A building interaction resource library construction method and system for power grid regulation

Abstract

The present application relates to the technical field of power grid regulation, in particular to a building interaction resource library construction method and system for power grid regulation. The present application firstly obtains building interaction daily data and building comparison daily data on the building side, and power grid interaction daily data and power grid comparison daily data on the power grid side, obtains actual working conditions capable of power grid physical operation, establishes coupling correlation between building regulation behavior and power grid operation state, and establishes data basis for quantitatively evaluating the ability of buildings facing power grid regulation. Then, the present application selects a plurality of buildings by calculating building side indexes and power grid side indexes of the buildings to construct an interaction resource library, establishes a multi-dimensional quantitative system based on building side resource regulation quality, power grid side regulation risk improvement gain, and performance comparability between buildings to construct the building interaction resource library, so that the power grid side can select and call interactive building resources for collaborative regulation according to the real regulation value of the buildings and the power grid operation state.

Description

Technical Field

[0001] This invention relates to the field of power grid control technology, and in particular to a method and system for constructing a building interactive resource database for power grid control. Background Technology

[0002] Buildings are core electricity consumers in the power system, accounting for a significant portion of total electricity consumption. They are characterized by concentrated load distribution, significant resource economies of scale, and well-defined operational patterns. With the increasing electrification of end-use energy equipment, more and more energy-consuming scenarios in buildings are driven by electricity, such as the widespread adoption of electric vehicle charging facilities, the use of heat pumps or electric boilers to replace traditional gas heating, and the continuous growth of data center loads. This trend has led to the integration of numerous electrical devices with flexible adjustment potential within buildings, primarily including HVAC systems, intelligent lighting systems, power systems, distributed photovoltaic and energy storage systems, and electric vehicle charging and discharging facilities with bidirectional interactive capabilities. These controllable resources constitute the physical basis for intelligent buildings to participate in grid interaction.

[0003] Under current technological conditions, smart buildings, based on building automation systems and IoT infrastructure, already possess the ability to monitor and control internal resources. Buildings can participate in grid interaction operations either as individual units or through aggregation mechanisms such as load aggregators and virtual power plants. Currently, common interaction mechanisms between building regulation behaviors and grid operating status mainly include peak shaving and valley filling based on grid commands. In this type of interaction model, the grid company publishes an ideal target load curve, i.e., a load baseline, based on the grid operating status. The building side adjusts the operating strategies of its internal electrical equipment to make its actual electricity consumption curve at the metering point as close as possible to this baseline. During the effect verification phase, the response power and tracking accuracy of the smart building are calculated by comparing the degree of fit between the actual load curve and the load baseline, which serves as the basis for evaluating the performance of the smart building.

[0004] However, existing smart building evaluation technologies primarily focus on the energy-saving and low-carbon attributes of the building itself. Even when evaluating the interaction between buildings and the power grid, they are often limited to individual building-side indicators such as adjustable capacity and total response. Such evaluation methods are detached from the actual operating conditions of the power grid and cannot establish a coupling relationship between building regulation behavior and the power grid's operating state. In particular, they lack quantitative representation methods for the actual contribution of building access locations and their regulation behavior to key control needs such as alleviating line congestion and voltage exceedances, promoting renewable energy consumption, and coordinated peak shaving and valley filling. Consequently, the power grid cannot verify the interactive effects of buildings' participation in power grid control based on their true regulatory value, and cannot conduct effective power grid operation state control based on building regulation behavior. Summary of the Invention

[0005] This invention aims to provide a method and system for constructing a building interaction resource database for power grid regulation. It establishes a multi-dimensional quantitative system based on the building-side resource regulation quality, the power grid-side regulation risk improvement gain, and the performance comparability between buildings to build a building interaction resource database. This provides support for the power grid side to verify the interaction effect of buildings in power grid regulation based on the actual regulation value of buildings, and to carry out effective power grid operation status regulation based on building regulation behavior.

[0006] To achieve the above objectives, the first aspect of the present invention provides a method for constructing a building interactive resource library for power grid regulation, comprising the following steps: Identify building clusters; For any building in the building cluster, the net power input curve of the power grid and the sub-metering curves of various power generation and consumption units of the building are obtained as building interaction day data for each interaction day in a preset historical period, thereby obtaining the building interaction day data for each building. The power distribution network area of ​​the building cluster is determined, and then the physical topology data and operation constraint data of the power distribution network area for each interaction day in the historical period are obtained as power grid interaction day data. Generate building reference day data corresponding to each of the building interaction day data, and generate power grid reference day data corresponding to the power grid interaction day data; Based on the power grid interaction day data, the power grid comparison day data, and the building interaction day data and the building comparison day data for each building, obtain the building-side indicators and power grid-side indicators for each building; The building-side indicators and the power grid-side indicators of each building are normalized to obtain multiple normalized indicators for each building. The power grid interaction evaluation index for each of the buildings is obtained based on all the normalized indices of each building; Based on the power grid interaction evaluation index of each building, several buildings are selected from the building cluster to construct an interaction resource library, which is used for coordinated regulation when power grid control is carried out in the distribution network area.

[0007] The aforementioned method for constructing a building interaction resource database for power grid regulation first obtains daily interaction data and comparison data for buildings on the building side, as well as daily interaction data and comparison data for the power grid on the power grid side. This yields the actual operating conditions of the power grid's physical operation, establishing a coupled correlation between building regulation behavior and power grid operating status, providing a data foundation for quantitatively evaluating the building's ability to regulate the power grid. Then, by calculating building-side and power grid-side indicators, several buildings are selected to construct the interaction resource database. This database establishes a multi-dimensional quantitative system based on building-side resource regulation quality, power grid-side regulation risk improvement gains, and inter-building performance comparability. This allows the power grid side to select and utilize interactive building resources for coordinated regulation based on the actual regulation value of buildings and the power grid's operating status, significantly improving the targeting and effectiveness of power grid regulation. This provides effective support for the power grid side to verify the interactive effects of buildings' participation in power grid regulation based on their actual regulation value and to conduct effective power grid operating status regulation based on building regulation behavior.

[0008] Further, the step of obtaining building-side indicators and grid-side indicators for each building based on the grid interaction day data, the grid reference day data, and the building interaction day data and the building reference day data for each building includes: The response potential index, interaction capability index, interaction effect index, environmental friendliness index, and interaction power cost index of each building are obtained as multiple building-side indicators for each building. The power flow heavy load adjustment sensitivity, overvoltage adjustment sensitivity, undervoltage adjustment sensitivity, renewable energy consumption sensitivity, peak shaving and valley filling sensitivity, and carbon emission reduction sensitivity of each building are obtained as multiple grid-side indicators for each building.

[0009] This implementation establishes a two-sided evaluation system from the building side to the grid side. On the one hand, building-side indicators include response potential, interaction capability, interaction effect, environmental friendliness, and interaction power cost indicators. This quantifies the deliverable physical capabilities, implementation effects, and costs of buildings as flexible resource clusters, addressing the problem that a single type of indicator cannot comprehensively evaluate building interaction performance. On the other hand, the grid-side indicator group includes power flow heavy load adjustment sensitivity, overvoltage adjustment sensitivity, undervoltage adjustment sensitivity, renewable energy absorption sensitivity, peak shaving and valley filling sensitivity, and carbon emission reduction sensitivity. This extends the evaluation perspective from the building itself to the physical operation level of the distribution network, quantifying the actual marginal contribution of building regulation behavior to mitigating different grid operation risks. By simultaneously acquiring these two types of indicators, a coupled correlation between building regulation behavior and grid operation status is established. In particular, it provides a quantitative representation of the actual contribution of building access location and its regulation behavior to key control needs such as alleviating line congestion and voltage exceedances, promoting renewable energy absorption, and coordinated peak shaving and valley filling. This provides a quantitative basis for the grid side to achieve differentiated incentives and precise resource selection.

[0010] Furthermore, the acquisition of response potential indicators, interaction capability indicators, interaction effect indicators, environmental friendliness indicators, and interaction power cost indicators for each of the buildings as multiple building-side indicators for each building includes: The adjustable load percentage, downward adjustment capacity, and upward adjustment capacity of each building are obtained as multiple response potential indicators for each building; The uphill speed threshold, downhill speed threshold, continuous discharge time threshold, and continuous charging time threshold of each building are obtained as multiple interactive capability indicators for each building. The load profile similarity, total load response, load peak-to-valley difference rate, and load fluctuation coefficient of each building are obtained as multiple interaction effect indicators for each building. The carbon emission intensity per unit area and the proportion of renewable energy electricity consumption for each building are obtained as multiple environmental friendliness indicators for each building; The unit response incentive revenue, unit response utility revenue, and unit response regulation cost of each building are obtained as multiple interactive power cost indicators for each building.

[0011] This implementation constructs a clearly structured and physically meaningful quantifiable indicator system. Taking the adjustable load ratio, downward adjustment capacity, and upward adjustment capacity under the response potential indicator as examples, they characterize the adjustment margin that a building can provide under specific conditions from the perspectives of relative proportion and absolute capacity, respectively. The ramp-up speed threshold and continuous charge / discharge time threshold under the interaction capability indicator accurately describe the dynamic characteristics and sustainability of resource response, ensuring that the mobilized resources can meet the dual requirements of rapid grid response and continuous support. The load baseline similarity indicator in the interaction effect indicator is directly related to the grid's control objectives, while indicators such as total load response verify the actual implementation effectiveness. The environmental friendliness and interactive power cost indicators supplement the evaluation of interactive behavior from the perspectives of social benefits and the power cost of building-grid interactive control, respectively.

[0012] Furthermore, obtaining the power grid interaction evaluation index for each building based on all the normalized indices of each building includes: For any of the buildings, obtain the benchmark index weight of each of the normalized indicators of the building, and then perform a weighted summation of all the normalized indicators based on the benchmark index weight of each of the normalized indicators to obtain the power grid interaction evaluation index.

[0013] This implementation method pre-sets a benchmark index weight for each normalized secondary indicator. Based on the benchmark index weight of each normalized indicator, it performs a weighted summation of all normalized indicators, thereby comprehensively evaluating a building's performance across multiple dimensions, including its response potential, interaction capabilities, regulation effects, environmental attributes, and regulation costs, into a single grid interaction evaluation index. Sorting and filtering based on this single comprehensive index efficiently establishes a normalized interaction resource library applicable to most scenarios without specific regulation needs. When the grid does not identify a problem requiring specific handling, it prioritizes the use of smart buildings from the interaction resource library to participate in building-grid interaction, improving the overall efficiency of building resources participating in grid regulation.

