A control method and system for a multi-split air conditioner
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-03-30
- Publication Date
- 2026-06-16
Smart Images
Figure CN122216769A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of multi-split air conditioning technology, and more specifically, to a control method and system for multi-split air conditioning. Background Technology
[0002] Multi-split air conditioners, as HVAC equipment adaptable to multi-space and individual temperature control, are widely used in various scenarios such as large office buildings, mixed-use commercial and residential complexes, multi-room residences, and long and narrow functional areas. Current control methods for multi-split air conditioners primarily rely on the deviation between indoor temperature and set temperature as the main control basis, adjusting the compressor's operating status based on single-point temperature sampling results. However, due to differences in airflow organization at different spatial locations, a single temperature feedback cannot reflect the degree of heat exchange in each spatial area, nor can it reflect the heat exchange relationship of primary air in different areas. If uniform control parameters are still used for adjustment, it is easy to experience excessive or insufficient heat exchange in local areas, resulting in repeated adjustments or response lags during operation. Furthermore, during outdoor heat exchange, the surface condition of the heat exchanger and the airflow heat exchange state are subject to multi-dimensional environmental interference. The lack of a mechanism to handle the coupling relationship between environmental factors makes it impossible for the control process to distinguish the influence of the external environment on heat exchange, leading to deviations in the adjustment process. In scenarios involving multi-zone heat exchange and environmental disturbances, multi-split air conditioners cannot match their heat exchange operating power, resulting in insufficient reliability of the closed-loop control of air heat exchange.
[0003] Therefore, it is necessary to design a control method and system for multi-split air conditioners to solve the problems existing in the current technology. Summary of the Invention
[0004] In view of this, the present invention proposes a control method and system for multi-split air conditioners, aiming to solve the above-mentioned problems.
[0005] In one aspect, the present invention proposes a control method for a multi-split air conditioner, comprising: Determine the heat exchange area and air supply area of the multi-split air conditioner, and set up several heat exchange data areas in the heat exchange area. Obtain the heat exchange attribute data corresponding to each heat exchange data area, and construct a heat exchange attribute set by heat exchange attribute data of the same type. Based on the association rule algorithm, the heat exchange attribute data that are associated in each heat exchange attribute set are merged to determine the heat exchange association set, and the unmerged heat exchange attribute data are constructed into the heat exchange disturbance set. The heat exchange influence value is determined based on the heat exchange association set and the heat exchange disturbance set. Obtain the heat exchange area and corresponding space area of all space heat exchange areas within the air supply area; determine the heat exchange ratio based on the heat exchange area and space area; determine the heat exchange type of each space heat exchange area based on the heat exchange ratio; and determine the heat exchange space value based on the heat exchange type. The temperature value of the air supply area is obtained, and the temperature difference between the temperature value and the set temperature value, the heat exchange influence value, and the heat exchange space value are analyzed in a historical database. If there are similar suspected data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on a clustering algorithm. If there are identical data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on the historical database. If there are no identical data in the historical database, a heat exchange model is determined based on the historical database, and the heat exchange operating power of the multi-split air conditioner is determined based on the heat exchange model.
[0006] Furthermore, in determining the heat exchange association set and the heat exchange disturbance set, the process includes: dividing the heat exchange attribute data of each heat exchange attribute set into several data items, determining the support of each data item, constructing a frequent data item header table based on the support, determining a starting point based on the frequent data item header table, constructing a corresponding conditional pattern base from bottom to top to determine the frequent data item set, determining the association and non-association results among all heat exchange attribute data based on the frequent data item set, merging all heat exchange attribute data with association results into the heat exchange association set, and merging all heat exchange attribute data with non-association results into the heat exchange disturbance set.
[0007] Furthermore, in determining the heat exchange impact value, the process includes: preprocessing the heat exchange association set and the heat exchange disturbance set, wherein the preprocessing includes data denoising and data normalization; determining the initial weight of each heat exchange attribute data in the preprocessed heat exchange disturbance set based on the type of the corresponding heat exchange attribute data; using the improved initial weight as the target weight of each heat exchange attribute data in the preprocessed heat exchange association set; and performing a weighted calculation on all preprocessed heat exchange attribute data and the corresponding initial weight or target weight to determine the heat exchange impact value.
[0008] Furthermore, in determining the heat exchange type, the process includes: obtaining the heat exchange area of each spatial heat exchange region and determining the corresponding spatial area; determining the ratio of the heat exchange area to the corresponding spatial area as the heat exchange ratio; if the heat exchange ratio is less than or equal to a heat exchange ratio threshold, then the heat exchange type of the spatial heat exchange region is determined to be a closed exchange region; if the heat exchange ratio is greater than the heat exchange ratio threshold, then the heat exchange type of the spatial heat exchange region is determined to be an open exchange region.
