Building controllable load peak-shaving scheduling method and device based on operation intention recognition

Abstract

The application relates to the technical field of load scheduling, in particular to a building controllable load peak-shaving arrangement method and device based on operation intention recognition, wherein the method comprises the following steps: acquiring building total incoming line power, floor or loop sub-item power, air conditioning system power, lighting loop power and socket loop power to determine power consumption data, acquiring building space attributes and equipment basic information to determine space semantic information, then extracting load operation characteristics from a historical power curve and real-time power data of a target building, recognizing a current load operation intention, and then generating a plurality of scheduling groups to generate a peak-shaving execution plan strategy of the scheduling groups according to electricity prices, demand thresholds, power grid demand response instructions, carbon emission intensity or photovoltaic output prediction. Therefore, the problems that, in the related art, load grouping usually depends on static equipment categories, manual configuration or fixed priority tables, leading to unsuitable control strategies, high maintenance costs and limited execution effects are solved.

Description

Technical Field

[0001] This application relates to the field of load scheduling technology, and in particular to a method and apparatus for staggered scheduling of controllable loads in buildings based on operational intent recognition. Background Technology

[0002] Among related technologies, there are methods that classify loads into air conditioning loads, lighting loads, socket loads, charging loads, and other power loads according to equipment category or control interface, and generate control strategies based on electricity prices or demand response commands.

[0003] However, load grouping in related technologies typically relies on static device categories, manual configuration, or fixed priority tables. Grouping by static device categories and fixed priority tables fails to adequately consider the actual operational purpose of the load at different times, easily conflating the adjustability of the same type of equipment in different scenarios. This leads to control strategies that are either overly conservative, missing adjustable potential, or overly aggressive, impacting user experience and business continuity. Manually configured grouping is costly to maintain, and in large buildings with numerous circuits, tenants, and variable space uses, property management personnel struggle to update the controllability attributes of each circuit and device in a timely manner. When tenant schedules change, meetings and activities increase, seasons change, or equipment is updated, the original groupings may become invalid, resulting in a decline in the effectiveness of demand response strategies. Summary of the Invention

[0004] This application provides a method and apparatus for staggered scheduling of controllable loads in buildings based on operational intent recognition, in order to solve the problems in related technologies where load grouping usually relies on static equipment categories, manual configuration or fixed priority tables, resulting in unsuitable control strategies, high maintenance costs and limited execution effects.

[0005] The first aspect of this application provides a method for controllable load staggering in buildings based on operational intent recognition, comprising the following steps: acquiring the total incoming power, floor or circuit sub-item power, air conditioning system power, lighting circuit power, socket circuit power, hot water equipment power, elevator auxiliary system power, and target non-intrusive device-level identification results of the target building to determine the power consumption data of the target building; acquiring the building spatial attributes and equipment basic information of the target building to determine the spatial semantic information of the target building; based on the power consumption data, extracting at least one load operation feature from the historical power curve and real-time power data of the target building, and identifying the current load operation intent of the target building according to at least one load operation feature and the spatial semantic information; generating multiple scheduling groups according to the current load operation intent, and generating staggered execution plan strategies for at least one scheduling group based on electricity price, demand threshold, grid demand response instructions, carbon emission intensity, or photovoltaic output prediction.

[0006] Based on the above technical means, this application embodiment can achieve comprehensive collection of equipment-level power consumption data by acquiring power consumption data, fully restoring the real power consumption conditions of various loads in the building. By acquiring spatial semantic information, it provides scenario support for distinguishing load adjustment restrictions. By extracting load operation characteristics, it can quantify the load power consumption behavior patterns in multiple dimensions, providing sufficient quantitative basis for intent recognition. By combining load operation characteristics and spatial semantic information to identify the current load operation intent, it can accurately determine the load adjustment attributes by integrating power consumption patterns and scenario information, reducing the identification bias caused by single information. By dividing multiple scheduling groups based on the current load operation intent, it can get rid of the limitations of fixed equipment classification and dynamically group according to the real-time load adjustment needs. By combining electricity price, demand threshold, grid demand response instructions, carbon emission intensity or photovoltaic output prediction to generate peak-shaving execution plan strategies for scheduling groups, it can adapt to various energy and grid control scenarios and output a refined peak-shaving solution that takes into account operational safety, user experience and energy-saving benefits.

[0007] Optionally, in one embodiment of this application, obtaining the building space attributes and equipment basic information of the target building to determine the spatial semantic information of the target building includes: obtaining the floors, areas, room uses, tenant types, business hours, personnel density, equipment names, rated power, control interfaces, comfort constraints, business importance, and manual control rules of the target building to determine the building space attributes and the equipment basic information; and generating the spatial semantic information based on the building space attributes and the equipment basic information.

[0008] Based on the above technical means, this application embodiment generates spatial semantic information according to the building space attributes and equipment basic information, which makes it easier to understand whether a certain power circuit belongs to a conference room, office area, computer room, public corridor, kitchen, toilet or equipment room, and provides semantic basis for subsequent judgment of the current load operation intention.

[0009] Optionally, in one embodiment of this application, the step of extracting at least one load operation feature from the historical power curve and real-time power data of the target building includes: generating the historical power curve based on the electricity consumption data, and collecting the real-time power data of the target building; based on the historical power curve and the real-time power data, extracting at least one of the following load operation features: start time, stop time, duration, intraday repeatability, weekday / holiday difference, peak power, average power, load ramp rate, low power standby ratio, correlation with outdoor temperature, correlation with personnel activity, correlation with business hours, sensitivity to electricity price changes, and performance of historical response.

[0010] Based on the above technical means, this application embodiment constructs historical power curves based on the target building's electricity consumption data and simultaneously collects real-time power data. It extracts various load operation characteristics, including start time, stop time, duration, intraday repeatability, weekday / holiday differences, peak power, average power, load ramp rate, low-power standby ratio, correlation with outdoor temperature, correlation with personnel activity, correlation with business hours, sensitivity to electricity price changes, and historical response effects. This can completely and meticulously depict the full-dimensional electricity consumption behavior patterns of the load, providing comprehensive quantitative data support for subsequent integration of spatial semantic information to identify the load's current operating intention. This improves the accuracy and reliability of the operating intention identification results and ensures that subsequent scheduling groups and staggered execution plans based on operating intentions can match the actual electricity consumption scenarios of the building.