[0014] Furthermore, obtaining the power grid interaction evaluation index for each building based on all the normalized indices of each building includes: For any of the aforementioned buildings: Obtain the baseline index weight for each of the normalized indices of the building; Several constraint types corresponding to different operating conditions in the power distribution network area are obtained, and then the severity coefficient of each constraint type under a preset power grid dispatch time window is obtained. For any constraint type, based on the severity coefficient and the baseline index weight of each normalized index, the dynamic index weight of each normalized index corresponding to the constraint type is obtained. Then, based on the dynamic index weight of each normalized index, all normalized indices are weighted and summed to obtain the power grid interaction evaluation index corresponding to the constraint type.

[0015] In scenarios where grid interaction resources are abundant and grid-side operational demands may vary across different dates, relying solely on a fixed comprehensive score ranking is insufficient to guarantee the relevance to specific operational constraints. Therefore, this implementation introduces a dynamic weight adjustment mechanism based on the current grid operating conditions and constraint severity. By incorporating constraint types and corresponding severity coefficients, and based on the correlation between each normalized indicator and a specific constraint type, the weights of indicators that contribute significantly to resolving that type of constraint are dynamically increased, generating targeted dynamic indicator weights. The resulting grid interaction evaluation index, obtained through weighted summation, is essentially a goal-oriented priority score that quantitatively characterizes the suitability and regulatory value of the system within the current specific grid risk scenario.

[0016] Furthermore, based on the power grid interaction evaluation index of each building, a number of buildings are selected from the building cluster to construct an interaction resource library. This interaction resource library is used for coordinated regulation during power grid control in the distribution network area, including: For any of the constraint types, based on the power grid interaction evaluation index corresponding to each building for that constraint type, several buildings are selected from the building cluster to construct a dynamic resource library corresponding to that constraint type, thereby obtaining a dynamic resource library for each constraint type, and all the dynamic libraries are used as the interaction resource library; The dynamic resource library is used for coordinated adjustment when power grid regulation is carried out in the distribution network area with the corresponding constraint type as the regulation target.

[0017] In this implementation, based on the power grid interaction evaluation indicators corresponding to different constraint types, a building dynamic resource library for different constraint types is selected and constructed. When the power grid is regulated in the distribution network area with the corresponding constraint type as the regulation target, the smart buildings in the corresponding dynamic resource library can be called first for coordinated regulation. This avoids the problem that a single resource library is not targeted enough when facing complex power grid operation states, and greatly improves the building-power grid interaction's refined support capability for the safe and stable operation and efficient absorption of the distribution network.

[0018] A second aspect of the present invention provides a building interactive resource database construction system for power grid regulation, comprising: The object determination module is used to determine building clusters; The building data acquisition module is used to acquire, for any building in the building cluster, the net power input curve of the power grid and the sub-metering curves of various power generation and consumption units for each interaction day in a preset historical period as building interaction day data, and then obtain the building interaction day data for each building. The power grid data acquisition module is used to determine the distribution network area of ​​the building cluster, and then acquire the physical topology data and operation constraint data of the distribution network area for each interaction day in the historical period as power grid interaction day data; The comparison data generation module is used to generate building comparison day data corresponding to each of the building interaction day data, and to generate power grid comparison day data corresponding to the power grid interaction day data; The indicator value acquisition module is used to acquire building-side indicators and grid-side indicators for each building based on the grid interaction day data, the grid reference day data, and the building interaction day data and the building reference day data for each building. The index value normalization module is used to normalize the building-side index and the power grid-side index of each building, thereby obtaining multiple normalized indexes for each building. The indicator value synthesis module is used to obtain the power grid interaction evaluation index of each building based on all the normalized indicators of each building; The resource library construction module is used to select several buildings in the building cluster to construct an interactive resource library based on the power grid interaction evaluation index of each building. The interactive resource library is used for coordinated regulation when power grid control is carried out in the distribution network area. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating a method for constructing a building interactive resource database for power grid regulation, as provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of a framework structure for intelligent building and power grid interaction indicators provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a building interactive resource database construction system for power grid regulation provided in an embodiment of the present invention. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that the following detailed descriptions are exemplary and intended to provide further detailed explanation of the invention. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein in the specification is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings are used to distinguish different objects, not to describe a particular order.

[0021] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0022] Before describing this application in detail with reference to the accompanying drawings and embodiments, the terminology used in this application will first be explained.

[0023] Interaction Day: Refers to the date on which a smart building actually participates in grid interaction and has complete and verifiable data records.

[0024] Comparison date: refers to a historical date that is comparable to the interaction date in terms of date type, operating status, and external conditions, selected to construct a reference level under conditions where the interaction date was not involved.

[0025] Load profile: refers to the power shape curve formulated and published by the grid side based on operational objectives, used to evaluate the performance of an object in fulfilling grid regulation targets within a specified period. The load profile should clearly specify the applicable dates and the set of applicable time periods.

[0026] Building-grid interaction refers to the coordinated adjustment behavior of intelligent buildings, guided by load guidelines, dispatch instructions, or price signals issued by the grid, in order to change their net input power or energy consumption behavior to meet the grid's operational objectives.

[0027] Adjustable resources refer to the power generation and consumption units within a smart building that, under constraints of safety, comfort, process, and equipment operation, can adjust their power or energy exchange within a specified time scale through control strategies, thereby altering the net input power curve at the metering boundary. Adjustable resources include adjustable loads on the consumption side, controllable resources on the generation side, and energy storage equipment.

[0028] Buildings are core electricity consumers in the power system, accounting for a significant portion of total electricity consumption. They are characterized by concentrated load distribution, significant resource economies of scale, and well-defined operational patterns. With the increasing electrification of end-use energy equipment, more and more energy-consuming scenarios in buildings are driven by electricity, such as the widespread adoption of electric vehicle charging facilities, the use of heat pumps or electric boilers to replace traditional gas heating, and the continuous growth of data center loads. This trend has led to the integration of numerous electrical devices with flexible adjustment potential within buildings, primarily including HVAC systems, intelligent lighting systems, power systems, distributed photovoltaic and energy storage systems, and electric vehicle charging and discharging facilities with bidirectional interactive capabilities. These controllable resources constitute the physical basis for intelligent buildings to participate in grid interaction.

[0029] Under current technological conditions, smart buildings, based on building automation systems and IoT infrastructure, already possess the ability to monitor and control internal resources. Buildings can participate in grid interaction operations either as individual units or through aggregation mechanisms such as load aggregators and virtual power plants. Currently, common interaction mechanisms between building regulation behaviors and grid operating status mainly include peak shaving and valley filling based on grid commands. In this type of interaction model, the grid company publishes an ideal target load curve, i.e., a load baseline, based on the grid operating status. The building side adjusts the operating strategies of its internal electrical equipment to make its actual electricity consumption curve at the metering point as close as possible to this baseline. During the effect verification phase, the response power and tracking accuracy of the smart building are calculated by comparing the degree of fit between the actual load curve and the load baseline, which serves as the basis for evaluating the performance of the smart building.

[0030] From the perspective of power grid physical operation, intelligent buildings can directly alter the power flow distribution and voltage state of the distribution system at the metering boundary by flexibly adjusting their power consumption and execution timing. By changing the load injection at nodes to directly affect the power flow of upstream feeders and transformers, the risk of overload and congestion on local lines of the distribution network may be mitigated. By utilizing the impact of power changes on node voltage, the risk of overvoltage or undervoltage exceeding limits in the distribution network under specific operating conditions can be mitigated through reasonable adjustment of load levels. By adding transferable loads such as cold and heat storage and orderly charging of electric vehicles during periods of high renewable energy generation to achieve "source-load matching," the local consumption level and utilization rate of clean energy can be improved. By utilizing the time-shifting characteristics of loads, through methods such as energy storage charging during off-peak hours and load reduction or energy storage discharge during peak hours, the power grid load curve can be effectively smoothed and the peak load pressure of the system can be reduced.

[0031] However, existing smart building evaluation technologies primarily focus on the energy-saving and low-carbon attributes of the building itself. Even when evaluating the interaction between buildings and the power grid, they are often limited to individual building-side indicators such as adjustable capacity and total response. Such evaluation methods are detached from the actual operating conditions of the power grid and cannot establish a coupling relationship between building regulation behavior and the power grid's operating state. In particular, they lack quantitative representation methods for the actual contribution of building access locations and their regulation behavior to key control needs such as alleviating line congestion and voltage exceedances, promoting renewable energy consumption, and coordinated peak shaving and valley filling. Consequently, the power grid cannot verify the interactive effects of buildings' participation in power grid control based on their true regulatory value, and cannot conduct effective power grid operation state control based on building regulation behavior.

[0032] Specifically, there is a commonly used quantitative characterization scheme for building performance based on green building indicators and operational performance. This scheme establishes a calculation model that covers multiple dimensions of indicators, including energy consumption per unit area, carbon emission intensity, renewable energy utilization ratio, equipment energy efficiency level, thermal performance of building envelope, and indoor environmental parameters (such as thermal comfort, air quality, and illuminance). This model is used to deeply characterize and quantitatively analyze the long-term operating level or typical operating conditions of buildings. It is mainly applied to scenarios such as energy efficiency supervision, green building classification certification, and low-carbon operation performance verification on a building life cycle or annual time scale. Its technical logic focuses on characterizing the static energy efficiency level and green attributes of the building itself to guide the energy-saving operation and low-carbon transformation of buildings.

[0033] This method characterizes a building's long-term operational level or typical operating condition performance using indicators such as energy consumption per unit area, carbon intensity, renewable electricity ratio, equipment energy efficiency, building envelope thermal performance, and indoor environmental parameters. Its main applications are energy conservation monitoring, green grading certification, and low-carbon operation verification. Therefore, it focuses on quantifying the building's static energy efficiency and green attributes, making it more suitable for answering whether a building is efficient, low-carbon, and comfortable, completely detached from the dynamic interaction between the building and the power grid. It lacks methods for characterizing the building's flexible adjustment characteristics and cannot quantify the actual technical contribution of the building's connection location to mitigating grid operation risks.

[0034] Another commonly used assessment method focuses solely on building-side adjustability and response performance, treating smart buildings as a set of controllable, flexible resources. It emphasizes the quantitative analysis of the physical response characteristics and performance of resources under controlled conditions. In terms of resource regulation capacity characterization, indicators typically include power regulation depth, response rate, ramp-up speed, maximum sustained regulation time, and resource availability coverage. Regarding response performance evaluation, this type of approach is mostly based on baseline load verification after a specific response event, focusing on assessing the scale of the response power and response reliability. Its core technical characteristic is that the calculation of evaluation indicators depends entirely on the resource's own electricity consumption behavior and physical constraints, without considering the grid topology characteristics of the resource access point, nor distinguishing the differences in contribution to grid operation (such as voltage and power flow) at different electrical locations.

