Method and device for scheduling based on building source network load storage adjustable resource aggregation

Abstract

The application relates to the technical field of energy management, in particular to a scheduling method and device based on building source network load storage adjustable resource aggregation, wherein the method comprises the following steps: acquiring aggregated electric parameters at a total incoming line and a key branch of a target building and comprehensive working condition parameters of the target building, analyzing a load interpretation result of a target device group based on the aggregated electric parameters and the comprehensive working condition parameters; correcting a pre-constructed digital twin model according to the load interpretation result, obtaining a corrected digital twin model, and calculating adjustable capacities of multiple types of resources in the target building; generating a source network load storage scheduling strategy based on preset constraint conditions and the adjustable capacities, and controlling the target building to execute the source network load storage scheduling strategy. The application can reduce the dependence on a large number of sub-metering and device protocol opening, improve the field consistency of the building energy digital twin model, and improve the engineering application ability of non-intrusive detection results from energy consumption identification to adjustable resource aggregation and collaborative scheduling.

Description

Technical Field

[0001] This application relates to the field of energy management technology, and in particular to a scheduling method and device based on the aggregation of adjustable resources of building source, grid, load and storage. Background Technology

[0002] As the energy consumption of urban commercial buildings, public buildings, and park-type buildings continues to expand, building energy systems have gradually evolved from single power loads into complex source-grid-load-storage systems composed of multiple devices and systems. Meanwhile, commercial buildings, office buildings, complexes, and park buildings are gradually transforming from simple power consumers into flexible resources with certain regulation capabilities.

[0003] Among related technologies, intelligent building energy source-grid-load-storage collaborative management includes three mainstream solutions: The first is a non-intrusive load identification system, which typically identifies the operating status and power curves of several target devices in a building by collecting aggregated signals such as total power, voltage, and current; the second is a building energy management or energy source-grid-load-storage optimization platform, which typically generates control strategies using sensor data, predictive models, and optimization algorithms; and the third is a building digital twin modeling and simulation system, which reproduces the operating status of the building energy system in digital space by establishing building models and visualization platforms.

[0004] However, in related technologies, the output of non-intrusive load identification systems usually remains at the level of equipment identification or sub-item energy consumption, lacking the ability to interpret the flexible capabilities for source-grid-load-storage scheduling; building energy management or source-grid-load-storage optimization platforms usually rely on initial models and on-site sensor data for state mapping, and after equipment replacement, sensor loss, or operational changes, the model is prone to deviating from the actual situation on site; the model establishment and visualization platform establishment of building digital twin modeling and simulation systems are mostly aimed at single types of loads or single equipment groups, with poor representativeness.

[0005] In summary, the building source-network-load-storage collaborative management schemes in related technologies are unable to fully identify the characteristics of the load, resulting in a lack of direct basis for management decisions. Optimization often relies on complete metering and ledgers, and when these are missing, only rough processing can be done. Rigid and flexible loads cannot be distinguished, and the models are prone to inaccuracy as the operation changes, and lack dynamic correction. A single indicator cannot reflect the true capacity of the building's overall adjustable resources, which urgently needs to be addressed. Summary of the Invention

[0006] This application provides a scheduling method and device based on the aggregation of adjustable resources of building source network load storage, in order to solve the problems in the related technology that the building source network load storage collaborative management scheme is difficult to fully identify the characteristics of the load, resulting in a lack of direct basis for management decisions. Optimization often relies on complete metering and ledgers, and when these are missing, only rough processing can be carried out. Rigid and flexible loads cannot be distinguished, and the model is prone to inaccuracy with changes in operation and lacks dynamic correction. A single indicator cannot reflect the true capacity of the building's overall adjustable resources.

[0007] The first aspect of this application provides a scheduling method based on the aggregation of adjustable resources in a building's source-grid-load-storage system, comprising the following steps: obtaining the aggregated electrical parameters at the total incoming lines and branch lines meeting preset important conditions of a target building, as well as the comprehensive operating condition parameters of the target building; analyzing the load interpretation results of the target equipment group in the target building based on the aggregated electrical parameters and the comprehensive operating condition parameters; correcting the pre-constructed digital twin model corresponding to the target building according to the load interpretation results to obtain the corrected digital twin model; calculating the adjustability of multiple types of resources in the target building based on the corrected digital twin model; generating a source-grid-load-storage scheduling strategy for the target building based on preset constraints and the adjustability, and controlling the target building to execute the source-grid-load-storage scheduling strategy.

[0008] Optionally, in one embodiment of this application, obtaining the aggregated electrical parameters at the main incoming line and branch lines that meet preset important conditions of the target building, as well as the comprehensive operating parameters of the target building, includes: obtaining the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar corresponding to the target building; and determining the comprehensive operating parameters of the target building based on the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar corresponding to the target building.

[0009] Optionally, in one embodiment of this application, the step of analyzing the load interpretation results of the target equipment group in the target building based on the aggregated electrical parameters and the comprehensive operating condition parameters includes: analyzing the power contribution and operating status of the target equipment group, as well as the equipment flexibility label and identification confidence level of the target equipment group, based on the aggregated electrical parameters and the comprehensive operating condition parameters; and determining the load interpretation results of the target equipment group based on the power contribution and operating status of the target equipment group, as well as the equipment flexibility label and identification confidence level of the target equipment group.

[0010] Optionally, in one embodiment of this application, the step of correcting the pre-built digital twin model corresponding to the target building based on the load interpretation result to obtain the corrected digital twin model includes: comparing the pre-built digital twin model and the load interpretation result to obtain a comparison result; when the comparison result indicates that there is a continuous deviation between the power of the equipment group in the pre-built digital twin model and the load interpretation result, and the identification confidence of the target equipment group reaches a preset confidence threshold, the model parameters of the pre-built digital twin model are corrected based on the confidence of the target equipment group to complete the correction of the pre-built digital twin model and obtain the corrected digital twin model.

[0011] Optionally, in one embodiment of this application, calculating the adjustability of multiple resources in the target building based on the modified digital twin model includes: calculating the power reduction and duration of the air conditioning load in the target building based on the modified digital twin model and the indoor temperature, comfort zone, thermal inertia, and equipment power of the target building; calculating the power reduction of the lighting load in the target building based on the modified digital twin model and the lower limit of illumination and lighting area occupancy status of the target building; calculating the delay capability of the charging load in the target building based on the modified digital twin model and the expected departure time and charging demand of the current charging equipment in the target building; calculating the charge and discharge capability of the energy storage device based on the modified digital twin model and the state of charge, upper power limit, and lifetime constraint of the energy storage device in the target building; and determining the adjustability of multiple resources in the target building based on the power reduction and duration of the air conditioning load, the power reduction of the lighting load, the delay capability of the charging load, and the charge and discharge capability of the energy storage device.

[0012] Optionally, in one embodiment of this application, after controlling the target building to execute the source-grid-load-storage scheduling strategy, the method further includes: obtaining changes in grid-connected power, equipment status, and comfort level from the execution result of the source-grid-load-storage scheduling strategy; and adjusting the scheduling order and call priority of the multiple types of resources based on the changes in grid-connected power, equipment status, and comfort level.

[0013] A second aspect of this application provides a scheduling device based on the aggregation of adjustable resources in a building's source-grid-load-storage configuration, comprising: an analysis module for acquiring aggregated electrical parameters at the total incoming lines and branch lines meeting preset important conditions of a target building, as well as comprehensive operating parameters of the target building; analyzing the load interpretation results of target equipment groups in the target building based on the aggregated electrical parameters and the comprehensive operating parameters; a correction module for correcting a pre-constructed digital twin model corresponding to the target building according to the load interpretation results, obtaining a corrected digital twin model, and calculating the adjustability of multiple types of resources in the target building based on the corrected digital twin model; and a scheduling module for generating a source-grid-load-storage scheduling strategy for the target building based on preset constraints and the adjustability, and controlling the target building to execute the source-grid-load-storage scheduling strategy.

[0014] Optionally, in one embodiment of this application, the analysis module includes: an acquisition unit, configured to acquire the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar corresponding to the target building; and a first determination unit, configured to determine the comprehensive operating parameters of the target building based on the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar corresponding to the target building.