[0011] Optionally, in one embodiment of this application, identifying the current load operation intention of the target building based on at least one load operation feature and the spatial semantic information includes: identifying at least one load feature of the target building based on at least one load operation feature and the spatial semantic information; determining the current load operation intention based on at least one load feature, wherein the current load operation intention is a rigid guarantee load operation intention, a comfort maintenance load operation intention, a service support load operation intention, a peak shifting load operation intention, an interruptible load operation intention, or a low-priority standby load operation intention.

[0012] Based on the above technical means, the embodiments of this application identify load characteristics and determine six types of current load operation intentions based on load characteristics: rigid guarantee type, comfort maintenance type, service support type, peak shifting type, interruptible type, and low priority standby type. By complementing and verifying the load operation characteristics in the dimension of electricity consumption behavior and the spatial semantic information in the dimension of spatial scene, it is possible to accurately distinguish the operation attributes and adjustment constraints of different loads, avoid the deviation caused by relying on a single information, and achieve a refined determination of the current load operation intention of each load in the target building. This provides an accurate and reliable classification basis for subsequent division of scheduling groups and hierarchical arrangement of peak shifting execution plans based on the current load operation intention, so that the peak shifting control scheme fits the actual load operation constraints.

[0013] Optionally, in one embodiment of this application, the method further includes: generating a staggered scheduling result based on the staggered execution plan strategy of at least one of the scheduling groups; sending the staggered scheduling result to the corresponding terminal of the target building; collecting the execution feedback result of the corresponding terminal, and generating adaptive learning data of the peak execution plan strategy based on the execution feedback result for adaptive learning.

[0014] Based on the above technical means, the embodiments of this application construct adaptive learning data and complete adaptive learning by collecting execution feedback results generated by real regulation. This can dynamically correct the load operation intention judgment and scheduling group division logic, continuously optimize the peak-shifting execution plan strategy and peak-shifting arrangement results, and eliminate the need for manual continuous maintenance of load adjustable parameters. While reducing manual operation and maintenance costs, it continuously improves the adaptability of peak-shifting regulation and the demand response effect.

[0015] A second aspect of this application provides a building controllable load staggering orchestration device based on operational intent recognition, comprising: an acquisition module, configured to acquire the total incoming power of a target building, the power of each floor or circuit, the power of the air conditioning system, the power of the lighting circuit, the power of the socket circuit, the power of the hot water equipment, the power of the elevator auxiliary system, and the target non-intrusive device-level identification results, to determine the electricity consumption data of the target building, and to acquire the building spatial attributes and basic equipment information of the target building, to determine the spatial semantic information of the target building; an identification module, configured to extract at least one load operation feature from the historical power curve and real-time power data of the target building based on the electricity consumption data, and to identify the current load operation intent of the target building according to at least one load operation feature and the spatial semantic information; and an orchestration module, configured to generate multiple scheduling groups according to the current load operation intent, and to generate a staggered execution plan strategy for at least one scheduling group based on electricity price, demand threshold, grid demand response command, carbon emission intensity, or photovoltaic output prediction.

[0016] Based on the above technical means, this application embodiment can achieve comprehensive collection of equipment-level power consumption data by acquiring power consumption data, fully restoring the real power consumption conditions of various loads in the building. By acquiring spatial semantic information, it provides scenario support for distinguishing load adjustment restrictions. By extracting load operation characteristics, it can quantify the load power consumption behavior patterns in multiple dimensions, providing sufficient quantitative basis for intent recognition. By combining load operation characteristics and spatial semantic information to identify the current load operation intent, it can accurately determine the load adjustment attributes by integrating power consumption patterns and scenario information, reducing the identification bias caused by single information. By dividing multiple scheduling groups based on the current load operation intent, it can get rid of the limitations of fixed equipment classification and dynamically group according to the real-time load adjustment needs. By combining electricity price, demand threshold, grid demand response instructions, carbon emission intensity or photovoltaic output prediction to generate peak-shaving execution plan strategies for scheduling groups, it can adapt to various energy and grid control scenarios and output a refined peak-shaving solution that takes into account operational safety, user experience and energy-saving benefits.

[0017] Optionally, in one embodiment of this application, the acquisition module includes: an acquisition unit, configured to acquire the floors, areas, room uses, tenant types, business hours, personnel density, equipment names, rated power, control interfaces, comfort constraints, business importance, and manual control rules of the target building, so as to determine the building space attributes and the equipment basic information; and a generation unit, configured to generate the spatial semantic information based on the building space attributes and the equipment basic information.

[0018] Based on the above technical means, this application embodiment generates spatial semantic information according to the building space attributes and equipment basic information, which makes it easier to understand whether a certain power circuit belongs to a conference room, office area, computer room, public corridor, kitchen, toilet or equipment room, and provides semantic basis for subsequent judgment of the current load operation intention.

[0019] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to implement the building controllable load staggering scheduling method based on running intention recognition as described in the above embodiments.

[0020] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described building controllable load staggering method based on runtime intent recognition.

[0021] The fifth aspect of this application provides a computer program product, including a computer program that, when executed, implements the above-described method for staggered peak load scheduling of buildings based on operational intent recognition.

