Photovoltaic air conditioner and control method and system thereof

Abstract

The invention discloses a photovoltaic air conditioner and a control method and system thereof, and belongs to the technical field of photovoltaic air conditioners. The control method comprises the steps that geographic information system data and satellite image data of a target area are obtained; classifying the buildings in the target area; for different types of classified buildings, processing the satellite image data by using corresponding deep learning models so as to identify roof areas of the various types of buildings; evaluating the annual total photovoltaic power generation potential of the target area; corresponding building energy consumption models are constructed according to the time features and the space features of all the types of buildings, the annual total air conditioner system power consumption of all the types of buildings is calculated and aggregated, and the annual total air conditioner system power consumption of the target area is obtained; and the photovoltaic distribution proportion is calculated, and a photovoltaic air conditioner system control strategy of the target area is determined based on the photovoltaic distribution proportion. The problem that in the prior art, photovoltaic system control is disjointed with the actual energy consumption requirement of a building is solved, and intelligent control over the photovoltaic air conditioning system is achieved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of photovoltaic air conditioners, and more particularly relates to a photovoltaic air conditioner and a control method and system thereof. BACKGROUND

[0002] With the increasing global energy demand and the aggravation of environmental pollution, solar energy, as a clean and renewable energy, has gradually become an important choice for building energy supply systems. Among them, the combination of photovoltaic systems and air conditioning systems has important application value in building energy management.

[0003] However, in actual engineering applications, due to the spatiotemporal variability of solar radiation intensity, the difference in air conditioning operation mode of different building types, and the unevenness of available roof area, etc., the potential evaluation of solar photovoltaic systems and the prediction accuracy of building energy consumption are limited, and it is difficult to achieve optimal configuration of energy utilization efficiency.

[0004] Although the prior art has a certain foundation in building photovoltaic potential evaluation, its analysis is mostly limited to static parameters such as installable roof area and solar radiation resources, and lacks dynamic response mechanism to actual energy demand of buildings, especially fails to effectively couple modeling of operation characteristics of high-energy-consuming equipment (such as air conditioning system) and photovoltaic power generation capacity, resulting in that the evaluation results are difficult to directly support the collaborative optimization and intelligent operation decision of energy systems.

[0005] Therefore, there is an urgent need for a method that can integrate multi-dimensional information such as building characteristics, solar radiation, and air conditioning operation mode, to realize intelligent and refined control of photovoltaic air conditioning systems. SUMMARY

[0006] To solve the problems in the prior art, the present application provides a photovoltaic air conditioner and a control method and system thereof.

[0007] The present application adopts the following technical solutions.

[0008] The first aspect of the present application provides a photovoltaic air conditioner control method, comprising: obtaining geographic information system data and satellite image data of a target area; classifying buildings in the target area based on the geographic information system data; for different categories of classified buildings, respectively processing the satellite image data using corresponding deep learning models to identify the roof areas of each category of buildings; based on all the identified roof areas, evaluating the annual total photovoltaic power generation potential of the target area; constructing a building energy consumption model corresponding to each category of buildings according to time characteristics and space characteristics of the category of buildings, calculating and aggregating annual total air conditioning system power consumption of all categories of buildings based on the building energy consumption model to obtain annual total air conditioning system power consumption of the target region; calculating a photovoltaic distribution ratio based on the annual total photovoltaic power generation potential and the annual total air conditioning system power consumption of the target region, and determining a photovoltaic air conditioning system control strategy of the target region based on the photovoltaic distribution ratio.

[0009] Optionally, the classification of the buildings in the target region comprises: dividing the buildings in the target region into civil buildings, residential buildings and industrial buildings based on a use attribute in the geographic information system data; wherein the civil buildings include buildings for commercial, office or entertainment use; the residential buildings are buildings for residential use; the industrial buildings are buildings for production and manufacturing use.

[0010] Optionally, the identification of the roof area of each category of buildings comprises: calling a deep learning model preset for each category of buildings to perform semantic segmentation on the satellite image data; wherein the deep learning model is a U2-Net network which extracts image features through an encoder and reconstructs image size through a decoder and outputs a segmentation map for distinguishing roof areas and non-roof areas in the image.

