Super high-rise commercial user-oriented power transaction scheduling device and method, and storage medium

Abstract

The invention provides a super high-rise commercial user-oriented power transaction scheduling device and method and a storage medium, and relates to the technical field of power transaction scheduling processing. According to the super high-rise commercial user-oriented power transaction scheduling device, the basic data layer collects equipment parameters to extract the total load response capability, and the circuit loss is combined to obtain the preliminary net load; the transaction data screening layer screens out tradable net loads according to user signing states and comfort constraints, the transaction decision-making layer calculates the operation and maintenance cost of a signing area and determines the optimal quotation, and the value distribution layer determines income distribution based on the optimal quotation and clarifies settlement results of cold overflow and hot overflow scenes. The method has the advantages that the decentralized function is modularly integrated, the problems of fuzzy load boundary and extensive scheduling of a traditional system are solved, link logic coherence is preserved by means of interlayer data linkage, a super high-rise complex scene is adapted, a clear framework is provided, and the overall collaboration and operability are improved.

Description

TECHNICAL FIELD

[0001] The present application mainly relates to the technical field of power transaction scheduling processing, and in particular relates to a power transaction scheduling device for super high-rise commercial users, a method and a storage medium. BACKGROUND

[0002] With the continuous expansion of super high-rise commercial building scale, there are various complex power consumption load types inside the building, including public area lighting, cooling and heating systems, shop and office building power consumption loads, etc. These loads have the characteristics of diversified ownership, complex space-time distribution and obvious load curve difference, which makes it difficult for traditional load response identification and power scheduling methods to accurately depict and distribute.

[0003] The existing technology has the following outstanding problems: (1) Fuzzy load boundary: public load and private load are often intertwined in the same circuit or system, and it is difficult to identify and schedule them separately; (2) Unfair income distribution: In the existing power market transaction, the energy interaction between contracted users and non-contracted users cannot be quantified, resulting in some users being unable to obtain income matching their contribution; (3) Lack of comfort protection: Non-contracted regional users may passively bear environmental fluctuations during power reduction, lack of comfort protection mechanism, and reduce user acceptance; (4) Insufficient prediction and scheduling accuracy: Traditional load prediction methods are difficult to cope with the characteristics of multi-region, nonlinearity and dynamic change of super high-rise buildings, resulting in large deviation between transaction price and actual response; (5) Lack of adaptive mechanism: The existing system is usually one-way scheduling, lacks feedback learning link, and cannot dynamically correct model parameters according to historical transaction and response data.

[0004] Therefore, there is an urgent need for a power transaction and scheduling device and method that can finely identify and classify the internal load of super high-rise commercial buildings, achieve fair and reasonable income distribution, protect the comfort of non-contracted users, and have adaptive optimization capability. SUMMARY

[0005] The technical problem solved by the present application is to provide a power transaction scheduling device for super high-rise commercial users, a method and a storage medium.

[0006] The technical solution of the present application to solve the above technical problems is as follows: a power transaction scheduling device for super high-rise commercial users, comprising: a basic data layer for collecting equipment parameters of equipment in a commercial building, extracting total load response capacity of the commercial building from the equipment parameters, and obtaining a preliminary net load based on the total load response capacity of the commercial building and circuit loss; a transaction data screening layer, configured to acquire a user subscription state and set a comfort constraint condition, and screen a net load response value for transaction from the preliminary net load based on the user subscription state and the comfort constraint condition; a transaction decision layer, configured to calculate an operation and maintenance cost of a subscription area in combination with a response frequency record of the subscription area, and determine an optimal offer based on the net load response value and the operation and maintenance cost; a value distribution layer, configured to determine a quantitative result of income distribution based on the optimal offer, and determine a corresponding income or settlement result of each subscription user in the commercial building in an over-cooling or over-heating scenario based on the quantitative result of income distribution.

[0007] Another technical solution of the present application to solve the above technical problems is as follows: a power transaction scheduling method for super-high-rise commercial users, applied to a power transaction scheduling device, comprising: acquiring equipment parameters of equipment in a commercial building, extracting a total load response capability of the commercial building from the equipment parameters, and obtaining a preliminary net load based on the total load response capability of the commercial building and circuit loss; acquiring a user subscription state and setting a comfort constraint condition, and screening a net load response value for transaction from the preliminary net load based on the user subscription state and the comfort constraint condition; calculating an operation and maintenance cost of a subscription area in combination with a response frequency record of the subscription area, and determining an optimal offer based on the net load response value and the operation and maintenance cost; determining a quantitative result of income distribution based on the optimal offer, and determining a corresponding income or settlement result of each subscription user in the commercial building in an over-cooling or over-heating scenario based on the quantitative result of income distribution.

[0008] Another technical solution of the present application to solve the above technical problems is as follows: a power transaction scheduling device for super-high-rise commercial users, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the power transaction scheduling method for super-high-rise commercial users as described above.

[0009] Another technical solution of the present application to solve the above technical problems is as follows: a computer readable storage medium, storing a computer program, wherein when the computer program is executed by a processor, the power transaction scheduling method for super-high-rise commercial users as described above is implemented.

[0010] The beneficial effects of the present application are: through the four-layer architecture design of "basic data layer, transaction data screening layer, transaction decision layer, value distribution layer", the full-process closed-loop management of the electricity transaction scheduling of the super-high-rise commercial building is realized; the core advantage lies in the modular integration of the dispersed function modules such as load identification, loss calculation, user state management, quotation optimization and income settlement, which not only solves the problems of fuzzy load boundary and extensive scheduling of traditional systems, but also ensures the logical coherence of "the whole link logic from load potential mining to transactionable load screening, optimal quotation generation and fair settlement" through interlayer data linkage, and at the same time adapts to the complex scene of super-high-rise multi-ownership and multi-user, provides clear framework support for the accurate execution of subsequent links, and significantly improves the overall collaboration and operability of the system. BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1 The structure connection diagram of each functional layer of the electricity transaction scheduling device provided by the embodiment of the present application is provided. Figure 2 The module schematic diagram of each functional layer provided by the embodiment of the present application is provided. Figure 3 The schematic diagram of the processing flow of the load response identification module and the circuit loss evaluation module provided by the embodiment of the present application is provided. Figure 4 The schematic diagram of the processing flow of the user response state management module and the flexible ownership mapping module provided by the embodiment of the present application is provided. Figure 5 The schematic diagram of the processing flow of the multi-target quotation optimization module and the transaction interface module provided by the embodiment of the present application is provided. Figure 6 The schematic diagram of the processing flow of the feedback learning module provided by the embodiment of the present application is provided. DETAILED DESCRIPTION

[0012] The principles and characteristics of the present application are described below in conjunction with the drawings, and the examples are only used to explain the present application and are not used to limit the scope of the present application.