[0015] Optionally, in one embodiment of this application, the analysis module includes: an analysis unit, configured to analyze the power contribution and operating status of the target equipment group, as well as the equipment flexibility label and identification confidence level of the target equipment group, based on the aggregated electrical parameters and the comprehensive operating condition parameters; and a second determination unit, configured to determine the load interpretation result of the target equipment group based on the power contribution and operating status of the target equipment group, as well as the equipment flexibility label and identification confidence level of the target equipment group.

[0016] Optionally, in one embodiment of this application, the correction module includes: a comparison unit, configured to compare the pre-built digital twin model and the load interpretation result to obtain a comparison result; and a correction unit, configured to, when the comparison result indicates that there is a continuous deviation between the power of the device group and the load interpretation result in the pre-built digital twin model, and the identification confidence of the target device group reaches a preset confidence threshold, correct the model parameters of the pre-built digital twin model according to the confidence of the target device group, so as to complete the correction of the pre-built digital twin model and obtain the corrected digital twin model.

[0017] Optionally, in one embodiment of this application, the correction module includes: a first calculation unit, configured to calculate the reduceable power and duration of the air conditioning load in the target building based on the corrected digital twin model and the indoor temperature, comfort zone, thermal inertia, and equipment power of the target building; a second calculation unit, configured to calculate the reduceable power of the lighting load in the target building based on the corrected digital twin model and the lower limit of illumination and lighting area occupancy status of the target building; a third calculation unit, configured to calculate the delay capability of the charging load in the target building based on the corrected digital twin model and the expected departure time and charging demand of the current charging equipment in the target building; a fourth calculation unit, configured to calculate the chargeable / discharge capability of the energy storage device based on the corrected digital twin model and the state of charge, upper power limit, and lifetime constraint of the energy storage device in the target building; and a fifth calculation unit, configured to determine the adjustability of multiple resources in the target building based on the reduceable power and duration of the air conditioning load, the reduceable power of the lighting load, the delay capability of the charging load, and the chargeable / discharge capability of the energy storage device.

[0018] Optionally, in one embodiment of this application, it further includes: an acquisition module, configured to acquire the grid-connected power, equipment status, and comfort changes in the execution result of the source-grid-load-storage scheduling strategy after controlling the target building to execute the strategy; and an adjustment module, configured to adjust the scheduling order and call priority of the multiple types of resources based on the grid-connected power, equipment status, and comfort changes.

[0019] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the program to implement the scheduling method based on adjustable resource aggregation of building source network load storage as described in the above embodiments.

[0020] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the scheduling method based on the aggregation of adjustable resources from building sources, networks, loads, and storage as described above.

[0021] A fifth aspect of this application provides a computer program product, including a computer program that, when executed, is used to implement the scheduling method described above based on the aggregation of adjustable resources from building sources, networks, loads, and storage.

[0022] This application's embodiments can complete equipment-level load status identification and flexibility attribute interpretation through a small number of aggregated measurement points. The identification results are used to dynamically correct the building's digital twin model, and the comprehensive adjustability of load, photovoltaics, and energy storage is calculated under the constraints of the twin model. This generates scheduling recommendations for electricity price optimization, demand response, peak shaving and valley filling, photovoltaic integration, and low-carbon operation. Furthermore, this application can reduce the cost of retrofitting existing buildings, decrease reliance on numerous sub-metering items and open equipment protocols, enable the twin model to continuously self-correct during field operation, improve the field consistency of the building energy digital twin model, and elevate non-intrusive detection results from "energy consumption identification" to engineering application capabilities of "adjustable resource aggregation and collaborative scheduling," thereby improving the reliability, interpretability, and engineering implementation capability of scheduling strategies. This solves the problems in related technologies, such as the difficulty in fully identifying the characteristics of loads in building source-network-load-storage collaborative management schemes, which leads to a lack of direct basis for management decisions, optimization often relies on complete measurement and ledgers, and can only be handled in a rough manner when they are missing, rigid and flexible loads cannot be distinguished, and the model is prone to inaccuracy with changes in operation and lacks dynamic correction. A single indicator cannot reflect the true capacity of the building's overall adjustable resources.

[0023] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0024] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart illustrating a scheduling method based on the aggregation of adjustable resources from building sources, networks, loads, and storage, according to an embodiment of this application. Figure 2This is a schematic diagram of the structure of a building source-network-load-storage adjustable resource scheduling system based on non-intrusive load flexibility interpretation and digital twin dynamic correction according to an embodiment of this application; Figure 3 This is a schematic diagram illustrating a non-invasive load flexibility interpretation according to an embodiment of this application; Figure 4 This is a schematic diagram illustrating the dynamic correction of a digital twin model according to an embodiment of this application; Figure 5 This is a schematic diagram illustrating the generation of adjustable resource aggregation and response packets according to an embodiment of this application; Figure 6 This is a flowchart illustrating the source-grid-load-storage coordinated scheduling of one embodiment of this application; Figure 7 A flowchart illustrating a building source-network-load-storage adjustable resource aggregation and scheduling method based on non-intrusive load flexibility interpretation and digital twin dynamic correction, according to an embodiment of this application; Figure 8 This is a schematic diagram of the structure of a scheduling device based on adjustable resource aggregation of building source network load storage according to an embodiment of this application; Figure 9 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application.

[0025] Figure label: 10-Scheduling device based on adjustable resource aggregation of building source network load storage; 100-Analysis module, 200-Correction module and 300-Scheduling module; 901-Memory, 902-Processor and 903-Communication interface. Detailed Implementation

[0026] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0027] The following describes a scheduling method and apparatus based on the aggregation of adjustable resources (source, grid, load, and storage) in a building, with reference to the accompanying drawings. Addressing the problems mentioned in the background section, this application provides a scheduling method based on the aggregation of adjustable resources (source, grid, load, and storage) in a building. In this method, equipment-level load status identification and flexible attribute interpretation can be completed using a small number of aggregated measurement points. The identification results are used to dynamically correct the building's digital twin model, and the comprehensive adjustability of load, photovoltaic, and energy storage is calculated under the constraints of the twin model. This generates scheduling recommendations for electricity price optimization, demand response, peak shaving and valley filling, photovoltaic absorption, and low-carbon operation. Furthermore, this application can reduce the cost of retrofitting existing buildings, reduce reliance on numerous sub-metering items and open equipment protocols, enable the twin model to continuously self-correct during on-site operation, improve the on-site consistency of the building energy digital twin model, and elevate the non-intrusive detection results from "energy consumption identification" to the engineering application capability of "adjustable resource aggregation and collaborative scheduling," thereby improving the reliability, interpretability, and engineering implementation capability of the scheduling strategy. This solves the problems in related technologies, such as the difficulty in fully identifying the characteristics of loads in building source-network-load-storage collaborative management schemes, which leads to a lack of direct basis for management decisions, optimization often relies on complete measurement and ledgers, and can only be handled in a rough manner when they are missing, rigid and flexible loads cannot be distinguished, and the model is prone to inaccuracy with changes in operation and lacks dynamic correction. A single indicator cannot reflect the true capacity of the building's overall adjustable resources.

[0028] Before explaining the scheduling method based on the aggregation of adjustable resources of building source network load storage in the embodiments of this application, the relevant technical principles and other related contents involved in the embodiments of this application will be explained first.

[0029] As the energy consumption of urban commercial buildings, public buildings, and industrial parks continues to expand, building energy systems have gradually evolved from single-load electrical systems into complex source-grid-load-storage systems composed of distributed photovoltaics, heating and cooling equipment, air conditioning and ventilation systems, lighting systems, charging facilities, energy storage devices, and building automation systems. Under the requirements of "dual carbon" targets, time-of-use pricing, demand response, and green electricity consumption, buildings are no longer just passive electricity consumers, but are gradually becoming flexible resources that can participate in grid interaction and possess certain regulation capabilities and energy management value.

[0030] In actual buildings, the source side typically includes rooftop photovoltaics, facade photovoltaics, combined cooling, heating and power (CCHP) systems, or other distributed energy sources; the grid side includes grid connection points, transformers, distribution cabinets, branch metering, internal building power distribution networks, and power quality monitoring; the load side includes a large number of devices such as air conditioning units, chilled water pumps, cooling pumps, fan coil units, fresh air handling units, lighting, elevators, office sockets, data centers, and charging piles; and the storage side includes electrochemical energy storage, cold and heat storage, backup power supplies, or adjustable charging facilities. There are clear coupling relationships among these entities. For example, air conditioning load is related to outdoor temperature and humidity, population density, and comfort constraints; photovoltaic output is related to weather and building shading; energy storage charging and discharging are related to electricity prices, demand, and safety margins; and some loads can be temporarily shifted or reduced, but this can affect comfort or production activities.

