is defined, including the module identification information and building physical information, and the data receiving and sending functions applicable to the corresponding data format are written; furthermore, the MQTT message server supporting the data receiving and sending function is deployed to realize the receiving, sending and reading of data; correspondingly, the data interface software, data format translation and data visualization program are deployed; after the communication and data interaction configuration are completed, in the task definition and library call link, the cultivation plan of the module is specifically set. First, the calculation (cultivation) task is defined, including the basic cultivation information such as the area where the building complex is located, the building thermal environment control standard and the building energy system configuration, as well as the specific task forms of optional building thermal environment simulation, building energy consumption simulation and energy system optimization control; according to the specific task, the corresponding model is written and the variables required by the model are defined, the strategy library containing strategies such as energy system operation and cluster optimization is deployed, and at the same time, the development of custom Apps is supported. The custom Apps can be encapsulated and downloaded into the deployed strategy library to perform personalized module cultivation; finally, in this link, the call signal of the calculation (cultivation) task is defined for the subsequent link to evoke the corresponding model and strategy. The next link is for the cultivation and post-evaluation of the module. A start signal is sent from the PC side, and this signal is transmitted to the distributed computing module cluster via the server. The cluster performs iterative calculations according to the preset cultivation plan; monitors the data streams of key nodes (such as cultivation object nodes, energy system nodes) to evaluate the module training (cultivation) effect; analyzes and visualizes the cluster calculation results, evaluates the overall effect of the calculation (cultivation) task, and saves the optimal model and strategy for the module.

[0063] Step 204, monitor the data streams of the new module and the computing module associated with the new module, and perform energy regulation based on the monitoring results and the dynamic regulation strategy until the cultivation result of the new module reaches the preset expectation.

[0064] After functional development on the platform, each module can be directly deployed locally in individual buildings, performing corresponding calculations. The calculation results, while conforming to physical constraints, also consider interference from other buildings and energy demands. The modules support multiple communication protocols, including RS485, Modbus, Zigbee, and BACnet. During module migration, the building thermal environment calculation model and energy consumption calculation model can be directly replaced by actual data after integration into the building automation system, or the model can be adaptively corrected using actual data. The output data of the control strategy can be transmitted as control signals to specific controlled objects, directly taking over the daily management of these objects. Compared to the simple algorithms and complex localization of traditional control modules, the modules, through the semi-realistic physical environment development of the simulation network in this invention, possess calculation and decision-making basis consistent with actual engineering, and the modular data interface enables rapid localization deployment and strategy execution.

[0065] In this embodiment, on the one hand, an energy consumption simulation network is established. Through modular design, it can be flexibly configured according to the actual spatial distribution and functional requirements of the building complex. Each module represents a single building and completes functional verification and development in a semi-realistic physical environment. The developed modules can be directly transferred to actual engineering applications, providing localized load calculation and energy optimization and control methods to ensure consistency with actual applications. On the other hand, when calculating energy consumption, the network splits the calculation task into multiple calculation modules, improving simulation efficiency through parallel and iterative calculations, and solving the bottleneck problem of computing resources in large-scale building complex energy consumption simulation. In addition, unlike traditional simulation methods, this invention considers factors such as the shading effect and heat island effect between buildings, making the energy consumption simulation more accurate and better reflecting the actual energy consumption characteristics of the building complex.

[0066] In one embodiment, triggering the access of a new module to the energy consumption simulation network and initiating the cultivation of the new module based on the basic cultivation information and task type included in the module cultivation task includes: determining the location information, attribute information, and neighboring computing modules of the new module, and accessing the energy consumption simulation network based on the determined information; setting target thresholds for cultivation includes: setting the energy consumption prediction error rate to less than 5%, increasing the energy flexibility of the regional energy system by 20%, and increasing the implementation accuracy of dynamic control strategies to over 90%.

[0067] In one embodiment, the dynamic control strategy of reinforcement learning includes a cluster-level strategy, a single-unit-level strategy, and a real-time update mechanism; wherein, the cluster-level strategy allocates the total energy consumption limit of the building group based on a deep deterministic policy gradient algorithm, the single-unit-level strategy applies a Q-learning algorithm to optimize the air conditioning setpoint temperature, and the real-time update mechanism updates the strategy parameters through transfer learning every preset time interval.

[0068] Furthermore, the execution of the cluster-level strategy includes: coordinating the energy allocation between the new module and related modules through a multi-agent deep deterministic strategy gradient algorithm; dynamically regulating the energy system of the building cluster through a nonlinear optimization algorithm in conjunction with the energy allocation strategy; comparing the output of the new module with the preset expected standard deviation, and triggering incremental training if the standard deviation is not met.

[0069] In one embodiment, the building thermal environment calculation model considers the shading effect by: calculating the solar shading coefficient and heat island effect weight based on the building height and spacing data sent by adjacent calculation modules; using a building thermal process model to solve for the indoor temperature, and correcting the temperature change results according to the weights.

[0070] In one embodiment, the method further includes: migrating the new module to the actual building automation system after the new module has been developed.