[0013] The prior art lacks fine identification and scheduling framework for complex loads in super-high-rise buildings. Due to the high coupling of building systems, traditional scheduling schemes cannot effectively distinguish between public area loads managed by property management, private loads authorized by users, and systematic loads with shared attributes. This fuzzy load boundary makes it impossible to carry out targeted regulation and control, making it difficult to effectively aggregate and release the flexible load potential hidden in the building.

[0014] This lack of fine-grained differentiation in the mechanism directly leads to difficulties in revenue distribution. In a multi-user environment, if the contribution of each user to load response cannot be accurately calculated, a fair and reasonable revenue distribution cannot be achieved. This not only dampens the participation enthusiasm of subscribed users but may also trigger a "free-rider effect," where non-participating users enjoy the benefits of "public goods" such as grid stability brought about by load regulation without contributing any adjustable load. This further hinders large-scale user participation.

[0015] In highly coupled building systems, indiscriminate load regulation can also negatively impact the comfort of uninvolved or unauthorized users. For example, reducing chiller load in response to market signals may cause an increase in indoor temperature throughout the building, affecting the comfort of uninsured users. Current technologies lack effective "response impact shielding strategies," failing to guarantee the energy comfort of unauthorized users. This not only leads to user complaints but also contradicts the service principles of commercial buildings.

[0016] The purpose of this invention is to propose an adjustable load identification and power trading dispatch system and method for ultra-high-rise commercial users. Through modular design and algorithm optimization, the following objectives are achieved: (1) Transparent load identification: By collecting data from multiple sources and combining LSTM and support vector regression (SVR) for modeling, public, private and systemic loads are distinguished and a clear load mapping relationship is formed based on the contract status; (2) Rationalization of revenue distribution: Establish a response contribution coefficient calculation and two-way overflow cooling / overheating settlement mechanism to ensure fair and reasonable revenue distribution among different ownership and contracted users; (3) Comfort protection standardization: In the scheduling process, dynamic comfort constraints and response impact shielding strategies are introduced to avoid excessive disturbance to non-contracted users, while quantitative compensation is made for load loss in contracted areas; (4) Optimize intelligent decision-making: Construct a multi-objective pricing optimization model that takes into account maximizing benefits, minimizing comfort disturbances, and optimizing response speed to ensure that the system operates at the best balance between economy and acceptability; (5) Adaptive system operation: Through the feedback learning module, the parameters of the prediction model and the optimization model are continuously corrected, so that the system can continuously evolve in long-term operation, improving the accuracy of transactions and overall stability.

[0017] The following detailed description uses several examples.

[0018] Example 1: As Figure 1 As shown, this embodiment of the invention provides a power trading and dispatching device for ultra-high-rise commercial users, comprising: The basic data layer is configured to collect equipment parameters of equipment in the commercial building, extract a total load response capability of the commercial building from the equipment parameters, and obtain a preliminary net load based on the total load response capability and a circuit loss. The transaction data screening layer is configured to obtain a user subscription state, set a comfort constraint condition, and screen a net load response value for transaction from the preliminary net load based on the user subscription state and the comfort constraint condition. The transaction decision layer is configured to calculate an operation and maintenance cost of a subscription area in combination with a response frequency record of the subscription area, and determine an optimal offer based on the net load response value and the operation and maintenance cost. The value distribution layer is configured to determine a quantitative result of income distribution based on the optimal offer, and determine a corresponding income or settlement result of each subscription user in the commercial building in an over-cooling or over-heating scenario based on the quantitative result of income distribution.

[0019] In the above embodiment, the four-layer architecture design of the basic data layer, the transaction data screening layer, the transaction decision layer, and the value distribution layer realizes full-process closed-loop management of power transaction scheduling of the super-high-rise commercial building. The core advantage is that the dispersed function modules of load identification, loss calculation, user state management, offer optimization, and income settlement are modularly integrated, which not only solves the problems of fuzzy load boundary and extensive scheduling of the traditional system, but also ensures the logical coherence of the full-link logic from load potential mining to transactionable load screening, optimal offer generation, and fair settlement through inter-layer data linkage. At the same time, it adapts to the complex scenarios of super-high-rise multi-ownership and multi-user, provides clear framework support for accurate execution of subsequent links, and significantly improves the overall collaboration and operability of the system.

[0020] Preferably, the collection of the equipment parameters of the equipment in the commercial building and the extraction of the total load response capability of the commercial building from the equipment parameters include: The collection of the equipment parameters of the equipment in the commercial building and the extraction of the total load response capability of the commercial building from the equipment parameters include:

[0021] In the above embodiment, the focus is on the accurate extraction of the total load response capability, and the core advantage is that a pre-trained long short-term memory (LSTM) neural network model is used to process the equipment parameters. The LSTM model is good at capturing long-term dependencies in time series data and can effectively mine the seasonal and daily periodic characteristics of equipment energy consumption data such as air conditioners and elevators in super-high-rise commercial buildings. Compared with traditional load prediction methods (such as single regression model), the prediction accuracy of the total load response potential is greatly improved, and the load estimation deviation caused by complex equipment parameters and variable energy consumption modes is avoided, providing high-quality basic data for subsequent net load calculation and transaction decision.

[0022] In particular, as shown in Figure 3 the base data layer includes a load response identification module and a circuit loss evaluation module.

[0023] The load response identification module and the circuit loss evaluation module are used to calculate the net load response value of the commercial building, and the net load response value is the load response potential value of the commercial building. Since not all users in the commercial building are willing to participate in power trading, the net load response value is not the actual load response value of the .