[0022] This application's embodiments achieve comprehensive collection of device-level power consumption data by acquiring power consumption data, fully restoring the actual power consumption conditions of various loads in a building. By acquiring spatial semantic information, it provides scenario support for distinguishing load regulation restrictions. By extracting load operation characteristics, it can quantify load power consumption behavior patterns in multiple dimensions, providing sufficient quantitative basis for intent recognition. By combining load operation characteristics and spatial semantic information to identify the current load operation intent, it can accurately determine load regulation attributes by integrating power consumption patterns and scenario information, reducing recognition bias caused by single information. By dividing multiple scheduling groups based on the current load operation intent, it can overcome the limitations of fixed equipment classification and dynamically group according to real-time load regulation needs. By combining electricity price, demand threshold, grid demand response instructions, carbon emission intensity, or photovoltaic output prediction to generate peak-shaving execution plan strategies for scheduling groups, it can adapt to various energy and grid control scenarios, outputting refined peak-shaving solutions that take into account operational safety, user experience, and energy-saving benefits. Thus, it solves the problem in related technologies where load grouping usually relies on static equipment categories, manual configuration, or fixed priority tables, resulting in unsuitable control strategies, high maintenance costs, and limited execution effects.

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

[0024] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart illustrating a method for staggered peak load scheduling of buildings based on operational intent recognition, according to an embodiment of this application. Figure 2 This is a schematic diagram illustrating the identification of operational intent according to an embodiment of this application; Figure 3 This is a schematic diagram of load grouping according to an embodiment of this application; Figure 4 This is a schematic diagram illustrating the generation of a staggered execution plan strategy according to an embodiment of this application; Figure 5 This is a flowchart of a building controllable load staggering method based on operation intent recognition according to an embodiment of this application; Figure 6 This is a block diagram of a building controllable load staggering scheduling device based on operation intent recognition provided in an embodiment of this application; Figure 7 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application.

[0025] Figure label: 60-Building controllable load staggering scheduling device based on operation intention recognition; 100-Acquisition module, 200-Identification module, 300-Scheduling module; 701-Memory, 702-Processor, 703-Communication interface. Detailed Implementation

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

[0027] Commercial buildings, office buildings, mixed-use complexes, and industrial parks typically contain various loads, including air conditioning, lighting, electrical outlets, elevator auxiliary systems, hot water equipment, fresh air systems, water pumps, charging facilities, kitchen equipment, and data centers. These loads differ significantly in terms of energy consumption patterns, comfort impact, start-stop flexibility, interruptibility, and equipment protection requirements. For example, fire protection, security, network communication, data centers, and critical office equipment are generally considered rigid protection loads and cannot be arbitrarily reduced in principle; air conditioning, fresh air systems, hot water systems, and some ventilation equipment have a certain thermal inertia or energy storage capacity, and their power can be moderately advanced, delayed, or reduced without significantly affecting user experience; lighting, landscaping, public area outlets, and some standby equipment are highly adjustable during certain periods; while elevator auxiliary systems, water supply and drainage pumps, and kitchen equipment have obvious time-dependent and service-related characteristics. Therefore, building demand response and peak shaving cannot simply treat all loads as the same type of resource; it is necessary to understand the operational purposes and constraints behind different loads.

[0028] Building energy management systems typically collect data on individual power consumption, total incoming power, air conditioning operating status, and environmental parameters, and execute certain load controls based on electricity prices, demand, or demand response commands. However, in actual buildings, many loads are not connected to refined control systems, or only have limited measurement capabilities such as total incoming power, floor-level individual power consumption, and circuit-level power. Managers often rely on experience to categorize loads into coarse-grained categories such as air conditioning, lighting, power, and sockets, and then manually formulate peak-shaving strategies. This approach has significant shortcomings: firstly, different operating purposes may exist within the same equipment category; for example, air conditioning may serve both comfort maintenance and support for conference rooms, computer rooms, or special areas. Secondly, rigid loads and adjustable loads may be mixed within the same circuit, making it difficult for manual grouping to accurately reflect actual controllable capacity.

[0029] With the development of non-intrusive load analysis, building energy consumption semantic modeling, and demand response technologies, load management based solely on equipment names is no longer sufficient to meet the needs of refined building operation. A more reasonable approach is to identify the "operational intent" of loads from multiple sources of information, including historical power curves, operating periods, durations, start-stop patterns, correlations with temperature or occupancy, correlations with weekdays / holidays, and user comfort feedback. Operational intent refers to the functional purpose and adjustment boundaries of a load or circuit within a specific time period, such as rigid protection, comfort maintenance, service support, peak-shifting operation, interruptible operation, and low-priority standby. Operational intent is not entirely equivalent to equipment category but is closer to a comprehensive description of "why it operates, whether it can be adjusted, how much it can be adjusted, how long it can be adjusted, and what impact adjustment will have."

[0030] In the context of demand response, demand control, and carbon-electricity synergy, buildings need to automatically generate staggered peak load scheduling schemes without affecting critical business operations and basic comfort. Therefore, establishing a method for automatic grouping and staggered peak load scheduling of controllable building loads based on operational intent recognition has significant engineering value.

[0031] Among related technologies, there are methods that classify loads into air conditioning loads, lighting loads, socket loads, charging loads, and other power loads according to equipment category or control interface, and generate control strategies based on electricity prices or demand response commands. For example, peak periods can be identified based on historical electricity consumption data, and load reduction ratios can be set for certain loads; air conditioning setpoints can be increased or scheduled for operation based on indoor temperature constraints; energy storage, hot water, or charging facilities can be operated according to peak-valley electricity prices; and load reduction can be triggered based on demand thresholds. Such methods can reduce peak loads or operating costs to some extent, but their load grouping usually relies on static equipment categories, manual configuration, or fixed priority tables.

[0032] In addition, there are methods that statistically analyze the adjustable power, duration, response speed, and comfort impact of different loads and use this data to assess demand response capabilities. These methods have begun to focus on load adjustability, but they typically emphasize "assessing how much a certain type of load can respond" rather than necessarily considering changes in load operating intent under different scenarios. For example, the same air conditioning circuit might be a comfort-maintaining load during weekday office hours, but a load that can be reduced during low-occupancy nighttime hours; the same hot water unit might be a guaranteed load before service peak hours, but a load that can be shifted during service off-peak hours. Without identifying operating intent, this can easily lead to static grouping results and inefficient peak-shifting strategies.