[0011] Optionally, the evaluation of the annual total photovoltaic power generation potential of the target region comprises: based on all the identified roof areas, simulating the arrangement of photovoltaic modules through a preset filling algorithm to calculate the total installable area; in combination with annual total solar radiation data of the target region, evaluating the annual total photovoltaic power generation potential based on the total installable area.

[0012] Optionally, the engineering constraint rules followed by the filling algorithm in simulating the arrangement comprise: the starting position of the arrangement of photovoltaic modules is located at a preset safety distance inward from the boundary of the roof area; aligning photovoltaic modules in a row-column orthogonal grid; reserving a maintenance spacing of no less than a preset minimum value between each photovoltaic module; identifying and avoiding obstacles in the roof area.

[0013] Optionally, the time characteristics include a typical daily operating period and annual total operating time defined based on the use attribute of the buildings; The spatial features include building story count, roof area, and area ratio of each story requiring cooling or heating.

[0014] Optionally, the calculation of the photovoltaic distribution ratio comprises: obtaining the annual total photovoltaic power generation potential of the target area and the annual total power consumption of the air conditioning system; calculating a ratio of the annual total photovoltaic power generation potential to the annual total power consumption of the air conditioning system, the ratio being the photovoltaic distribution ratio.

[0015] Optionally, the determining of the photovoltaic air conditioning system control strategy of the target area based on the photovoltaic distribution ratio comprises: comparing the photovoltaic distribution ratio with a preset threshold interval, and determining a corresponding system architecture scheme according to the comparison result; The system architecture scheme comprises: when the photovoltaic distribution ratio is not lower than a first preset threshold, determining a system architecture scheme of off-grid operation; when the photovoltaic distribution ratio is between the first preset threshold and a second preset threshold, determining a system architecture scheme of photovoltaic and energy storage hybrid power supply; when the photovoltaic distribution ratio is not higher than the second preset threshold, determining a system architecture scheme of photovoltaic and grid complementation.

[0016] The second aspect of the application provides a photovoltaic air conditioning control system for operating the photovoltaic air conditioning control method of the first aspect of the application, comprising: a data acquisition module, a building classification module, a roof identification module, a photovoltaic potential evaluation module, a power consumption calculation module, and a control strategy generation module, wherein: The data acquisition module is configured to acquire geographic information system data and satellite image data of a target area; The building classification module is configured to classify buildings in the target area based on the geographic information system data; The roof identification module is configured to process the satellite image data using corresponding deep learning models for different categories of classified buildings to identify the roof areas of the different categories of buildings; The photovoltaic potential evaluation module is configured to evaluate the annual total photovoltaic power generation potential of the target area based on all the identified roof areas; The power consumption calculation module is configured to construct corresponding building energy consumption models according to the time features and spatial features of different categories of buildings, calculate and aggregate the annual total power consumption of the air conditioning system of all categories of buildings based on the building energy consumption models, and obtain the annual total power consumption of the air conditioning system of the target area; The control strategy generation module is configured to calculate a photovoltaic distribution ratio based on the annual total photovoltaic power generation potential and the annual total air conditioning system power consumption of the target region, and determine a photovoltaic air conditioning system control strategy for the target region based on the photovoltaic distribution ratio.

[0017] The third aspect of the present application provides a photovoltaic air conditioner adopting the photovoltaic air conditioner control method according to the first aspect of the present application.

[0018] Compared with the prior art, the present application has at least the following beneficial effects: 1. The present application obtains geographic information system data and satellite image data of a target region, classifies buildings based on the geographic information system data, and identifies the roof area of each type of building using a corresponding deep learning model, thereby evaluating the photovoltaic potential and air conditioning power consumption, and finally generating a control strategy, solving the problem that the photovoltaic system control in the prior art is not consistent with the actual energy demand of the building, and realizing intelligent control of the photovoltaic air conditioning system.

[0019] 2. The present application classifies buildings into civil, residential and industrial buildings based on the use attribute in the geographic information system data, solving the problem of ambiguous building classification standards and weak correlation with energy consumption characteristics, and laying a solid foundation for subsequent type-specific accurate identification and energy consumption modeling.