[0031] To improve building energy management, building management systems in related technologies typically rely on sub-meters, building automation systems, energy management platforms, and manual statistical ledgers to obtain equipment-level or branch-level energy consumption information. However, existing buildings often suffer from complex branch divisions, frequent historical renovations, insufficient metering points, incomplete equipment ledgers, and inconsistent communication protocols. Re-deploying high-precision sub-metering devices for every type of equipment and every circuit would not only be costly and difficult to construct, but would also disrupt the operation of the existing power distribution system. Therefore, relying solely on traditional intrusive sub-metering methods makes it difficult to obtain the equipment-level power status and flexible resource information required for coordinated scheduling of building energy sources, grid, load, and storage in a low-cost and rapid manner.

[0032] Non-intrusive load sensing technology provides a new technological foundation for addressing the aforementioned issues. This type of technology collects aggregated electrical parameters such as voltage, current, active power, reactive power, power factor, and harmonic summaries at a few measurement points on the main incoming line, floor distribution cabinets, or key branches. Load identification algorithms are then used to infer equipment type, operating status, or energy consumption contribution. Compared to installing sensors on each device, non-intrusive sensing significantly reduces retrofit costs and is suitable for rapid deployment in existing buildings. However, traditional non-intrusive load identification primarily aims to "identify whether a device is operating" or "estimate the energy consumption of a device," and has not yet fully addressed the "adjustable resource capabilities" required for coordinated scheduling of power generation, grid, load, and storage.

[0033] Digital twin technology provides another foundation for building energy source-grid-load-storage systems. By constructing a building energy system model in digital space and continuously integrating on-site operational data, online simulation and prediction of equipment status, energy flow, comfort constraints, energy storage status, grid-connected power, and carbon emissions can be performed. However, building digital twin systems in related technologies often rely heavily on existing equipment ledgers, BMS / BAS interfaces, and sub-metering data. When the underlying measurement points are incomplete, equipment operating status is not visible, or historical ledgers are inconsistent with reality, the digital twin model is prone to the problem of "the model looks complete, but deviates from the actual situation."

[0034] Therefore, in the collaborative management of power generation, grid, load and storage in smart buildings, a new technical solution is urgently needed: on the one hand, non-intrusive detection is used to recover the equipment-level operating status and load composition from a small number of aggregated measurement points; on the other hand, the identification results are transformed into "virtual sensing information" that can be used to correct digital twin models, further forming adjustable resource profiles such as equipment-level flexibility, response boundaries, sustainability time and comfort risks, ultimately supporting collaborative decision-making in electricity price optimization, demand response, photovoltaic consumption, energy storage scheduling and low-carbon operation.

[0035] Among related technologies, building energy source-grid-load-storage collaborative management solutions mainly fall into three categories: the first is a non-intrusive load identification system; the second is a building energy management or energy source-grid-load-storage optimization platform; and the third is a building digital twin modeling and simulation system. These three technologies respectively address some issues related to "seeing the load," "optimizing energy," and "mapping the system," but have not yet formed a unified closed loop for the aggregation of adjustable building resources. Specifically: The first type is the non-intrusive load identification system. This typically involves installing a data acquisition terminal at the user's main power line or distribution box. By collecting aggregated signals such as total power, voltage, and current, it uses event detection, steady-state feature recognition, hidden Markov models, convolutional neural networks, or other deep learning models to identify the operating status and power curves of several target devices in a household or building. This approach has proven the feasibility of non-intrusive detection in low-cost monitoring, but its objectives are usually device identification or energy consumption estimation, and it has not yet directly formed a flexible resource boundary for building-grid-load-storage scheduling.

[0036] The second category is building energy management and source-grid-load-storage collaborative optimization platforms. These platforms typically integrate data from photovoltaic (PV), energy storage, air conditioning, lighting, charging piles, electricity meters, and environmental sensors. They utilize predictive models and optimization algorithms to generate strategies for energy storage charging and discharging, PV consumption, air conditioning regulation, peak shaving and valley filling, or demand control. Some systems can formulate operating plans based on electricity prices, load forecasts, and energy storage status, and can also analyze building carbon emissions and overall energy efficiency. However, these platforms heavily rely on sub-metering, equipment control interfaces, and ledger data. When buildings lack sufficient sensors or equipment protocols are not open, the platform struggles to accurately determine "which devices are operating, which loads are adjustable, the adjustment range, the duration, and whether adjustments will rebound."

[0037] The third category is building digital twin systems. These systems replicate the building's energy system operation in a digital space by establishing building physical models, equipment mechanism models, energy consumption prediction models, and visualization platforms. In publicly available information and industry solutions, digital twins are commonly used for park energy monitoring, equipment operation simulation, energy and carbon management, and integrated display of energy sources, grids, loads, and storage. However, in engineering applications, digital twin models often suffer from problems such as reliance on manual input of initial parameter values, dependence on existing control systems for equipment status, and reliance on individual metering data for model calibration. Once there are insufficient on-site sensors or changes in equipment operating modes, the state of the twin model will gradually deviate from actual operation.

[0038] These technical solutions, while possessing non-intrusive detection, energy optimization, and digital twin modeling capabilities respectively, lack a comprehensive approach that integrates all three. Specifically, they lack the following closed loop: inferring equipment-level load status from a small number of aggregated electrical parameters; converting equipment-level identification results into virtual measurement points in the digital twin model; calculating the building's adjustable resource capacity based on the twin model's correction results; and then applying this adjustable capacity to coordinated scheduling of power generation, grid, load, and storage, as well as demand response.

[0039] Specifically, while non-intrusive load identification technology can estimate the operating status and energy consumption of some equipment from the total power curve, its output usually remains at the level of equipment identification or individual energy consumption, lacking the flexibility to interpret power generation, grid, load, and storage scheduling. For example, the system may identify that air conditioners, water pumps, or lighting are running, but it cannot further determine whether their power can be reduced, their shift time, the impact on comfort after adjustment, the risk of rebound load, or their reliability in participating in demand response. Therefore, such results are difficult to directly utilize by building energy management platforms.

[0040] Building-based energy storage and distribution (EHDD) management platforms often rely on numerous sub-meters and equipment interfaces. If existing buildings lack branch metering or have incomplete equipment records, the platform can only perform coarse-grained optimization based on the total load curve, making it difficult to distinguish between rigid loads, flexible loads, transferable loads, and uncontrollable loads. This can lead to conservative scheduling strategies, inaccurate assessment of peak-shaving potential, underutilization of energy storage capacity, and even the issuance of unreasonable control recommendations to unadjustable equipment.

[0041] Digital twin building energy systems typically rely on an initial model and on-site sensor data for state mapping. However, after equipment replacement, sensor loss, communication interruption, metering errors, or changes in operating modes, the digital twin model can easily deviate from the actual situation on-site. If there is a lack of a mechanism to dynamically correct the twin model through non-intrusive detection results, the twin platform may continuously output distorted energy consumption predictions and scheduling suggestions.

[0042] Furthermore, demand response potential assessment methods often target single load types or single equipment groups, such as air conditioning and temperature control loads, and rarely consider the comprehensive relationship between photovoltaics, energy storage, adjustable loads, comfort constraints, electricity safety boundaries, and carbon emission factors within a building. For commercial buildings, the potential of a single load is difficult to represent the true adjustable resource capacity of the entire building, and it is also difficult to form a standardized response package that can be called upon by the power grid or the park's energy platform.

[0043] Based on this, this application designs a building source-grid-load-storage adjustable resource aggregation method based on non-intrusive load flexibility interpretation and digital twin dynamic correction. The core idea is to use a small number of aggregated measurement points to complete equipment-level load status identification and flexibility attribute interpretation. The identification results are used to dynamically correct the building digital twin model, and the comprehensive adjustable capacity of load, photovoltaic and energy storage is calculated under the constraints of the twin model. In this way, scheduling suggestions for electricity price optimization, demand response, peak shaving and valley filling, photovoltaic consumption and low-carbon operation are generated.