[0035] This method uses power regulation depth, response rate, ramp rate, maximum duration, and available time period coverage to characterize how much regulation is possible, how quickly it can be regulated, and how long it can be regulated. The evaluation results reflect the building's deliverability and performance. However, problems in the distribution network such as power flow congestion, low voltage, overvoltage, and limited renewable energy absorption depend on the actual operating state of the grid and exhibit spatial heterogeneity. Therefore, the same regulation of 1kW or 1kWh will result in different improvements to these problems for different buildings. This scheme, based solely on building-side indicators, is insufficient to support precise grid-side regulation of specific issues.

[0036] Furthermore, another existing analytical method based on the power transfer distribution factor (PTDF) simplifies and abstracts intelligent buildings as a single static load injection item in the power grid topology, focusing on quantifying the marginal impact of node load changes on branch power flow using the PTDF. This method relies solely on the grid admittance matrix and physical topology parameters for calculation, failing to consider the operational constraints, thermal inertia characteristics, and flexible response space of building equipment. It also lacks the multi-dimensional quantification capability for complex control requirements such as node voltage fluctuations, the contribution of renewable energy absorption, and peak shaving and valley filling matching. Its core characteristic is that its evaluation perspective is one-sidedly limited to the physical power flow dimension, lacking a comprehensive indicator system and failing to truly represent the multi-dimensional interactive value of buildings as flexible resources under different operating conditions.

[0037] This method uses PTDF to characterize the marginal impact of nodal injection changes on branch power flow, i.e., a linear mapping from active power injection to power flow, thus the output dimension mainly focuses on power flow-related impacts. The interactive value of buildings as flexible resources typically presents multi-dimensional objectives (in addition to power flow, these include voltage management, improved absorption matching with renewable energy output timing, peak-valley improvement, etc.), and building-side deliverability is affected by internal equipment constraints, thermal inertia, and operational strategies. Therefore, when buildings are only abstracted as static load injections and the evaluation dimension mainly focuses on the marginal impact of power flow, the evaluation dimensions may be insufficient to cover the aforementioned multi-objective requirements, making it difficult to form a unified representation of comprehensive interactive value.

[0038] In building-grid interaction, the grid side is not only concerned with whether a building is energy-efficient, green, or responsive, or the quantity of that response, but also with the extent to which utilizing a building's adjustable resources can improve these constraints under given time periods, grid operating conditions, and constraint types (such as local line / transformer overload risks, node voltage exceedance risks, limited distributed renewable energy absorption, and excessive peak-valley differences). Furthermore, this improvement must be verifiable and retrospectively analyzed. Therefore, quantitative characterization schemes for smart building interaction performance should not be limited to individual building-side response indicators, but should establish a multi-dimensional quantitative system encompassing the quality of building-side resource regulation, the improvement gains from grid-side regulation risks, and the comparability of performance across buildings. This will provide support for the grid to implement differentiated incentives, precise resource selection, and effectiveness verification.

[0039] like Figure 1 As shown, in order to achieve the above objectives, the first aspect of the present invention provides a method for constructing a building interaction resource library for power grid regulation, comprising the following steps: S1. Identify building clusters; S2. For any building in the building cluster, obtain the net power input curve of the power grid and the sub-metering curves of various power generation and consumption units for each interaction day in a preset historical period as building interaction day data, and then obtain the building interaction day data for each building. S3. Determine the power distribution network area of ​​the building cluster, and then obtain the physical topology data and operation constraint data of the power distribution network area for each interaction day in the historical period as power grid interaction day data; S4. Generate building reference day data corresponding to each of the building interaction day data, and generate power grid reference day data corresponding to the power grid interaction day data; S5. Based on the power grid interaction day data, the power grid comparison day data, and the building interaction day data and the building comparison day data of each building, obtain the building-side indicators and power grid-side indicators of each building; S6. Normalize the building-side indicators and the power grid-side indicators of each building to obtain multiple normalized indicators for each building. S7. Obtain the power grid interaction evaluation index for each building based on all the normalized indices of each building; S8. Based on the power grid interaction evaluation index of each building, select several buildings in the building cluster to construct an interaction resource library, which is used for coordinated regulation when power grid control is carried out in the distribution network area.

[0040] The aforementioned method for constructing a building interaction resource library for power grid regulation first obtains daily interaction data and comparison data for buildings on the building side, as well as daily interaction data and comparison data for the power grid on the power grid side, to obtain the actual operating conditions of the power grid's physical operation. This establishes a coupling correlation between building regulation behavior and power grid operating status, providing a data foundation for quantitatively evaluating the building's ability to regulate the power grid. Then, by calculating building-side and power grid-side indicators, several buildings are selected to construct the interaction resource library. This establishes a multi-dimensional quantitative system based on building-side resource regulation quality, power grid-side regulation risk improvement gains, and inter-building performance comparability. This allows the power grid side to select and utilize interactive building resources for coordinated regulation based on the actual regulation value of buildings and the power grid's operating status, significantly improving the targeting and effectiveness of power grid regulation. This provides effective support for the power grid side to verify the interactive effects of buildings' participation in power grid regulation based on their actual regulation value and to conduct effective power grid operating status regulation based on building regulation behavior.

[0041] In one possible embodiment, determining the building cluster includes acquiring individual buildings or building clusters involved in building-grid interaction as a set of evaluation objects. ,common The buildings listed above must be equipped with building automation systems or energy management systems, configured to monitor and control the building's internal heating and cooling loads, distributed energy resources, and energy storage facilities, and have adjustable resources.

[0042] In one possible embodiment, for any building in the building cluster, the net power input curve of the power grid and the sub-metering curves of various power generation and consumption units for each interaction day within a preset historical period are obtained as building interaction day data, thereby obtaining the building interaction day data for each building, including obtaining complete time-series data records of the evaluation object within a preset historical period (e.g., the most recent 30 interaction days). The time resolution of the data is set to... (e.g., 15 minutes or 1 hour), and ensure that this resolution is consistent with the time granularity of the power grid dispatch instructions. Data acquisition includes: Obtain the net power input curve of the power grid at the metering boundary (usually the grid connection point) to characterize the real-time power exchange characteristics between the building and the power grid; Obtain the sub-metering curves for various power generation and consumption units. These curves are generated by aligning and summing the power data of individual units within the same unit type (e.g., air conditioners) along the time axis. The resource categories include at least: Rigid loads: Basic electrical loads that do not participate in flexible adjustment (such as lighting and basic power). Temperature-controlled loads: Environmental control loads with thermal inertia or adjustable power characteristics (such as air conditioners, heat pumps, etc.). Transferable loads: Loads that can be shifted over time while meeting total power consumption constraints (such as circulating water pumps, washing machines, etc.). Energy storage resources: Facilities with bidirectional power regulation capabilities for charging and discharging (such as electrochemical energy storage and electric vehicle charging and discharging piles) require separate acquisition of their charging power and discharging power records. Power generation side resources: building-side distributed power sources (such as photovoltaics, micro wind turbines, and backup generators).

[0043] The data collected above is from the building side, measured by the building side and reported to the power grid side. It includes the metering boundary and the sub-metering curves of various power generation and consumption units. These are all time-series data within the preset time period.

[0044] In one possible embodiment, determining the distribution network area of ​​the building cluster and then obtaining the physical topology data and operational constraint data of the distribution network area for each interaction day in the historical period as grid interaction day data includes obtaining grid-side line impedance parameters, node voltage safety upper and lower limits, transformer / line capacity limits, and output prediction data of distributed photovoltaics in the area, etc., for subsequent construction of a grid operation sensitivity analysis model.

[0045] The physical topology data and operational constraint data of the aforementioned distribution network area for each interactive day during the historical period are collected by the power grid side. The power grid operation itself uses this data, so it is only acquired rather than measured. It can be regarded as externally given during the evaluation process and subsequently used for indicator calculation.

[0046] In addition, to establish a quantitative benchmark for the effectiveness of smart buildings participating in grid interaction, the daily grid interaction data also includes a load baseline. Specifically, the grid-side system first generates a time-series curve, i.e., the load baseline, representing the ideal load trajectory of a specific power supply area during the interaction period, based on the operational objectives of the regional power supply system (including but not limited to renewable energy consumption, peak shaving and valley filling, fluctuation suppression, or emergency supply). This curve is then normalized to meet the constraint that the sum of the values ​​at each time point during the interaction period equals 1, thereby eliminating the influence of differences in the basic loads of different buildings on the waveform similarity calculation.

[0047] To quantify the impact of smart buildings on building-grid interaction, a unique baseline data point needs to be constructed for each interaction day. This baseline must include not only the net power input curve of the grid at the metering boundary, but also the power curves of each of the sub-regulation resources described in step one (such as temperature-controlled loads, energy storage, electric vehicles, etc.). The construction of the baseline day preferentially adopts the weighted combination method of historical similar days. When the historical samples are insufficient to support the calculation, the operational simulation method is used as a supplementary implementation method.

[0048] Therefore, in one possible embodiment, generating building reference day data corresponding to each of the building interaction day data includes: The weighted combination method for historical similar days works as follows: First, historical similar days are selected based on the principle of consistency of multi-dimensional features. Selection dimensions include: day type (must be a weekday, weekend, or similar holiday as the interaction day), operational status (must be consistent with the interaction day in terms of business, production, or holiday status), meteorological conditions (daily maximum / minimum / average temperature deviation controlled within a preset threshold, such as within 3℃), and electricity consumption behavior characteristics. Second, a feature vector containing net input power, sub-resource power, and meteorological data is constructed. The Euclidean distance algorithm is used to calculate the difference between candidate historical days and interaction days. The days are sorted from smallest to largest difference, and the top three historical dates that did not participate in the interaction are selected to form a set of similar days. Finally, the set of similar days is weighted and synthesized. Typically, the three dates with the smallest to largest difference are assigned decreasing weight coefficients (e.g., set to 0.5, 0.3, and 0.2 respectively). The power values ​​for the corresponding time periods of each similar day are multiplied by the weight coefficients and then algebraically summed to generate various power curves for the control day.

[0049] Operational simulation model method: When the number of similar days that meet the conditions is insufficient, a control day is constructed by establishing a building operation simulation model. This model takes the meteorological conditions, day type and operation boundary of the interactive day as input, and, while keeping the building's established operation strategy and comfort constraints unchanged, extrapolates the building's operation process in non-interactive scenarios, and outputs the corresponding net input power and sub-item resource power curves.