[0020] 3. The present application calls the U2-Net deep learning model pre-installed for each type of building to perform semantic segmentation, solving the problem of insufficient recognition accuracy of general models for diversified building roof shapes, materials and structural features, and realizing high-precision and automated extraction of roof areas of different types of buildings.

[0021] 4. The present application uses a filling algorithm to simulate the arrangement of photovoltaic modules based on the identified roof area to calculate the total installable area, and evaluates the annual total photovoltaic power generation potential in combination with solar radiation data, solving the problem that photovoltaic potential evaluation is too theoretical and does not consider actual installation feasibility, and realizing reliable and quantitative evaluation of regional photovoltaic power generation potential.

[0022] 5. The present application makes the filling algorithm follow engineering constraint rules including a preset safety distance, orthogonal grid arrangement, maintenance spacing and obstacle avoidance, solving the problem that the photovoltaic module arrangement scheme deviates from construction specifications and safety requirements, and ensuring the feasibility and safety of the control scheme in actual engineering.

[0023] 6. The present application introduces time and space characteristics including building operation period, number of floors, and area ratio to construct a building energy consumption model, solving the problem that building air conditioning energy consumption estimation is too rough and fails to reflect differences in building function and structure, and realizing a more realistic prediction of air conditioning system power consumption.

[0024] 7. This invention solves the problem of the inability to quantitatively match energy supply and demand by calculating the ratio of the total annual photovoltaic power generation potential to the total annual air conditioning system power consumption as the photovoltaic allocation ratio, and provides a key, quantitative decision indicator to guide the selection of system architecture.

[0025] 8. This invention dynamically determines system architecture schemes such as off-grid, hybrid, or grid-connected by comparing the photovoltaic allocation ratio with a preset threshold range. This solves the problem of a single control strategy that cannot adapt to the differences in energy endowment in different regions, and realizes flexible and adaptive matching of control strategies, thereby optimizing the system's economy and reliability.

[0026] 9. This invention solves the problems of lack of implementation and difficulty in process integration of the above methods by constructing a system that includes modules for data acquisition, building classification, roof identification, photovoltaic potential assessment, power consumption calculation and control strategy generation. It provides an efficient and integrated tool for automating the execution of the photovoltaic air conditioning control method.

[0027] 10. By employing the photovoltaic air conditioning control method described above, this invention solves the problems of blind and unscientific control of photovoltaic air conditioning, ensuring that the controlled photovoltaic air conditioning system can achieve optimal energy matching and system architecture based on regional assessment. Attached Figure Description

[0028] Figure 1 This is an overall framework diagram of the solution provided according to the embodiments of the present invention; Figure 2 This is a flowchart of a method provided according to an embodiment of the present invention. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0030] In Embodiment 1, the present invention provides a photovoltaic air conditioning control method, such as... Figure 2 As shown, it includes: Step 1: Obtain geographic information system data and satellite imagery data for the target area.

[0031] Step 2: Based on the geographic information system data, classify the buildings within the target area.

[0032] Preferably, classifying the buildings within the target area includes: Based on the usage attributes in the geographic information system data, the buildings within the target area are divided into civil buildings, residential buildings, and industrial buildings. The civil buildings mentioned include buildings used for commercial, office, or entertainment purposes; The residential building is a building used for residential purposes; The industrial building is a building used for production and manufacturing purposes.

[0033] Step 3: For the different categories of buildings after classification, the satellite image data is processed using the corresponding deep learning models to identify the roof areas of each category of buildings.

[0034] Preferably, identifying the roof areas of each type of building includes: The satellite image data is semantically segmented by calling a deep learning model pre-configured for each type of building. The deep learning model is a U2-Net network, which extracts image features through an encoder and reconstructs the image size through a decoder to output a segmentation map. The segmentation map is used to distinguish between roof areas and non-roof areas in the image.

[0035] For example, the deep learning model is a U2-Net network structure with an input and output size of 320×320. It extracts features through an encoder that includes five downsampling operations and performs size restoration and feature fusion through a decoder that includes five upsampling operations. The final output is a binarized segmentation result that distinguishes image pixels into roof areas and non-roof areas.