[0024] The load response identification module first collects the operating parameters of all electrical equipment inside the building, including the set temperature and indoor temperature and humidity of the air conditioning system, the fan speed and power, the on-off state and light intensity of the lighting equipment, the operation frequency and load number of the elevator, as well as historical load data and real-time power metering data. These data are uploaded to the central processing system in real time through the sensor network and the Internet of Things gateway. Based on the classification and feature extraction method, the data is divided into load-related data and comfort-related data. In the prediction part, long short-term memory neural network is used to train and predict time series load, and the response potential and time evolution law of each type of load are obtained, and support vector regression model is used to divide the response boundary range of each type of load. Specifically, it includes the following 4 steps: S11, obtaining the physical layer load equipment information of the commercial building: In step S11, the key information of the main load in the commercial building needs to be obtained. Including: each node of the air conditioning system air conditioning pipeline, fan power, air volume, air speed, supply air temperature, fresh air volume, etc., as well as the water pump power, flow, inlet and outlet water temperature of the cooling water pipeline and chilled water pipeline in the system; elevator operation system operation control logic, historical load data, floor elevator distribution topology diagram, etc.; water pump power, inlet and outlet water temperature, water pressure, etc. of the water pump water supply system; lighting and office equipment power; other flexible loads such as washing machine, sterilizer, etc. Load time, working power, etc.

[0025] The above collected information and the building plan of the commercial building are processed to construct a mapping matrix, realizing the mapping relationship between the information layer and the physical layer.

[0026] S12, load response capacity analysis: In the step S12, first, the original data is classified and arranged, and is divided into two categories of load-related data and comfort-related data. The load-related data mainly includes information such as fan power of the air conditioning system, pump power, inlet and outlet temperature of chilled water, temperature and humidity of supply and return air, elevator operation load, lighting loop state, and power of flexible electrical equipment. The data sources cover smart meter curve, building automation system (BAS), elevator group control log, and lighting and office equipment monitoring system.

[0027] In the load prediction and boundary modeling, the support vector regression model and the LSTM network work cooperatively, and are specifically divided into the following four parts: ①Subsystem feature cluster division and graph structure coding: The air conditioning air system, chilled water system, cooling water system, elevator system, water supply system, lighting and flexible device load are respectively divided into feature clusters. For the subsystems with physical or logical topology, such as elevator partition distribution, room air pipe and air pipe level, the graph structure coding method is used to represent the node features (power, flow, temperature) and edge features (valve opening, flow rate, etc.). In addition, for external disturbance and operation state change, lag features and statistical window features such as peak-valley electricity price, weekend / holiday label and meteorological information are introduced to capture the working condition switching characteristics.

[0028] ②Support vector regression modeling and feasible region extraction: support vector regression models are respectively constructed for regional load, fan power, pump power and other key indicators, the data-driven feasible region is extracted, and the upper and lower boundary conditions of each operation indicator are obtained.

[0029] ③LSTM prediction and soft constraint fusion: In the LSTM model, a rolling prediction strategy is adopted, and the feasible region information provided by the SVR is introduced into the loss function to construct a joint loss function of prediction error + soft boundary regularization. The model output is processed by a clipping layer to ensure that the predicted value falls within a reasonable boundary range. The model adopts a shared encoder + separate decoding structure to simultaneously predict the end load and device power, realizing multi-task learning.

[0030] ④Model evaluation and robustness analysis: the prediction accuracy is evaluated using error indicators, and the proportion of out-of-bound predicted values is calculated as the out-of-bound rate indicator. Through time-rolling cross-validation method, the time-varying robustness of the model under different working conditions and time periods is analyzed.

[0031] ⑤Online update and drift monitoring: after deployment, the data distribution drift and working condition change are continuously monitored. When data drift or operation strategy adjustment occurs, the feasible region model of SVR and the LSTM network parameters are dynamically updated to ensure the stability and physical reasonableness of the prediction model in long-term operation.

[0032] The feasible region information learned by the support vector regression model is introduced as a dynamic constraint into the LSTM network, so that the load prediction result has higher stability and physical interpretability while ensuring accuracy.

[0033] For indoor temperature and humidity, air quality, and other comfort-related data, they are introduced as dynamic constraint conditions in step S13 to further modify the load prediction and energy consumption adjustment strategy, ensuring energy saving while meeting the comfort needs of personnel.

[0034] S13, load response optimization model solving: The step S13 is based on the load response capability and adjustment boundary range output by the step S12, and the dynamic constraint conditions formed by combining comfort-related data, to construct and solve an optimization model. The goal of this model is to determine the optimal load response strategy and specific response under the premise of meeting all operation and comfort constraints.

[0035] S14, obtaining the load response : The step S13 is the final output of this module, and the result of the step S13 optimization model solving is the total load response potential that the commercial building can provide under the current conditions . This value will be passed to the subsequent circuit loss evaluation module and flexible ownership mapping module as the basic input of power transaction decision.

[0036] The circuit loss evaluation module is responsible for calculating the power loss under different operating conditions.

[0037] Preferably, based on the total load response capability of the commercial building and the circuit loss, a preliminary net load is obtained, including: Collecting physical parameters of the internal power distribution system of the commercial building, the physical parameters including effective current flowing through the cable, total line resistance, power supply duration, conduction loss of equipment, switching loss of elements, and other loss of equipment; Based on formula one and the physical parameters of the power distribution system, the power loss generated by the cable during operation is calculated, and the formula one is: , Wherein, is the cable loss, ; is the effective current flowing through the cable, A; is the total line resistance, ; is the power supply duration, .

[0038] The total power loss of all power distribution equipment during the load response cycle is calculated based on Formula 2 and the physical parameters of the power distribution system. Formula 2 is as follows: , in, This represents the total electrical energy loss of the equipment. The power loss during conduction for each device. The power consumption is due to the switching on and off of each device. Other power consumption of each device; Based on Formula 3, integrate cable loss and total power loss of equipment The total power loss of the commercial building's power distribution system during the load response cycle is obtained. Formula 3 is as follows: , Based on Formula 4, the total load response capability of the commercial building output by the load response identification module is calculated. Excluding total circuit power loss To obtain the initial net load Formula four is: .

[0039] The core technological advantage of the above embodiment lies in its refined design for circuit loss calculation. This embodiment collects physical parameters of the power distribution system by category, constructs calculation models for cable loss and equipment loss respectively, integrates them to obtain the total loss, and finally obtains the preliminary net load through the logic of "total load response capacity - total loss". This solves the problem of "large deviation between theoretical load potential and actual available load" caused by the traditional system ignoring circuit loss, ensuring that the preliminary net load is more in line with the actual operation of the ultra-high-rise power distribution system, and providing real and reliable basic data support for the subsequent selection of tradable load.