[0033] In addition, there are methods for scheduling building loads using predictive optimization or model predictive control. These methods typically assume a known set of controllable devices, controllable boundaries, and constraint parameters, and then obtain control commands through optimization solutions. Their drawback is that the set of controllable devices and constraint parameters often require manual input and maintenance, and load grouping and control priorities cannot be updated adaptively in a timely manner when building operating status, tenant activities, seasons, and equipment usage patterns change.

[0034] Therefore, related technologies mainly solve the problem of "how to adjust the predetermined load according to electricity price or demand response instructions," but they still lack a complete closed loop for "how to automatically understand the load operation intention, how to automatically group loads according to the operation intention, how to arrange peak-shaving strategies according to the grouping results, and how to write back and correct the intention recognition results after execution." This application, based on building demand response and load dispatching, further proposes a complete technical solution for operation intention recognition, automatic grouping, peak-shaving arrangement, and execution feedback correction.

[0035] The following describes, with reference to the accompanying drawings, a method and apparatus for staggered scheduling of controllable building loads based on operational intent recognition, according to embodiments of this application.

[0036] Figure 1 This is a flowchart of a building controllable load staggering method based on operation intent recognition, according to an embodiment of this application.

[0037] like Figure 1 As shown, the building controllable load staggering scheduling method based on operation intent recognition includes the following steps: In step S101, the total incoming power of the target building, the power of each floor or circuit, the power of the air conditioning system, the power of the lighting circuit, the power of the socket circuit, the power of the hot water equipment, the power of the elevator auxiliary system, and the target non-intrusive device-level identification results are obtained to determine the power consumption data of the target building. The building space attributes and basic equipment information of the target building are also obtained to determine the spatial semantic information of the target building.

[0038] In actual implementation, the embodiments of this application can obtain the total incoming power of the target building, the power of each floor or circuit, the power of the air conditioning system, the power of the lighting circuit, the power of the socket circuit, the power of the hot water equipment, the power of the elevator auxiliary system, and the target non-intrusive device-level identification results of the target building through smart meters, sub-meters, energy management platforms, building automation systems, edge acquisition gateways, or non-intrusive load monitoring terminals, so as to determine the electricity consumption data of the target building.

[0039] Furthermore, in one embodiment of this application, the building space attributes and equipment basic information of the target building are obtained to determine the spatial semantic information of the target building, including: obtaining the floors, areas, room uses, tenant types, business hours, personnel density, equipment names, rated power, control interfaces, comfort constraints, business importance and manual control rules of the target building to determine the building space attributes and equipment basic information; and generating spatial semantic information based on the building space attributes and equipment basic information.

[0040] In actual implementation, the embodiments of this application can obtain the floor, area, room use, tenant type, business hours, personnel density, equipment name, rated power, control interface, comfort constraints, business importance and manual control rules of the target building to determine the building's spatial attributes and basic equipment information, thereby determining the spatial semantic information of the target building.

[0041] This application embodiment generates spatial semantic information based on building space attributes and equipment basic information, which facilitates understanding whether a certain power circuit belongs to a conference room, office area, computer room, public corridor, kitchen, toilet or equipment room, and provides semantic basis for subsequent judgment of the current load operation intention.

[0042] In step S102, based on electricity consumption data, at least one load operation feature is extracted from the historical power curve and real-time power data of the target building, and the current load operation intention of the target building is identified according to the at least one load operation feature and spatial semantic information.

[0043] The following details how embodiments of this application extract at least one load operation feature from the historical power curves and real-time power data of a target building based on electricity consumption data, and identify the current load operation intention of the target building based on at least one load operation feature and spatial semantic information.

[0044] Specifically, in one embodiment of this application, at least one load operation feature is extracted from the historical power curve and real-time power data of the target building, including: generating a historical power curve based on the electricity consumption data of the target building and collecting real-time power data of the target building; and extracting at least one of the following load operation features based on the historical power curve and real-time power data: start time, stop time, duration, intraday repeatability, weekday / holiday difference, peak power, average power, load ramp rate, low power standby ratio, correlation with outdoor temperature, correlation with personnel activity, correlation with business hours, sensitivity to changes in electricity price, and performance of historical response.

[0045] In actual implementation, this embodiment first generates historical power curves for each circuit, device, and other load object based on the electricity consumption data of the target building, and collects corresponding real-time power data. Then, based on the historical power curves and real-time power data, load operation characteristics are extracted. Load operation characteristics include, but are not limited to, start time, stop time, duration, intraday repeatability, weekday / holiday differences, peak power, average power, load ramp rate, low-power standby ratio, correlation with outdoor temperature, correlation with personnel activity, correlation with business hours, sensitivity to electricity price changes, and historical response effect.

[0046] This application embodiment constructs historical power curves based on the target building's electricity consumption data and simultaneously collects real-time power data. It extracts various load operation characteristics, including start time, stop time, duration, intraday repeatability, weekday / holiday differences, peak power, average power, load ramp rate, low-power standby ratio, correlation with outdoor temperature, correlation with personnel activity, correlation with business hours, sensitivity to electricity price changes, and historical response effects. This can comprehensively and meticulously depict the full-dimensional electricity consumption behavior patterns of the load, providing comprehensive quantitative data support for subsequent integration of spatial semantic information to identify the load's current operating intent. This improves the accuracy and reliability of the operating intent identification results, ensuring that subsequent scheduling groups and staggered execution plans based on operating intent can match the actual electricity consumption scenarios of the building.

[0047] Furthermore, in one embodiment of this application, identifying the current load operation intention of a target building based on at least one load operation feature and spatial semantic information includes: identifying at least one load feature of the target building based on at least one load operation feature and spatial semantic information; determining the current load operation intention based on at least one load feature, wherein the current load operation intention is a rigid guarantee load operation intention, a comfort maintenance load operation intention, a service support load operation intention, a peak shifting load operation intention, an interruptible load operation intention, or a low-priority standby load operation intention.