[0036] It should be noted that this invention solves the problem of poor adaptability and low recognition accuracy of general models to different roof structures by calling pre-set deep learning models to process satellite image data for different types of buildings, thus achieving accurate and efficient recognition of roofs of different types of buildings.

[0037] Step 4: Based on all identified rooftop areas, assess the total annual photovoltaic power generation potential of the target area.

[0038] Preferably, the assessment yields the total annual photovoltaic power generation potential of the target area, including: Based on all identified roof areas, photovoltaic modules are simulated and arranged using a preset filling algorithm to calculate the total installable area; The total annual solar radiation data for the target area is used to assess the total annual photovoltaic power generation potential based on the total installable area.

[0039] More preferably, the filling algorithm follows the following engineering constraints for simulation layout: The photovoltaic modules are placed at a predetermined safe distance inward from the boundary of the roof area; Photovoltaic modules are arranged in an orthogonal grid with row and column alignment; A maintenance spacing of not less than the preset minimum value should be reserved between each photovoltaic module; Identify and avoid obstacles within the roof area.

[0040] It should be noted that this invention uses a preset filling algorithm to simulate the arrangement of photovoltaic modules based on the identified entire roof area to calculate the total installable area, and combines the annual total solar radiation data to evaluate the annual total photovoltaic power generation potential. This solves the problem that photovoltaic potential assessment only considers static parameters and does not take into account the constraints of actual engineering layout, and realizes a scientific and refined assessment of the regional photovoltaic power generation potential.

[0041] Step 5: Construct corresponding building energy consumption models based on the temporal and spatial characteristics of each type of building. Calculate and aggregate the total annual air conditioning system power consumption of all types of buildings based on the building energy consumption models to obtain the total annual air conditioning system power consumption of the target area.

[0042] Preferably, the time characteristics include typical daily operating periods and total annual operating time defined based on building use attributes; The spatial characteristics include the number of building floors, roof area, and the percentage of each floor that requires cooling or heating.

[0043] It should be noted that this invention solves the problem of the disconnect between building energy consumption prediction and actual building operation characteristics by constructing a building energy consumption model based on the time and spatial characteristics of various types of buildings, calculating and aggregating the annual total power consumption of the air conditioning system, and realizing an accurate and realistic prediction of the total power consumption of the regional air conditioning system.

[0044] Step 6: Calculate the photovoltaic allocation ratio based on the total annual photovoltaic power generation potential and the total annual air conditioning system power consumption of the target area, and determine the photovoltaic air conditioning system control strategy of the target area based on the photovoltaic allocation ratio.

[0045] Preferably, the calculation of the photovoltaic allocation ratio includes: Obtain the total annual photovoltaic power generation potential and the total annual air conditioning system power consumption of the target area; Calculate the ratio of the total annual photovoltaic power generation potential to the total annual air conditioning system power consumption, where the ratio is the photovoltaic allocation ratio.

[0046] It should be noted that this invention calculates the photovoltaic allocation ratio based on the total annual photovoltaic power generation potential and the total annual air conditioning system power consumption, and determines the photovoltaic air conditioning system control strategy accordingly. This solves the problem of the photovoltaic system control being disconnected from the energy demand side and the difficulty in achieving optimal energy allocation. It realizes the dynamic generation of the optimal system architecture scheme based on different energy matching degrees, and improves the scientific and economic nature of system control.

[0047] Preferably, the control strategy for determining the photovoltaic air conditioning system in the target area includes: The photovoltaic allocation ratio is compared with a preset threshold range, and the corresponding system architecture scheme is determined based on the comparison result. The system architecture scheme includes: When the photovoltaic allocation ratio is not lower than the first preset threshold, the off-grid operation system architecture scheme is determined; When the photovoltaic allocation ratio is between the first preset threshold and the second preset threshold, a system architecture scheme for hybrid power supply of photovoltaic and energy storage is determined. When the photovoltaic allocation ratio is not higher than the second preset threshold, a system architecture scheme that complements photovoltaics and the power grid is determined.

[0048] It should be noted that this invention determines system architecture schemes such as off-grid operation, photovoltaic and energy storage hybrid, or photovoltaic and grid complementarity by comparing the photovoltaic allocation ratio with a preset threshold range. This solves the problem that a single system architecture cannot adapt to different energy supply and demand scenarios, realizes flexible and precise matching of photovoltaic air conditioning system control strategies, and ensures the economy and reliability of system operation.