[0044] Furthermore, the embodiments of this application can further reduce the cost of retrofitting existing buildings, reduce reliance on a large number of sub-metering and equipment protocol openness, improve the on-site consistency of building energy digital twin models, and upgrade non-intrusive detection results from "energy consumption identification" to engineering application capabilities of "adjustable resource aggregation and collaborative scheduling".

[0045] Specifically, Figure 1 This is a flowchart illustrating a scheduling method based on the aggregation of adjustable resources from building sources, networks, loads, and storage, provided in an embodiment of this application.

[0046] like Figure 1 As shown, the scheduling method based on the aggregation of adjustable resources from building sources, networks, loads, and storage includes the following steps: In step S101, the aggregated electrical parameters of the main incoming line and the branch lines that meet the preset important conditions of the target building are obtained, as well as the comprehensive operating parameters of the target building. Based on the aggregated electrical parameters and the comprehensive operating parameters, the load interpretation results of the target equipment group in the target building are analyzed.

[0047] It is understood that the target building here can be understood as the building object to be scheduled for source-grid-load-storage, that is, the implementation object of the scheduling method based on the aggregation of adjustable resources of building source-grid-load-storage in this application embodiment. It includes, but is not limited to, commercial buildings, public buildings, office parks, hospitals, schools and complexes.

[0048] Preset critical conditions can be understood here as pre-defined standards or rules used to screen out key branches from the building's power distribution network. For example, in this application embodiment, certain screening rules can be pre-defined as preset critical conditions based on factors such as branch load capacity, load type adjustability, and contribution to total power consumption. Branches that meet the conditions are considered key branches.

[0049] In other words, this application embodiment can designate branches in the target building that meet certain important conditions as key branches of the target building, thereby installing acquisition units and / or prioritizing load monitoring on these branches. Moreover, this application embodiment only sets up a small number of measuring points on the main incoming line and key branches, and does not require the permanent installation of sensors on every device; for data involving user behavior, only statistical summaries, adjustable resource packages, and model error indicators can be uploaded, avoiding the upload of complete fine-grained power consumption curves, thus balancing energy management and privacy protection.

[0050] In the embodiments of this application, these key branches include, but are not limited to, building main incoming line, floor distribution cabinet, air conditioning power cabinet, lighting distribution cabinet, charging facility branch or energy storage grid connection point, etc.

[0051] It should be noted that the specific preset conditions can be set and adjusted by those skilled in the art according to actual needs and application scenarios. The embodiments in this application are only illustrative and do not impose specific limitations.

[0052] In actual implementation, this application can obtain the aggregated electrical parameters of the main incoming line and these key branches in the target building, as well as the comprehensive operating parameters of the target building, in order to analyze the load interpretation results of the target equipment group in the target building.

[0053] Here, the target equipment group refers to at least one type (one or more types) of main equipment groups in the target building that are currently subject to adjustable capacity assessment or scheduling control, including but not limited to air conditioning equipment groups, lighting equipment groups, charging facility equipment groups, and water pump equipment groups.

[0054] In addition, the main incoming line here refers to the main power supply inlet used by the building when it draws power from the external power grid, which generally corresponds to the incoming line terminal of the building's main distribution cabinet.

[0055] Aggregated electrical parameters can be understood here as a summary of various electrical parameters obtained from collection points on the main incoming line or key branches. Comprehensive operating parameters can be understood here as a multi-dimensional set of information on the current operating environment and scheduling conditions of the target building. Unlike aggregated electrical parameters, comprehensive operating parameters include, but are not limited to, information such as photovoltaic output, energy storage status of charge, indoor and outdoor environmental parameters, and building operation calendar.

[0056] Optionally, in one embodiment of this application, obtaining the aggregated electrical parameters at the main incoming line and branch lines that meet preset important conditions of the target building, as well as the comprehensive operating parameters of the target building, includes: obtaining the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar corresponding to the target building; and determining the comprehensive operating parameters of the target building based on the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar corresponding to the target building.

[0057] In some embodiments, the aggregated electrical parameters obtained by this application include, but are not limited to, data such as the active power, reactive power, voltage, current, power factor, electrical energy, and optional harmonic summaries of the target building.

[0058] Furthermore, in the embodiments of this application, the comprehensive operating parameters of the target building include, but are not limited to, the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar parameters corresponding to the target building.

[0059] For example, Figure 2 This is a schematic diagram of the structure of a building source-network-load-storage adjustable resource scheduling system based on non-intrusive load flexibility interpretation and digital twin dynamic correction, according to one embodiment of this application. Figure 2 As shown, the system includes, but is not limited to, an aggregated electrical parameter acquisition unit, an environmental and comfort acquisition unit, an edge load interpretation unit, a digital twin dynamic correction unit, a source-grid-load-storage resource modeling unit, an adjustable resource aggregation unit, a collaborative scheduling decision-making unit, an execution feedback unit, and a user interaction unit. These units can be deployed in the same building energy management server, or they can adopt a layered architecture of "field edge terminal + building energy management platform + cloud model service". The embodiments in this application are only illustrative and do not impose specific limitations.

[0060] The aggregated electrical parameter acquisition unit is installed at the building's main incoming line, floor distribution cabinets, air conditioning power cabinets, lighting distribution cabinets, charging facility branch lines, or energy storage grid connection points. It is used to collect data such as active power, reactive power, voltage, current, power factor, electrical energy, and optional harmonic summaries. This unit can employ smart meters, power quality acquisition devices, power quality terminals, current transformers, Rogowski coils, Hall effect sensors, or other non-intrusive sampling devices that meet electrical safety requirements (building automation systems already have measurement points). The embodiments described in this application are merely illustrative and do not impose specific limitations.

[0061] Therefore, this embodiment of the application can utilize the aggregated electrical parameter acquisition unit in the system to collect aggregated electrical parameters of the main incoming line and key branches in the building in real time, and simultaneously collect photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar. After the collected data enters the edge buffer, this embodiment of the application can further perform preprocessing operations such as clock synchronization, outlier removal, and baseline separation on the collected data.

[0062] Here, clock synchronization refers to the unified calibration of the local clocks of the acquisition units at the main incoming line and each branch line to ensure that the data from all acquisition points belong to the same time base timestamp and that data from different sources are aligned in time.

[0063] Anomaly removal here refers to filtering or thresholding the collected electrical parameter data sequence to remove sampling points that are outside the reasonable range or do not conform to physical laws (such as negative power values, sudden change amplitude exceeding the set threshold, etc.) to avoid affecting subsequent calculations.

[0064] Baseline separation here refers to decomposing the acquired actual power curve into two parts: baseline load and event component. The baseline load is mainly used to characterize the normal power consumption trend during the period that is not affected by response control, while the event component is mainly used to characterize the deviation caused by response actions or load fluctuations.

[0065] Optionally, in one embodiment of this application, the load interpretation results of the target equipment group in the target building are analyzed based on aggregated electrical parameters and comprehensive operating condition parameters, including: analyzing the power contribution and operating status of the target equipment group, as well as the equipment flexibility label and identification confidence level of the target equipment group, based on aggregated electrical parameters and comprehensive operating condition parameters; and determining the load interpretation results of the target equipment group based on the power contribution and operating status of the target equipment group, as well as the equipment flexibility label and identification confidence level of the target equipment group.

[0066] In some embodiments, when analyzing the load interpretation results of the main equipment groups in the target building based on the aggregated electrical parameters and comprehensive operating condition parameters of the target building, this application mainly analyzes the power contribution and operating status of the main equipment groups in the target building.

[0067] For example, this application can use an edge load interpretation unit to receive aggregated electrical parameters and comprehensive operating condition parameters of the target building, and decompose these data to obtain information such as equipment category, equipment group operating status, power contribution, identification confidence level, and flexible tags.

[0068] Unlike traditional non-intrusive load identification, this application, in addition to focusing on "how much electricity the equipment uses", also focuses on "whether the equipment or equipment group is adjustable, how much it can be adjusted, how long it can be adjusted, whether the adjustment is reliable, and whether the adjustment affects comfort".

[0069] Unlike traditional NILM (non-intrusive load monitoring) which only outputs power, the embodiments of this application can generate flexible tags and credibility (confidence level) for devices, such as determining whether a certain type of air conditioning load is in a short-term reduction range, whether a certain lighting branch has space for reduced illuminance, whether a certain charging load can be delayed, and whether a certain device belongs to a safety protection load.