[0050] It's important to note that the interaction day is for evaluation purposes and is a real day. To demonstrate the building's interactive effect, a baseline is needed for comparison; this baseline is the control day, which is a hypothetical date. However, the control day has a problem: its building data, power grid data, and meteorological data differ from the interaction day. Therefore, a control day needs to be constructed, and there are two methods. The first method is the weighted combination of historical similar days. By finding historical dates that are close to the interaction day, these are considered similar days. The power data of these similar days are then weighted and combined to obtain the power data for the control day. The second method is the operational simulation model method. This method first requires building an operational simulation model of the building, and then constructing a control day whose external boundaries are consistent with the interaction day. Simulating this model yields the building's power data on the control day.

[0051] It can be assumed that the interactive day is real, while the comparison day is constructed. The meteorological external conditions of the two are very similar, and the latter serves as the comparison benchmark for the former.

[0052] Furthermore, in generating the grid reference day data corresponding to the grid interaction day data, the acquisition of the grid state variables for the reference day cannot be performed using a weighted combination as with building power data. Therefore, a preferred embodiment is to directly use the grid state variables of the similar day with the smallest difference as the grid state variables for the reference day.

[0053] After constructing the control day, it is necessary to determine the weighting coefficients for the interaction day or control day. This is used to characterize the relative importance of each interaction day or control day within the evaluation period, ensuring that the sum of the weights of all interaction days is 1. When no specific demand signal is issued by the grid side, and the importance of each interaction day is not significantly different, all interaction days are assigned equal weight values. When the grid company issues an importance coefficient for a specific date based on grid operation needs (such as critical days for peak summer / winter demand, major supply guarantee days, etc.), the system calculates the weight of each interaction day based on this coefficient to reflect the core position of regulation capability during critical periods in the comprehensive evaluation.

[0054] It should also be noted that there are several potential methods for constructing the building operation simulation model to generate the control day using the simulation model method. In a preferred embodiment, an operation optimization model is constructed, inputting data such as electricity price, demand response incentive price, and external weather conditions, with the objective function of maximizing utility / revenue, and with the operational boundaries of various loads as constraints, such as the SOC constraint of distribution network energy storage. Solving this optimization model yields the power curve. However, this method would require extensive discussion and is not our focus. In another possible embodiment, a simulation model or a digital twin system is constructed, and external data is input into the simulation model or digital twin system to automatically generate the relevant control day power curve data.

[0055] Further, the step of obtaining building-side indicators and grid-side indicators for each building based on the grid interaction day data, the grid reference day data, and the building interaction day data and the building reference day data for each building includes: The response potential index, interaction capability index, interaction effect index, environmental friendliness index, and interaction power cost index of each building are obtained as multiple building-side indicators for each building. The power flow heavy load adjustment sensitivity, overvoltage adjustment sensitivity, undervoltage adjustment sensitivity, renewable energy consumption sensitivity, peak shaving and valley filling sensitivity, and carbon emission reduction sensitivity of each building are obtained as multiple grid-side indicators for each building.

[0056] This implementation establishes a two-sided evaluation system from the building side to the grid side. On the one hand, building-side indicators include response potential, interaction capability, interaction effect, environmental friendliness, and interaction power cost indicators. This quantifies the deliverable physical capabilities, implementation effects, and costs of buildings as flexible resource clusters, addressing the problem that a single type of indicator cannot comprehensively evaluate building interaction performance. On the other hand, the grid-side indicator group includes power flow heavy load adjustment sensitivity, overvoltage adjustment sensitivity, undervoltage adjustment sensitivity, renewable energy absorption sensitivity, peak shaving and valley filling sensitivity, and carbon emission reduction sensitivity. This extends the evaluation perspective from the building itself to the physical operation level of the distribution network, quantifying the actual marginal contribution of building regulation behavior to mitigating different grid operation risks. By simultaneously acquiring these two types of indicators, a coupled correlation between building regulation behavior and grid operation status is established. In particular, it provides a quantitative representation of the actual contribution of building access location and its regulation behavior to key control needs such as alleviating line congestion and voltage exceedances, promoting renewable energy absorption, and coordinated peak shaving and valley filling. This provides a quantitative basis for the grid side to achieve differentiated incentives and precise resource selection.

[0057] Furthermore, the acquisition of response potential indicators, interaction capability indicators, interaction effect indicators, environmental friendliness indicators, and interaction power cost indicators for each of the buildings as multiple building-side indicators for each building includes: The adjustable load percentage, downward adjustment capacity, and upward adjustment capacity of each building are obtained as multiple response potential indicators for each building; The uphill speed threshold, downhill speed threshold, continuous discharge time threshold, and continuous charging time threshold of each building are obtained as multiple interactive capability indicators for each building. The load profile similarity, total load response, load peak-to-valley difference rate, and load fluctuation coefficient of each building are obtained as multiple interaction effect indicators for each building. The carbon emission intensity per unit area and the proportion of renewable energy electricity consumption for each building are obtained as multiple environmental friendliness indicators for each building; The unit response incentive revenue, unit response utility revenue, and unit response regulation cost of each building are obtained as multiple interactive power cost indicators for each building.

[0058] This implementation constructs a clearly structured and physically meaningful quantifiable indicator system. Taking the adjustable load ratio, downward adjustment capacity, and upward adjustment capacity under the response potential indicator as examples, they characterize the adjustment margin that a building can provide under specific conditions from the perspectives of relative proportion and absolute capacity, respectively. The ramp-up speed threshold and continuous charge / discharge time threshold under the interaction capability indicator accurately describe the dynamic characteristics and sustainability of resource response, ensuring that the mobilized resources can meet the dual requirements of rapid grid response and continuous support. The load baseline similarity indicator in the interaction effect indicator is directly related to the grid's control objectives, while indicators such as total load response verify the actual implementation effectiveness. The environmental friendliness and interactive power cost indicators supplement the evaluation of interactive behavior from the perspectives of social benefits and the power cost of building-grid interactive control, respectively.

[0059] like Figure 2 As shown, in a specific embodiment, an indicator framework for the comprehensive evaluation of smart buildings participating in building-grid interaction is established based on the above-mentioned multiple indicators. The indicator framework adopts a two-level structure, including six primary indicators and their corresponding secondary indicator sets, which are used for subsequent indicator value calculation, normalization processing, weight assignment and comprehensive score calculation.

[0060] like Figure 2As shown, the six primary indicators include, as building-side indicators, response potential indicators, interaction capability indicators, interaction effect indicators, environmental friendliness indicators, and interaction electricity cost indicators, as well as, as grid-side indicators, regulation sensitivity indicators. The secondary indicators include ramp-up speed threshold, ramp-down speed threshold, continuous discharge time threshold, continuous charging time threshold, load baseline similarity, total load response, load peak-valley difference rate, load fluctuation coefficient, carbon emission intensity per unit area, renewable energy consumption ratio, unit response incentive benefit, unit response utility benefit, and unit response regulation cost.

[0061] In this embodiment, the uphill speed threshold is specifically the maximum uphill speed, the downhill speed threshold is specifically the maximum downhill speed, the continuous discharge time threshold is specifically the maximum continuous discharge time, and the continuous charging time threshold is specifically the maximum continuous charging time.

[0062] This indicator framework divides the interactive characteristics of smart buildings into two groups: one group consists of building-side indicators that quantify building-side characteristics, describing resource deliverability and interactive performance, including response potential, interactive capability, interactive effect, environmental friendliness, and interactive economy; the other group consists of grid-side indicators that quantify grid-side control sensitivity, describing the degree of impact of resources on critical grid states under different access locations and operating conditions, including six indicators under control sensitivity. This indicator framework is used to characterize the deliverability and interactive performance of building-side resources, as well as the differences in the control value of buildings under different grid operating objectives, thus providing a unified data field and evaluation basis for subsequently constructing the calculation process for building-side indicator values ​​and the calculation process for grid-side control sensitivity.

[0063] After establishing the evaluation index framework, based on the power grid interaction day data, the power grid comparison day data, and the building interaction day data and comparison day data for each building, the building-side indicators and power grid-side indicators for each building are obtained. Based on the given metering boundaries and data, the values ​​of each secondary indicator on the building side are calculated, including all secondary indicators under response potential, interaction capability, interaction effect, environmental friendliness, and interaction economy. The specific calculation process includes the following steps.

[0064] 1. Response potential indicators.

[0065] a) Adjustable load percentage: In the formula: Adjustable load percentage; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; It is a collection of adjustable electrical loads within intelligent buildings, including temperature-controlled loads, transferable loads, energy storage devices, and charging loads for electric vehicles; To regulate resources On the day of comparison Time period The power; Rigid loads within the building On the day of comparison Time period The power consumption.

[0066] It should be noted that the above The superscript RP represents the primary indicator of response potential, and similarly... The superscript RC represents the regulation capability of the primary indicator; its subscripts 1, 2, 3 have no specific physical meaning, but simply indicate which one it is. The meanings of related symbols shown below are all deduced in the same way.

[0067] Adjustable load percentage refers to the ratio of the total power of adjustable electrical loads within a smart building to the total electrical power of the building within the applicable time period defined by the load baseline. It characterizes the relative scale of load resources available for flexible adjustment during grid interaction periods. This indicator reflects the proportion of loads with adjustment potential in the building's electrical structure; a higher value indicates a stronger load base capable of participating in grid regulation during interaction periods, and a more abundant resource pool available for dispatch when the grid requires load transfer or power adjustment. By combining the weighted average with the interaction day weights, this indicator can comprehensively reflect the overall sufficiency of the building's adjustable resources within the evaluation period.

[0068] b) Adjust capacity downwards: in: In the formula: To adjust capacity downwards; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; For smart buildings at the interactive day Time period Maximum downward adjustment capacity; It is a collection of controllable power generation resources within intelligent buildings, including backup generators, distributed photovoltaics, micro wind turbines, energy storage discharge, and electric vehicle discharge power supplies, etc. It is a collection of adjustable electrical loads within intelligent buildings, including temperature-controlled loads, transferable loads, energy storage devices, and charging loads for electric vehicles; To regulate resources On the day of comparison Time period The power; To regulate resources Interactive Day Time period The upper limit of power; To regulate resources Interactive Day Time period The lower bound of the power.

[0069] Downward adjustment capacity refers to the maximum net reduction in input power that a smart building can achieve by reducing electrical load or increasing generation output within the applicable time period defined by the load baseline. It characterizes the building's power reduction potential when the grid needs to lower its load level. This indicator uses the power of the control day as a benchmark, comprehensively considering the lower bound constraint of adjustable electrical load and the upper bound constraint of controllable resources on the generation side, quantifying the building's maximum downward adjustment potential for each time period on the interactive day. A larger downward adjustment capacity indicates a stronger power reduction support capability that the building can provide when the grid faces overload, overvoltage, or needs to reduce peak load, and has a direct potential to contribute to alleviating the grid's supply and demand tension.