[0049] This invention acquires geographic information system (GIS) data and satellite imagery data of the target area, classifies buildings based on the GIS data, and uses a deep learning model to identify roof areas. This solves the problems of low automation in building feature identification and inaccurate estimation of available roof area in traditional methods, and achieves comprehensive and automated extraction of regional building features, providing an accurate data foundation for photovoltaic potential assessment.

[0050] It should be noted that, in one embodiment, it can be achieved through... Figure 1 The overall framework of the scheme shown implements the photovoltaic air conditioning control method provided by this invention, specifically including: Obtain geographic information system data of the target area, and classify the building types in the target area based on the usage attributes in the geographic information system data, including civil buildings, residential buildings and industrial buildings; Satellite image data of the target area is acquired, and combined with the building type classification results, the satellite image data is processed using deep learning models corresponding to different building types to identify the roof areas of each type of building, and a preset filling algorithm is used to simulate the arrangement of photovoltaic modules to calculate the total installable area. Based on the annual total solar radiation data and total installable area of ​​the target area, calculate the annual total photovoltaic power generation potential of the target area; Based on the data from the geographic information system, the building height within the target area is extracted. The corresponding building energy consumption model is constructed by combining the time and spatial characteristics of different building types. Based on the building energy consumption model, the annual total power consumption of the air conditioning system of all types of buildings is calculated and aggregated to obtain the annual total power consumption of the air conditioning system in the target area. The photovoltaic allocation ratio is calculated based on the total annual photovoltaic power generation potential and the total annual air conditioning system power consumption of the target area, and the photovoltaic air conditioning system control strategy of the target area is determined based on the photovoltaic allocation ratio.

[0051] In Embodiment 2, the present invention provides a photovoltaic air conditioning control system for executing a photovoltaic air conditioning control method described in Embodiment 1, comprising: The module includes a data acquisition module, a building classification module, a roof identification module, a photovoltaic potential assessment module, a power consumption calculation module, and a control strategy generation module, among which: The data acquisition module is used to acquire geographic information system data and satellite image data of the target area; The building classification module classifies buildings within the target area based on the geographic information system data; The roof recognition module is used to process the satellite image data using corresponding deep learning models for different categories of buildings after classification, so as to identify the roof area of ​​each category of building; The photovoltaic potential assessment module is used to assess the total annual photovoltaic power generation potential of the target area based on all identified rooftop areas. The power consumption calculation module constructs a corresponding building energy consumption model based on the time and spatial characteristics of each type of building, calculates and aggregates the annual total air conditioning system power consumption of all types of buildings based on the building energy consumption model, and obtains the annual total air conditioning system power consumption of the target area. The control strategy generation module is used to calculate the photovoltaic allocation ratio based on the total annual photovoltaic power generation potential and the total annual air conditioning system power consumption of the target area, and to determine the photovoltaic air conditioning system control strategy of the target area based on the photovoltaic allocation ratio.

[0052] It should be noted that this invention solves the problem of fragmented processes and difficulty in efficient collaboration when the above methods are implemented by constructing an integrated system that includes modules for data acquisition, building classification, roof identification, photovoltaic potential assessment, power consumption calculation, and control strategy generation. It provides a physical tool with clear functions and smooth processes, ensuring the efficient execution of the control method.

[0053] In Example 3, the present invention provides a photovoltaic air conditioner, which adopts a photovoltaic air conditioner control method as described in Example 1.

[0054] It should be noted that by using the aforementioned control method to control photovoltaic air conditioning, this invention solves the problem of lack of scientific guidance and high degree of blindness in the early control of photovoltaic air conditioning projects, ensuring that the final controlled photovoltaic air conditioning system matches the regional energy conditions from the source, and achieving the optimization of the overall energy efficiency and economy of the system.