[0070] Figure 3 This is a schematic diagram illustrating a non-invasive load flexibility interpretation according to an embodiment of this application. Figure 3 As shown, the embodiments of this application can first identify the main load categories and power contributions from aggregated electrical parameters, and then combine equipment operating patterns, regional attributes, and historical profiles to provide flexible labels. For air conditioning loads, the system focuses on their power cycle, temperature deviation, and area occupancy; for lighting loads, the system focuses on illuminance redundancy and human activity; for charging loads, the system focuses on charging power, remaining time, and user demand.

[0071] Flexible tags include at least one of the following types: reduceable, transferable, delayable, replaceable, non-adjustable, and require protection. For example, lighting loads can be reduced when the illuminance limit is met, air conditioning loads can temporarily reduce power within the temperature comfort range, energy storage can be charged and discharged, and elevator and security loads are considered loads requiring protection.

[0072] It should be noted that the specific flexible tags and their corresponding equipment groups can be set and adjusted by those skilled in the art according to actual needs. The embodiments in this application are only illustrative and do not constitute specific limitations. Furthermore, the load interpretation algorithm may, but is not limited to, employ event detection, hidden Markov models, convolutional neural networks, temporal convolutional networks, lightweight Transformers, regular models, or multi-model fusion; the input may also be expanded from active power to multi-channel features such as active power, reactive power, current, voltage, power factor, and harmonic summaries.

[0073] The specific load interpretation algorithm and input data can be set and adjusted by those skilled in the art according to the actual situation. The embodiments in this application are only illustrative and do not limit the specific model.

[0074] Step S102: Based on the load interpretation results, the pre-constructed digital twin model corresponding to the target building is corrected to obtain the corrected digital twin model. Based on the corrected digital twin model, the adjustability of multiple resources in the target building is calculated.

[0075] In some embodiments, after obtaining the load interpretation results of the main equipment group, this application can use the load interpretation results to correct the pre-built digital twin model corresponding to the target building to obtain the corrected digital twin model.

[0076] The pre-constructed digital twin model corresponding to the target building (hereinafter referred to as the existing twin model) can be understood here as a virtual mapping model that has been constructed to reflect the electrical and physical characteristics of the target building, that is, a dynamic mapping model of the on-site building energy system in digital space. This model includes, but is not limited to, the power response characteristics, energy consumption baseline curves, and initial values ​​of operating parameters of each equipment group in the target building, as an initial reference benchmark for subsequent load interpretation result correction.

[0077] In actual operation, the embodiments of this application can continuously and dynamically modify the existing digital twin model based on the load interpretation results of the main equipment group. That is to say, the pre-built digital twin model in the embodiments of this application refers to the "pre-built" digital twin model for the current moment. From the perspective of runtime, the digital twin model library is constantly updated.

[0078] Furthermore, based on the modified digital twin model, embodiments of this application can calculate the adjustability (currently adjustable resources) of multiple types of resources in the target building.

[0079] Optionally, in one embodiment of this application, the process of correcting a pre-built digital twin model corresponding to the target building based on the load interpretation results to obtain a corrected digital twin model includes: comparing the pre-built digital twin model and the load interpretation results to obtain a comparison result; when the comparison result indicates that there is a continuous deviation between the power of the equipment group and the load interpretation results in the pre-built digital twin model, and the identification confidence of the target equipment group reaches a preset confidence threshold, the model parameters of the pre-built digital twin model are corrected based on the confidence of the target equipment group to complete the correction of the pre-built digital twin model and obtain a corrected digital twin model.

[0080] In some embodiments, when correcting a pre-built digital twin model based on the load interpretation results, this application may, but is not limited to, compare the load interpretation results with the state of the pre-built digital twin model.

[0081] For example, embodiments of this application can use non-intrusive load interpretation results as virtual measurement points, compare them with the predicted states in the digital twin model, and correct equipment parameters, load distribution ratios, or regional energy consumption models based on confidence levels and residual magnitudes. Thus, embodiments of this application can compensate for the shortcomings of existing building sub-metering, enabling the twin model to continuously self-correct during on-site operation.

[0082] Figure 4 This is a schematic diagram illustrating the dynamic correction of a digital twin model according to an embodiment of this application. Figure 4As shown, in practical applications, the digital twin dynamic correction unit can proactively maintain the digital twin model of the building energy system. This model includes, but is not limited to, building area models, main equipment models, power distribution network models, photovoltaic output models, energy storage status models, load comfort models, and carbon emission accounting models.

[0083] Then, the digital twin dynamic correction unit can use the non-intrusive load interpretation results as virtual sensor inputs to compare with the existing twin model state, and perform online correction on the equipment state, load distribution ratio, model parameters and energy flow in the twin model.

[0084] If there is a persistent deviation between the power of a device group and the load interpretation result in the existing twin model, that is, if there is a persistent deviation between the power of a certain device group predicted by the existing twin model and the non-intrusive load interpretation result, and the identification confidence of the power of the device group in the load interpretation result is greater than or equal to the preset confidence threshold, this application can correct the twin model parameters according to the confidence.

[0085] Specific methods for correcting digital twin models can include, but are not limited to, Kalman filtering, recursive least squares, Bayesian updates, residual correction, gray-box model correction, or data-driven proxy model updates, as long as they can improve the consistency between the twin model and the field using non-intrusive load interpretation results. Specific settings and adjustments can be made by those skilled in the art based on actual circumstances. The embodiments in this application are merely illustrative and do not impose specific limitations.

[0086] If the interpretation confidence of the load interpretation result is low (e.g., below a preset confidence threshold), this embodiment of the application can retain the existing digital twin model and mark that period as pending confirmation. In this way, the digital twin model can continue to operate close to the field even when the measurement points are incomplete.

[0087] Here, the pre-set confidence threshold can be understood as the minimum boundary limit for the recognition confidence level. If the recognition confidence level of a certain equipment group's power in the load interpretation results is lower than this value, it proves that the recognition reliability is low, and the existing twin model cannot be modified.

[0088] The specific preset reliability threshold can be set and adjusted by those skilled in the art according to the actual situation. The embodiments in this application are only illustrative and do not impose specific limitations.

[0089] Optionally, in one embodiment of this application, the adjustability of multiple resources in a target building is calculated based on a modified digital twin model, including: calculating the power reduction and duration of the air conditioning load in the target building based on the modified digital twin model and the indoor temperature, comfort zone, thermal inertia, and equipment power of the target building; calculating the power reduction of the lighting load in the target building based on the modified digital twin model and the lower limit of illumination and lighting area occupancy status of the target building; calculating the delay capability of the charging load in the target building based on the modified digital twin model and the expected departure time and charging demand of the current charging equipment in the target building; calculating the charge and discharge capability of the energy storage device based on the modified digital twin model and the state of charge, upper power limit, and lifetime constraint of the energy storage device in the target building; and determining the adjustability of multiple resources in the target building based on the power reduction and duration of the air conditioning load, the power reduction of the lighting load, the delay capability of the charging load, and the charge and discharge capability of the energy storage device.

[0090] In some embodiments, based on a modified digital twin model, this application can calculate the adjustability of various resources in a target building.

[0091] For example, for air conditioning load, this application can estimate the power reduction and duration by combining the current room temperature of the target building (or the average temperature of the rooms if there are multiple rooms), the indoor temperature comfort range, thermal inertia, and equipment power.

[0092] For lighting loads, embodiments of this application can calculate the power reduction that can be achieved by combining the lower limit of illumination and the occupancy status of the lighting area.

[0093] For charging facilities, embodiments of this application can calculate their delay capability by combining the current estimated departure time of the current charging equipment being charged in the target building and the charging demand.

[0094] For energy storage devices, the embodiments of this application can calculate their charge and discharge capacity by combining their state of charge, power limit and lifetime constraints.

[0095] By combining the power reduction and duration of these air conditioning loads, the power reduction of lighting loads, the delay capability of charging loads, and the charge and discharge capability of energy storage devices, the embodiments of this application can determine the adjustability of multiple types of resources in the target building.