[0071] In one embodiment, the four-way standardized data interface includes: a power interface, a data communication interface, a clock synchronization interface, and an analog signal interface; wherein, the power interface supports 24V DC power supply and PoE dual-mode input, the data communication interface adopts RS-485 / CAN bus dual protocol stack, the clock synchronization interface achieves microsecond-level synchronization through the IEEE 1588 protocol, and the analog signal interface collects cultivation task information data and module information data.

[0072] Figure 5 This is a schematic structural diagram of a device provided in an exemplary embodiment. Please refer to... Figure 5 At the hardware level, the device includes a processor 502, an internal bus 504, a network interface 506, memory 508, and non-volatile memory 510, and may also include other hardware required for its functions. One or more embodiments of the present invention can be implemented in software, for example, the processor 502 reads the corresponding computer program from the non-volatile memory 510 into memory 508 and then runs it. Of course, in addition to software implementation, one or more embodiments of the present invention do not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0073] Please refer to Figure 6 A building complex energy consumption simulation device can be applied to, for example... Figure 6 The device shown, in order to implement the technical solution of the present invention, may include:

[0074] The acquisition unit 601 is used to acquire the spatial topology data of the target building group and generate a topology map containing building locations, attributes and connection relationships based on the spatial topology data.

[0075] The construction unit 602 is used to establish an energy consumption simulation network based on the topology map. The energy consumption simulation network includes multiple computing modules, each computing module corresponds to a single building node in the topology map, and the computing modules are connected based on a four-way standardized data interface. Each computing module is equipped with a building thermal environment calculation model that considers the shading effect and an energy load prediction model for outputting the electricity load curve. The energy consumption simulation network performs energy regulation based on a dynamic regulation strategy of reinforcement learning.

[0076] Triggering unit 603 is used to respond to a module cultivation task issued by the user, trigger the access of a new module to the energy consumption simulation network, and start the cultivation of the new module according to the basic cultivation information and task type contained in the module cultivation task;

[0077] The control unit 604 is used to monitor the data flow of the new module and the computing module associated with the new module, and to perform energy control based on the monitoring results and the dynamic control strategy until the cultivation result of the new module reaches the preset expectation.

[0078] Optionally, the triggering unit 603 is specifically used for:

[0079] The location information, attribute information, and neighboring computing modules of the new module are determined, and the new module is connected to the energy consumption simulation network based on the determined information;

[0080] The target thresholds for cultivation include: setting the energy consumption prediction error rate to less than 5%, increasing the energy flexibility of the regional energy system by 20%, and improving the implementation accuracy of dynamic control strategies to over 90%.

[0081] Optionally, the dynamic control strategy of reinforcement learning includes a cluster-level strategy, a single-unit strategy, and a real-time update mechanism; wherein, the cluster-level strategy allocates the total energy consumption limit of the building group based on a deep deterministic policy gradient algorithm, the single-unit strategy applies the Q-learning algorithm to optimize the air conditioning setpoint temperature, and the real-time update mechanism updates the strategy parameters through transfer learning every preset time interval.

[0082] Optionally, the execution of the cluster-level policy includes:

[0083] The energy allocation between new modules and associated modules is coordinated through a multi-agent deep deterministic policy gradient algorithm.

[0084] By combining energy allocation strategies, the energy system of the building complex is dynamically regulated through nonlinear optimization algorithms;

[0085] Compare the output of the new module with the preset standard deviation; if the standard deviation is not met, incremental training is triggered.

[0086] Optionally, the method by which the building thermal environment calculation model corrects the temperature field includes:

[0087] Calculate the solar shading coefficient and heat island effect weight based on the building height and spacing data sent by the adjacent calculation module;

[0088] The indoor temperature is solved using a building thermal process model, and the temperature change results are corrected according to weights.

[0089] Optionally, the device further includes:

[0090] The migration unit 605 is used to migrate the new module to the actual building automation system after the new module has been cultivated.

[0091] Optionally, the four-way standardized data interface includes: a power interface, a data communication interface, a clock synchronization interface, and an analog signal interface; wherein, the power interface supports 24V DC power supply and PoE dual-mode input, the data communication interface adopts RS-485 / CAN bus dual protocol stack, the clock synchronization interface achieves microsecond-level synchronization through the IEEE 1588 protocol, and the analog signal interface collects cultivation task information data and module information data.

[0092] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, which can take the form of a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email sending and receiving device, game console, tablet computer, wearable device, or any combination of these devices.

[0093] In a typical configuration, a computer includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0094] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0095] Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0096] For any other form of computer-readable medium (or computer-readable storage medium) as described above, computer instructions may be stored thereon, which, when executed by a processor, implement one or more of the above embodiments, thereby realizing the technical solution of the present invention.

[0097] The present invention also proposes a computer program that, when executed by a processor, implements one or more of the embodiments described above, thereby realizing the technical solution of the present invention. This computer program may be specifically recorded on the above-described or other computer-readable media, and the present invention does not impose any limitations on this.