[0055] In Example 4, this invention provides an application example of a photovoltaic air conditioning control method, including: Step 1: Obtain basic data and train a deep learning model for identifying roof areas. The specific process is as follows: Step 1.1: Based on the Geographic Information System (GIS) data of the target area, perform a preliminary classification of the buildings within the area. The classification system includes: Civil buildings: mainly include buildings used for commercial, office, and entertainment purposes; Residential buildings: Buildings primarily used for residential purposes; Industrial buildings: Buildings primarily used for production and manufacturing.

[0056] Meanwhile, image label data required for model training was created based on an open-source satellite image dataset. The labels are in grayscale format, dividing image pixels into two main categories: "roof area" and "non-roof area".

[0057] Step 1.2: Based on the above building classification results, independently construct training sets and test sets for the three types of buildings: civil, residential, and industrial, and set the sample size ratio of the training set and the test set to 7:3. A dedicated U2-Net deep learning model is trained for each type of building. The key parameters for model training include: (1) Network structure: The U2-Net model is adopted, which mainly consists of a six-segment encoder, a five-segment decoder and a saliency map fusion module connected to the decoder level and the last segment encoder. The input satellite image is processed by five downsampling and five upsampling in sequence. (2) Input / output: The input and output image sizes of the model are fixed at 320×320 pixels; (3) Hyperparameter settings: as shown in Table 1: Table 1. Hyperparameter settings for the U2-Net model

[0058] Step 2: Based on the optimal model trained in Step 1, identify rooftop areas and accurately assess the total annual photovoltaic power generation potential of those areas.

[0059] Preferably, step 2 includes: Step 2.1: Using satellite data of the target area as input, call the optimal U2-Net model trained for civilian, residential, and industrial buildings respectively for prediction; The model output is a grayscale image, where white pixels represent the identified roof area and black pixels represent the non-roof area. By summarizing all the identification results, a complete roof distribution map of the target area is obtained.

[0060] Step 2.2: Based on the identified roof area, a filling algorithm is used to simulate the actual arrangement of photovoltaic modules in order to calculate the accurate actual installable area.

[0061] The filling algorithm follows the following key engineering constraints for simulation layout: (1) Layout start rule: The arrangement of photovoltaic modules must start from the boundary of the roof graphic and be no less than a preset safety distance (e.g., 1 meter) inward. It is forbidden to place them close to the edge of the roof.

[0062] (2) Arrangement rules: The rows and columns are aligned using a regular orthogonal grid, and misalignment is prohibited to ensure construction feasibility and the continuity of maintenance channels.

[0063] (3) Spacing rules: Each photovoltaic module must maintain a maintenance distance of not less than the preset minimum value (e.g., 2 meters) to meet the needs of maintenance, wiring and reserved passage.

[0064] (4) Obstacle avoidance rules: During the layout process, the algorithm automatically identifies and eliminates obstacles (such as skylights, ventilation equipment, etc.) in the roof area, avoiding these areas.

[0065] Step 2.3: The algorithm calculates the actual area of ​​the roof where photovoltaic panels can be installed. A PV Combined with local annual solar radiation data R y (Considering characteristics such as photovoltaic tilt angle, cloud cover, and altitude), calculate the region's total annual photovoltaic power generation potential. S PV See equation (1).

[0066] S PV= Ry ×A PV X η (1) in: S PV This refers to the region's total annual photovoltaic power generation potential, in kWh. R y This refers to the local annual solar radiation, in kWh / m². 2 ; A PV This refers to the actual area of ​​the roof where photovoltaic panels can be installed, in meters. 2 ; η The photovoltaic conversion efficiency is typically taken as 15%, and it depends on the type of material.

[0067] Step 3: Construct a building energy consumption model to accurately estimate the total annual power consumption of the air conditioning system in the target area. This model takes into account both the temporal and spatial characteristics of the building.

[0068] Step 3.1: Based on the building classification in Step 1, define the typical operating mode and space utilization parameters for each type of building: Building heights are extracted based on GIS information, while also considering the time characteristics of different buildings. Table 2 is used as an example to define the regular operating hours and total annual operating time of different types of buildings.

