A multi-split configuration adaptive method, device, system and program product

Abstract

The application discloses a multi-connected machine configuration adaptive method, device, system and program product, and relates to the technical field of energy-saving air conditioners. The multi-connected machine configuration adaptive method comprises the following steps: obtaining building plan information and configuration requirement information, verifying the configuration data; generating an optimal configuration scheme through preset artificial intelligence algorithm based on multivariate objective function iterative optimization; constructing a three-dimensional installation model to generate visual installation guide information; and dynamically adjusting the operation parameters of the temperature adjusting equipment according to the collected operation state data after the installation of the temperature adjusting equipment is completed. The application performs multivariate optimization calculation on total energy consumption, installation cost and energy efficiency loss through the preset artificial intelligence algorithm, automatically generates an optimal configuration scheme, and dynamically adjusts the operation parameters according to the operation state data after the installation is completed, so that intelligent configuration and dynamic adaptive adjustment of the multi-connected machine system are realized, and the overall energy efficiency of the multi-connected machine system is improved.

Description

Technical Field

[0001] This invention relates to the field of energy-saving air conditioning technology, and in particular to a method, apparatus, system and program product for adaptive configuration of multi-split air conditioning units. Background Technology

[0002] A multi-split air conditioning system is a temperature-regulating equipment system consisting of an outdoor unit connected to multiple indoor units. It is widely used in various building scenarios such as residences and offices. In related technologies, the configuration of a multi-split air conditioning system usually requires on-site design and scheme formulation by professional engineers based on the building structure and usage requirements. Once the configuration scheme is determined, the multi-split air conditioning system operates according to a fixed configuration mode.

[0003] However, the above configuration method relies on manual experience and lacks intelligent and dynamic adaptive capabilities. On the one hand, it is difficult to comprehensively consider the impact of various factors such as pipe length and indoor unit layout on system operation during the configuration process, which can easily lead to unreasonable pipe length and unscientific indoor unit layout. In particular, when the pipe is too long, the cooling efficiency can be reduced by more than 15%. On the other hand, after the temperature control equipment is installed, it is impossible to dynamically adjust the operating parameters according to the actual operating status, resulting in a low overall energy efficiency of the multi-split system. Summary of the Invention

[0004] This invention provides a multi-split air conditioning configuration adaptive method, device, system, and program product, aiming to solve the technical problem that the configuration of multi-split air conditioning systems in the prior art relies on manual experience and lacks intelligent and dynamic adaptive capabilities, resulting in low overall energy efficiency of multi-split air conditioning systems.

[0005] In a first aspect, embodiments of the present invention provide a multi-unit configuration adaptive method, comprising: The building floor plan information and the corresponding configuration requirements information are obtained separately, and then verified and processed according to preset rules to obtain configuration data. Based on the configuration data, multivariate optimization calculations are performed using a preset artificial intelligence algorithm, and iterative optimization is performed based on a pre-constructed multivariate objective function to generate an optimal configuration scheme that meets preset constraints; wherein, the variables of the multivariate objective function include total energy consumption, installation cost, and energy efficiency loss; A three-dimensional installation model is constructed, and the installation location and path of the temperature control equipment are planned and marked based on the optimal configuration scheme to generate visual installation guidance information; After the temperature control device is installed based on the visual installation guide information, the operating status data of the temperature control device is collected, and the operating parameters of the temperature control device are dynamically adjusted according to the operating status data.

[0006] Secondly, embodiments of the present invention provide a multi-unit configuration adaptive device, comprising: The data configuration unit is used to acquire building floor plan information and corresponding configuration requirements information, and perform verification processing in combination with preset rules to obtain configuration data. The scheme configuration unit is used to perform multivariate optimization calculations based on the configuration data using a preset artificial intelligence algorithm, and to perform iterative optimization based on a pre-constructed multivariate objective function to generate an optimal configuration scheme that meets preset constraints; wherein, the variables of the multivariate objective function include total energy consumption, installation cost, and energy efficiency loss; The model building unit is used to build a three-dimensional installation model and plan and mark the installation position and path of the temperature control equipment based on the optimal configuration scheme, and generate visual installation guidance information; The parameter adjustment unit is used to collect the operating status data of the temperature control device after the installation of the temperature control device is completed based on the visual installation guidance information, and to dynamically adjust the operating parameters of the temperature control device according to the operating status data.

[0007] Thirdly, embodiments of the present invention provide a multi-unit configuration adaptive system, including a multi-unit configuration adaptive device as described in the second aspect, a terminal device, multiple temperature control devices, a cloud service platform, an intelligent sensor network, and an edge gateway; The multi-unit configuration adaptive device is deployed on the cloud service platform; The terminal device is communicatively connected to the cloud service platform and is used to upload the building floor plan information and configuration requirements information, and to display the visual installation guidance information. Multiple temperature control devices are deployed on the building site to perform operational adjustments based on issued operating parameters; wherein, the temperature control devices include indoor units and outdoor units; The intelligent sensor network is deployed on multiple temperature control devices to collect the operating status data; The edge gateway is communicatively connected to the smart sensor network, multiple temperature control devices, and the cloud service platform, respectively, and is used to aggregate and convert the operating status data.

[0008] Fourthly, embodiments of the present invention provide a computer program product, the computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, which, when executed by a computer, cause the computer to perform the multi-unit configuration adaptive method according to any one of claims 1-7.

[0009] This invention provides a multi-split air conditioning system configuration adaptive method, including acquiring building floor plan information and corresponding configuration requirements, verifying them using preset rules to obtain configuration data; performing multivariate optimization calculations based on the configuration data using a preset artificial intelligence algorithm, and iteratively optimizing based on a pre-constructed multivariate objective function to generate an optimal configuration scheme that meets preset constraints; wherein the variables of the multivariate objective function include total energy consumption, installation cost, and energy efficiency loss; constructing a three-dimensional installation model, and planning and marking the installation location and path of the temperature control equipment based on the optimal configuration scheme to generate visual installation guidance information; after the installation of the temperature control equipment is completed based on the visual installation guidance information, collecting the operating status data of the temperature control equipment, and dynamically adjusting the operating parameters of the temperature control equipment according to the operating status data. This invention uses a preset artificial intelligence algorithm to perform multivariate optimization calculations on total energy consumption, installation cost, and energy efficiency loss, automatically generating an optimal configuration scheme, and dynamically adjusting the operating parameters according to the operating status data after installation, realizing intelligent configuration and dynamic adaptive adjustment of the multi-split air conditioning system, thereby improving the overall energy efficiency of the multi-split air conditioning system.

[0010] This invention also provides a multi-unit configuration adaptive device, a multi-unit configuration adaptive system, and a computer program product, which have the same beneficial effects as described above. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating a multi-unit configuration adaptive method provided in an embodiment of the present invention; Figure 2 This is another flowchart illustrating a multi-unit configuration adaptive method provided in an embodiment of the present invention; Figure 3 This is a schematic block diagram of a multi-unit configuration adaptive device provided in an embodiment of the present invention; Figure 4 A schematic diagram of a multi-unit configuration adaptive system provided in an embodiment of the present invention. Figure 1 ; Figure 5 A schematic diagram of a multi-unit configuration adaptive system provided in an embodiment of the present invention. Figure 2 .

[0013] Explanation of reference numerals in the attached figures: 300. Multi-unit configuration adaptive device; 301. Data configuration unit; 302. Scheme configuration unit; 303. Model building unit; 304. Parameter adjustment unit. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0016] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0017] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0018] Please see below. Figure 1 , Figure 1 The flowchart of a multi-unit configuration adaptive method provided in an embodiment of the present invention specifically includes steps S101 to S104.