[0070] c) Adjust capacity upwards: in: In the formula: To adjust capacity upwards; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; For smart buildings at the interactive day Time period Maximum downward adjustment capacity; It is a collection of controllable power generation resources within intelligent buildings, including backup generators, distributed photovoltaics, micro wind turbines, energy storage discharge, and electric vehicle discharge power supplies, etc. It is a collection of adjustable electrical loads within intelligent buildings, including temperature-controlled loads, transferable loads, energy storage devices, and charging loads for electric vehicles; To regulate resources On the day of comparison Time period The power; To regulate resources Interactive Day Time period The upper limit of power; To regulate resources Interactive Day Time period The lower bound of the power.

[0071] Upward adjustment capacity refers to the maximum net increase in input power that a smart building can achieve by increasing electrical load or reducing power generation output within the applicable time period defined by the load baseline. It characterizes the building's power upsizing potential when the grid needs to increase load levels. This indicator also uses the power of the control day as a benchmark, comprehensively considering the upper limit constraint of adjustable electrical load and the lower limit constraint of controllable resources on the generation side, quantifying the building's maximum upward adjustment potential for each time period on the interactive day. A larger upward adjustment capacity indicates a stronger ability to provide power upsizing support when the grid faces pressure to absorb new energy sources, needs to increase load to balance excess power generation, or needs to increase off-peak load levels. This is of significant value in promoting the local consumption of clean energy.

[0072] 2. Interaction capability indicators.

[0073] a) Climbing speed threshold / Maximum climbing speed: In the formula: Maximum uphill speed; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; Interactive Day The number of time periods requiring a response; It is a collection of controllable power generation resources within intelligent buildings, including backup generators, distributed photovoltaics, micro wind turbines, energy storage discharge, and electric vehicle discharge power supplies, etc. It is a collection of adjustable electrical loads within intelligent buildings, including temperature-controlled loads, transferable loads, energy storage devices, and charging loads for electric vehicles; To regulate resources Interactive Day Time period Maximum downhill climbing speed; To regulate resources Interactive Day Time period Maximum uphill climbing speed.

[0074] Maximum ramp rate refers to the upper limit of the net input power increase rate that a smart building can achieve by adjusting various adjustable resources within the applicable time period defined by the load baseline. It characterizes the building's upward power response agility in grid interaction. This indicator combines the maximum downward ramp rate of controllable resources on the generation side with the maximum upward ramp rate of adjustable electrical load, reflecting the building's ultimate ability to rapidly increase net electrical power per unit time. A higher maximum ramp rate indicates that the building can more quickly execute power increase commands when the grid needs to rapidly increase load to absorb sudden renewable energy output, meeting the grid's technical requirements for rapid response resources.

[0075] b) Downhill speed threshold / maximum downhill speed: In the formula: For maximum downhill / uphill speed, The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; Interactive Day The number of time periods requiring a response; It is a collection of controllable power generation resources within intelligent buildings, including backup generators, distributed photovoltaics, micro wind turbines, energy storage discharge, and electric vehicle discharge power supplies, etc. It is a collection of adjustable electrical loads within intelligent buildings, including temperature-controlled loads, transferable loads, energy storage devices, and charging loads for electric vehicles; To regulate resources Interactive Day Time period Maximum downhill climbing speed; To regulate resources Interactive Day Time period Maximum uphill climbing speed.

[0076] Maximum downhill ramp rate refers to the upper limit of the rate at which a smart building can reduce net input power by adjusting various adjustable resources within the applicable time period defined by the load baseline. It characterizes the building's agility in responding to downward power demands in grid interactions. This indicator combines the maximum uphill ramp rate of controllable resources on the generation side with the maximum downhill ramp rate of adjustable loads, reflecting the building's ultimate ability to rapidly reduce net power consumption per unit time. A higher maximum downhill ramp rate indicates that the building can more quickly execute power reduction commands when the grid experiences an emergency power shortage or needs to rapidly reduce load to ensure system safety, providing timely and effective power support to the grid.

[0077] c) Threshold for continuous discharge time / Maximum continuous discharge time: In the formula: This is the maximum continuous discharge time; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; It is a collection of energy storage devices, including electric energy storage systems and electric vehicles; For energy storage equipment Interactive Day Time period Maximum discharge power; For energy storage equipment Interactive Day Time period Maximum available energy; For energy storage equipment Interactive Day Time period The minimum available energy.

[0078] Maximum continuous discharge time refers to the longest period within the applicable time frame defined by the load baseline during which a collection of energy storage devices within a smart building continuously releases electrical energy to the power grid or the building interior at maximum discharge power. It characterizes the building's discharge endurance in providing continuous power support during grid interactions. This indicator is calculated based on the difference between the maximum and minimum available energy of the energy storage devices at different times on an interaction day, combined with the maximum discharge power. It reflects the longest continuous discharge capacity of the building's energy storage resources under fully satisfied operational constraints. A longer maximum continuous discharge time indicates a longer duration for which the building can stably provide discharge services during periods when the grid requires continuous power support, which is of significant value in ensuring power balance of the grid over longer periods. d) Continuous charging time threshold / maximum continuous charging time: In the formula: For maximum continuous charging time; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; It is a collection of energy storage devices, including electric energy storage systems and electric vehicles; For energy storage equipment Interactive Day Time period Maximum charging power; For energy storage equipment Interactive Day Time period Maximum available energy; For energy storage equipment Interactive Day Time period The minimum available energy.

[0079] Maximum continuous charging time refers to the longest period within the applicable time frame defined by the load baseline for which a collection of energy storage devices within a smart building can continuously absorb electrical energy from the grid or the building itself at maximum charging power. It characterizes the building's charging endurance in absorbing excess power during grid interactions. This indicator is calculated based on the difference between the maximum and minimum available energy of the energy storage devices at different times on an interaction day, combined with the maximum charging power. It reflects the longest continuous charging capacity of a building's energy storage resources under fully satisfied operational constraints. A longer maximum continuous charging time indicates a longer duration for which the building can continuously absorb power during periods of high grid renewable energy generation and when increased load is needed to absorb excess electricity. This has significant value in improving the grid's ability to absorb fluctuating renewable energy.

[0080] 3. Interaction effectiveness metrics.

[0081] a) Similarity of load lines: in: In the formula: For load line similarity; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; For smart buildings at the interactive day Similarity of load guidelines; These are shape parameter coefficients; For smart buildings at the interactive day Time period Net input power; Interactive Day for Power Grid Companies Time period The load baseline is the set of periods requiring a response. The definition above satisfies ; For smart buildings at the interactive day The total net input energy required during the response period, in kWh, is calculated as follows: .

[0082] Load profile similarity refers to the degree of waveform shape matching between the actual net input power curve of a smart building and the load profile published by the power grid within the applicable time period defined by the load profile. It characterizes the accuracy with which the building's actual electricity consumption behavior tracks the ideal load trajectory of the power grid. This indicator eliminates the influence of differences in the basic load scale of different buildings on the comparison results by calculating the weighted similarity between the actual power curve and the load profile at a normalized energy scale, focusing on the consistency evaluation of waveform shape. Higher load profile similarity indicates a better alignment between the building's power regulation strategy and the power grid dispatch objectives, and a better coupling between its regulation behavior and the power grid's operational needs in the time dimension. It is one of the core indicators for measuring the quality of building interaction.

[0083] b) Total load response: in: In the formula: This represents the total load response. The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; For smart buildings at the interactive day Time period Net input power; For smart buildings in comparison day Time period Net input power; For smart buildings at the interactive day Effective response power.

[0084] Total load response refers to the sum of absolute deviations between the actual net input power of a smart building and the baseline power on the control day within the applicable time period defined by the load baseline. It characterizes the total scale of power regulation actually performed by the building during the grid interaction period. This indicator uses the control day as a benchmark in non-interactive scenarios. By comparing the net input power differences between the interactive day and the control day on a time-by-time basis and summing the absolute values, it comprehensively quantifies the total power transfer generated by the building during the regulation process. A larger total load response indicates a larger scale of actual power dispatched by the building in grid interaction, and a more significant contribution to grid regulation. It is a fundamental indicator for measuring the effectiveness of building interaction.

[0085] c) Load peak-valley difference rate: In the formula: This refers to the load peak-to-valley difference rate. The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; For smart buildings at the interactive day Time period The net input power.

[0086] The load peak-valley difference ratio refers to the ratio of the difference between the peak and valley values ​​of the net input power of a smart building to the peak power within the applicable time period defined by the load baseline. It characterizes the fluctuation range and smoothness of the building's load level during the interaction period. This indicator directly reflects the degree of peak-valley difference in the building's electricity consumption curve during the interaction period; the smaller the value, the smoother the building's load curve and the more balanced the power distribution. The load peak-valley difference ratio is an important indicator for evaluating the building's own electricity consumption characteristics. A lower peak-valley difference ratio means that the building's load has a smaller impact on the power grid, which is more conducive to the safe and stable operation of the power grid. It also reflects the positive effect of the building in regulating its own electricity consumption behavior and smoothing power fluctuations.

[0087] d) Load fluctuation coefficient: In the formula: This is the load fluctuation coefficient; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; For smart buildings at the interactive day Time period Net input power; For smart buildings at the interactive day The total net input energy required during the response period, in kWh, is calculated as follows: .

[0088] The load fluctuation coefficient refers to the dispersion of a smart building's net input power relative to its average power within the applicable time period defined by the load baseline. It characterizes the time-series stationarity of the building's load power during the interaction period. This index is calculated by taking the square root of the weighted average of the squares of the differences between the power in each time period and the average power, and then comparing it with the average power to eliminate the influence of dimensions. The smaller the load fluctuation coefficient, the smoother the power change of the building throughout the interaction period, the more stable its electricity consumption behavior, and the smaller the impact on the power grid. This index complements the load peak-valley difference rate, jointly characterizing the smoothness of the building's load curve from different perspectives, and is of reference value for the power grid side in assessing the grid-friendliness of the building.

[0089] 4. Environmental friendliness indicators.

[0090] a) Carbon emission intensity per unit area: In the formula: Carbon emission intensity per unit area; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; Interactive Day The set of all scheduling periods; This refers to the building area, in square meters (m2). They represent the interaction days respectively. Time period The standby generator output power and the gas-converted power; For smart buildings at the interactive day Time period Net input power; These represent the carbon emission factors of standby generators, gas, and regional power grids, respectively.