[0069] Table 2 Setting of Building Time Characteristics

[0070] Taking Table 3 as an example, the spatial characteristics define the proportion of electricity consumption for air conditioning and the proportion of area requiring cooling / heating on each floor for different types of buildings: Table 3 Architectural Spatial Characteristics Settings

[0071] Step 3.2: Based on the above characteristic parameters, calculate the annual total power consumption of the air conditioning system of a single building using formula (2), and aggregate all buildings to obtain the annual total power consumption of the air conditioning system of the target area.

[0072] For the building's total annual power consumption of air conditioning systems Q sum The calculation is shown in equation (2): Q sum = A x r x n x Q x t (2) in, Q sum This refers to the building's total annual power consumption for its air conditioning system, expressed in kWh. A The roof area of ​​a building, in meters. 2 ; r This refers to the percentage of area requiring cooling / heating on each floor. This parameter can be calculated based on architectural design drawings by excluding areas that do not require temperature control from the total building area (such as equipment rooms in civil buildings, kitchens and bathrooms in residential buildings, and warehouses in industrial buildings). n This refers to the number of floors; Q This refers to the average power consumption per square meter of a building, in kW / m². 2 ; t This refers to the total number of hours (52 weeks) of operation per year, in hours (h).

[0073] Step 4: Calculate the photovoltaic allocation ratio based on the total annual photovoltaic power generation potential and the total annual air conditioning system power consumption of the target area, and determine the photovoltaic air conditioning system control strategy of the target area based on the photovoltaic allocation ratio.

[0074] Step 4.1, Calculate the photovoltaic allocation ratio R See equation (3): R = S PV X λ / Q sum (3) in, λ The percentage of electricity consumption used for air conditioning (values ​​for the corresponding building type are taken from Table 3).

[0075] Step 4.2, based on the photovoltaic allocation ratio R The numerical range is used to automatically match and generate the optimal control strategy for the photovoltaic air conditioning system, as shown in Table 4: Table 4 Photovoltaic Air Conditioning Control Strategy

[0076] It should be noted that the threshold setting for this strategy is primarily based on ensuring the economy and reliability of system operation. When photovoltaic power generation can cover the vast majority (≥80%) of air conditioning electricity consumption, the system is feasible for off-grid operation; When the coverage is insufficient (≤50%), off-grid or large-scale energy storage costs are too high, and grid-connected mode is more economical; When the situation falls between these two extremes, a hybrid power supply mode combining photovoltaics and energy storage is adopted to achieve a balance.

[0077] It should be noted that the core of this invention lies in the early control of photovoltaic air conditioning systems, which aims to select system architecture and make preliminary capacity configuration through annual total assessment, so as to avoid over-configuration waste or under-configuration failure.

[0078] The present invention proposes a static, annual total-based rapid evaluation method, which has the advantages of high computational efficiency and convenient data acquisition, and is particularly suitable for feasibility studies and preliminary screening of schemes in the early stage of projects.

[0079] It is understandable that this control method does not involve real-time operation control on a daily or seasonal basis.

[0080] In subsequent optimization implementation, based on the control strategy determined in this invention, hourly data of typical days can be further introduced for simulation verification to accurately calculate the energy storage capacity; or dynamic optimization control can be achieved through machine learning models. These advanced applications are all extensions and expansions under the control framework of this invention.

[0081] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A photovoltaic air conditioning control method, characterized in that, include: Acquire geographic information system data and satellite imagery data for the target area; Based on the geographic information system data, the buildings within the target area are classified; For different categories of buildings after classification, the satellite image data is processed using corresponding deep learning models to identify the roof areas of each category of buildings; Based on all identified rooftop areas, the total annual photovoltaic power generation potential of the target area is assessed. Based on the temporal and spatial characteristics of each type of building, a corresponding building energy consumption model is constructed. Based on the building energy consumption model, the annual total power consumption of the air conditioning system of all types of buildings is calculated and aggregated to obtain the annual total power consumption of the air conditioning system of the target area. The photovoltaic allocation ratio is calculated based on the total annual photovoltaic power generation potential and the total annual air conditioning system power consumption of the target area, and the photovoltaic air conditioning system control strategy of the target area is determined based on the photovoltaic allocation ratio.