[0048] like Figure 2 As shown, in actual implementation, this embodiment of the application can first construct a load operation feature vector based on load operation characteristics and spatial semantic information. The expression of the load operation feature vector can be, but is not limited to, as follows: , in, For load objects During the period The load operation characteristic vector, For real-time power, For statistical periodic energy consumption, For duration, To be relevant to business hours, For intraday repetition, For spatial semantic information, To be relevant to personnel activities, To correlate with outdoor temperature, To be related to personnel occupancy.

[0049] The load characteristic set includes rigid guarantee type, comfort maintenance type, service support type, peak shifting type, interruptible type, and low priority standby type. The expression for the load characteristic set can be, but is not limited to, as follows: , in, For load characteristic set, For rigid guarantee type, For comfort maintenance, For service support, It is a shiftable peak type. It is interruptible. This is a low-priority standby mode.

[0050] Furthermore, embodiments of this application can calculate the score of a load object belonging to various load characteristics based on the load operation feature vector. The expression for the load feature score can be, but is not limited to, as follows: , in, For load objects This belongs to the load characteristics The higher the score, the greater the load on the object. The more it matches the load characteristics ; It is the characteristic transformation function; Load characteristics The weight vector; This is a bias term.

[0051] For example, if the load is located in a security, fire protection, or communication area and its operating time highly overlaps with the service guarantee period, the rigid guarantee type scores higher; if the load's power is strongly correlated with indoor temperature, personnel activity, or comfort settings, the comfort maintenance type scores higher; if the load's operating time and power level change synchronously with service activities, the service support type scores higher; if the load's power can be shifted within a certain time range and has a low impact on services, the peak shifting type scores higher; if a short-term power interruption of the load has a low impact on services, the interruptible type scores higher; if the load is in a low-power state for a long time and makes no significant contribution to services, the low-priority standby type scores higher.

[0052] Furthermore, in this embodiment, the load feature category with the highest score is selected, and this category is determined as the identified load feature. Its expression can be, but is not limited to, as follows: , in, To identify load characteristics, To obtain the highest-scoring load characteristic category operation, for , , , , or .

[0053] Furthermore, to avoid low-confidence misgrouping, embodiments of this application calculate an identification confidence level. When the identification confidence level is greater than or equal to a preset confidence threshold, the current load operation intention is determined based on the identified load characteristics. Conversely, when the identification confidence level is lower than the preset confidence threshold, the load is placed into a pending confirmation group and does not directly participate in automatic peak shaving control. The expression for the identification confidence level can be, but is not limited to, as follows: , in, To identify confidence levels, The score corresponding to the identified load characteristics.

[0054] This application's embodiments identify load characteristics and, based on these characteristics, determine six categories of current load operation intentions: rigid guarantee type, comfort maintenance type, service support type, peak shifting type, interruptible type, and low priority standby type. By complementing and verifying the load operation characteristics in the dimension of electricity consumption behavior with the spatial semantic information in the dimension of spatial scene, it is possible to accurately distinguish the operation attributes and adjustment constraints of different loads, avoiding the deviation caused by relying on a single piece of information for judgment. This enables a refined determination of the current load operation intentions of each load in the target building, providing an accurate and reliable classification basis for subsequent division of scheduling groups and hierarchical arrangement of peak shifting execution plans based on the current load operation intentions, so that the peak shifting control scheme fits the actual load operation constraints.

[0055] In step S103, multiple scheduling groups are generated based on the current load operation intention, and at least one scheduling group's peak-shaving execution plan strategy is generated based on electricity price, demand threshold, grid demand response instructions, carbon emission intensity, or photovoltaic output forecast.

[0056] like Figure 3 As shown, in actual execution, this embodiment generates multiple scheduling groups based on the current load operation intention. The scheduling groups include a rigid guarantee group, a comfort maintenance group, a service support group, a peak shifting group, an interruptible group, and a low-priority standby group.

[0057] Each dispatch group includes not only the load object, but also the corresponding adjustable window, maximum power reduction, duration, recovery time, comfort impact level, business impact level, and priority. This grouping can change dynamically over time. For example, the same air conditioning circuit may belong to the comfort maintenance group during office hours but switch to the interruptible group during low-occupancy nighttime hours; the same hot water equipment may belong to the service support group before peak service but to the peak-shifting group during non-service hours.

[0058] In some cases, embodiments of this application may generate a scheduling group based on a type of current load operation intention. During the period The expression for the maximum power reduction can be, but is not limited to, as:

[0059] in, For the scheduling group During the period Maximum power reduction; For load objects During the period The maximum adjustment ratio; For the scheduling group Load objects within .

[0060] like Figure 4 As shown, in this embodiment, based on electricity price, demand threshold, grid demand response command, carbon emission intensity, or photovoltaic output forecast, the weight coefficients corresponding to electricity cost, peak value, and carbon emission in the multi-objective optimization function are adjusted to minimize the overall cost. This is achieved by overlaying upper and lower bound constraints on load regulation and global comfort impact constraints to generate a peak-shifting execution plan strategy for the scheduling group. The expression of the multi-objective optimization function can be, but is not limited to, as follows: , in, For overall cost, For electricity costs, Peak demand cost For carbon emission costs, For the sake of comfort, punishment was imposed. Penalties for impacting business operations This is the weighting factor for electricity costs. This is the weighting factor for peak demand cost. This is a weighting factor for the cost of carbon emissions. The weighting coefficient for the impact of comfort on the penalty. The weighting coefficient for penalties that affect business operations.

[0061] The expressions for the upper and lower bound constraints of load regulation can be, but are not limited to: , in, For load objects During the period The actual adjustment ratio ranges from 0 to 1, where 0 indicates no adjustment and 1 indicates adjustment within the maximum allowable range.

[0062] The expression for the global comfort effect constraint can be, but is not limited to: , in, For load objects The comfort impact coefficient For time period The maximum allowable comfort impact on the budget.