[0091] Carbon emission intensity per unit area refers to the total carbon emissions generated by a smart building due to electricity consumption and backup power generation within the entire scheduling period of an interactive day, expressed as a unit of building area. It characterizes the overall carbon emission level of a building during grid interaction. This indicator comprehensively considers indirect grid-side carbon emissions corresponding to the building's net input power, direct carbon emissions from backup generators, and carbon emissions from gas consumption. Through weighted calculation of various carbon emission factors, it comprehensively reflects the degree of carbon impact of building operation on the environment. The lower the carbon emission intensity per unit area, the higher the level of decarbonization in the building's overall operation while participating in grid interaction; it is a core quantitative indicator for measuring the environmental friendliness of buildings.

[0092] b) Percentage of electricity consumption from renewable energy sources: In the formula: The proportion of electricity consumption from renewable energy sources; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; Interactive Day The set of all scheduling periods; This refers to the power grid's dispatching time interval; Distributed renewable energy within smart buildings at the interactive day Time period The self-generated and self-consumed power; For the regional power grid on the interactive day Time period The proportion of renewable energy in electricity generation; For smart buildings at the interactive day Time period Net input power; Interactive Day Time period The total power of all electrical loads in the building.

[0093] The renewable energy share refers to the proportion of renewable energy electricity consumed by a smart building during the entire dispatch period on an interactive day, representing the building's actual utilization of clean energy during grid interaction. This indicator comprehensively considers the building's self-generated and self-consumed distributed renewable energy, as well as the renewable energy component in electricity purchased from the grid, fully reflecting the greening level of the building's energy structure. A higher renewable energy share indicates a lower dependence on fossil fuels when meeting its own electricity needs. Its participation in grid interaction not only provides power regulation services but also simultaneously promotes the overall absorption of renewable energy, demonstrating the environmental synergistic benefits of building interaction.

[0094] 5. Interactive electricity cost indicators.

[0095] a) Incentive benefits per unit response: In the formula: Incentive benefits per unit of response; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; This refers to the power grid's dispatching time interval; Interactive Day The unit response incentive price; For smart buildings at the interactive day Similarity of load guidelines; For smart buildings at the interactive day The total net input energy required during the response period, in kWh, is calculated as follows: ; For smart buildings at the interactive day Effective response power.

[0096] Unit response incentive revenue refers to the incentive compensation revenue obtained by a smart building for each unit of effective response electricity in grid interaction, used to characterize the direct economic return level of a building's participation in grid interaction. This indicator is based on the unit response incentive price published by the grid side, and adjusted by incorporating response quality reflected by load profile similarity, allowing buildings with high-quality responses to receive a higher weighting in the unit electricity revenue calculation. Higher unit response incentive revenue indicates more substantial economic compensation the building can obtain during grid interaction, which plays a crucial guiding role in assessing the economic feasibility of building participation in interactive services and incentivizing buildings to continuously optimize their regulation strategies.

[0097] b) Utility gain per unit response: In the formula: Utility benefit per unit response; Interactive Day Time period Changes in the effectiveness of temperature-controlled load; Interactive Day Time period Change in the utility of transferable loads.

[0098] In the formula: Utility benefit per unit response; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; Interactive Day The set of all scheduling periods; The time interval for power grid dispatching; where Let be the utility function of the temperature-controlled load. The utility function for transferable loads; Interactive Day Time period Building indoor temperature For comparison date Time period Building indoor temperature Interactive Day Time period Transferable load power, For the control day s period Transferable load power; For smart buildings at the interactive day Effective response power.

[0099] Unit response utility gain refers to the change in user utility corresponding to each unit of effective response electricity in a smart building's interaction with the power grid. It characterizes the impact of a building's adjusted electricity consumption behavior on user comfort or productivity. This indicator incorporates the non-economic impacts of internal building regulation behavior into the evaluation system by quantifying changes in indoor environmental comfort caused by temperature-controlled load regulation and changes in electricity consumption time-series utility caused by transferable load regulation. A positive and larger unit response utility gain indicates that the building, while fulfilling its grid regulation tasks, has a smaller negative impact on user utility, and may even achieve utility improvement. This reflects the level of synergistic optimization between meeting grid demands and protecting user interests in building regulation strategies.

[0100] c) Unit response control cost: In the formula: Cost per unit of response and control; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The definition of the load baseline or the set of applicable time periods; Interactive Day The set of all scheduling periods; This refers to the power grid's dispatching time interval; The unit power generation cost of the standby generator; For the first day Electricity purchase cost for a given period of time; Interactive Day Time period Change in the power output of the standby generator; Interactive Day Time period Changes in the input power of intelligent buildings.

[0101] Unit response control cost refers to the additional control cost incurred by a smart building for each unit of effective response power in grid interaction. It characterizes the cost incurred by the building in executing grid control commands. This indicator comprehensively considers changes in generation costs due to the activation of standby generators and changes in electricity purchase costs due to changes in electricity purchase strategies, fully reflecting the incremental cost pressure faced by the building during the control process. The lower the unit response control cost, the more economically the building can complete the power control tasks required by the grid, the lower its cost burden in participating in grid interaction, and the stronger its competitiveness. It has positive reference value for the grid side to obtain control resources at a lower cost.

[0102] 6. Calculation of grid-side indicators / control sensitivity indicators.

[0103] Regulation sensitivity aims to quantify the marginal grid benefits brought about by smart buildings' participation in building-grid interaction. These indicators are related to the building's connection location and the grid's operating status, and are used to characterize the unit regulation value of building flexibility resources in alleviating specific grid problems. Regulation sensitivity includes power flow heavy load adjustment sensitivity, overvoltage adjustment sensitivity, undervoltage adjustment sensitivity, renewable energy absorption sensitivity, peak shaving and valley filling sensitivity, and carbon emission reduction sensitivity.

[0104] a) Adjust sensitivity under heavy load: This refers to the reduction in active power flow in heavily loaded branches per unit of effective power regulation in a smart building, used to characterize the effectiveness of the building in alleviating line congestion or overload pressure. It is calculated using the following formula: in: In the formula: Adjust sensitivity for heavy loads; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The set of time periods in which branch overload occurs; Interactive Day Time period The set of heavily loaded branches, with a threshold of 85%; branch road On the day of comparison Time period The meritorious trend; branch road Interactive Day Time period The meritorious trend; For smart buildings at the interactive day Time period Effective regulation of power.

[0105] b) Overvoltage adjustment sensitivity: This refers to the voltage drop at over-limit nodes caused by each unit of effective power regulation in a smart building, used to characterize the voltage regulation value of a building in scenarios with high voltage. It is calculated using the following formula: In the formula: Sensitivity is adjusted for overvoltage. The number of building-grid interaction days; Interactive Day The weights; Interactive Day The set of time periods in which node voltages exceed the warning threshold; Interactive Day Time period The set of nodes whose node voltage exceeds the warning threshold, which can be set to 1.05 pu; For nodes Interactive Day Time period The voltage amplitude; For nodes Interactive Day Time period The voltage amplitude; For smart buildings at the interactive day Time period Effective regulation of power.

[0106] c) Low voltage adjustment sensitivity: This refers to the voltage boost at over-limit nodes resulting from each unit of effective power regulation in a smart building, used to characterize the voltage support value of a building in low-voltage scenarios. It is calculated using the following formula: In the formula: Sensitivity is adjusted for low voltage. The number of building-grid interaction days; Interactive Day The weights; Interactive Day The set of time periods in which node voltages fall below the warning threshold; Interactive Day Time period The set of nodes whose node voltage is below a warning threshold, which can be set to 0.95 pu; For nodes Interactive Day Time period The voltage amplitude; For nodes Interactive Day Time period The voltage amplitude; For smart buildings at the interactive day Time period Effective regulation of power.

[0107] d) Sensitivity to renewable energy consumption: This refers to the increase in renewable energy consumption on the system side resulting from each unit of effective regulation in a smart building. It characterizes the marginal contribution of building regulation behavior to promoting renewable energy consumption. It is calculated using the following formula: In the formula: Sensitivity to new energy consumption; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The set of all scheduling periods; This refers to the power grid's dispatching time interval; For the regional power grid on the interactive day Time period The proportion of renewable energy in electricity generation; For smart buildings at the interactive day Time period Net input power; For smart buildings in comparison day Time period Net input power; For smart buildings at the interactive day Time period Effective regulation of power.

[0108] e) Peak shaving and valley filling sensitivity: This refers to the sum of effective peak-shaving and effective valley-filling power generated by each unit of effective regulation power in a smart building within the peak and valley periods defined by the net load of the power grid. It characterizes the degree of matching between the building's regulation power and the critical peak and valley periods of the power grid. It is calculated using the following formula: In the formula: To improve peak shaving and valley filling sensitivity; The number of building-grid interaction days; Interactive Day The weights; This refers to the power grid's dispatching time interval; Interactive Day The set of peak net load periods; Interactive Day The collection of net load trough periods; For smart buildings at the interactive day Time period Effective regulation of power; For smart buildings at the interactive day Time period Net input power; For smart buildings in comparison day Time period The net input power.

[0109] f) Carbon emission reduction sensitivity: This refers to the reduction in carbon emissions per unit of effective power regulation in a smart building, used to characterize the unit emission reduction benefit of a building's regulatory behavior. It is calculated using the following formula: In the formula: Sensitivity to carbon emission reduction; The number of building-grid interaction days; Interactive Day The weights; Interactive Day The set of all scheduling periods; For smart buildings in comparison day Time period Carbon emissions; For smart buildings at the interactive day Time period Carbon emissions; For smart buildings at the interactive day Time period Effective regulation of power.

[0110] In one possible embodiment, the normalization of the building-side indicators and the power grid-side indicators for each building to obtain multiple normalized indicators for each building includes: after obtaining the original values ​​of all secondary indicators for each smart building, performing normalization on all secondary indicator values ​​to eliminate dimensional differences and achieve comparability across buildings. For any smart building... With any secondary indicator Its original value is denoted as The normalized index value is denoted as For each secondary indicator Determine the upper and lower limits of normalization The upper and lower limits can be determined based on the minimum / maximum values ​​of the samples in the same batch of smart buildings, or by using the reference upper and lower limits published by the power grid or the upper and lower limits of publicly available historical statistical data. Normalization is calculated separately according to the direction of the indicators, divided into positive and negative indicators. Negative indicators include load peak-valley difference rate, load fluctuation coefficient, carbon emission intensity per unit area, and unit response control cost. Positive indicators include all other secondary indicators except for load peak-valley difference rate, load fluctuation coefficient, carbon emission intensity per unit area, and unit response control cost.

[0111] For positive indicators: For contrarian indicators: The normalization result is 0 when it is less than 0 and 1 when it is greater than 1.