[0096] Data such as indoor and outdoor temperature, humidity, carbon dioxide concentration, illuminance, personnel density, and building area occupancy status can be collected by the environmental and comfort acquisition unit (used to constrain load adjustment from exceeding comfort and functional boundaries). For areas that cannot be directly collected, auxiliary estimation can be performed by building automation systems, access control systems, reservation systems, or historical occupancy models. This embodiment is only illustrative and does not impose specific limitations.

[0097] Additionally, Figure 5 This is a schematic diagram illustrating the generation of adjustable resource aggregation and response packets according to one embodiment of this application. Figure 5 As shown, for ease of application, embodiments of this application can aggregate the adjustable capabilities of multiple devices or groups of devices into a building-level resource package through an adjustable resource aggregation unit.

[0098] For example, this application can calculate the adjustable power, adjustable power, duration, response speed, and comfort risk for each type of equipment group separately, and then summarize them into a building-level adjustable resource package. That is, it summarizes the adjustable or de-adjustable capabilities that a building can provide within a certain period of time and their constraints into a building-level adjustable resource package. This resource package includes, but is not limited to, adjustable power, adjustable duration, response speed, comfort risk, rebound risk, confidence level, and carbon emission reduction potential, which can be called by the park energy management platform, the grid demand response platform, or the building's internal optimization scheduling module.

[0099] Step S103: Based on preset constraints and adjustable capabilities, generate a source-grid-load-storage scheduling strategy for the target building, and control the target building to execute the source-grid-load-storage scheduling strategy.

[0100] As one possible approach, after determining the adjustability of various resources in the target building, this application can generate a source-grid-load-storage scheduling strategy for the target building under preset constraints, so as to control the target building to execute the source-grid-load-storage scheduling strategy for actual scheduling.

[0101] In this context, the preset constraints can be understood as the constraints that need to be met when generating the control strategy. Specific preset constraints can be set and adjusted by those skilled in the art based on actual circumstances. The embodiments in this application are merely illustrative and do not constitute specific limitations.

[0102] In the embodiments of this application, the preset constraints include, but are not limited to, power boundaries and state constraints for various resources, such as upper and lower limits of power, energy state boundaries, and operational state logic constraints for various resources (source side (distributed energy), grid side (distribution network), load side (electricity load), and storage side (energy storage resources)); they also include system-level unified constraints on source-grid-load-storage, such as grid connection point power limits, demand constraints, carbon emission limits, and power balance constraints between source, load, and storage. These constraints together constitute the feasible domain boundary for the source-grid-load-storage scheduling decision of the target building.

[0103] For example, Figure 6 This is a flowchart illustrating the source-grid-load-storage coordinated scheduling of one embodiment of this application. Figure 6 As shown, this application can use a source-grid-load-storage resource modeling unit to pre-establish and store descriptions of the power boundaries and state constraints of each resource within a building. The source side includes, but is not limited to, photovoltaic or other distributed energy sources; the grid side includes, but is not limited to, grid connection point power, demand constraints, and electricity price information; the load side includes, but is not limited to, air conditioning, lighting, office sockets, charging equipment, and other flexible or rigid loads; and the storage side includes, but is not limited to, battery energy storage, cold and heat storage, or other energy storage media.

[0104] Next, the collaborative scheduling decision-making unit can generate energy storage charging and discharging, flexible load adjustment, photovoltaic absorption and grid-connected power control strategies under certain constraints, based on factors such as electricity price, demand, photovoltaic forecast, carbon emission factor, energy storage status, load flexibility, carbon emission factor and user constraints, while ensuring comfort and safety boundaries.

[0105] For example, if there is a surplus of photovoltaic power and the energy storage is not fully utilized, priority should be given to energy storage charging and flexible load absorption; if the current demand is close to the peak, priority should be given to calling up loads with high confidence and low comfort risk that can be reduced and energy storage discharge; if participating in demand response, response power, response duration and reserve resources should be generated based on the building's adjustable resource package; if the carbon emission factor is high, priority should be given to reducing electricity purchases during high-carbon periods.

[0106] Finally, the execution feedback unit sends the data to the energy storage converter, air conditioning controller, lighting controller, or building energy management system to execute the scheduling strategy.

[0107] In terms of adjustable resource aggregation and scheduling methods, rule priority, linear programming, mixed integer programming, model predictive control, reinforcement learning or robust optimization methods can be adopted; the resource objects can also be expanded from air conditioning, lighting and energy storage to charging piles, elevator groups, data centers, cold and heat storage and other controllable loads. The embodiments in this application are only illustrative and do not impose specific limitations.

[0108] Optionally, in one embodiment of this application, after controlling the target building to execute the source-grid-load-storage scheduling strategy, the method further includes: obtaining changes in grid-connected power, equipment status, and comfort level from the execution results of the source-grid-load-storage scheduling strategy; and adjusting the scheduling order and call priority of multiple types of resources based on the changes in grid-connected power, equipment status, and comfort level.

[0109] In other embodiments, after the execution feedback unit sends the scheduling instruction to the controllable device or building control system and controls the target building to execute the source-grid-load-storage scheduling strategy, the embodiments of this application can collect the changes in grid-connected power, equipment status and comfort after execution, and then adjust the scheduling order and calling priority of multiple types of resources based on the changes in grid-connected power, equipment status and comfort.

[0110] If the actual response volume is lower than the planned value, this application embodiment can automatically adjust the subsequent resource sorting and correct the adjustable capability profile of the device; if the comfort or safety boundary is close to the threshold, this application embodiment can reduce the subsequent call priority of the device.

[0111] In simple terms, the embodiments of this application can record the actual power changes, comfort changes, and resource response deviations before and after each scheduling, and then update the flexibility profile of the equipment group. After long-term operation, an adjustable capability baseline can be formed for different seasons, different time periods, and different occupancy conditions.

[0112] This application embodiment can obtain load interpretation results for various types of loads by analyzing the aggregated electrical parameters and comprehensive operating condition parameters of a building, and further transform these load interpretation results into flexible resource information that can be used for scheduling. Compared with traditional solutions that typically output equipment power or equipment status, which are difficult to directly serve the building's source-grid-load-storage coordinated control, this application embodiment can further determine whether the equipment group can be reduced, transferred, delayed, or needs protection after identification, and provide adjustable power, duration, comfort risk, and confidence level, so that the non-intrusive detection results can be directly incorporated into the demand response, peak shaving and valley filling, and energy storage scheduling processes.

[0113] Furthermore, compared with traditional building energy management or digital twin platforms, the embodiments of this application can also continuously and dynamically correct the digital twin model through non-intrusive virtual measurement points in existing buildings with insufficient sub-metering, incomplete equipment ledgers, or missing sensors, thereby reducing the deviation between the model and the on-site operation and enabling the model to adapt to changes in existing building equipment and insufficient measurement points.

[0114] The corrected twin model is then used to calculate building-level adjustable resource packages, and the resource profile is continuously updated through execution feedback. This reduces hardware modification costs while improving the reliability, interpretability, and engineering implementation capabilities of scheduling strategies. It is applicable to scenarios such as existing commercial buildings, public buildings, office parks, hospitals, schools, and complexes, and is especially suitable for platform construction projects that have insufficient sub-metering but need to carry out source-grid-load-storage collaborative optimization, demand response, or energy and carbon management.

[0115] The following is a detailed explanation of the scheduling method based on the aggregation of adjustable resources of building source network load storage in this application, using a specific embodiment.

[0116] Figure 7 This is a flowchart illustrating an embodiment of a building source-network-load-storage adjustable resource aggregation and scheduling method based on non-intrusive load flexibility interpretation and digital twin dynamic correction, according to one embodiment of this application. Figure 7 As shown: The first step involves the system collecting aggregated electrical parameters from the building's main incoming line and key branch lines, while simultaneously collecting photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and the building's operating calendar. After the collected data enters the edge buffer, the system performs time synchronization, outlier removal, missing value imputation, noise smoothing, and power baseline separation to form a multi-channel input sequence for load interpretation.

[0117] The second step involves the edge load interpretation unit decomposing the aggregated signal and outputting the power contribution and operating status of the main equipment groups. Unlike traditional NILMs that only output power, this invention further generates flexible labels and credibility for the equipment, such as determining whether a certain type of air conditioning load is in a short-term reduction range, whether a certain lighting branch has room for reducing illuminance, whether a certain charging load can be delayed, and whether a certain device belongs to a safety protection load.