[0063] Off-peak execution strategies include, but are not limited to, short-term reduction plans, recovery plans, early operation plans, and delayed operation plans of the scheduling group.

[0064] In some cases, embodiments of this application configure differentiated weighting coefficients and maximum adjustment ratios for different scheduling groups. When solving for the minimum overall cost, the multi-objective optimization model prioritizes selecting load groups with lower reduction costs, thereby achieving tiered peak-shifting control: it prioritizes reducing low-priority standby and interruptible loads, then schedules peak-shifting loads, only slightly adjusts comfort-maintaining and service-supporting loads when peak pressure is high, and sets the maximum adjustment ratio of rigid guarantee loads to 0, so they do not participate in reduction at all.

[0065] Furthermore, in one embodiment of this application, the method further includes: generating staggered scheduling results based on the staggered execution plan strategy of at least one scheduling group; sending the staggered scheduling results to the corresponding terminal of the target building; collecting the execution feedback results of the corresponding terminal, and generating adaptive learning data of the peak execution plan strategy based on the execution feedback results for adaptive learning.

[0066] In actual execution, this embodiment first generates staggered scheduling results based on the scheduling group's staggered execution plan strategy, and then sends the staggered scheduling results to the corresponding terminals in the target building. Corresponding terminals include, but are not limited to, building automation systems, air conditioning controllers, lighting controllers, socket controllers, hot water equipment controllers, or manual maintenance terminals. If some terminals do not have automatic control interfaces, this embodiment can generate a maintenance operation list or a manual confirmation task.

[0067] Furthermore, this application embodiment collects the execution feedback results from the corresponding terminal and generates adaptive learning data for the off-peak execution plan strategy based on the execution feedback results for adaptive learning. The execution feedback results include, but are not limited to, actual power changes, load rebounds, comfort complaints, equipment alarms, and manual intervention after the off-peak strategy is implemented. The adaptive learning data includes, but is not limited to, actual power reduction, execution success rate, execution feedback score, and correction suggestions.

[0068] In some cases, embodiments of this application can correct the load characteristic score based on the execution feedback score. The correction formula may be, but is not limited to, the following: , in, To correct the score, This is the original score. To implement feedback scoring, Load characteristics Expected feedback baseline To learn step length.

[0069] This application embodiment constructs adaptive learning data and completes adaptive learning by collecting execution feedback results generated by real control. It can dynamically correct the load operation intention judgment and scheduling group division logic, continuously optimize the peak-shifting execution plan strategy and peak-shifting arrangement results, and eliminate the need for manual continuous maintenance of load adjustable parameters. While reducing manual operation and maintenance costs, it continuously improves the adaptability of peak-shifting control and demand response effect.

[0070] like Figure 5 As shown below, the principle of the building controllable load staggering method based on operation intention recognition will be illustrated with a specific embodiment.

[0071] S501 collects electricity consumption data and spatial semantic information of the target building.

[0072] In this embodiment, the total incoming power of the target building, the power of each floor or circuit, the power of the air conditioning system, the power of the lighting circuit, the power of the socket circuit, the power of the hot water equipment, the power of the elevator auxiliary system, and the target non-intrusive device-level identification results are obtained to determine the power consumption data of the target building. The floor, area, room use, tenant type, business hours, personnel density, equipment name, rated power, control interface, comfort constraints, business importance, and manual control rules of the target building are also obtained to determine the building space attributes and basic equipment information, so as to generate spatial semantic information based on the building space attributes and basic equipment information.

[0073] S502, Extract load operation characteristics.

[0074] In this embodiment, historical power curves are generated based on electricity consumption data, and real-time power data of the target building is collected. Then, based on the historical power curves and real-time power data, the following parameters are extracted: start time, stop time, duration, intraday repeatability, weekday / holiday difference, peak power, average power, load ramp rate, low power standby ratio, correlation with outdoor temperature, correlation with personnel activity, correlation with business hours, sensitivity to electricity price changes, and historical response effect.

[0075] S503 identifies load characteristics.

[0076] In this embodiment, based on load operation characteristics and spatial semantic information, a load operation feature vector is first constructed, and then the score of the load object belonging to various load features is calculated based on the load operation feature vector. Subsequently, the load feature category with the highest score is selected and the category is determined as the identified load feature.

[0077] S504, determine the confidence level of recognition.

[0078] In this embodiment of the application, for the identified load characteristics, the identification confidence level is calculated, and it is determined whether the identification confidence level is greater than or equal to a preset confidence threshold. When the identification confidence level is greater than or equal to the preset confidence threshold, the process proceeds to S505, and when the identification confidence level is lower than the preset confidence threshold, the process proceeds to S510.

[0079] S505 generates multiple scheduling groups.

[0080] In this embodiment, the current load operation intention is determined based on the identified load characteristics, and then multiple scheduling groups are generated based on the current load operation intention.

[0081] S506, Generate off-peak execution plan strategy.

[0082] In this embodiment, the electricity price, demand threshold, grid demand response command, carbon emission intensity or photovoltaic output forecast are first obtained, and the weight coefficients corresponding to electricity price, peak value and carbon emission in the multi-objective optimization function are adjusted to minimize the overall cost. The upper and lower bound constraints of load regulation and the global comfort impact constraints are superimposed to generate the peak shifting execution plan strategy of the dispatching group.

[0083] S507, send the staggered scheduling results.

[0084] In this embodiment, the staggered scheduling result is first generated according to the staggered execution plan strategy of the scheduling group, and then the staggered scheduling result is sent to the corresponding terminal of the target building.

[0085] S508, Adaptive Learning.

[0086] In this embodiment, the execution feedback results of the corresponding terminal are first collected, then adaptive learning data for the off-peak execution plan strategy is generated based on the execution feedback results, then an execution feedback score is generated based on the adaptive learning data, and finally the load characteristic score is corrected based on the execution feedback score.

[0087] S509, entering the next cycle.

[0088] In this embodiment, S501 is returned.