2. The photovoltaic air conditioning control method according to claim 1, characterized in that: The classification of buildings within the target area includes: Based on the usage attributes in the geographic information system data, the buildings within the target area are divided into civil buildings, residential buildings, and industrial buildings. The civil buildings mentioned include buildings used for commercial, office, or entertainment purposes; The residential building is a building used for residential purposes; The industrial building is a building used for production and manufacturing purposes.

3. The photovoltaic air conditioning control method according to claim 1, characterized in that: The identification of roof areas for each type of building includes: The satellite image data is semantically segmented by calling a deep learning model pre-configured for each type of building. The deep learning model is a U2-Net network, which extracts image features through an encoder and reconstructs the image size through a decoder to output a segmentation map. The segmentation map is used to distinguish between roof areas and non-roof areas in the image.

4. The photovoltaic air conditioning control method according to claim 1, characterized in that: The assessment concluded that the total annual photovoltaic power generation potential of the target area includes: Based on all identified roof areas, photovoltaic modules are simulated and arranged using a preset filling algorithm to calculate the total installable area; The total annual solar radiation data for the target area is used to assess the total annual photovoltaic power generation potential based on the total installable area.

5. A photovoltaic air conditioning control method according to claim 4, characterized in that: The engineering constraints followed by the filling algorithm during simulated layout include: The photovoltaic modules are placed at a predetermined safe distance inward from the boundary of the roof area; Photovoltaic modules are arranged in an orthogonal grid with row and column alignment; A maintenance spacing of not less than the preset minimum value should be reserved between each photovoltaic module; Identify and avoid obstacles within the roof area.

6. The photovoltaic air conditioning control method according to claim 1, characterized in that: The time characteristics include typical daily operating periods and total annual operating time defined based on building use attributes; The spatial characteristics include the number of building floors, roof area, and the percentage of each floor that requires cooling or heating.

7. The photovoltaic air conditioning control method according to claim 1, characterized in that: The calculation of the photovoltaic allocation ratio includes: Obtain the total annual photovoltaic power generation potential and the total annual air conditioning system power consumption of the target area; Calculate the ratio of the total annual photovoltaic power generation potential to the total annual air conditioning system power consumption, where the ratio is the photovoltaic allocation ratio.

8. The photovoltaic air conditioning control method according to claim 1, characterized in that: The photovoltaic air conditioning system control strategy for determining the target area based on the photovoltaic allocation ratio includes: The photovoltaic allocation ratio is compared with a preset threshold range, and the corresponding system architecture scheme is determined based on the comparison result; The system architecture scheme includes: When the photovoltaic allocation ratio is not lower than the first preset threshold, the off-grid operation system architecture scheme is determined; When the photovoltaic allocation ratio is between the first preset threshold and the second preset threshold, a system architecture scheme for hybrid power supply of photovoltaic and energy storage is determined. When the photovoltaic allocation ratio is not higher than the second preset threshold, a system architecture scheme that complements photovoltaics and the power grid is determined.

9. A photovoltaic air conditioning control system, used to operate the photovoltaic air conditioning control method according to any one of claims 1-8, characterized in that, include: The module includes a data acquisition module, a building classification module, a roof identification module, a photovoltaic potential assessment module, a power consumption calculation module, and a control strategy generation module, among which: The data acquisition module is used to acquire geographic information system data and satellite image data of the target area; The building classification module classifies buildings within the target area based on the geographic information system data; The roof recognition module is used to process the satellite image data using corresponding deep learning models for different categories of buildings after classification, so as to identify the roof area of ​​each category of building; The photovoltaic potential assessment module is used to assess the total annual photovoltaic power generation potential of the target area based on all identified rooftop areas. The power consumption calculation module constructs a corresponding building energy consumption model based on the time and spatial characteristics of each type of building, calculates and aggregates the annual total air conditioning system power consumption of all types of buildings based on the building energy consumption model, and obtains the annual total air conditioning system power consumption of the target area. The control strategy generation module is used to calculate the photovoltaic allocation ratio based on the total annual photovoltaic power generation potential and the total annual air conditioning system power consumption of the target area, and to determine the photovoltaic air conditioning system control strategy of the target area based on the photovoltaic allocation ratio.

10. A photovoltaic air conditioner, characterized in that, The photovoltaic air conditioning control method described in any one of claims 1-8 is adopted.

Images