[0040] Preferably, the user's contract status is obtained, and comfort constraints are set, starting from the initial net load. Filter out the net load response value used for trading. ,include: Obtain user subscription status, determine subscribed and non-subscribed regions based on user subscription status, define comfort constraints for non-subscribed regions, and determine non-tradable loads for non-subscribed regions based on comfort constraints. , Formula 5 represents the comfort constraints in the non-contractual area: , in, This is a real-time comfort index for non-contracted areas. This is the minimum permissible comfort standard; The maximum load response value of the contracted area was determined based on historical power data without comfort constraints. And the maximum load response value of the contracted area after considering comfort constraints Based on Formula 6, the maximum load response value and maximum load response value By performing interpolation calculations, we obtain the loss of the contracted area sacrificed to meet comfort constraints. Formula six is: = - , Based on Formula 7, from the initial net load Excluding non-tradable loads in non-contracted areas And losses in the contracted area This yields the net load response value ultimately used for electricity trading. Formula seven is: .

[0041] In the above embodiments, a two-dimensional tradable load screening mechanism based on "user subscription status + comfort constraints" is constructed. On the one hand, the user subscription status distinguishes between subscribed and non-subscribed areas, clarifying the boundaries of load scheduling; on the other hand, the comfort constraint defines the minimum experience standard for non-subscribed areas and quantifies the non-tradable load in non-subscribed areas and the losses in subscribed areas. Finally, the tradable load is obtained through the net load response value. This avoids the problem of "non-subscribed users passively bearing the comfort loss" in traditional scheduling, and ensures that the load entering the trading process is real and effective by accurately deducting non-tradable load, thus taking into account both user experience and trading feasibility.

[0042] like Figure 4 As shown, specifically, the transaction data filtering layer includes a user response status management module and a flexible ownership mapping module.

[0043] The user response status management module is used to implement user response status management at the execution level, while the flexible ownership mapping module is used to clarify the management logic of user response status.

[0044] The user response status management module includes the following steps: S31, Response Authorization Status Registration Submodule: The permission authorization status recording function confirms whether different tenants, owners or property units agree to participate in load response, and records their authorization scope and validity period.

[0045] S32, Enter the comfort protection strategy execution submodule: The system sets minimum comfort thresholds (such as minimum allowable indoor temperature and minimum lighting brightness) based on user comfort protection strategies to ensure that user experience is not significantly degraded during load adjustment. Simultaneously, it monitors comfort indicators in the office areas of non-contracted users in real time to ensure that their energy usage experience is not affected.

[0046] S33, Response Impact Shielding Strategy Scheduling Execution: Based on the strategy of the flexible ownership mapping module, specific physical isolation measures are implemented to ensure that load response-related instructions only apply to the contracted areas.

[0047] The physical isolation measures include: ① In conjunction with the building floor division, setting up zone control points in vertical shafts or power distribution rooms to achieve layered isolation and zoned scheduling, avoiding cross-floor return air or power supply coupling; ② The control logic of all valves and switches is bound to the flexible ownership mapping module in real time, and uncontracted areas do not receive any load response commands, and changes in the contracted status will automatically update the control permissions.

[0048] S34. Assessment of load response contribution capacity of each contracted region: The load response contribution capacity of each contracted area is determined by the load response contribution coefficient of each contracted area. The load response contribution coefficient of each contracted area is determined. The response load capacity, load response speed, and execution accuracy were taken into account.

[0049] S35, Response Frequency Recording: Record the number of times and the time of each contracted region's participation in load response, which will be used to calculate the response maintenance cost of each contracted region in the subsequent multi-objective pricing optimization module. .

[0050] Specifically, it is necessary to count the number of times each contracted region participates in load response within different scheduling cycles. and the start and end times and duration of each response. (No. (Response duration). In addition, the distribution characteristics of the response time periods, such as peak periods, flat periods, and off-peak periods, are recorded to facilitate subsequent assessment of their impact on equipment lifespan and maintenance costs. Furthermore, load response values ​​for each contracted region during the load response period are collected. The effect of reaction response intensity on equipment wear.

[0051] S36. Aggregate the net load response values ​​of all contracted areas and participate in the unified bidding and transaction clearing of the electricity market through the transaction interface.

[0052] The flexible ownership mapping module includes the following steps: S41. Classification of Net Load Response Values: In commercial buildings with multiple stakeholders, net load response values ​​need to be categorized across multiple dimensions to ensure the controllability and transparency of subsequent zoning management and electricity trading. Therefore, net load response values ​​are categorized based on ownership and contract status. Based on ownership, net load is divided into public load, private load, and shared systemic load. Based on contract status, net load is divided into contracted area net load and non-contracted area net load.

[0053] The aforementioned public loads are managed centrally by the property management company, such as public lighting, elevator group control, water pump group control, and central air conditioning units. These loads have centralized dispatch characteristics, large response potential, and are usually traded as a whole in the electricity market.

[0054] These private loads are managed individually by commercial users, such as shop air conditioning, electrical equipment, and display lighting. These loads are highly decentralized and differentiated, requiring unified management through user-side willingness modeling and contractual constraints.

[0055] The shared systemic loads refer to loads that cross the boundaries between public and private sectors, such as heating and cooling systems and losses in main power supply and distribution lines. These loads have a systemic impact on overall operation and require unified allocation and control at the system level.

[0056] The net load of the contracted area refers to the area that has signed a contract with the power trading platform or load aggregator. Its load response value can be directly entered into the trading mechanism and has reliable execution constraints.

[0057] The net load of non-contracted areas refers to areas that have not signed response contracts. Their response behavior does not have market effect, but may be indirectly affected by public load dispatching, and needs to be screened and corrected.

[0058] S42. Zoning and hierarchical division of commercial building areas: First, create spatial mapping relationships for high-rise buildings and perform grid-based processing on physical or logical spaces.

[0059] Then, regions are divided according to the load response value classification method and marked to achieve one-to-one mapping.

[0060] S43. Response Impact Shielding Strategy: First, define dynamic comfort constraints for non-contracted areas. This ensures that the area remains within acceptable comfort levels even when indirectly affected by public load dispatching.

[0061] Then, it was clarified that load reductions should be implemented in non-contracted areas. and the response load loss value of the contracted area due to the need to meet constraints. .