[0118] The third step involves the digital twin dynamic correction unit comparing the load interpretation results with the existing twin model status. If the power prediction of a certain equipment group by the twin model shows a persistent deviation from the non-intrusive interpretation results, the system corrects the twin model parameters based on the confidence level. If the interpretation confidence level is low, the original model is retained, and that period is marked as a pending confirmation state. In this way, the digital twin model can continuously operate close to the field even when measurement points are incomplete.

[0119] The fourth step involves the adjustable resource aggregation unit calculating the adjustable capabilities of various resources based on the corrected twin model. For air conditioning loads, the system calculates the power reduction and duration based on current room temperature, comfort zone, thermal inertia, and equipment power estimation. For lighting loads, the system calculates the power reduction based on the lower limit of illuminance and area occupancy status. For charging facilities, the system calculates the delay capability based on the expected departure time and charging demand. For energy storage, the system calculates the charge and discharge capability based on the state of charge, power upper limit, and lifetime constraints.

[0120] The fifth step involves the collaborative scheduling decision-making unit generating control strategies under the unified constraints of power generation, grid, load, and storage. If there is a surplus of photovoltaic power and the energy storage is not fully utilized, the system prioritizes energy storage charging and flexible load absorption. If the current demand is close to peak, the system prioritizes calling up loads with high confidence and low comfort risk that can be reduced and energy storage discharging. If participating in demand response, the system generates response power, response duration, and reserve resources based on the building's adjustable resource package. If the carbon emission factor is high, priority is given to reducing electricity purchases during high-carbon periods.

[0121] The sixth step involves the execution feedback unit sending scheduling instructions to controllable devices or building control systems, and collecting data on grid-connected power, device status, and comfort changes after execution. If the actual response is lower than the planned value, the system automatically adjusts the subsequent resource prioritization and corrects the adjustable capacity profile of the device; if the comfort or safety boundary is close to the threshold, the system lowers the subsequent call priority of the device.

[0122] The key model expressions involved in the process can be, but are not limited to, represented as follows: The building aggregate load decomposition relationship can be represented, but is not limited to, as follows: , in, Aggregated load power for buildings, For the first Estimated power of the group of equipment, For unknown or unmodeled loads, This represents the measurement error or model residual. This formula is used to decompose the total load into known equipment groups, unknown loads, and error terms.

[0123] The state correction of a digital twin model can be, but is not limited to, expressed as:

[0124] in, For the building digital twin state vector, To control the input, Disturbances caused by weather, occupancy, and electricity prices, The virtual measurement quantity is obtained through non-intrusive interpretation. This is for dynamic gain correction. This formula reflects how the embodiments of this application use non-invasive detection results to correct the digital twin model.

[0125] Device-level downscaling capabilities can be, but are not limited to, expressed as:

[0126] Device-level upscaling capabilities can be, but are not limited to, expressed as:

[0127] in, and They represent the first The adjustable and adjustable power of the resource type. This represents the reliability of the resource's capabilities. By multiplying by the reliability, embodiments of this application avoid directly using low-reliability identification results for scheduling.

[0128] The overall confidence level can be, but is not limited to, expressed as:

[0129] The four components represent model output stability, energy conservation consistency, digital twin consistency, and device profile consistency, respectively.

[0130] The objectives of coordinated scheduling of building energy sources, grids, loads, and storage can be, but are not limited to, expressed as: , This objective function comprehensively considers electricity purchase costs, electricity sales revenue, carbon emission costs, comfort deviations, and energy storage degradation costs. In practice, objective items can be added or removed based on project requirements.

[0131] The power balance between power generation, grid, load, and storage can be expressed, but is not limited to, as:

[0132] Among them, the energy balance is maintained in each dispatch period by photovoltaic power generation, grid-side power purchase and sales, load-side power consumption, and energy storage charging and discharging.

[0133] The scheduling method based on adjustable resource aggregation of building energy sources, grid, load, and storage proposed in this application can complete equipment-level load status identification and flexible attribute interpretation through a small number of aggregated measurement points. The identification results are used to dynamically correct the building digital twin model, and the comprehensive adjustable capacity of load, photovoltaic, and energy storage is calculated under the constraints of the twin model. This generates scheduling recommendations for electricity price optimization, demand response, peak shaving and valley filling, photovoltaic integration, and low-carbon operation. Furthermore, this application can reduce the cost of retrofitting existing buildings, reduce reliance on numerous sub-metering items and open equipment protocols, enable the twin model to continuously self-correct during field operation, improve the field consistency of the building energy digital twin model, and elevate the non-intrusive detection results from "energy consumption identification" to the engineering application capability of "adjustable resource aggregation and collaborative scheduling," thereby improving the reliability, interpretability, and engineering implementation capability of the scheduling strategy. This solves the problems in related technologies, such as the difficulty in fully identifying the characteristics of loads in building source-network-load-storage collaborative management schemes, which leads to a lack of direct basis for management decisions, optimization often relies on complete measurement and ledgers, and can only be handled in a rough manner when they are missing, rigid and flexible loads cannot be distinguished, and the model is prone to inaccuracy with changes in operation and lacks dynamic correction. A single indicator cannot reflect the true capacity of the building's overall adjustable resources.

[0134] Next, referring to the accompanying drawings, a scheduling device based on adjustable resource aggregation of building source network load storage proposed in the embodiments of this application is described.

[0135] Figure 8 This is a schematic diagram of the scheduling device based on the aggregation of adjustable resources of building source network load storage according to an embodiment of this application.

[0136] like Figure 8 As shown, the scheduling device 10 based on the aggregation of adjustable resources of building source network load storage includes: analysis module 100, correction module 200 and scheduling module 300.

[0137] The analysis module 100 is used to acquire the aggregated electrical parameters of the main incoming line and the branch lines that meet the preset important conditions of the target building, as well as the comprehensive operating parameters of the target building. Based on the aggregated electrical parameters and comprehensive operating parameters, it analyzes the load interpretation results of the target equipment group in the target building. The correction module 200 is used to correct the pre-constructed digital twin model corresponding to the target building according to the load interpretation results, to obtain the corrected digital twin model, and to calculate the adjustability of multiple resources in the target building based on the corrected digital twin model. The scheduling module 300 is used to generate the source-grid-load-storage scheduling strategy of the target building based on preset constraints and adjustability, and to control the target building to execute the source-grid-load-storage scheduling strategy.

[0138] Optionally, in one embodiment of this application, the analysis module 100 includes: an acquisition unit, used to acquire the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar of the target building; and a first determination unit, used to determine the comprehensive operating parameters of the target building based on the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar of the target building.

[0139] Optionally, in one embodiment of this application, the analysis module 100 includes: an analysis unit, used to analyze the power contribution and operating status of the target equipment group, as well as the equipment flexibility label and identification confidence level of the target equipment group, based on aggregated electrical parameters and comprehensive operating condition parameters; and a second determination unit, used to determine the load interpretation result of the target equipment group based on the power contribution and operating status of the target equipment group, as well as the equipment flexibility label and identification confidence level of the target equipment group.

[0140] Optionally, in one embodiment of this application, the correction module 200 includes: a comparison unit, used to compare a pre-built digital twin model with a load interpretation result to obtain a comparison result; and a correction unit, used to correct the model parameters of the pre-built digital twin model according to the confidence level of the target device group when the comparison result shows that there is a continuous deviation between the power of the device group and the load interpretation result in the pre-built digital twin model, and the identification confidence level of the target device group reaches a preset confidence threshold, so as to complete the correction of the pre-built digital twin model and obtain a corrected digital twin model.

[0141] Optionally, in one embodiment of this application, the correction module 200 includes: a first calculation unit, used to calculate the reduceable power and duration of the air conditioning load in the target building based on the corrected digital twin model and the indoor temperature, comfort zone, thermal inertia, and equipment power of the target building; a second calculation unit, used to calculate the reduceable power of the lighting load in the target building based on the corrected digital twin model and the lower limit of illumination and the occupancy status of the lighting area of ​​the target building; a third calculation unit, used to calculate the delay capability of the charging load in the target building based on the corrected digital twin model and the expected departure time and charging demand of the current charging equipment in the target building; a fourth calculation unit, used to calculate the chargeable and dischargeable capacity of the energy storage device based on the corrected digital twin model and the state of charge, upper limit of power, and lifetime constraints of the energy storage device in the target building; and a fifth calculation unit, used to determine the adjustability of multiple resources in the target building based on the reduceable power and duration of the air conditioning load, the reduceable power of the lighting load, the delay capability of the charging load, and the chargeable and dischargeable capacity of the energy storage device.