[0019] S101. Obtain the building floor plan information and the corresponding configuration requirements information respectively, and perform verification processing in combination with preset rules to obtain configuration data; S102. Based on the configuration data, multivariate optimization calculations are performed using a preset artificial intelligence algorithm, and iterative optimization is performed based on a pre-constructed multivariate objective function to generate an optimal configuration scheme that meets preset constraints; wherein, the variables of the multivariate objective function include total energy consumption, installation cost, and energy efficiency loss; S103. Construct a three-dimensional installation model, and based on the optimal configuration scheme, plan and mark the installation location and path of the temperature control equipment to generate visual installation guidance information; S104. After the installation of the temperature control device is completed based on the visual installation guidance information, the operating status data of the temperature control device is collected, and the operating parameters of the temperature control device are dynamically adjusted according to the operating status data.

[0020] In step S101, the building floor plan information reflects the spatial structure and layout of the building, while the configuration requirement information reflects the actual configuration requirements for the multi-split system. After acquiring the building floor plan information and the corresponding configuration requirement information, the acquired information is verified using preset rules to ensure its completeness and rationality, thereby obtaining configuration data and providing a data foundation for subsequent optimization calculations.

[0021] In one embodiment, step S101 includes: Obtain the building floor plan information uploaded by the user, identify the room boundaries, area and layout in the building floor plan information, and integrate them to obtain the room layout information; Missing room layout information is detected, corresponding prompts are generated, and the reasonableness is verified based on preset rules to obtain the verification results. Obtain configuration requirement information input by the user; wherein, the configuration requirement information includes usage scenario information and personalized requirement information; Match the corresponding preset heat load template based on the usage scenario information; Based on the geographic information system, the system matches preset environmental parameters of the building's location and generates initial heat load information based on these preset environmental parameters. The room layout information, configuration requirements information, preset heat load template, initial heat load information, and verification results are encrypted and stored, and then integrated to obtain configuration data.

[0022] In this embodiment, users can upload architectural floor plan information through an application (APP) installed on a mobile phone or other terminal. The uploaded architectural floor plan information supports multiple file formats, including Portable Document Format (PDF), DWG format (Drawing, a computer-aided design drawing file format), and Portable Network Graphics (PNG), to ensure compatibility with architectural drawings from different sources and in different formats. After obtaining the architectural floor plan information uploaded by the user, image recognition and parsing are performed on the architectural floor plan information to automatically identify the room boundaries, areas, and layouts, that is, to identify the wall outlines, bay dimensions, and relative positional relationships between the rooms within the building, and then integrate the recognition results into structured room layout information. For example, for a 100㎡ office building, four offices, one meeting room, and one reception area can be identified from its building floor plan information, and the boundaries and areas of each room can be determined respectively; similarly, for a 150㎡ villa building (including three bedrooms, a living room, and a kitchen), users can directly upload building floor plan information in DWG format, and the system can also identify the boundaries, areas, and layouts of each room.

[0023] After obtaining the room layout information, a missing information detection process is performed. This involves real-time analysis to identify incomplete information such as unmarked room areas, unclosed room boundaries, or missing key dimensions. When a missing item is detected, a corresponding prompt is generated and provided to the user through graphical guidance. For example, a prompt like "Living room area not filled in, please confirm dimensions" is generated to guide the user in completing the missing information, thereby ensuring the completeness and accuracy of the user's input. Simultaneously, the room layout information is validated for reasonableness based on preset rules. These preset rules can be judgment rules pre-set according to building design specifications and experience in configuring temperature control equipment. For example, a reasonable lower limit for bedroom area is set; when the identified bedroom area is less than 5 square meters, a warning is triggered, prompting the user to confirm whether the room information is correct. The resulting validation results characterize whether the room layout information is complete and reasonable, serving as the data quality basis for subsequent optimization calculations.

[0024] Furthermore, the system acquires configuration requirements input by the user. These requirements include usage scenario information and personalized requirements. Usage scenario information characterizes the actual purpose of the building or room; users can select a usage scenario through the graphical interface provided by the app, such as a high-end office, smart home bedroom, or commercial retail. Personalized requirements information characterizes the user's personalized requirements for the multi-split system's performance, such as temperature control modes (e.g., zoned constant temperature, energy-saving mode), indoor-outdoor temperature difference thresholds, and budget range, where the budget range can be an optional input. For example, for the aforementioned 100㎡ office building, the user can select the high-end office usage scenario and set a zoned constant temperature combined with an energy-saving mode, along with a moderate budget range; for the aforementioned 150㎡ villa building, the user can select the smart home bedroom usage scenario and set personalized requirements such as a nighttime sleep mode. After obtaining the usage scenario information, a corresponding preset heat load template is matched based on the usage scenario information. The preset heat load template is a heat load calculation template pre-built for different usage scenarios. Since the heating characteristics and comfort requirements inside the building differ under different usage scenarios, for example, the preset heat load template corresponding to the office scenario needs to consider the heat generation of office equipment such as computers and the heat generation of lights, while the preset heat load template corresponding to the bedroom scenario focuses on human comfort. Therefore, by matching the preset heat load template, the heat load calculation can be made to fit the actual usage of the building.

[0025] Simultaneously, based on the Geographic Information System (GIS), preset environmental parameters of the building's location are matched. These preset environmental parameters include climate parameters of the building's location. For example, when the building is located in Beijing, climate parameters such as "average winter temperature in Beijing -5℃" can be automatically matched. On this basis, information on the building's insulation materials (such as whether glass curtain walls are used, insulation layer thickness, etc.) can be combined to correct for climate impacts, thereby generating initial heat load information based on the preset environmental parameters. The initial heat load information is used to characterize the building's basic heating and cooling load requirements under local climate conditions, providing a load benchmark for subsequent multivariate optimization calculations.

[0026] Finally, the room layout information, configuration requirements, preset heat load template, initial heat load information, and verification results are encrypted and stored. Specifically, blockchain encryption can be used to ensure the privacy and security of user input data by leveraging the immutability and traceability of blockchain data. The encrypted information is then integrated into a structured data packet, such as a JSON (JavaScript Object Notation, a lightweight structured data exchange format), to obtain configuration data. This configuration data can then be used as input to a preset artificial intelligence algorithm for multivariate optimization calculations via a unified data interface. In this way, through automatic identification, missing data detection, and rationality verification of building floor plan information and configuration requirements, the integrity, accuracy, and security of the configuration data are ensured, laying the data foundation for generating the optimal configuration scheme.

[0027] In step S102, the configuration data is used as input to a preset artificial intelligence algorithm for multivariate optimization calculation. During the calculation, iterative optimization is performed based on a pre-constructed multivariate objective function. Since the variables of the multivariate objective function include total energy consumption, installation cost, and energy efficiency loss, a comprehensive balance can be made between these factors during the iterative optimization process. The configuration scheme obtained through iterative optimization is then verified against preset constraints, thereby generating an optimal configuration scheme that satisfies the preset constraints. Compared to relying on human experience for on-site design and scheme formulation, automatically generating the optimal configuration scheme through a preset artificial intelligence algorithm achieves intelligent configuration of the multi-split air conditioning system.