[0089] S510 allows load objects to enter the pending confirmation group and not directly participate in automatic peak shaving control.

[0090] According to the building controllable load staggering scheduling method based on operation intention recognition proposed in the embodiments of this application, by acquiring electricity consumption data, comprehensive collection of equipment-level electricity consumption data can be achieved, fully restoring the real electricity consumption conditions of various loads in the building. By acquiring spatial semantic information, scenario support is provided for distinguishing load adjustment restrictions. By extracting load operation characteristics, the load electricity consumption behavior pattern can be quantified in multiple dimensions, providing sufficient quantitative basis for intention recognition. By combining load operation characteristics and spatial semantic information to identify the current load operation intention, the load adjustment attributes can be accurately determined by integrating electricity consumption patterns and scenario information, reducing the identification bias caused by single information. By dividing multiple scheduling groups based on the current load operation intention, the limitations of fixed equipment classification can be overcome, and dynamic grouping can be performed according to the real-time adjustment needs of the load. By combining electricity price, demand threshold, grid demand response instructions, carbon emission intensity or photovoltaic output prediction to generate staggering execution plan strategies for scheduling groups, it can be adapted to various energy and grid control scenarios, and output a refined staggering scheme that takes into account operation safety, user experience and energy-saving benefits. This solves the problem that load grouping in related technologies usually relies on static equipment categories, manual configuration, or fixed priority tables, resulting in unsuitable control strategies, high maintenance costs, and limited execution effects.

[0091] Next, referring to the accompanying drawings, a building controllable load staggering device based on operation intention recognition is described according to an embodiment of this application.

[0092] Figure 6 This is a block diagram of a building controllable load staggering device based on operation intent recognition provided in an embodiment of this application.

[0093] like Figure 6 As shown, the building controllable load staggering scheduling device 60 based on operation intention recognition includes: acquisition module 100, recognition module 200 and scheduling module 300.

[0094] The acquisition module 100 is used to acquire the total incoming power of the target building, the power of each floor or circuit, the power of the air conditioning system, the power of the lighting circuit, the power of the socket circuit, the power of the hot water equipment, the power of the elevator auxiliary system, and the target non-intrusive device-level identification results, so as to determine the power consumption data of the target building, and to acquire the building space attributes and equipment basic information of the target building, so as to determine the spatial semantic information of the target building.

[0095] The identification module 200 is used to extract at least one load operation feature from the historical power curve and real-time power data of the target building based on electricity consumption data, and to identify the current load operation intention of the target building based on at least one load operation feature and spatial semantic information.

[0096] The orchestration module 300 is used to generate multiple scheduling groups based on the current load operation intention, and to generate at least one scheduling group's peak-shaving execution plan strategy based on electricity price, demand threshold, grid demand response instructions, carbon emission intensity or photovoltaic output forecast.

[0097] Optionally, in one embodiment of this application, the acquisition module 100 includes: an acquisition unit and a first generation unit.

[0098] The acquisition unit is used to acquire information such as the floor, area, room usage, tenant type, business hours, personnel density, equipment name, rated power, control interface, comfort constraints, business importance, and manual control rules of the target building in order to determine the building's spatial attributes and basic equipment information.

[0099] The first generation unit is used to generate spatial semantic information based on building space attributes and equipment basic information.

[0100] Optionally, in one embodiment of this application, the identification module 200 includes a second generation unit and an extraction unit.

[0101] The second generation unit is used to generate historical power curves based on electricity consumption data and collect real-time power data of the target building.

[0102] The extraction unit is used to extract at least one of the following load operation characteristics based on historical power curves and real-time power data: start time, stop time, duration, intraday repeatability, weekday / holiday difference, peak power, average power, load ramp rate, low power standby ratio, correlation with outdoor temperature, correlation with personnel activity, correlation with business hours, sensitivity to electricity price changes, and historical response performance.

[0103] Optionally, in one embodiment of this application, the identification module 200 includes an identification unit and a determination unit.

[0104] The identification unit is used to identify at least one load feature of the target building based on at least one load operation feature and spatial semantic information.

[0105] The determining unit is used to determine the current load operation intention based on at least one load characteristic, wherein the current load operation intention is a rigid guarantee load operation intention, a comfort maintenance load operation intention, a service support load operation intention, a peak shifting load operation intention, an interruptible load operation intention, or a low-priority standby load operation intention.

[0106] Optionally, in one embodiment of this application, it further includes: a generation module, a sending module, and a learning module.

[0107] The generation module is used to generate staggered scheduling results based on the staggered execution plan strategy of at least one scheduling group.

[0108] The sending module is used to send the staggered scheduling results to the corresponding terminals in the target buildings.

[0109] The learning module is used to collect the execution feedback results of the corresponding terminal and generate adaptive learning data for the peak execution plan strategy based on the execution feedback results, so as to carry out adaptive learning.

[0110] It should be noted that the foregoing explanation of the embodiment of the building controllable load staggering scheduling method based on operation intention recognition also applies to the building controllable load staggering scheduling device based on operation intention recognition in this embodiment, and will not be repeated here.

[0111] According to the building controllable load staggering scheduling device based on operation intention recognition proposed in the embodiments of this application, by acquiring electricity consumption data, comprehensive collection of equipment-level electricity consumption data can be achieved, fully restoring the real electricity consumption conditions of various loads in the building. By acquiring spatial semantic information, scenario support is provided for distinguishing load adjustment restrictions. By extracting load operation characteristics, the load electricity consumption behavior pattern can be quantified in multiple dimensions, providing sufficient quantitative basis for intention recognition. By combining load operation characteristics and spatial semantic information to identify the current load operation intention, the load adjustment attributes can be accurately determined by integrating electricity consumption patterns and scenario information, reducing the identification bias caused by single information. By dividing multiple scheduling groups based on the current load operation intention, the limitations of fixed equipment classification can be overcome, and dynamic grouping can be performed according to the real-time adjustment needs of the load. By combining electricity price, demand threshold, grid demand response instructions, carbon emission intensity or photovoltaic output prediction to generate staggering execution plan strategies for scheduling groups, it can be adapted to various energy and grid control scenarios, and output a refined staggering scheme that takes into account operation safety, user experience and energy-saving benefits. This solves the problem that load grouping in related technologies usually relies on static equipment categories, manual configuration, or fixed priority tables, resulting in unsuitable control strategies, high maintenance costs, and limited execution effects.