[0062] The non-contractual area reduced load This refers to the response electricity that cannot be included in the transaction due to non-contractual area constraints.

[0063] The contracted area response loss This refers to the amount of electricity supplied in contracted areas that is sacrificed to meet comfort constraints in non-contracted areas.

[0064] The Data-driven approximation calculations are used to perform regression / curve fitting on historical data: ① Calculate the real-time comfort index feature matrix for each event and label it as follows. ; ② Collect historical power data and label it. The label is: (Maximum power reduction observed at the comfort boundary) and (Maximum power reduction when there are no boundaries or the boundaries are greatly relaxed); ③ Calculate based on the feature matrix after the label = - ; ④ Fit the function and define the real-time comfort index. and The relationship determines the fitting function. .

[0065] Subsequently, the net load response value for electricity trading is obtained according to Formula 7.

[0066] Preferably, the operation and maintenance cost of the contracted area is calculated by combining the response frequency records of the contracted area, based on the net load response value. Determine the optimal quote based on operation and maintenance costs, including: The response duration, number of responses, and response load value of the contracted area are obtained from the response frequency records. The operation and maintenance cost of the contracted area is calculated based on Formula 8 and the response frequency records of the contracted area. Formula 8 is: , in, For the first The response and maintenance costs for each contracted region The maintenance unit price corresponding to the response time. For the recorded number The first of the contracted areas The duration of each response. The maintenance unit price is the number of start-stop cycles. For the recorded number Number of responses in each contracted region The additional maintenance unit price resulting from the load response intensity, For the recorded number The first of the contracted areas The load response value of the second response. For the first Fixed maintenance costs for each contracted area; Construct a multi-objective optimization function that maximizes commercial user benefits, minimizes comfort disturbances, and optimizes response speed. Equation 9 is the multi-objective optimization function: , in, The normalization factor for each term can be determined using the entropy weight method because different terms have different dimensions. For the first Market transaction prices for a given period of time For the first Load response of commercial buildings participating in market transactions during a given time period This is a comfort disturbance term, which can be represented by a quadratic deviation of comfort indices such as indoor temperature or illuminance. For the first The response and maintenance costs for each contracted region For response speed; Based on the premise that the total transaction revenue is not less than the total operation and maintenance cost, a constraint is set, which is: , The optimal price was obtained by solving the multi-objective optimization function and constraints using MATLAB.

[0067] like Figure 5 As shown, specifically, the transaction decision layer includes a multi-objective pricing optimization module.

[0068] The multi-objective pricing optimization module includes the following steps: S51. Construct a multi-objective optimization function: First, calculate the response maintenance cost for each region based on the response frequency records in the user response status management module. .

[0069] S52. Specify the constraints: Constraints on load response for electricity trading: ① Constrain the load response before clearing, and each contracted area within a certain time period. Application volume No more than available capacity: .in, It takes into account the equipment's rated capacity, comfort level, and adjustable capacity after non-contractual overflow. .

[0070] ② Constrain the load response after clearing: .in, For market acceptance, .

[0071] Electricity market clearing price constraints: Aggregators are required to ensure that total revenue covers maintenance costs. .

[0072] Various load response time constraints: Set a minimum response time constraint; after participating in an event, the response time must be maintained for the shortest possible duration. : ,in It is a binary variable used to represent time periods. Is it in an "on response" state?

[0073] Output constraints of various equipment: ; Slope constraints for various types of equipment: Contract / non-contractual region restrictions, setting non-contractual regions as... ; Overall response constraints: ,in This is the upper limit for safe power distribution.

[0074] S53. Obtain the optimal bid volume and electricity trading price for each time period.

[0075] Build the relevant model in MATLAB or other specialized software, and use intelligent algorithms or commercial solvers such as GUROBI and CPLEX to complete the solution.

[0076] In the above embodiments, the operation and maintenance costs of the contracted area are first accurately calculated by combining parameters such as response time, frequency, and intensity, which solves the problem of "insufficient revenue coverage" caused by the traditional bidding ignoring equipment loss costs. Then, by constructing a multi-objective optimization function that includes "maximizing revenue, minimizing comfort disturbance, and optimizing response speed" and setting the constraint condition of "total revenue ≥ total operation and maintenance cost", compared with the single-objective bidding model, it can find the optimal balance between economy, user experience and response efficiency. At the same time, the scientific nature of the optimization results is ensured by using MATLAB tools and professional solvers, providing a more competitive and risk-controllable bidding scheme for electricity market transactions.

[0077] Preferably, the quantitative result of revenue allocation is determined based on the optimal price, and the revenue or liability settlement result for each contracted user in the commercial building under the scenarios of overflow cooling and overflow heating is determined based on the quantitative result of revenue allocation, including: Based on the optimal bid, the contribution coefficient of each contracted region is quantified and used as the basis for revenue distribution. The contribution coefficient is: , in, For the first Load response contribution coefficient of each region For adjustable capacity factor, for the first Net load response value of each region The ratio of its baseline load, The response speed factor is the ratio of the system's response speed after receiving an instruction. The accuracy factor is the ratio of the actual response load to the dispatch response command. , and These are the weight coefficients of each factor, and they satisfy... + + ; Formula 10 is constructed based on the heat transfer principle of the building envelope. Formula 10 is then used to analyze adjacent... The heat transfer in the region is calculated using Formula 10: , in, For the first Time period, from the Region to the first Heat transfer through the building envelope in the area The heat transfer coefficient of the building envelope. For contact area, For the first Time period Indoor temperature in the area For the first Time period Indoor temperature in the area Formula 11 is constructed based on the principle of heat transfer caused by infiltration and gap air. Formula 11 is then used to analyze adjacent... The calculation of heat transfer in the region is given by Formula 11: , in, For the first Time period, from the Region to the first Heat transfer in the region occurs through infiltration and gap airflow. For the first The infiltration air volume for a given period is calculated based on pressure difference, door opening status, or duct operating conditions. The isobaric specific heat capacity of air. For the first Time period Indoor temperature in the area For the first Time period Indoor temperature in the area Based on Formula Twelve, heat transfer... and heat transfer Summing gives the total heat flow. Formula 12 is: , Based on total heat flow Determine and define the symbols for cooling and heating benefits. During the cooling season: If and This indicates that the cooling capacity comes from Regional flow area, income, During the heating season: If and This indicates that heat comes from Regional flow area, income; Based on Formula Thirteen, the total heat flow The cumulative cooling and heating load is obtained by calculating the net cooling and heating overflow during the statistical period, as described in Formula Thirteen: , in, Settlement period From the first Region to the first The cumulative heat and cold passing through the region, For the settlement period, The determination coefficient is when Indicates the determination of cooling revenue. Indicates the determination of heating revenue; Based on formulas fourteen and fifteen, accumulated cold and heat The overflow energy is converted into equivalent electrical energy and monetary value for settlement of cold and heat overflow compensation. Formula fourteen is as follows: , in, For from the first Region to the first The region utilizes the equivalent electrical energy of cooling and heating. For the first The performance coefficient of the air conditioning system during different time periods. Formula 15 is as follows: , in, For from the first Region to the first The price of the equivalent electrical energy for cooling and heating in a region. For the first Electricity price during specific time periods Through bilateral settlement and the price of equivalent electricity Determine the corresponding revenue or settlement results for each contracted user in a commercial building under the scenarios of overflow cooling and overflow heating.