[0028] In one embodiment, step S102 includes: The current configuration status is used as the state space, and the indoor unit model adjustment and pipe routing adjustment are used as the action space. Based on the preset heat load model and the multivariate objective function, the state space and action space are iteratively trained to generate multiple candidate configuration schemes. A fitness function is constructed based on the total energy consumption, installation cost, and preset differential pressure exceeding penalty, and the fitness function is used to filter and retain each of the candidate configuration schemes; Perform cross-mutation operations on the selected and retained candidate configuration schemes to generate new candidate configuration schemes and iteratively filter them. The candidate configuration schemes after the iteration are verified according to the preset constraints, and the optimal configuration scheme is selected from them based on the multivariate objective function.

[0029] In this embodiment, the preset artificial intelligence algorithm can be a combination of deep reinforcement learning (DRL) and genetic algorithm (GA), and a constraint satisfaction problem (CSP) verification mechanism is introduced.

[0030] Specifically, the current configuration state is used as the state space, which includes the currently selected indoor unit and the determined pipe routing. Simultaneously, the indoor unit model adjustment and pipe routing adjustment are used as the action space, where "adding or replacing a certain model of indoor unit" and "adjusting the pipe routing" are the set of executable actions for the algorithm. Based on this, the Proximal Policy Optimization (PPO) algorithm from deep reinforcement learning can be used to iteratively train the state space and action space based on a preset heat load model and the multivariate objective function. The preset heat load model can be calculated based on preset standards and is used to characterize the heating and cooling load requirements of each room under different operating conditions. During iterative training, the algorithm continuously executes configuration adjustment actions in the action space within the state space, and uses the value of the multivariate objective function as reward feedback to guide the algorithm towards configurations with lower total energy consumption, lower installation costs, and less energy efficiency loss. For example, the state space and action space can be iteratively trained for 1000 rounds, generating more than 1000 candidate configuration schemes, forming a candidate scheme space.

[0031] Furthermore, a fitness function is constructed based on the total energy consumption, installation cost, and a preset penalty for exceeding the differential pressure limit. The penalty for exceeding the differential pressure limit refers to the penalty applied when the differential pressure in the pipeline corresponding to a candidate configuration exceeds a preset range. Since exceeding the differential pressure limit means excessive flow resistance of the refrigerant in the pipeline, which will cause a decrease in energy efficiency, candidate configurations with excessive differential pressure need to be penalized during the scheme selection stage. For example, the fitness function can be constructed as follows: in, The fitness of candidate configuration schemes; The total energy consumption corresponding to the candidate configuration scheme; The installation cost corresponding to the candidate configuration options; The preset penalty for exceeding the differential pressure limit is denoted by 0.5 and 0.3, respectively, representing the weighting coefficients for installation cost and the penalty for exceeding the differential pressure limit. This fitness function shows that the lower the total energy consumption, the lower the installation cost, and the smaller the degree of differential pressure exceeding the limit in the candidate configuration, the higher its fitness. The higher the fitness, the better. The fitness of each candidate configuration is calculated using the fitness function and then filtered and retained. For example, in each generation of candidate configurations, the top 20% of high-fitness individuals are retained, while the remaining low-fitness individuals are eliminated, to ensure that the retained candidate configurations are of higher overall quality.

[0032] Furthermore, crossover and mutation operations are performed on the selected candidate configuration schemes. Crossover and mutation are basic evolutionary operations in genetic algorithms: crossover involves exchanging and recombining some configuration features (such as indoor unit models in some rooms or pipe routing in some sections) between two candidate configuration schemes; mutation involves randomly changing a configuration feature in a candidate configuration scheme with a certain probability. New candidate configuration schemes are generated through crossover and mutation operations, and the fitness calculation and selection process is repeated on the newly generated candidate configuration schemes to achieve iterative selection. This allows the overall quality of candidate configuration schemes to improve with each iteration, avoiding the optimization process from getting trapped in local optima.

[0033] Finally, the iterative candidate configuration schemes are verified according to preset constraints. These preset constraints are engineering constraints required to ensure the normal and safe operation of the multi-split system. For example, preset constraints can be set as follows: total pipe length less than or equal to 120m, coverage area of ​​each indoor unit less than or equal to 50㎡, compressor frequency fluctuation within ±10%, and outdoor unit load rate less than or equal to 90%. Candidate configuration schemes that do not meet any of these preset constraints are directly eliminated. Among the candidate configuration schemes that meet all preset constraints, the objective function value of each candidate configuration scheme is calculated based on the multivariate objective function. The candidate configuration scheme that best balances total energy consumption, installation cost, and energy efficiency loss (i.e., has the optimal objective function value) is selected as the optimal configuration scheme. The optimal configuration scheme includes indoor unit model selection, indoor unit quantity configuration, pipe length allocation, and power matching, and can be output as a configuration report in JSON format.

[0034] For example, for the aforementioned 100㎡ office building, in the multivariate optimization calculation process, the range of selectable indoor unit models (e.g., 3HP, 5HP) and outdoor unit models (e.g., 12HP) can be initially determined based on the heat load requirements, where "HP" is a commonly used power measurement unit in the air conditioning industry. Then, through deep reinforcement learning iterative training, more than 1000 candidate configuration schemes are generated. After fitness screening, cross-mutation iteration, and verification under preset constraints, the optimal configuration scheme can be: three 9K indoor units (covering the conference room, reception area, and main office area respectively) and one 18K unit (covering the secondary office area), where K represents the cooling capacity in British Thermal Units per hour (BTU / h), and 9K means a cooling capacity of approximately 9000 BTU / h; one 48K outdoor unit; a total pipe length of 98m, meeting the preset constraint of less than or equal to 120m; a compressor frequency of 55Hz with a fluctuation range within ±10%; and a coefficient of performance (COP) of 120%. Performance (COP, the ratio of cooling capacity to power consumption) reaches 4.5; simultaneously, a risk warning can be output in the configuration report: "There are 6 pipe bends, it is recommended to add supports." For example, for the aforementioned 150㎡ villa, the optimal configuration could be: three 12K indoor units (covering three bedrooms) and two 18K units (covering the living room and kitchen respectively), and one 55K outdoor unit. The total pipe length is 112m, close to the upper limit of the preset constraints. In this case, the total pipe length can be marked in the configuration report for focused monitoring during subsequent operation. Thus, through the combined application of deep reinforcement learning, genetic algorithms, and constraint satisfaction verification, the optimal configuration that balances energy consumption, cost, and energy efficiency is automatically searched in the configuration scheme space, replacing the on-site design process relying on manual experience, significantly shortening configuration time and improving configuration efficiency.

[0035] In one embodiment, step S102 further includes: Acquire performance data of multi-split air conditioning units with different temperature control devices; the performance data of multi-split air conditioning units includes outdoor unit power consumption, pipe pressure drop loss, pipe length, pipe diameter and refrigerant pressure drop parameters; The total energy consumption is calculated based on the outdoor unit power consumption and pipeline pressure drop loss. Obtain the number of indoor units and calculate the installation cost based on the number of indoor units and the length of the pipes; Energy efficiency loss is calculated based on the pipe length, pipe diameter, and refrigerant pressure drop parameters. The total energy consumption, installation cost, and energy efficiency loss are weighted according to preset weights to construct a multivariate objective function.

[0036] In this embodiment, the multivariate objective function is constructed based on the actual performance parameters of different temperature control devices to ensure the scientific validity and engineering feasibility of the optimization calculation results. Specifically, the performance data of multi-split air conditioning units of different temperature control devices are first obtained. The multi-split air conditioning unit performance data can be obtained from a pre-built multi-split air conditioning unit performance database, which can include the performance parameters of more than 200 brands and models of temperature control devices. The multi-split air conditioning unit performance data includes the outdoor unit power consumption, pipe pressure drop loss, pipe length, pipe diameter, and refrigerant pressure drop parameters of each model of temperature control device. In addition, it can also include parameters such as cooling capacity, energy efficiency ratio, and maximum pipe length of each model of temperature control device, thereby covering various types of temperature control devices that can be selected in actual engineering and providing complete data support for the construction of the objective function.