[0112] Figure 7 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. The electronic device may include: The memory 701, the processor 702, and the computer program stored on the memory 701 and capable of running on the processor 702.

[0113] When the processor 702 executes the program, it implements the building controllable load staggering method based on running intent recognition provided in the above embodiments.

[0114] Furthermore, electronic devices also include: Communication interface 703 is used for communication between memory 701 and processor 702.

[0115] The memory 701 is used to store computer programs that can run on the processor 702.

[0116] The memory 701 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0117] If the memory 701, processor 702, and communication interface 703 are implemented independently, then the communication interface 703, memory 701, and processor 702 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized into address buses, data buses, control buses, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0118] Optionally, in a specific implementation, if the memory 701, processor 702, and communication interface 703 are integrated on a single chip, then the memory 701, processor 702, and communication interface 703 can communicate with each other through an internal interface.

[0119] The processor 702 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0120] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for controllable load staggering of buildings based on operational intent recognition.

[0121] This application also provides a computer program product, including a computer program that, when executed, implements the above-described method for staggered peak load scheduling of buildings based on operational intent recognition.

[0122] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0123] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0124] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0125] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0126] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0127] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0128] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0129] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for staggered peak scheduling of controllable building loads based on operational intent recognition, characterized in that, Includes the following steps: The system acquires the total incoming power, floor or circuit power, air conditioning system power, lighting circuit power, socket circuit power, hot water equipment power, elevator auxiliary system power, and target non-intrusive device-level identification results of the target building to determine the power consumption data of the target building. It also acquires the building space attributes and basic equipment information of the target building to determine the spatial semantic information of the target building. Based on the electricity consumption data, at least one load operation feature is extracted from the historical power curve and real-time power data of the target building, and the current load operation intention of the target building is identified according to at least one load operation feature and the spatial semantic information; Multiple scheduling groups are generated based on the current load operation intention, and at least one scheduling group's peak-shaving execution plan strategy is generated based on electricity price, demand threshold, grid demand response instructions, carbon emission intensity, or photovoltaic output forecast.

2. The method according to claim 1, characterized in that, The step of obtaining the building space attributes and basic equipment information of the target building to determine the spatial semantic information of the target building includes: The floor, area, room usage, tenant type, business hours, personnel density, equipment name, rated power, control interface, comfort constraints, business importance, and manual control rules of the target building are obtained to determine the building's spatial attributes and the equipment's basic information. The spatial semantic information is generated based on the building space attributes and the equipment basic information.

3. The method according to claim 1, characterized in that, The extraction of at least one load operation feature from the historical power curves and real-time power data of the target building includes: The historical power curve is generated based on the electricity consumption data, and real-time power data of the target building is collected. Based on the historical power curve and the real-time power data, extract at least one of the following load operation characteristics: start time, stop time, duration, intraday repeatability, weekday / holiday difference, peak power, average power, load ramp rate, low power standby ratio, correlation with outdoor temperature, correlation with personnel activity, correlation with business hours, sensitivity to electricity price changes, and historical response effect.

4. The method according to claim 1, characterized in that, The step of identifying the current load operation intention of the target building based on at least one of the load operation characteristics and the spatial semantic information includes: Based on at least one of the load operation characteristics and the spatial semantic information, identify at least one load characteristic of the target building; The current load operation intention is determined based on at least one of the load characteristics, wherein the current load operation intention is a rigid guarantee load operation intention, a comfort maintenance load operation intention, a service support load operation intention, a peak shifting load operation intention, an interruptible load operation intention, or a low-priority standby load operation intention.

5. The method according to claim 1, characterized in that, Also includes: Generate staggered scheduling results based on the staggered execution plan strategy of at least one of the scheduling groups; The staggered scheduling results are sent to the corresponding terminals in the target building; The execution feedback results of the corresponding terminal are collected, and adaptive learning data for the peak execution plan strategy is generated based on the execution feedback results for adaptive learning.

6. A building controllable load staggering scheduling device based on operational intent recognition, characterized in that, include: The acquisition module is used to acquire the total incoming power of the target building, the power of each floor or circuit, the power of the air conditioning system, the power of the lighting circuit, the power of the socket circuit, the power of the hot water equipment, the power of the elevator auxiliary system, and the target non-intrusive device-level identification results, so as to determine the power consumption data of the target building, and to acquire the building space attributes and basic equipment information of the target building, so as to determine the spatial semantic information of the target building; The identification module is used to extract at least one load operation feature from the historical power curve and real-time power data of the target building based on the electricity consumption data, and to identify the current load operation intention of the target building according to at least one load operation feature and the spatial semantic information; The orchestration module is used to generate multiple scheduling groups based on the current load operation intention, and to generate at least one scheduling group's peak-shaving execution plan strategy based on electricity price, demand threshold, grid demand response instructions, carbon emission intensity or photovoltaic output forecast.

7. The apparatus according to claim 6, characterized in that, The acquisition module includes: The acquisition unit is used to acquire the floor, area, room use, tenant type, business hours, personnel density, equipment name, rated power, control interface, comfort constraints, business importance and manual control rules of the target building, so as to determine the building space attributes and the basic information of the equipment; The generation unit is used to generate the spatial semantic information based on the building space attributes and the equipment basic information.

8. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to implement the building controllable load staggering method based on operation intention recognition as described in any one of claims 1-5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the building controllable load staggering method based on running intent recognition as described in any one of claims 1-5.

10. A computer program product, comprising a computer program, characterized in that, The computer program is executed to implement the building controllable load staggering method based on operation intent recognition as described in any one of claims 1-5.

Images