[0112] Furthermore, obtaining the power grid interaction evaluation index for each building based on all the normalized indices of each building includes: For any of the buildings, obtain the benchmark index weight of each of the normalized indicators of the building, and then perform a weighted summation of all the normalized indicators based on the benchmark index weight of each of the normalized indicators to obtain the power grid interaction evaluation index.

[0113] This implementation method pre-sets a benchmark index weight for each normalized secondary indicator. Based on the benchmark index weight of each normalized indicator, it performs a weighted summation of all normalized indicators, thereby comprehensively evaluating a building's performance across multiple dimensions, including its response potential, interaction capabilities, regulation effects, environmental attributes, and regulation costs, into a single grid interaction evaluation index. Sorting and filtering based on this single comprehensive index efficiently establishes a normalized interaction resource library applicable to most scenarios without specific regulation needs. When the grid does not identify a problem requiring specific handling, it prioritizes the use of smart buildings from the interaction resource library to participate in building-grid interaction, improving the overall efficiency of building resources participating in grid regulation.

[0114] In one possible embodiment, in normal mode, the system predetermines a set of benchmark indicator weights for normal mode. Its satisfaction The weight of this benchmark index can be determined and fixed as a system parameter through methods such as the analytic hierarchy process (AHP). The "normal mode" refers to the situation where the power grid side has not identified any specific constraint type or operational control objective requiring priority handling.

[0115] Under normal operating conditions, the system calculates the power grid interaction evaluation index for each evaluation object based on the benchmark weight. This is used to characterize its overall performance level in participating in building-grid interaction during the evaluation period, and is calculated as follows: The system follows All smart buildings are sorted from highest to lowest priority to obtain a normal call priority sequence. Based on this sequence, the system constructs a normal resource library. Its contents include the first position in the sorting results. Intelligent buildings, or those with a comprehensive score of not less than Intelligent buildings, among which and This is a configurable parameter. When the power grid does not identify a problem requiring specific handling (such as heavy power flow), the interactive resource library will be invoked first. Smart buildings participate in building-grid interaction.

[0116] Furthermore, obtaining the power grid interaction evaluation index for each building based on all the normalized indices of each building includes: For any of the aforementioned buildings: Obtain the baseline index weight for each of the normalized indices of the building; Several constraint types corresponding to different operating conditions in the power distribution network area are obtained, and then the severity coefficient of each constraint type under a preset power grid dispatch time window is obtained. For any constraint type, based on the severity coefficient and the baseline index weight of each normalized index, the dynamic index weight of each normalized index corresponding to the constraint type is obtained. Then, based on the dynamic index weight of each normalized index, all normalized indices are weighted and summed to obtain the power grid interaction evaluation index corresponding to the constraint type.

[0117] In scenarios where grid interaction resources are abundant and grid-side operational demands may vary across different dates, relying solely on a fixed comprehensive score ranking is insufficient to guarantee the relevance to specific operational constraints. Therefore, this implementation introduces a dynamic weight adjustment mechanism based on the current grid operating conditions and constraint severity. By incorporating constraint types and corresponding severity coefficients, and based on the correlation between each normalized indicator and a specific constraint type, the weights of indicators that contribute significantly to resolving that type of constraint are dynamically increased, generating targeted dynamic indicator weights. The resulting grid interaction evaluation index, obtained through weighted summation, is essentially a goal-oriented priority score that quantitatively characterizes the suitability and regulatory value of the system within the current specific grid risk scenario.

[0118] Furthermore, based on the power grid interaction evaluation index of each building, a number of buildings are selected from the building cluster to construct an interaction resource library. This interaction resource library is used for coordinated regulation during power grid control in the distribution network area, including: For any of the constraint types, based on the power grid interaction evaluation index corresponding to each building for that constraint type, several buildings are selected from the building cluster to construct a dynamic resource library corresponding to that constraint type, thereby obtaining a dynamic resource library for each constraint type, and all the dynamic libraries are used as the interaction resource library; The dynamic resource library is used for coordinated adjustment when power grid regulation is carried out in the distribution network area with the corresponding constraint type as the regulation target.

[0119] In this implementation, based on the power grid interaction evaluation indicators corresponding to different constraint types, a building dynamic resource library for different constraint types is selected and constructed. When the power grid is regulated in the distribution network area with the corresponding constraint type as the regulation target, the smart buildings in the corresponding dynamic resource library can be called first for coordinated regulation. This avoids the problem that a single resource library is not targeted enough when facing complex power grid operation states, and greatly improves the building-power grid interaction's refined support capability for the safe and stable operation and efficient absorption of the distribution network.

[0120] In one specific embodiment, in scenarios where grid interaction resources are abundant and grid-side operational demands may vary across different dates, relying solely on a fixed comprehensive score ranking is insufficient to guarantee the relevance to specific operational constraints. Therefore, based on the comprehensive evaluation results and grid operating status, multiple dynamic resource libraries are constructed for different constraint types, and executable call priorities and call parameter boundaries are output for interactive calls between the grid side or load aggregators / virtual power plants.

[0121] First, constraint type identification and severity quantification are performed. Before the start of each scheduling cycle or interactive execution window, i.e., within the preset power grid scheduling time window, the current constraint type set is identified based on power grid monitoring or forecasting information. The constraint types include at least: branch overload / congestion risk, overvoltage risk, undervoltage risk, renewable energy consumption demand, peak shaving and valley filling demand, and carbon emission reduction demand. Regarding constraint types... Calculate the corresponding severity coefficient. This is used to characterize the urgency of the constraint in the current time period. The severity coefficient is... Including the severity of branch overload / congestion risk. Overvoltage risk type and corresponding overvoltage severity Low voltage risk type and corresponding low voltage severity The severity of new energy consumption demand corresponding to the type of new energy consumption demand Peak shaving and valley filling demand severity corresponding to different demand types And the severity of carbon reduction demand corresponding to the type of carbon reduction demand. .

[0122] If no obvious constraints are identified, it is recorded as the normal mode, and the above-mentioned benchmark index weights are directly used to calculate the power grid interaction evaluation index and construct the interaction resource library. Severity coefficient The calculation can be found below. When deploying the project, the corresponding formula can be selected based on the available monitoring / prediction data.

[0123] Branch road overload severity : Let the power grid dispatch time window be For each time period With each branch road Obtain the branch load rate ( , (These are the branch power flows and their maximum allowable values, respectively). Let the warning threshold be... (If we take 0.85), then the maximum overload amplitude is: The normalized branch overload severity is: Overvoltage severity : Let the power grid dispatch time window be For each time period With each node Obtain its voltage amplitude Let the warning threshold be... (e.g., 1.05 pu), then the maximum over-limit range is: The overvoltage severity is obtained by normalization as follows: In the formula: This is a configurable upper limit for normalization.

[0124] Low voltage severity : Let the power grid dispatch time window be For each time period With each node Obtain its voltage amplitude Let the low voltage warning threshold be (e.g., 0.95 pu), and the maximum over-limit of low voltage be: Normalization yields the low voltage severity as follows: In the formula: This is a configurable upper limit for normalization.

[0125] Severity of New Energy Consumption Demand : The calculation of the proportion of new energy power output is as follows: ( (These are the predicted renewable energy output and power consumption, respectively), and the severity of renewable energy consumption demand is: In the formula: The trigger threshold (e.g., 0.7).

[0126] Peak shaving and valley filling demand severity : Let the power grid dispatch time window be Let the net load within the system be... The severity of peak shaving and valley filling demand is calculated as follows: In the formula: The upper limit is configurable. For each time period The new energy source is contributing power.

[0127] Severity of carbon emission reduction demand : Let the target carbon emission reduction within the scheduling window be... The baseline carbon emissions predicted without additional interactive resources are: The severity of carbon emission reduction demand is calculated as follows: In the formula: This is a configurable upper limit.

[0128] Secondly, for any constraint type, based on the power grid interaction evaluation index corresponding to each building for that constraint type, several buildings are selected from the building cluster to construct a dynamic resource library corresponding to that constraint type, thereby obtaining a dynamic resource library for each constraint type. All of these dynamic resource libraries are then used as the interaction resource library, including dynamically generated index weights for power grid operation and control objectives. For each constraint type... Predefine an indicator correlation vector Used to indicate the secondary indicators Types of control constraints The degree of correlation is determined by taking a higher value (e.g., 1) for the control sensitivity index directly corresponding to the constraint, a medium value (e.g., 0.3) for the building-side index that matches it, and 0 or a smaller value for the rest.

[0129] For any constraint type Generate dynamic index weights under this constraint type. : In the formula, This is a configurable enhancement coefficient used to control the magnitude of the enhancement of the benchmark indicator weight by the regulatory target; when hour, The system predetermines a set of benchmark indicator weights for the normal mode. Its satisfaction .

[0130] Then, the goal-oriented priority score is calculated and a dynamic resource library is constructed. For each constraint type... Calculate intelligent buildings based on dynamic indicator weights Goal-oriented priority score As an indicator for evaluating power grid interaction: In the formula, for any smart building With any secondary indicator , For constraint type The j-th normalized index corresponds to the dynamic index weight. Let j be the j-th normalized index.

[0131] Based on this score, a constraint-oriented type is generated. Dynamic resource libraries and call priority queues: by Sort the sequence from highest to lowest. Take the front Each intelligent building is constrained by type. dynamic resource library Used in the distribution network area with corresponding constraint types When performing power grid regulation to achieve regulatory objectives, this dynamic resource pool should be used first. Intelligent buildings in the system can coordinate and regulate each other.

[0132] In a preferred embodiment, when multiple constraint types need to be adjusted simultaneously, adjustments can be made based on each constraint type. and its severity coefficient Smart buildings Goal-oriented priority score The priority scores of each constraint type are combined according to severity to obtain a comprehensive score. As an evaluation indicator for power grid interaction and ranked accordingly: The system follows All smart buildings are ranked from highest to lowest priority to obtain a comprehensive priority sequence. Based on this sequence, the system constructs an interactive resource library, which contains the top-ranked buildings from the ranking results. Intelligent buildings, or those with a comprehensive score of not less than Intelligent buildings, among which and These are configurable parameters. When multiple constraint types require adjustment simultaneously, the smart buildings in this interactive resource library will be prioritized for building-grid interaction.