[0037] After acquiring the performance data of the multi-split air conditioning system, the total energy consumption is calculated based on the outdoor unit's power consumption and pipe pressure drop loss. For example, the total energy consumption can be expressed as: in, Total energy consumption; For outdoor unit power consumption; This refers to the pressure drop loss in the pipeline. The pressure drop loss in the pipeline is determined by the pipeline length, the number of bends, and the pipeline material. That is, the longer the pipeline and the more bends there are, the greater the pressure drop loss caused by the refrigerant flowing in the pipeline. The more energy is required to overcome this pressure drop loss, and the higher the total energy consumption will be.

[0038] Furthermore, the number of indoor units is obtained, and the installation cost is calculated based on the number of indoor units and the pipe length. For example, the installation cost can be expressed as: in, For installation costs; This refers to the number of indoor units; This is the unit price of the indoor unit; Total length of the pipeline; This represents the unit price per unit length of the pipe. It is evident that the more indoor units there are and the longer the total pipe length, the higher the corresponding equipment purchase and installation costs.

[0039] Furthermore, energy efficiency loss is calculated based on the pipe length, pipe diameter, and refrigerant pressure drop parameters. Energy efficiency loss is mainly caused by the refrigerant pressure drop due to excessive pipe length, which can be calculated using the following empirical formula: in, For refrigerant pressure drop; This is the drag coefficient; This refers to the length of the pipe. Density of the refrigerant; This refers to the refrigerant flow rate; Pipe diameter; , , All of these parameters pertain to the refrigerant pressure drop. As can be seen from this formula, the longer the pipe length and the smaller the pipe diameter, the greater the refrigerant pressure drop. The larger the value, the greater the energy loss, which is consistent with the actual engineering phenomenon that cooling efficiency can be reduced by more than 15% when the pipeline is too long; based on this formula, the energy loss under each candidate configuration can be quantified.

[0040] Finally, the total energy consumption, installation cost, and energy efficiency loss are weighted according to preset weights to construct a multivariate objective function. For example, the multivariate objective function can be expressed as: in, The objective function value; Total energy consumption; For installation costs; Energy loss; , , These are preset weights corresponding to total energy consumption, installation cost, and energy efficiency loss, respectively. For example, the preset weights can be set to... =1、 =0.6、 =0.4, that is, the multivariate objective function is +0.6 +0.4 This indicates that the primary optimization objective during the optimization process is to reduce total energy consumption, while also considering the control of installation costs and energy efficiency losses. The specific values ​​of the preset weights can be flexibly adjusted according to actual project needs. For example, in application scenarios with tight budgets, the preset weight corresponding to installation costs can be appropriately increased. By iteratively optimizing the multivariate objective function, the generated configuration scheme can achieve the best balance between total energy consumption, installation cost, and energy efficiency loss, fundamentally avoiding the problem that manual configuration cannot comprehensively consider multiple factors.

[0041] In step S103, a three-dimensional installation model is constructed to present the installation environment of the temperature control equipment. Based on the optimal configuration scheme, the installation position and path of the temperature control equipment are planned and marked in the three-dimensional installation model, thereby generating visual installation guidance information. This makes the installation process of the temperature control equipment visible, making it easier to accurately complete the installation of the temperature control equipment according to the visual installation guidance information and reducing installation errors.

[0042] In one embodiment, step S103 includes: Import architectural space data into a 3D model to construct a 3D installation model; The installation location of the temperature control device is marked in the three-dimensional installation model, and the path of the temperature control device is planned and marked using a preset path planning algorithm to obtain the installation planning result; The pressure difference distribution is calculated by calling a preset fluid dynamics model, and the risk level of the pressure difference distribution is divided according to a preset pressure difference threshold, generating a pressure difference heat map superimposed on the three-dimensional installation model; Visualized installation guidance information is generated using the installation planning results and differential pressure heat map.

[0043] In this embodiment, building space data is imported into a 3D model to construct a 3D installation model. Specifically, building space data can be imported from a Building Information Modeling (BIM) model in Industry Foundation Classes (IFC) format. This building space data includes structural information such as beams, columns, ceilings, and wall thicknesses, thereby reconstructing the real building space in the 3D model and obtaining a 3D installation model that is highly consistent with the actual building site. This provides an accurate spatial basis for planning and marking the installation location and path of the temperature control equipment.

[0044] Furthermore, the installation location of the temperature control equipment is marked in the three-dimensional installation model. When determining the installation point of the indoor unit, areas with a clear height of less than 2.2m under beams can be automatically avoided, and the center point of the ceiling is preferentially selected as the installation location to ensure the installation space and air supply effect of the indoor unit. When determining the location of the outdoor unit base, geographic information system data can be used to automatically avoid areas with direct sunlight and duct blockages, and the suggested installation location is dynamically marked in coordinate form in the three-dimensional installation model, for example, marked "Suggested installation location (X=12.3m, Y=8.7m)". At the same time, the path of the temperature control equipment (i.e., the pipe path connecting the indoor and outdoor units) is planned and marked by a preset path planning algorithm. The preset path planning algorithm can adopt an artificial intelligence path planning algorithm, which automatically avoids building structures such as beams and columns during the planning process, and uses minimizing the number of bends as the planning objective to reduce the pressure drop loss caused by pipe bends. Thus, the installation planning result including the installation location marking of the temperature control equipment and the pipe path marking is obtained.

[0045] Furthermore, a preset fluid dynamics model is invoked to calculate the pressure differential distribution. This preset fluid dynamics model can be a simplified Computational Fluid Dynamics (CFD) model, calculating the pressure differential distribution at various points in the piping system based on parameters such as pipe material, pipe diameter, refrigerant type, and flow velocity. Then, the pressure differential distribution is classified into risk levels according to preset pressure differential thresholds. For example, areas with a pressure differential greater than 0.15 MPa are classified as high-risk and marked in red; areas with a pressure differential between 0.10 and 0.15 MPa are classified as medium-risk and marked in yellow; and areas with a pressure differential less than 0.10 MPa are classified as safe and marked in green. The pressure differential distribution after risk level classification is then superimposed on the three-dimensional installation model in the form of a color gradient to generate a pressure differential heat map, allowing installers to intuitively identify pressure differential risk areas in the piping system.

[0046] Finally, the installation planning results and differential pressure heat map are used to generate visualized installation guidance information. This visualized installation guidance information may include a 3D installation simulation diagram, which shows the layout and connection methods of the indoor unit, outdoor unit, and piping, and overlays the differential pressure heat map. For example, in the aforementioned 100㎡ office building, the generated 3D installation simulation diagram shows that the piping route avoids beam and column structures, and the differential pressure heat map shows that the red high-risk areas are concentrated at the end of the pipes. In this case, the visualized installation guidance information can also provide an installation suggestion to "install a buffer tank at the end" to buffer pressure fluctuations of the refrigerant at the end. Furthermore, the visualized installation guidance information can also be presented through augmented reality (AR) interaction. When a user turns on their terminal camera, Simultaneous Localization and Mapping (SLAM) technology is used to locate the current viewpoint in real time, overlaying a virtual pipeline model onto the real scene in an AR manner. This allows for rotation and zooming of the virtual pipeline model, and users can click on a pipeline to view its corresponding parameters (e.g., material PPR (Polypropylene Random), pressure resistance 0.8 MPa). Users can also compare the virtual path with the actual wall in real time to identify and avoid potential installation conflicts before formal construction. In addition, a printable 3D installation manual can be automatically generated as a deliverable based on the installation planning results and pressure differential thermal map. This manual includes the installation location coordinates of each temperature control device, a pipeline length verification table, a tool requirement list, and installation step diagrams, facilitating installation by personnel. In this way, by using various forms of visual installation guidance information such as 3D installation models, differential pressure heat maps, and AR interaction, the installation process of temperature control equipment can be visualized and standardized, effectively reducing installation errors and lowering installation costs.