[0078] like Figure 5 As shown, the value allocation layer includes a transaction interface module.

[0079] The transaction interface module includes the following steps: S61. Enter the market trading submodule.

[0080] It connects with the electricity spot market, ancillary services market, or virtual power plant platform to complete identity registration and completes transaction bidding based on the electricity trading price obtained by the multi-objective pricing optimization module.

[0081] S62. Enter the cost-sharing and settlement management submodule.

[0082] The total revenue after electricity trading is calculated based on the load response contribution coefficient of each contracted area obtained from the user response status management module. The specific revenue value for contracted users is calculated. Then, using a two-way overflow cooling / overflowing heat identification and pricing settlement mechanism, the revenue is determined to the affected entity based on the measured and model-calculated heat flow direction, realizing calculations such as compensation for the beneficiary and return of revenue to the affected entity.

[0083] The specific details of the bidirectional cold / heat overflow identification and pricing settlement mechanism are as follows: In multi-tenant high-rise commercial buildings operating under cooling (or heating) conditions, adjacent areas may experience non-contractual spillover of cold / heat due to heat transfer from the building envelope, air infiltration through door gaps / pipe shafts, and the redistribution of airflow from the ceiling return air and shared ducts. To ensure fairness in transactions, this embodiment does not pre-determine the beneficiary; instead, it determines the beneficiary and the injured party based on the measured and model-calculated net heat flow direction, thereby achieving compensation for the beneficiary and a return of benefits to the injured party.

[0084] For example, under cooling conditions, when an uncontracted area maintains a lower set temperature while an adjacent contracted area raises its set temperature during the response phase, the net transfer of cooling capacity from the uncontracted area to the contracted area is calculated using a regional thermal coupling model. This net transfer is then converted into a consideration amount based on the air conditioning system's coefficient of performance and time-of-use electricity pricing, identifying the contracted area as the beneficiary. Based on this, the system generates revenue payment or revenue return records in the allocation and settlement module, thereby ensuring settlement fairness for entities with different contracted statuses when overflow cooling / heat occurs.

[0085] Specifically, the following steps are included: S621, Topology Modeling; The building is divided into a set of regions, and a region-neighborhood topology matrix is ​​established (adjacent is 1, otherwise is 0), and the physical parameters of the separating components (walls, floors, curtain walls, doors) and ventilation paths (gap, air duct, return air plenum) are labeled.

[0086] S622, thermodynamics and airflow coupling; S623, Determine and define symbols for cold / heat revenue; S624. Converting cooling / heating into electricity and monetization.

[0087] Quantified economic benefits The system compensates for any cooling capacity loss during contract execution through bilateral settlement. The compensation amount can be automatically settled on a periodic basis (such as daily, weekly, or monthly), preventing contracted users from reducing their participation due to free cooling.

[0088] The bilateral settlement refers to the settlement within a specific region during the settlement period. since The amount of profit is .

[0089] ①If For contracted users, For users who have not signed a contract, the contract stipulates that regardless of whether a contract is signed or not, the beneficiary pays and the loser receives payment. Payment Give Other settlement methods can also be adopted. For example, if the non-signing entity does not participate in direct settlement, then... It enters the adjustment pool to offset public losses in the current period or to be returned according to the ownership ratio.

[0090] ②If both parties are contracted users, then according to Direct point-to-point settlement.

[0091] ③ If both parties are unsigned users: only statistical display or signing guidance will be provided, and no fund settlement will be carried out.

[0092] If an unsigned user In continuous Each cycle generates significant spillover cooling to adjacent signing areas ( The system will automatically push contract suggestions and revenue simulations; if rejected, the optimization will reduce the scheduling amplitude of the surrounding area or adjust the air supply allocation to reduce the overflow without a contract. If a contracted user... For companies that continuously supply cold to other companies (providing cooling to others), compensation weight should be given in the pricing optimization process, or they should be advised to adjust their boundary wind pressure / gating strategies to reduce unexpected spillovers.

[0093] Finally, the settlement results for the revenue or liability of each contracted user in the commercial building under the scenarios of overflow cooling and overflow heating are completed.

[0094] The above embodiments implement a fair settlement mechanism of "contribution quantification + energy overflow compensation". On the one hand, the actual contribution of each contracted area is quantified from three dimensions: adjustable capacity, response speed, and execution accuracy through contribution coefficients, avoiding the extensive model of "distributing benefits according to area or fixed proportion". On the other hand, in response to the unique problems of "overflowing cold and heat" in super high-rise buildings, the energy interaction between regions is calculated through thermodynamic models, and then the benefiting party compensates the losing party through energy monetization. This solves the "free-rider" problem of "unfair distribution of benefits due to energy overflow" in traditional settlement, ensuring that the benefits of each contracted user are accurately matched with the actual contribution and energy interaction results, significantly improving the user's enthusiasm for participating in electricity trading and the credibility of the system settlement.

[0095] like Figure 6 As shown, the power trading and dispatching device for ultra-high-rise commercial users also includes a feedback learning module.

[0096] The feedback learning module is mainly used to process the operational results of the transaction interface module and the user response status management module. Specifically, the feedback learning module receives electricity market feedback parameters such as actual transaction price, transaction volume, and total revenue output by the transaction interface module, and simultaneously receives operational feedback parameters such as actual response load, response deviation, response delay, and comfort disturbance index output by the user response status management module.