[0142] Optionally, in one embodiment of this application, it further includes: an acquisition module, configured to acquire the grid-connected power, equipment status, and comfort changes in the execution result of the source-grid-load-storage scheduling strategy after the target building is controlled to execute the source-grid-load-storage scheduling strategy; and an adjustment module, configured to adjust the scheduling order and call priority of multiple types of resources based on the grid-connected power, equipment status, and comfort changes.

[0143] It should be noted that the foregoing explanation of the scheduling method embodiment based on the aggregation of adjustable resources of building source, network, load and storage also applies to the scheduling device based on the aggregation of adjustable resources of building source, network, load and storage in this embodiment, and will not be repeated here.

[0144] The scheduling device based on adjustable resource aggregation of building energy sources, grid, load, and storage proposed in this application can complete equipment-level load status identification and flexible attribute interpretation through a small number of aggregated measurement points. The identification results are used to dynamically correct the building digital twin model, and the comprehensive adjustable capacity of load, photovoltaic, and energy storage is calculated under the constraints of the twin model. This generates scheduling recommendations for electricity price optimization, demand response, peak shaving and valley filling, photovoltaic absorption, and low-carbon operation. Furthermore, this application can reduce the cost of retrofitting existing buildings, reduce reliance on numerous sub-metering items and open equipment protocols, enable the twin model to continuously self-correct during field operation, improve the field consistency of the building energy digital twin model, and elevate the non-intrusive detection results from "energy consumption identification" to the engineering application capability of "adjustable resource aggregation and collaborative scheduling," thereby improving the reliability, interpretability, and engineering implementation capability of the scheduling strategy. This solves the problems in related technologies, such as the difficulty in fully identifying the characteristics of loads in building source-network-load-storage collaborative management schemes, which leads to a lack of direct basis for management decisions, optimization often relies on complete measurement and ledgers, and can only be handled in a rough manner when they are missing, rigid and flexible loads cannot be distinguished, and the model is prone to inaccuracy with changes in operation and lacks dynamic correction. A single indicator cannot reflect the true capacity of the building's overall adjustable resources.

[0145] Figure 9 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 901, the processor 902, and the computer program stored on the memory 901 and capable of running on the processor 902.

[0146] When the processor 902 executes the program, it implements the scheduling method based on the aggregation of adjustable resources of building source network load storage provided in the above embodiments.

[0147] Furthermore, electronic devices also include: Communication interface 903 is used for communication between memory 901 and processor 902.

[0148] The memory 901 is used to store computer programs that can run on the processor 902.

[0149] The memory 901 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0150] If the memory 901, processor 902, and communication interface 903 are implemented independently, then the communication interface 903, memory 901, and processor 902 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 9 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0151] Optionally, in a specific implementation, if the memory 901, processor 902, and communication interface 903 are integrated on a single chip, then the memory 901, processor 902, and communication interface 903 can communicate with each other through an internal interface.

[0152] The processor 902 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0153] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the scheduling method based on the aggregation of adjustable resources from building sources, networks, loads, and storage as described above.

[0154] This application also provides a computer program product, including a computer program that can run computer instructions. When the computer instructions are executed by a processor, they implement the scheduling method based on adjustable resource aggregation of building source network load storage provided in this application.

[0155] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0156] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0157] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0158] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0159] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0160] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0161] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0162] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A scheduling method based on the aggregation of adjustable resources from building sources, networks, loads, and storage, characterized in that, Includes the following steps: The aggregated electrical parameters of the main incoming line and the branch lines that meet the preset important conditions of the target building are obtained, as well as the comprehensive operating parameters of the target building. Based on the aggregated electrical parameters and the comprehensive operating parameters, the load interpretation results of the target equipment group in the target building are analyzed. Based on the load interpretation results, the pre-constructed digital twin model corresponding to the target building is corrected to obtain the corrected digital twin model. Based on the corrected digital twin model, the adjustability of multiple resources in the target building is calculated. Based on preset constraints and the adjustable capabilities, a source-grid-load-storage scheduling strategy for the target building is generated, and the target building is controlled to execute the source-grid-load-storage scheduling strategy.

2. The method according to claim 1, characterized in that, The acquisition of aggregated electrical parameters at the main incoming line and branch lines meeting preset important conditions of the target building, as well as the comprehensive operating parameters of the target building, includes: Obtain the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar of the target building; Based on the photovoltaic output, energy storage state of charge, indoor and outdoor environmental parameters, and building operation calendar of the target building, the comprehensive operating parameters of the target building are determined.

3. The method according to claim 1, characterized in that, The analysis of the load interpretation results of the target equipment group in the target building based on the aggregated electrical parameters and the comprehensive operating condition parameters includes: Based on the aggregated electrical parameters and the comprehensive operating condition parameters, the power contribution and operating status of the target equipment group, as well as the equipment flexible tags and identification confidence of the target equipment group, are analyzed. Based on the power contribution and operating status of the target equipment group, as well as the equipment flexibility tags and identification confidence of the target equipment group, the load interpretation result of the target equipment group is determined.

4. The method according to claim 1, characterized in that, The step of correcting the pre-constructed digital twin model corresponding to the target building based on the load interpretation results to obtain the corrected digital twin model includes: By comparing the pre-constructed digital twin model with the load interpretation results, a comparison result is obtained; When the comparison result shows that there is a continuous deviation between the power of the device group and the load interpretation result in the pre-built digital twin model, and the identification confidence of the target device group reaches a preset confidence threshold, the model parameters of the pre-built digital twin model are corrected according to the confidence of the target device group to complete the correction of the pre-built digital twin model and obtain the corrected digital twin model.

5. The method according to claim 1, characterized in that, The calculation of the adjustability of multiple resources in the target building based on the modified digital twin model includes: Based on the modified digital twin model and the indoor temperature, comfort zone, thermal inertia, and equipment power of the target building, the power reduction and duration of the air conditioning load in the target building are calculated. Based on the modified digital twin model and the lower limit of illumination and the occupancy status of the lighting area of ​​the target building, the power reduction of the lighting load in the target building is calculated; Based on the modified digital twin model and the estimated departure time and charging demand of the current charging equipment in the target building, the delay capability of the charging load in the target building is calculated; Based on the modified digital twin model and the state of charge, power limit, and lifetime constraints of the energy storage devices in the target building, the charge and discharge capacity of the energy storage devices is calculated. Based on the power reduction and duration of the air conditioning load, the power reduction of the lighting load, the delay capability of the charging load, and the charging and discharging capability of the energy storage device, the adjustability of multiple resources in the target building is determined.

6. The method according to claim 1, characterized in that, After controlling the target building to execute the source-grid-load-storage scheduling strategy, the process also includes: Obtain the changes in grid-connected power, equipment status, and comfort level from the execution results of the source-grid-load-storage scheduling strategy; Based on the changes in grid-connected power, equipment status, and comfort level, the scheduling order and call priority of the various types of resources are adjusted.

7. A scheduling device based on the aggregation of adjustable resources from building sources, networks, loads, and storage, characterized in that, include: The analysis module is used to obtain the aggregated electrical parameters of the total incoming line and the branch lines that meet the preset important conditions of the target building, as well as the comprehensive operating parameters of the target building. Based on the aggregated electrical parameters and the comprehensive operating parameters, the load interpretation results of the target equipment group in the target building are analyzed. The correction module is used to correct the pre-constructed digital twin model corresponding to the target building based on the load interpretation results, to obtain the corrected digital twin model, and to calculate the adjustability of multiple resources in the target building based on the corrected digital twin model. The scheduling module is used to generate a source-grid-load-storage scheduling strategy for the target building based on preset constraints and the adjustable capacity, and to control the target building to execute the source-grid-load-storage scheduling strategy.

8. An electronic device, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the scheduling method based on adjustable resource aggregation of building source network load storage as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the scheduling method based on the aggregation of adjustable resources of building source network load storage as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed, it is used to implement the scheduling method based on the aggregation of adjustable resources of building source network load storage as described in any one of claims 1-6.

Images