[0133] Please refer to Figure 3 The second aspect of the present invention provides a building interactive resource database construction system for power grid regulation, comprising: Object determination module 100 is used to determine building clusters; The building data acquisition module 200 is used to acquire, for any building in the building cluster, the net power input curve of the power grid and the sub-metering curves of various power generation and consumption units for each interactive day in a preset historical period as building interactive day data, and then obtain the building interactive day data for each building. The power grid data acquisition module 300 is used to determine the distribution network area of ​​the building cluster, and then acquire the physical topology data and operation constraint data of the distribution network area for each interaction day in the historical period as power grid interaction day data; The comparison data generation module 400 is used to generate building comparison day data corresponding to each of the building interaction day data, and to generate power grid comparison day data corresponding to the power grid interaction day data; The indicator value acquisition module 500 is used to acquire building-side indicators and grid-side indicators for each building based on the grid interaction day data, the grid reference day data, and the building interaction day data and the building reference day data for each building. The index value normalization module 600 is used to normalize the building-side index and the power grid-side index of each building, thereby obtaining multiple normalized indexes for each building. The indicator value integration module 700 is used to obtain the power grid interaction evaluation index of each building based on all the normalized indicators of each building; The resource library construction module 800 is used to select several buildings in the building cluster to construct an interactive resource library based on the power grid interaction evaluation index of each building. The interactive resource library is used for coordinated regulation when power grid control is carried out in the distribution network area.

[0134] The present invention provides a method and system for constructing a building interactive resource database for power grid regulation, which has at least the following advantages compared to the prior art: First, this invention constructs a two-level evaluation index framework for smart buildings to participate in building-grid interaction, which includes five primary indicators on the building side and control sensitivity indicators on the grid side, to evaluate the comprehensive value of smart buildings participating in grid interaction.

[0135] Based on distribution network topology and operational constraint data, this invention performs power grid calculations under control and interactive operating conditions, identifies risk objects such as heavily loaded branches and voltage over-limit nodes, and quantifies the unit regulation value of buildings under different access locations and operating conditions by using the ratio of risk improvement to effective regulation power.

[0136] Third, after obtaining the normalized values ​​of each secondary indicator, this invention uses benchmark weights to form a normal comprehensive score and call priority. When there are specific constraint types, it introduces severity coefficients and indicator correlation to generate goal-oriented dynamic weights and priority scores, thereby constructing a dynamic resource library and call order oriented to constraint types.

[0137] In summary, when power grid operation objectives change on different dates, a single fixed scoring ranking is insufficient to maintain relevance. This invention explicitly maps the current constraint type of the power grid to evaluation weights and priority scores through severity coefficients and dynamic weighting rules, thereby generating a target-oriented dynamic resource library and calling order, and simultaneously outputting callable boundary parameters. Therefore, the evaluation system can not only be used for post-event verification and horizontal comparison of the coordinated regulation behavior of smart buildings and power grid control, but also form executable and interactive building resource selection and calling strategies before / during power grid control, achieving closed-loop support for differentiated incentives, precise control, and effect verification within the same framework.

[0138] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0139] The term "embodiment" as used herein means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described; however, any combination of these technical features that does not contradict each other should be considered within the scope of this specification.

[0140] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the concept of this application, and these improvements and substitutions should also be considered within the scope of protection of this invention. Therefore, the scope of protection of this application should be determined by the appended claims.

Claims

1. A method for constructing a building interactive resource database for power grid regulation, characterized in that, include: Identify building clusters; For any building in the building cluster, the net power input curve of the power grid and the sub-metering curves of various power generation and consumption units of the building are obtained as building interaction day data for each interaction day in a preset historical period, thereby obtaining the building interaction day data for each building. The power distribution network area of ​​the building cluster is determined, and then the physical topology data and operation constraint data of the power distribution network area for each interaction day in the historical period are obtained as power grid interaction day data. Generate building reference day data corresponding to each of the building interaction day data, and generate power grid reference day data corresponding to the power grid interaction day data; Based on the power grid interaction day data, the power grid comparison day data, and the building interaction day data and the building comparison day data for each building, obtain the building-side indicators and power grid-side indicators for each building; The building-side indicators and the power grid-side indicators of each building are normalized to obtain multiple normalized indicators for each building. The power grid interaction evaluation index for each of the buildings is obtained based on all the normalized indices of each building; Based on the power grid interaction evaluation index of each building, several buildings are selected from the building cluster to construct an interaction resource library, which is used for coordinated regulation when power grid control is carried out in the distribution network area.

2. The method for constructing a building interactive resource database for power grid regulation according to claim 1, characterized in that, The process of obtaining building-side indicators and grid-side indicators for each building based on the grid interaction day data, the grid reference day data, and the building interaction day data and the building reference day data for each building includes: The response potential index, interaction capability index, interaction effect index, environmental friendliness index, and interaction power cost index of each building are obtained as multiple building-side indicators for each building. The power flow heavy load adjustment sensitivity, overvoltage adjustment sensitivity, undervoltage adjustment sensitivity, renewable energy consumption sensitivity, peak shaving and valley filling sensitivity, and carbon emission reduction sensitivity of each building are obtained as multiple grid-side indicators for each building.

3. The method for constructing a building interactive resource database for power grid regulation according to claim 2, characterized in that, The acquisition of response potential indicators, interaction capability indicators, interaction effect indicators, environmental friendliness indicators, and interaction electricity cost indicators for each building as multiple building-side indicators for each building includes: The adjustable load percentage, downward adjustment capacity, and upward adjustment capacity of each building are obtained as multiple response potential indicators for each building; The uphill speed threshold, downhill speed threshold, continuous discharge time threshold, and continuous charging time threshold of each building are obtained as multiple interactive capability indicators for each building. The load profile similarity, total load response, load peak-to-valley difference rate, and load fluctuation coefficient of each building are obtained as multiple interaction effect indicators for each building. The carbon emission intensity per unit area and the proportion of renewable energy electricity consumption for each building are obtained as multiple environmental friendliness indicators for each building; The unit response incentive revenue, unit response utility revenue, and unit response regulation cost of each building are obtained as multiple interactive power cost indicators for each building.

4. The method for constructing a building interactive resource database for power grid regulation according to claim 1, characterized in that, The process of obtaining the power grid interaction evaluation index for each building based on all the normalized indices of each building includes: For any of the buildings, obtain the benchmark index weight of each of the normalized indicators of the building, and then perform a weighted summation of all the normalized indicators based on the benchmark index weight of each of the normalized indicators to obtain the power grid interaction evaluation index.

5. The method for constructing a building interactive resource database for power grid regulation according to claim 1, characterized in that, The process of obtaining the power grid interaction evaluation index for each building based on all the normalized indices of each building includes: For any of the aforementioned buildings: Obtain the baseline index weight for each of the normalized indices of the building; Several constraint types corresponding to different operating conditions in the power distribution network area are obtained, and then the severity coefficient of each constraint type under a preset power grid dispatch time window is obtained. For any constraint type, based on the severity coefficient and the baseline index weight of each normalized index, the dynamic index weight of each normalized index corresponding to the constraint type is obtained. Then, based on the dynamic index weight of each normalized index, all normalized indices are weighted and summed to obtain the power grid interaction evaluation index corresponding to the constraint type.

6. The method for constructing a building interactive resource database for power grid regulation according to claim 5, characterized in that, The step involves selecting several buildings from the building cluster based on the power grid interaction evaluation index of each building to construct an interaction resource library. This interaction resource library is used for coordinated regulation during power grid control in the distribution network area, including: For any of the constraint types, based on the power grid interaction evaluation index corresponding to each building for that constraint type, several buildings are selected from the building cluster to construct a dynamic resource library corresponding to that constraint type, thereby obtaining a dynamic resource library for each constraint type, and all the dynamic libraries are used as the interaction resource library; The dynamic resource library is used for coordinated adjustment when power grid regulation is carried out in the distribution network area with the corresponding constraint type as the regulation target.

7. A building interactive resource database construction system for power grid regulation, characterized in that, include: The object determination module is used to determine building clusters; The building data acquisition module is used to acquire, for any building in the building cluster, the net power input curve of the power grid and the sub-metering curves of various power generation and consumption units for each interaction day in a preset historical period as building interaction day data, and then obtain the building interaction day data for each building. The power grid data acquisition module is used to determine the distribution network area of ​​the building cluster, and then acquire the physical topology data and operation constraint data of the distribution network area for each interaction day in the historical period as power grid interaction day data; The comparison data generation module is used to generate building comparison day data corresponding to each of the building interaction day data, and to generate power grid comparison day data corresponding to the power grid interaction day data; The indicator value acquisition module is used to acquire building-side indicators and grid-side indicators for each building based on the grid interaction day data, the grid reference day data, and the building interaction day data and the building reference day data for each building. The index value normalization module is used to normalize the building-side index and the power grid-side index of each building, thereby obtaining multiple normalized indexes for each building. The indicator value synthesis module is used to obtain the power grid interaction evaluation index of each building based on all the normalized indicators of each building; The resource library construction module is used to select several buildings in the building cluster to construct an interactive resource library based on the power grid interaction evaluation index of each building. The interactive resource library is used for coordinated regulation when power grid control is carried out in the distribution network area.

8. A building interactive resource database construction system for power grid regulation according to claim 7, characterized in that, The process of obtaining building-side indicators and grid-side indicators for each building based on the grid interaction day data, the grid reference day data, and the building interaction day data and the building reference day data for each building includes: The response potential index, interaction capability index, interaction effect index, environmental friendliness index, and interaction power cost index of each building are obtained as multiple building-side indicators for each building. The power flow heavy load adjustment sensitivity, overvoltage adjustment sensitivity, undervoltage adjustment sensitivity, renewable energy consumption sensitivity, peak shaving and valley filling sensitivity, and carbon emission reduction sensitivity of each building are obtained as multiple grid-side indicators for each building.

9. A building interactive resource database construction system for power grid regulation according to claim 8, characterized in that, The acquisition of response potential indicators, interaction capability indicators, interaction effect indicators, environmental friendliness indicators, and interaction electricity cost indicators for each building as multiple building-side indicators for each building includes: The adjustable load percentage, downward adjustment capacity, and upward adjustment capacity of each building are obtained as multiple response potential indicators for each building; The uphill speed threshold, downhill speed threshold, continuous discharge time threshold, and continuous charging time threshold of each building are obtained as multiple interactive capability indicators for each building. The load profile similarity, total load response, load peak-to-valley difference rate, and load fluctuation coefficient of each building are obtained as multiple interaction effect indicators for each building. The carbon emission intensity per unit area and the proportion of renewable energy electricity consumption for each building are obtained as multiple environmental friendliness indicators for each building; The unit response incentive revenue, unit response utility revenue, and unit response regulation cost of each building are obtained as multiple interactive power cost indicators for each building.

10. A building interactive resource database construction system for power grid regulation according to claim 7, characterized in that, The process of obtaining the power grid interaction evaluation index for each building based on all the normalized indices of each building includes: For any of the buildings, obtain the benchmark index weight of each of the normalized indicators of the building, and then perform a weighted summation of all the normalized indicators based on the benchmark index weight of each of the normalized indicators to obtain the power grid interaction evaluation index.

Images