[0097] The feedback learning module inputs the aforementioned input parameters into the error learning model for processing. The error learning model can be a deep reinforcement learning model, used to achieve dynamic policy adjustment under the objective of long-term return optimization; it can also be a Bayesian error correction model, used to correct uncertainties in price prediction and response capability prediction under small sample conditions; or it can be an adaptive weighted ensemble model (combining LSTM, support vector regression, and XGBoost, etc.), used to improve the fitting ability to nonlinear features and random fluctuations.

[0098] Optimized output conditions can be obtained through training and correction of the error learning model. For the pricing optimization module, the feedback learning module can dynamically adjust parameters in the constraints, such as the price elasticity coefficient, load response boundary, and maintenance cost weights, thereby improving the matching accuracy between electricity trading prices and actual response volumes. For the load response identification module, the feedback learning module can adaptively adjust parameters such as the time window length, number of hidden layer units, and learning rate of the LSTM neural network, while simultaneously optimizing the kernel function parameters and penalty factor of the support vector regression model, thereby improving the accuracy of load response capability predictions.

[0099] The feedback learning module can also output a user response reliability coefficient to measure the consistency and deviation patterns of different regions or users in multiple historical response processes; output a dynamic comfort threshold correction amount to achieve a dynamic balance between benefits and comfort in subsequent scheduling processes; and generate a confidence interval for the prediction results to provide an uncertainty boundary reference for power trading scheduling, thereby achieving adaptive optimization and long-term stability assurance for the overall system operation.

[0100] Example 2: This embodiment of the invention also provides a power trading and dispatching method for ultra-high-rise commercial users, applied to the power trading and dispatching device, including: Collect equipment parameters of equipment in commercial buildings, extract the total load response capacity of commercial buildings from equipment parameters, and obtain the preliminary net load based on the total load response capacity of commercial buildings and circuit losses; Obtain the user's contract status and set comfort constraints; and based on the user's contract status and comfort constraints, filter out the net load response value for transactions from the initial net load. The operation and maintenance cost of the contracted area is calculated by combining the response frequency records of the contracted area, and the optimal quotation is determined based on the net load response value and the operation and maintenance cost. Based on the optimal price, the quantitative result of revenue distribution is determined, and based on the quantitative result of revenue distribution, the corresponding revenue or settlement result of each contracted user in the commercial building under the scenarios of overflow cooling and overflow heating is determined.

[0101] Preferably, the equipment parameters of the equipment within the commercial building are collected, and the total load response capacity of the commercial building is extracted from the equipment parameters, including: The system collects equipment parameters from the equipment within the commercial building, extracts the total load response potential of the commercial building from the equipment parameters based on a pre-trained LSTM neural network model, and outputs the total load response capability of the commercial building.

[0102] Example 3: This embodiment of the invention also provides a power trading and dispatching device for ultra-high-rise commercial users, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the power trading and dispatching method for ultra-high-rise commercial users as described above.

[0103] Example 4: This embodiment of the invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the power trading and scheduling method for ultra-high-rise commercial users as described above.

[0104] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0105] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and unit can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0106] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0107] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0108] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A power trading and dispatching device for ultra-high-rise commercial users, characterized in that, include: The basic data layer is used to collect equipment parameters of equipment in commercial buildings, extract the total load response capacity of commercial buildings from equipment parameters, and obtain the preliminary net load based on the total load response capacity of commercial buildings and circuit losses. The transaction data filtering layer is used to obtain the user's contract status and set comfort constraints. Based on the user's contract status and comfort constraints, net load response values ​​for transactions are selected from the initial net load. The transaction decision-making layer is used to calculate the operation and maintenance cost of the contracted area by combining the response frequency records of the contracted area, and to determine the optimal price based on the net load response value and the operation and maintenance cost. The value allocation layer is used to determine the quantitative result of revenue allocation based on the optimal price, and to determine the corresponding revenue or settlement result of each contracted user in the commercial building under the scenarios of overflow cooling and overflow heating based on the quantitative result of revenue allocation.

2. The power trading and dispatching device according to claim 1, characterized in that, Collect equipment parameters from the equipment within the commercial building, and extract the total load response capacity of the commercial building from these parameters, including: The system collects equipment parameters from the equipment within the commercial building, extracts the total load response potential of the commercial building from the equipment parameters based on a pre-trained LSTM neural network model, and outputs the total load response capability of the commercial building.