[0047] In step S104, after the installation of the temperature control device is completed based on the visual installation guidance information, the operating status data of the temperature control device during actual operation is collected, and the operating parameters of the temperature control device are dynamically adjusted according to the operating status data, so that the temperature control device can compensate for energy efficiency loss in a timely manner according to the actual operating status, maintain a stable and efficient operating status, and realize the dynamic adaptive adjustment of the multi-split system.

[0048] In one embodiment, step S104 includes: The pipeline pressure difference, temperature, energy efficiency ratio, and operating current are collected separately and integrated to obtain the operating status data; The operating status data is analyzed in real time to identify excessive pipeline pressure difference and excessive room temperature. The compressor operating frequency of the temperature control equipment is adjusted for excessive pipeline pressure difference, and the air volume of the indoor unit of the temperature control equipment is adjusted for excessive room temperature.

[0049] In this embodiment, after the temperature control equipment is installed based on the visualized installation guidance information, the pipeline pressure difference, temperature, energy efficiency ratio, and operating current can be collected by sensors deployed on the temperature control equipment. Specifically, a sensor node can be deployed on both the indoor and outdoor units of the temperature control equipment. Each sensor node can include a pressure sensor, a temperature probe, and a current transformer. The pressure sensor is used to collect the pipeline pressure difference with an accuracy of ±0.01 MPa to ensure the accuracy of pressure difference monitoring. The temperature probe is used to collect the temperature, which can include both indoor and outdoor temperatures. The current transformer is used to collect the compressor's operating current. Simultaneously, the energy efficiency ratio during the operation of the temperature control equipment is collected. According to a preset collection cycle, the above parameters are collected at regular intervals, and the collected pipeline pressure difference, temperature, energy efficiency ratio, and operating current are integrated and aligned by time to obtain operating status data that comprehensively reflects the actual operating status of the temperature control equipment.

[0050] After obtaining the operating status data, the data is analyzed in real time to identify situations where the pipeline pressure difference exceeds the limit and the room temperature exceeds the limit. The pipeline pressure difference exceeding the limit refers to a situation where the pipeline pressure difference exceeds a preset upper limit, and the room temperature exceeding the limit refers to a situation where the room temperature exceeds a set temperature range. For situations where the pipeline pressure difference exceeds the limit, the compressor operating frequency of the temperature control device is adjusted. For example, when the pipeline length exceeds a threshold (e.g., the total pipeline length is greater than 100m) resulting in a pipeline pressure difference greater than 0.15MPa, the compressor operating frequency is automatically reduced by 5%. By reducing the compressor operating frequency, the refrigerant flow rate is reduced, and the pipeline pressure drop is reduced, thereby compensating for the energy efficiency loss caused by the excessive pipeline length. As another example, for a 150㎡ villa building with a total pipeline length of 112m, close to the upper limit of the constraint, when real-time analysis detects a pipeline pressure difference close to 0.15MPa, the compressor operating frequency can be automatically reduced by 3% during operation to compensate for energy efficiency loss and ensure stable system energy efficiency. Meanwhile, in response to the room temperature exceeding the limit, the airflow of the indoor unit of the temperature control device is adjusted: for example, when real-time analysis detects that a certain room frequently exceeds the temperature limit, the airflow of the indoor unit corresponding to that room is dynamically adjusted to bring the room temperature back to the set range, avoiding overload of the entire unit due to temperature imbalance in a single room; as another example, after a week of operation, real-time analysis of the aforementioned 150㎡ villa building found that the temperature fluctuation of a certain bedroom was large, so the airflow of the indoor unit corresponding to that bedroom was automatically adjusted to improve the stability of the room temperature control.

[0051] Through the above methods, the temperature control equipment can continuously and dynamically adjust itself based on the actual collected operating status data after installation. That is, when the pipeline pressure difference exceeds the limit, the compressor operating frequency is adjusted; when the room temperature exceeds the limit, the air volume of the indoor unit is adjusted. This allows the multi-split system to compensate for energy loss and maintain a stable and efficient operating state even under non-ideal configuration conditions such as long pipelines. This realizes the transformation of the multi-split system from static configuration to dynamic self-adaptation, achieving the technical effect of improving overall energy efficiency.

[0052] Combination Figure 2 As shown, in one embodiment, the multi-unit configuration adaptive method further includes: Receive user feedback information and perform semantic parsing on the feedback information to determine the corresponding scenario category, and classify the parsing results into the optimization strategy library; When the number of times the same scene category is classified reaches the preset number, the corresponding scene will be added to the preset operation mode and the corresponding heat load model will be updated. Optimization reports are generated periodically based on the optimization strategy library and runtime status data. Simultaneously, the optimization reports are updated to the system configuration strategy and configuration suggestions are pushed.

[0053] In this embodiment, the multi-unit configuration adaptive method can be configured as an architecture consisting of a user interaction and solution generation layer, a monitoring, optimization and feedback layer, and a physical deployment and operation layer. In the user interaction and solution generation layer, users upload drawings or requirements through the user interaction module (APP / graphical interface), the AI ​​configuration engine module (core algorithm) outputs configuration solutions accordingly, and the 3D modeling and visualization module (BIM / AR) generates corresponding visualization content. In the monitoring, optimization and feedback layer, the remote control and feedback module (APP / cloud platform) is used to generate installation instructions, support user monitoring and feedback, and use user preference feedback for policy updates. The operation monitoring and optimization module (sensor / AI edge computing, where edge computing refers to a computing method that provides computing and analysis services at the network edge close to the data source) is used to collect real-time operation data and use the data feedback for model iteration. In the physical deployment and operation layer, on-site installation is completed according to the installation instructions. After deployment, the multi-unit system hardware performs operation adjustments according to the issued dynamic optimization instructions. Thus, the configuration, installation, operation, feedback, and optimization processes are seamlessly connected, forming a complete closed-loop optimization mechanism. The feedback optimization process in this embodiment is implemented under this closed-loop architecture.

[0054] Specifically, the system receives user feedback. During actual use, users can submit feedback through the app, expressing their genuine experience and adjustment needs, such as "The bedroom is too cold at night; it is recommended to lower the cooling capacity." Upon receiving feedback, the system performs semantic analysis, identifying semantic elements such as room location, time period, perceived temperature, and adjustment intention through natural language semantic analysis. Based on this, it determines the corresponding scenario category (e.g., bedroom cooling at night) and categorizes the analysis results into an optimization strategy library. This library aggregates and stores the analysis results and corresponding optimization strategies for various types of user feedback by scenario category. As user feedback accumulates, the scenario categories and strategy content covered by the optimization strategy library continuously expand.