[0098] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0099] The foregoing has described specific embodiments of the invention. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0100] The terminology used in one or more embodiments of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” used in one or more embodiments of the invention and in the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0101] It should be understood that although the terms first, second, third, etc., may be used to describe various information in one or more embodiments of the present invention, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of one or more embodiments of the present invention, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0102] The above description is merely a preferred embodiment of one or more embodiments of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of the present invention should be included within the protection scope of one or more embodiments of the present invention.

Claims

1. A method for simulating the energy consumption of a building complex, characterized in that, The method includes: Acquire spatial topology data of the target building complex, and generate a topology map containing building locations, attributes, and connection relationships based on the spatial topology data; An energy consumption simulation network is established based on the topology diagram. The energy consumption simulation network includes multiple computing modules. Each computing module corresponds to a single building node in the topology diagram. The computing modules are connected based on a four-way standardized data interface. Each computing module is equipped with a building thermal environment calculation model that considers the shading effect and an energy load prediction model for outputting the electricity load curve. The energy consumption simulation network performs energy regulation based on a dynamic regulation strategy of reinforcement learning. In response to a module cultivation task issued by the user, the energy consumption simulation network is triggered to access a new module, and the cultivation of the new module is started according to the basic cultivation information and task type contained in the module cultivation task. Monitor the data flow of the new module and the computing modules associated with the new module, and perform energy regulation based on the monitoring results and the dynamic regulation strategy until the cultivation result of the new module reaches the preset expectation.

2. The method according to claim 1, characterized in that, The triggering of the new module access to the energy consumption simulation network, and the initiation of the cultivation of the new module according to the basic cultivation information and task type included in the module cultivation task, includes: The location information, attribute information, and neighboring computing modules of the new module are determined, and the new module is connected to the energy consumption simulation network based on the determined information; The target thresholds for cultivation include: setting the energy consumption prediction error rate to less than 5%, increasing the energy flexibility of the regional energy system by 20%, and improving the implementation accuracy of dynamic control strategies to over 90%.

3. The method according to claim 1, characterized in that, The dynamic control strategy of reinforcement learning includes a cluster-level strategy, a single-entity strategy, and a real-time update mechanism. The cluster-level strategy allocates the total energy consumption limit of the building group based on a multi-agent deep deterministic policy gradient algorithm. The single-entity strategy applies the Q-learning algorithm to optimize the air conditioning setpoint temperature. The real-time update mechanism updates the strategy parameters through transfer learning every preset time interval.

4. The method according to claim 3, characterized in that, The execution of the cluster-level policy includes: The energy allocation between new modules and associated modules is coordinated through a multi-agent deep deterministic policy gradient algorithm. By combining energy allocation strategies, the energy system of the building complex is dynamically regulated through nonlinear optimization algorithms; Compare the output of the new module with the preset standard deviation; if the standard deviation is not met, incremental training is triggered.

5. The method according to claim 1, characterized in that, The building thermal environment calculation model considers the shading effect in the following ways: Calculate the solar shading coefficient and heat island effect weight based on the building height and spacing data sent by the adjacent calculation module; The indoor temperature is solved using a building thermal process model, and the temperature change results are corrected according to weights.

6. The method according to claim 1, characterized in that, The method further includes: Once the new module has been developed, it will be migrated to the actual building automation system.

7. The method according to claim 1, characterized in that, The four-way standardized data interface includes: a power interface, a data communication interface, a clock synchronization interface, and an analog signal interface. The power interface supports 24V DC power supply and PoE dual-mode input. The data communication interface adopts an RS-485 / CAN bus dual protocol stack. The clock synchronization interface achieves microsecond-level synchronization through the IEEE 1588 protocol. The analog signal interface collects cultivation task information data and module information data.

8. A building complex energy consumption simulation device, characterized in that, The device includes: Acquisition Unit: Acquires spatial topology data of the target building complex and generates a topology map containing building locations, attributes, and connection relationships based on the spatial topology data; Construction Unit: An energy consumption simulation network is established based on the topology diagram. The energy consumption simulation network includes multiple computing modules. Each computing module corresponds to a single building node in the topology diagram. The computing modules are connected based on a four-way standardized data interface. Each computing module is equipped with a building thermal environment calculation model that considers the shading effect and an energy load prediction model for outputting the electricity load curve. The energy consumption simulation network performs energy regulation based on a dynamic regulation strategy of reinforcement learning. Triggering unit: In response to the module cultivation task issued by the user, triggers the access of the new module to the energy consumption simulation network, and starts the cultivation of the new module according to the basic cultivation information and task type contained in the module cultivation task; Control unit: Monitors the data flow of the new module and the computing modules associated with the new module, and performs energy regulation based on the monitoring results and the dynamic control strategy until the cultivation result of the new module reaches the preset expectation.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor implements the steps of the method as described in any one of claims 1-7 by running the executable instructions.

10. A computer-readable storage medium storing computer instructions thereon, characterized in that, When executed by the processor, this instruction implements the steps of the method as described in any one of claims 1-7.

Images