3. The power trading and dispatching device according to claim 2, characterized in that, Based on the total load response capacity and circuit losses of the aforementioned commercial building, a preliminary net load is obtained, including: Collect physical parameters of the power distribution system inside the commercial building. The physical parameters include the effective current flowing through the cable, the total line resistance, the power supply duration, the conduction loss of the equipment, the switching loss of the components, and other losses of the equipment. The power loss generated by the cable during operation is calculated based on Formula 1 and the physical parameters of the power distribution system. Formula 1 is as follows: ， in, For cable loss, The effective current flowing through the cable. The total line resistance is... For power supply duration; The total power loss of all power distribution equipment during the load response cycle is calculated based on Formula 2 and the physical parameters of the power distribution system. Formula 2 is as follows: ， in, This represents the total power loss of the equipment. The power loss during conduction for each device. The power consumption is due to the switching on and off of each device. Other power consumption of each device; Based on Formula 3, integrate cable loss and total power loss of equipment The total power loss of the commercial building's power distribution system during the load response cycle is obtained. Formula 3 is as follows: ， Based on Formula 4, the total load response capability of the commercial building output by the load response identification module is calculated. Excluding total circuit power loss To obtain the initial net load Formula four is: 。 4. The power trading and dispatching device according to claim 3, characterized in that, Obtain the user's contract status and set comfort constraints. Filter the net load response values ​​for transactions from the initial net load, including: Obtain user subscription status, determine subscribed and non-subscribed regions based on user subscription status, define comfort constraints for non-subscribed regions, and determine non-tradable loads for non-subscribed regions based on comfort constraints. , Formula 5 represents the comfort constraints in the non-contractual area: ， in, This is a real-time comfort index for non-contracted areas. This is the minimum permissible comfort standard; The maximum load response value of the contracted area was determined based on historical power data without comfort constraints. And the maximum load response value of the contracted area after considering comfort constraints Based on Formula 6, the maximum load response value and maximum load response value By performing interpolation calculations, we obtain the loss of the contracted area sacrificed to meet comfort constraints. Formula six is: = - ， Based on Formula 7, from the initial net load Excluding non-tradable loads in non-contracted areas And losses in the contracted area This yields the net load response value ultimately used for electricity trading. Formula seven is: 。 5. The power trading and dispatching device according to claim 4, characterized in that, The operation and maintenance cost of the contracted area is calculated by combining the response frequency records of the contracted area, based on the net load response value. Determine the optimal quote based on operation and maintenance costs, including: The response duration, number of responses, and response load value of the contracted area are obtained from the response frequency records. The operation and maintenance cost of the contracted area is calculated based on Formula 8 and the response frequency records of the contracted area. Formula 8 is: ， in, For the first The response and maintenance costs for each contracted region The maintenance unit price corresponding to the response time. For the recorded number The first of the contracted areas The duration of each response. The maintenance unit price is the number of start-stop cycles. For the recorded number Number of responses in each contracted region The additional maintenance unit price resulting from load response intensity, For the recorded number The first of the contracted areas The load response value of the second response. For the first Fixed maintenance costs for each contracted area; Construct a multi-objective optimization function that maximizes commercial user benefits, minimizes comfort disturbances, and optimizes response speed. Equation 9 is the multi-objective optimization function: ， in, For each term, is the normalization factor. For the first Market transaction prices for a given period of time. For the first Load response of commercial buildings participating in market transactions during a given time period For comfort disturbances, For the first The response and maintenance costs for each contracted region For response speed; Based on the premise that the total transaction revenue is not less than the total operation and maintenance cost, a constraint is set, which is: ， The optimal price was obtained by solving the multi-objective optimization function and constraints using MATLAB.

6. The power trading and dispatching device according to claim 5, characterized in that, Based on the optimal bid, a quantitative result for revenue distribution is determined, and based on this quantitative result, the revenue or liability settlement results for each contracted user within the commercial building under the scenarios of overflow cooling and overflow heating are determined, including: Based on the optimal bid, the contribution coefficient of each contracted region is quantified and used as the basis for revenue distribution. The contribution coefficient is: ， in, For the first Load response contribution coefficient of each region For adjustable capacity factor, for the first Net load response value of each region The ratio of its baseline load, The response speed factor is the ratio of the system's response speed after receiving an instruction. The accuracy factor is the ratio of the actual response load to the dispatch response command. , and These are the weight coefficients of each factor, and they satisfy... + + ; Formula 10 is constructed based on the heat transfer principle of the building envelope. Formula 10 is then used to analyze adjacent... The heat transfer in the region is calculated using Formula 10: ， in, For the first Time period, from the Region to the first Heat transfer through the building envelope in the area The heat transfer coefficient of the building envelope. For contact area, For the first Time period Indoor temperature in the area For the first Time period Indoor temperature in the area Formula 11 is constructed based on the principle of heat transfer caused by infiltration and gap air. Formula 11 is then used to analyze adjacent... The calculation of heat transfer in the region is given by Formula 11: ， in, For the first Time period, from the Region to the first Heat transfer in the region occurs through infiltration and gap airflow. For the first Infiltration air volume during the period The isobaric specific heat capacity of air. For the first Time period Indoor temperature in the area For the first Time period Indoor temperature in the area Based on Formula Twelve, heat transfer... and heat transfer Summing gives the total heat flow. Formula 12 is: ， Based on total heat flow Determine and define the symbols for cooling and heating benefits. During the cooling season: If and This indicates that the cooling capacity comes from Regional flow area, income, During the heating season: If and This indicates that heat comes from Regional flow area, income; Based on Formula Thirteen, the total heat flow The cumulative cooling and heating load is obtained by calculating the net cooling and heating overflow during the statistical period, as described in Formula Thirteen: ， in, Settlement period From the first Region to the first The cumulative heat and cold passing through the region, For the settlement period, The determination coefficient is when Indicates the determination of cooling revenue. Indicates the determination of heating revenue; Based on formulas fourteen and fifteen, accumulated cold and heat The overflow energy is converted into equivalent electrical energy and monetary value for settlement of cold and heat overflow compensation. Formula fourteen is as follows: ， in, For from the first Region to the first The region utilizes the equivalent electrical energy of cooling and heating. For the first The performance coefficient of the air conditioning system during different time periods. Formula 15 is as follows: ， in, For from the first Region to the first The price of the equivalent electrical energy for cooling and heating in a region. For the first Electricity price during specific time periods Through bilateral settlement and the price of equivalent electricity Determine the corresponding revenue or settlement results for each contracted user in a commercial building under the scenarios of overflow cooling and overflow heating.

7. A power trading and dispatching method for ultra-high-rise commercial users, applied to the power trading and dispatching device according to any one of claims 1 to 6, characterized in that, include: Collect equipment parameters of equipment in commercial buildings, extract the total load response capacity of commercial buildings from equipment parameters, and obtain the preliminary net load based on the total load response capacity of commercial buildings and circuit losses; Obtain the user's contract status and set comfort constraints; Based on the user's contract status and comfort constraints, net load response values ​​for transactions are selected from the initial net load. The operation and maintenance cost of the contracted area is calculated by combining the response frequency records of the contracted area, and the optimal quotation is determined based on the net load response value and the operation and maintenance cost. Based on the optimal price, the quantitative result of revenue distribution is determined, and based on the quantitative result of revenue distribution, the corresponding revenue or settlement result of each contracted user in the commercial building under the scenarios of overflow cooling and overflow heating is determined.

8. The power trading and dispatching method according to claim 7, characterized in that, Collect equipment parameters from the equipment within the commercial building, and extract the total load response capacity of the commercial building from these parameters, including: The system collects equipment parameters from the equipment within the commercial building, extracts the total load response potential of the commercial building from the equipment parameters based on a pre-trained LSTM neural network model, and outputs the total load response capability of the commercial building.

9. A power trading and dispatching device for ultra-high-rise commercial users, characterized in that, The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the power trading and dispatching method for ultra-high-rise commercial users as described in any one of claims 1 to 6.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the power trading and scheduling method for ultra-high-rise commercial users as described in any one of claims 1 to 6.

Images