[0055] When the number of times the same scenario category is categorized reaches a preset number, the corresponding scenario is added to the preset operating mode, and the corresponding heat load model is updated. For example, the preset number can be set to 3 times: when a user submits feedback on discomfort caused by nighttime cooling 3 times consecutively, that is, when the number of times the bedroom nighttime cooling scenario category is categorized reaches 3 times, the scenario is added to the sleep mode preset operating mode, and the corresponding heat load model is updated accordingly, so that the heat load calculation of the bedroom during the nighttime period is more in line with the user's actual comfort needs; thereafter, the multi-split system can automatically execute the corresponding operating strategy according to the sleep mode during the nighttime period, without the need for the user to repeatedly submit feedback, realizing continuous learning and solidification of the user's personalized needs.

[0056] Furthermore, optimization reports are generated periodically based on the optimization strategy library and operational status data. For example, optimization reports can be generated quarterly, combining accumulated user feedback analysis results from the optimization strategy library with operational status data collected during operation. The content of the optimization reports may include user preference analysis conclusions, such as: over 70% of households prefer energy-saving mode, suggesting it be enabled by default. Simultaneously, the optimization reports are updated to the system configuration strategy and configuration suggestions are pushed: on the one hand, the analysis conclusions in the optimization reports are used to update the heat load model, optimize the system configuration strategy, and generate training data for a new version of the artificial intelligence algorithm, enabling the preset artificial intelligence algorithm to continuously evolve in subsequent multivariate optimization calculations, achieving data-driven continuous iteration; on the other hand, personalized configuration suggestions (such as suggesting that energy-saving mode be enabled by default) are pushed to users, thereby upgrading the multi-split system from passively responding to user needs to proactively predicting user needs, further improving the system's intelligent operation level and overall energy efficiency.

[0057] Combination Figure 3 As shown, Figure 3 This is a schematic block diagram of a multi-split air conditioner configuration adaptive device provided in an embodiment of the present invention. The multi-split air conditioner configuration adaptive device 300 includes: The data configuration unit 301 is used to acquire building floor plan information and corresponding configuration requirement information respectively, and perform verification processing in combination with preset rules to obtain configuration data. The scheme configuration unit 302 is used to perform multivariate optimization calculations based on the configuration data using a preset artificial intelligence algorithm, and to perform iterative optimization based on a pre-constructed multivariate objective function to generate an optimal configuration scheme that meets preset constraints; wherein, the variables of the multivariate objective function include total energy consumption, installation cost, and energy efficiency loss; The model building unit 303 is used to build a three-dimensional installation model and plan and mark the installation position and path of the temperature control equipment based on the optimal configuration scheme, and generate visual installation guidance information; The parameter adjustment unit 304 is used to collect the operating status data of the temperature control device after the installation of the temperature control device is completed based on the visual installation guidance information, and to dynamically adjust the operating parameters of the temperature control device according to the operating status data.

[0058] In this embodiment, the data configuration unit 301 acquires the building floor plan information and the corresponding configuration requirement information, and performs verification processing in combination with preset rules to obtain configuration data; the scheme configuration unit 302 performs multivariate optimization calculation based on the configuration data using a preset artificial intelligence algorithm, and performs iterative optimization based on a pre-constructed multivariate objective function to generate an optimal configuration scheme that meets preset constraints; wherein, the variables of the multivariate objective function include total energy consumption, installation cost, and energy efficiency loss; the model construction unit 303 constructs a three-dimensional installation model, and plans and marks the installation position and path of the temperature control equipment based on the optimal configuration scheme to generate visual installation guidance information; after the installation of the temperature control equipment is completed based on the visual installation guidance information, the parameter adjustment unit 304 collects the operating status data of the temperature control equipment, and dynamically adjusts the operating parameters of the temperature control equipment according to the operating status data.

[0059] In one embodiment, the data configuration unit 301 is specifically used for: Obtain the building floor plan information uploaded by the user, identify the room boundaries, area and layout in the building floor plan information, and integrate them to obtain the room layout information; Missing room layout information is detected, corresponding prompts are generated, and the reasonableness is verified based on preset rules to obtain the verification results. Obtain configuration requirement information input by the user; wherein, the configuration requirement information includes usage scenario information and personalized requirement information; Match the corresponding preset heat load template based on the usage scenario information; Based on the geographic information system, the system matches preset environmental parameters of the building's location and generates initial heat load information based on these preset environmental parameters. The room layout information, configuration requirements information, preset heat load template, initial heat load information, and verification results are encrypted and stored, and then integrated to obtain configuration data.

[0060] In one embodiment, the scheme configuration unit 302 is specifically used for: The current configuration status is used as the state space, and the indoor unit model adjustment and pipe routing adjustment are used as the action space. Based on the preset heat load model and the multivariate objective function, the state space and action space are iteratively trained to generate multiple candidate configuration schemes. A fitness function is constructed based on the total energy consumption, installation cost, and preset differential pressure exceeding penalty, and the fitness function is used to filter and retain each of the candidate configuration schemes; Perform cross-mutation operations on the selected and retained candidate configuration schemes to generate new candidate configuration schemes and iteratively filter them. The candidate configuration schemes after the iteration are verified according to the preset constraints, and the optimal configuration scheme is selected from them based on the multivariate objective function.

[0061] In one embodiment, the scheme configuration unit 302 is further specifically used for: Acquire performance data of multi-split air conditioning units with different temperature control devices; the performance data of multi-split air conditioning units includes outdoor unit power consumption, pipe pressure drop loss, pipe length, pipe diameter and refrigerant pressure drop parameters; The total energy consumption is calculated based on the outdoor unit power consumption and pipeline pressure drop loss. Obtain the number of indoor units and calculate the installation cost based on the number of indoor units and the length of the pipes; Energy efficiency loss is calculated based on the pipe length, pipe diameter, and refrigerant pressure drop parameters. The total energy consumption, installation cost, and energy efficiency loss are weighted according to preset weights to construct a multivariate objective function.

[0062] In one embodiment, the model building unit 303 is specifically used for: Import architectural space data into a 3D model to construct a 3D installation model; The installation location of the temperature control device is marked in the three-dimensional installation model, and the path of the temperature control device is planned and marked using a preset path planning algorithm to obtain the installation planning result; The pressure difference distribution is calculated by calling a preset fluid dynamics model, and the risk level of the pressure difference distribution is divided according to a preset pressure difference threshold, generating a pressure difference heat map superimposed on the three-dimensional installation model; Visualized installation guidance information is generated using the installation planning results and differential pressure heat map.

[0063] In one embodiment, the parameter adjustment unit 304 is specifically used for: The pipeline pressure difference, temperature, energy efficiency ratio, and operating current are collected separately and integrated to obtain the operating status data; The operating status data is analyzed in real time to identify excessive pipeline pressure difference and excessive room temperature. The compressor operating frequency of the temperature control equipment is adjusted for excessive pipeline pressure difference, and the air volume of the indoor unit of the temperature control equipment is adjusted for excessive room temperature.

[0064] In one embodiment, the multi-unit configuration adaptive device 300 is further configured to: Receive user feedback information and perform semantic parsing on the feedback information to determine the corresponding scenario category, and classify the parsing results into the optimization strategy library; When the number of times the same scene category is classified reaches the preset number, the corresponding scene will be added to the preset operation mode and the corresponding heat load model will be updated. Optimization reports are generated periodically based on the optimization strategy library and runtime status data. Simultaneously, the optimization reports are updated to the system configuration strategy and configuration suggestions are pushed.

[0065] Since the embodiments of the apparatus and the embodiments of the method correspond to each other, please refer to the description of the embodiments of the method for the embodiments of the apparatus, which will not be repeated here.

[0066] Combination Figure 4 and Figure 5 As shown, this embodiment of the invention also provides a multi-unit configuration adaptive system, including the multi-unit configuration adaptive device, terminal equipment, multiple temperature control devices, cloud service platform, intelligent sensor network and edge gateway as described above; The multi-unit configuration adaptive device is deployed on the cloud service platform; The terminal device is communicatively connected to the cloud service platform and is used to upload the building floor plan information and configuration requirements information, and to display the visual installation guidance information. Multiple temperature control devices are deployed on the building site to perform operational adjustments based on issued operating parameters; wherein, the temperature control devices include indoor units and outdoor units; The intelligent sensor network is deployed on multiple temperature control devices to collect the operating status data; The edge gateway is communicatively connected to the smart sensor network, multiple temperature control devices, and the cloud service platform, respectively, and is used to aggregate and convert the operating status data.

[0067] In this embodiment, combined with Figure 4As shown, from the perspective of functional architecture, the multi-unit configuration adaptive system can be divided into five layers: user interaction layer, AI configuration engine layer, 3D modeling and visualization layer, operation monitoring and optimization layer, and remote control and feedback layer. The user interaction layer supports uploading architectural floor plans, integrating Geographic Information System (GIS) data, providing graphical guidance, and receiving personalized input via a mobile app, corresponding to the aforementioned process of acquiring configuration data. The AI ​​configuration engine layer performs configuration optimization according to the process of data input, optimization of the objective function, application of combined algorithms, and output of results. The combined application of algorithms includes the combined application of Deep Reinforcement Learning (DRL), Genetic Algorithm (GA), and Constraint Satisfaction Optimization (CSP), corresponding to the aforementioned process of generating the optimal configuration scheme. The key technologies of the 3D modeling and visualization layer include reconstructing building space based on Building Information Modeling (BIM) data, dynamic annotation, pressure differential heat maps, AR augmented reality functionality, and generating deliverables, corresponding to the aforementioned process of generating visualized installation guidance information. The operation monitoring and optimization layer covers sensor network deployment, edge computing units, predictive maintenance, and optimization closed-loop mechanisms, corresponding to the aforementioned process of collecting operational status data and dynamically adjusting operational parameters. The remote control and feedback layer provides real-time monitoring via the app, a feedback closed-loop mechanism, and data-driven evolution capabilities, corresponding to the aforementioned feedback optimization process. These five layers work together to support the implementation of the multi-unit configuration adaptive method at the system level.

[0068] Combination Figure 5As shown, from a deployment architecture perspective, the multi-unit configuration adaptive device is deployed on the cloud service platform. Specifically, the cloud service platform may include modules such as an AI configuration engine and hot load model, a 3D / BIM model library and installation simulator, a big data analysis and AI model training platform, and a user and project management database. These modules interact via encrypted data channels to ensure the security of data transmission within the platform. Specifically, the AI ​​configuration engine and hot load model module performs multivariate optimization calculations using preset artificial intelligence algorithms to generate the optimal configuration scheme; the 3D / BIM model library and installation simulator module constructs a 3D installation model, simulates the installation process, and generates visual installation guidance information; the big data analysis and AI model training platform module performs big data analysis on operational status data and usage feedback information, and trains and iterates the algorithm model; and the user and project management database module encrypts and stores configuration data and management information for each project. The functions of the data configuration unit, scheme configuration unit, model building unit, and parameter adjustment unit of the multi-unit configuration adaptive device can be collaboratively implemented by the aforementioned modules within the cloud service platform. Furthermore, the big data analytics and AI model training platform can also analyze historically collected operating status data such as operating current and temperature (and further, it can collect and combine compressor vibration data) based on Long Short-Term Memory (LSTM) neural networks to train a fault prediction model to achieve predictive maintenance. For example, when the fault prediction model predicts that "the compressor bearing is about to wear out," it can generate a maintenance work order 72 hours in advance and push it to the operation and maintenance platform, thereby completing maintenance before the fault actually occurs and ensuring the continuous and stable operation of the multi-split system.

[0069] The terminal device is connected to the cloud service platform via the Internet. The terminal device can be a user's mobile phone, tablet, or other device with an installed APP, used to upload the building floor plan information and configuration requirements, display the visual installation guidance information (including interactive AR display of 3D installation models and virtual pipelines), and submit user feedback information. Furthermore, the terminal device can also provide real-time monitoring and remote control functions: for example, the APP displays daily / weekly / monthly energy consumption curves, equipment status information (e.g., outdoor unit operating, COP=4.2), and abnormal alarm information (e.g., high risk of pipeline leakage) in the form of a dynamic dashboard, and supports voice command control. Users can remotely adjust the temperature by saying voice commands such as "raise the living room temperature by 2℃," allowing users to monitor and control the operation of the multi-split system anytime, anywhere.

[0070] The temperature control equipment, deployed on-site and including indoor and outdoor units, is used to perform operational adjustments based on issued system operating parameters, such as compressor operating frequency and indoor unit airflow. It serves as the physical execution object for the multi-split air conditioning configuration adaptive method. The intelligent sensor network, deployed on the temperature control equipment, includes sensors for pressure, temperature, and current, used to collect operational status data according to a preset acquisition cycle, serving as a data source for dynamic optimization of the entire system.

[0071] The edge gateway is communicatively connected to the smart sensor network, the temperature control device, and the cloud service platform. Specifically, the operational status data collected by the smart sensor network can be transmitted to the edge gateway via Long Range Wide Area Network (LoRaWAN) or wired connection; the edge gateway and the temperature control device can communicate via Modbus protocol (a commonly used serial communication protocol in industrial fields) or BACnet protocol (Building Automation and Control networks data communication protocol) to adapt to the communication interfaces of different manufacturers' devices; in addition, the edge gateway can also communicate with terminal devices via local area network / Wi-Fi (Wireless Fidelity), facilitating the provision of real-time data to users in the local network environment. Edge gateways are used to aggregate and convert operational status data from smart sensor networks, that is, to unify and convert data of different protocols and formats into a standard format before uploading it to the cloud service platform. At the same time, edge gateways have built-in edge computing units to analyze the operational status data in real time and execute AI dynamic optimization strategies locally. For example, when real-time analysis identifies that the pipeline pressure difference or room temperature exceeds the limit, the edge gateway can directly send dynamic optimization commands to the temperature control equipment without waiting for a response from the cloud, adjusting the compressor operating frequency or the air volume of the indoor unit. This significantly reduces the dependence on the cloud link, shortens the response latency, and improves the real-time performance and reliability of dynamic adaptive adjustment.

[0072] Based on the above system architecture, after users upload building floor plans and configuration requirements via terminal devices, the multi-split air conditioning system configuration adaptive device deployed on the cloud service platform executes the aforementioned multi-split air conditioning system configuration adaptive method, generating the optimal configuration scheme and visual installation guidance information, which is then sent to the terminal devices for display. After installation personnel complete the installation of the temperature control equipment on-site according to the visual installation guidance information, the intelligent sensor network continuously collects operational status data. This data is then aggregated, converted, and analyzed in real time via the edge gateway, and dynamic optimization strategies are executed. Simultaneously, the operational status data is uploaded to the cloud service platform for big data analysis, model iteration, and predictive maintenance. During use, users submit feedback information via terminal devices, and the cloud service platform updates the optimization strategy library, heat load model, and system configuration strategy accordingly, pushing configuration suggestions. Thus, the entire multi-split air conditioning system configuration adaptive system forms a complete optimization scheme of configuration-installation-operation-feedback-evolution, realizing intelligent configuration and dynamic adaptive adjustment of the multi-split air conditioning system, effectively improving the overall energy efficiency of the multi-split air conditioning system.

[0073] This invention also provides a computer program product, which includes a computing program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions that, when executed by a computer, cause the computer to perform the multi-unit configuration adaptive method as described above.

[0074] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

[0075] It should also be noted that, in this specification, 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 a 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 a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A multi-unit configuration adaptive method, characterized in that, include: The building floor plan information and the corresponding configuration requirements information are obtained separately, and then verified and processed according to preset rules to obtain configuration data. Based on the configuration data, multivariate optimization calculations are performed using a preset artificial intelligence algorithm, and iterative optimization is performed based on a pre-constructed multivariate objective function to generate an optimal configuration scheme that meets preset constraints; wherein, the variables of the multivariate objective function include total energy consumption, installation cost, and energy efficiency loss; A three-dimensional installation model is constructed, and the installation location and path of the temperature control equipment are planned and marked based on the optimal configuration scheme to generate visual installation guidance information; After the temperature control device is installed based on the visual installation guide information, the operating status data of the temperature control device is collected, and the operating parameters of the temperature control device are dynamically adjusted according to the operating status data.

2. The multi-unit configuration adaptive method according to claim 1, characterized in that, The process involves acquiring building floor plan information and corresponding configuration requirements, verifying them using preset rules, and obtaining configuration data, including: Obtain the building floor plan information uploaded by the user, identify the room boundaries, area and layout in the building floor plan information, and integrate them to obtain the room layout information; Missing room layout information is detected, corresponding prompts are generated, and the reasonableness is verified based on preset rules to obtain the verification results. Obtain configuration requirement information input by the user; wherein, the configuration requirement information includes usage scenario information and personalized requirement information; Match the corresponding preset heat load template based on the usage scenario information; Based on the geographic information system, the system matches preset environmental parameters of the building's location and generates initial heat load information based on these preset environmental parameters. The room layout information, configuration requirements information, preset heat load template, initial heat load information, and verification results are encrypted and stored, and then integrated to obtain configuration data.

3. The multi-unit configuration adaptive method according to claim 1, characterized in that, The step of performing multivariate optimization calculations based on the configuration data using a preset artificial intelligence algorithm, and iteratively optimizing based on a pre-constructed multivariate objective function to generate an optimal configuration scheme that satisfies preset constraints, includes: The current configuration status is used as the state space, and the indoor unit model adjustment and pipe routing adjustment are used as the action space. Based on the preset heat load model and the multivariate objective function, the state space and action space are iteratively trained to generate multiple candidate configuration schemes. A fitness function is constructed based on the total energy consumption, installation cost, and preset differential pressure exceeding penalty, and the fitness function is used to filter and retain each of the candidate configuration schemes; Perform cross-mutation operations on the selected and retained candidate configuration schemes to generate new candidate configuration schemes and iteratively filter them. The candidate configuration schemes after the iteration are verified according to the preset constraints, and the optimal configuration scheme is selected from them based on the multivariate objective function.

4. The multi-unit configuration adaptive method according to claim 1, characterized in that, The multivariate objective function is constructed as follows: Acquire performance data of multi-split air conditioning units with different temperature control devices; the performance data of multi-split air conditioning units includes outdoor unit power consumption, pipe pressure drop loss, pipe length, pipe diameter and refrigerant pressure drop parameters; The total energy consumption is calculated based on the outdoor unit power consumption and pipeline pressure drop loss. Obtain the number of indoor units and calculate the installation cost based on the number of indoor units and the length of the pipes; Energy efficiency loss is calculated based on the pipe length, pipe diameter, and refrigerant pressure drop parameters. The total energy consumption, installation cost, and energy efficiency loss are weighted according to preset weights to construct a multivariate objective function.

5. The multi-unit configuration adaptive method according to claim 1, characterized in that, The process involves constructing a three-dimensional installation model and planning and marking the installation location and path of the temperature control equipment based on the optimal configuration scheme, generating visual installation guidance information, including: Import architectural space data into a 3D model to construct a 3D installation model; The installation location of the temperature control device is marked in the three-dimensional installation model, and the path of the temperature control device is planned and marked using a preset path planning algorithm to obtain the installation planning result; The pressure difference distribution is calculated by calling a preset fluid dynamics model, and the risk level of the pressure difference distribution is divided according to a preset pressure difference threshold, generating a pressure difference heat map superimposed on the three-dimensional installation model; Visualized installation guidance information is generated using the installation planning results and differential pressure heat map.

6. The multi-unit configuration adaptive method according to claim 1, characterized in that, After the temperature control device is installed based on the visual installation guidance information, the operating status data of the temperature control device is collected, and the operating parameters of the temperature control device are dynamically adjusted according to the operating status data, including: The pipeline pressure difference, temperature, energy efficiency ratio, and operating current are collected separately and integrated to obtain the operating status data; The operating status data is analyzed in real time to identify excessive pipeline pressure difference and excessive room temperature. The compressor operating frequency of the temperature control equipment is adjusted for excessive pipeline pressure difference, and the air volume of the indoor unit of the temperature control equipment is adjusted for excessive room temperature.

7. The multi-unit configuration adaptive method according to claim 1, characterized in that, Also includes: Receive user feedback information and perform semantic parsing on the feedback information to determine the corresponding scenario category, and classify the parsing results into the optimization strategy library; When the number of times the same scene category is classified reaches the preset number, the corresponding scene will be added to the preset operation mode and the corresponding heat load model will be updated. Optimization reports are generated periodically based on the optimization strategy library and runtime status data. Simultaneously, the optimization reports are updated to the system configuration strategy and configuration suggestions are pushed.

8. A multi-unit configuration adaptive device, characterized in that, include: The data configuration unit is used to acquire building floor plan information and corresponding configuration requirements information, and perform verification processing in combination with preset rules to obtain configuration data. The scheme configuration unit is used to perform multivariate optimization calculations based on the configuration data using a preset artificial intelligence algorithm, and to perform iterative optimization based on a pre-constructed multivariate objective function to generate an optimal configuration scheme that meets preset constraints; wherein, the variables of the multivariate objective function include total energy consumption, installation cost, and energy efficiency loss; The model building unit is used to build a three-dimensional installation model and plan and mark the installation position and path of the temperature control equipment based on the optimal configuration scheme, and generate visual installation guidance information; The parameter adjustment unit is used to collect the operating status data of the temperature control device after the installation of the temperature control device is completed based on the visual installation guidance information, and to dynamically adjust the operating parameters of the temperature control device according to the operating status data.

9. A multi-unit configuration adaptive system, characterized in that, Includes the multi-unit configuration adaptive device, terminal equipment, multiple temperature control devices, cloud service platform, intelligent sensor network and edge gateway as described in claim 8; The multi-unit configuration adaptive device is deployed on the cloud service platform; The terminal device is communicatively connected to the cloud service platform and is used to upload the building floor plan information and configuration requirements information, and to display the visual installation guidance information. Multiple temperature control devices are deployed on the building site to perform operational adjustments based on issued operating parameters; wherein, the temperature control devices include indoor units and outdoor units; The intelligent sensor network is deployed on multiple temperature control devices to collect the operating status data; The edge gateway is communicatively connected to the smart sensor network, multiple temperature control devices, and the cloud service platform, respectively, and is used to aggregate and convert the operating status data.

10. A computer program product, characterized in that, The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions that, when executed by a computer, cause the computer to perform the multi-unit configuration adaptive method according to any one of claims 1-7.

Images