[0009] Furthermore, in determining the heat exchange space value, the process includes: counting the number of closed exchanges in the closed exchange area, counting the number of open exchanges in the open exchange area, determining the sum of the number of exchanges in the closed and open exchange areas, and determining the ratio of the number of exchanges in the open and the sum of the number of exchanges as the heat exchange space value.
[0010] Furthermore, when analyzing the temperature difference between the temperature value and the set temperature value, the heat exchange influence value, and the heat exchange space value in a historical database, the following steps are taken: the historical database includes several historical data groups and several historical heat exchange operating powers, with each historical data group corresponding to a historical heat exchange operating power. Each historical data group includes historical temperature difference, historical heat exchange influence value, and historical heat exchange space value. The temperature difference, heat exchange influence value, and heat exchange space value are compared with each historical data group to determine the data similarity. If there is a data similarity of 1, it is determined that there is a historical data group in the historical database that is the same as each data group. If there is a data similarity greater than or equal to the data similarity threshold, and there is no data similarity of 1, it is determined that there is a suspected historical data group in the historical database that is similar to each data group. If the data similarity is less than the data similarity threshold, it is determined that there is no historical data group in the historical database that is the same as each data group.
[0011] Furthermore, when determining the heat exchange operating power of the multi-split air conditioner based on the clustering algorithm, the process includes: extracting suspected historical data groups with a similarity greater than or equal to a data similarity threshold and their corresponding historical heat exchange operating power, and using the temperature difference, heat exchange influence value, and heat exchange space value as the heat exchange set to be clustered; determining the desired number of clusters and initializing the parameters of the Gaussian distribution; determining the suspected heat exchange set corresponding to the temperature difference, heat exchange influence value, and heat exchange space value based on the probability that each data in the heat exchange set to be clustered belongs to each Gaussian distribution; and using the mean of the historical heat exchange operating power in the suspected heat exchange set as the heat exchange operating power of the multi-split air conditioner.
[0012] Furthermore, when determining the heat exchange operating power of the multi-split air conditioner based on the historical database, the method includes: extracting historical data groups with a data similarity of 1; if there is only one historical data group, the historical heat exchange operating power corresponding to that historical data group is used as the heat exchange operating power of the multi-split air conditioner; if there are multiple historical data groups, the average of the historical heat exchange operating power corresponding to each historical data group is used as the heat exchange operating power of the multi-split air conditioner.
[0013] Furthermore, when determining the heat exchange operating power of the multi-split air conditioner based on the heat exchange model, the process includes: acquiring a machine learning model, training the machine learning model based on the historical database, determining the heat exchange model based on the training results, substituting the temperature difference, heat exchange influence value, and heat exchange space value into the heat exchange model, and using the output of the heat exchange model as the heat exchange operating power of the multi-split air conditioner.
[0014] Compared with existing technologies, the beneficial effects of this invention are as follows: By determining the heat exchange area and air supply area of the multi-split air conditioner, and setting several heat exchange data areas in the heat exchange area, the heat exchange attribute data corresponding to each heat exchange data area is obtained and a heat exchange attribute set is constructed, thereby reflecting the heat exchange degree of each spatial area and the heat exchange relationship of primary air in different areas, laying a data foundation for precise control. By dividing the heat exchange association set and the heat exchange disturbance set through the association rule algorithm, the heat exchange influence value is determined, and a processing mechanism for the coupling relationship between environments is constructed, thereby distinguishing the influence of the multi-dimensional external environment on heat exchange and reducing the deviation of the adjustment process. By adapting the heat exchange ratio, heat exchange type and heat exchange space value to the heat exchange characteristics of different spaces, the heat exchange operating power is determined, ensuring the reliability of air heat exchange in complex scenarios of multi-zone heat exchange and environmental disturbances, and improving the reliability of multi-split air conditioner control.
[0015] On the other hand, this application also provides a control system for a multi-split air conditioner, for applying the above-described control method for a multi-split air conditioner, including: The data acquisition and analysis unit is configured to determine the heat exchange area and air supply area of the multi-split air conditioner, and set up several heat exchange data areas in the heat exchange area, acquire the heat exchange attribute data corresponding to each heat exchange data area, and construct a heat exchange attribute set by the heat exchange attribute data of the same type. The first heat exchange unit is configured to merge the associated heat exchange attribute data in each heat exchange attribute set based on the association rule algorithm to determine the heat exchange association set, and construct the unmerged heat exchange attribute data into a heat exchange disturbance set, and determine the heat exchange influence value based on the heat exchange association set and the heat exchange disturbance set; The second heat exchange unit is configured to acquire the heat exchange area and corresponding space area of all space heat exchange areas within the air supply area, determine the heat exchange ratio based on the heat exchange area and space area, determine the heat exchange type of each space heat exchange area based on the heat exchange ratio, and determine the heat exchange space value based on the heat exchange type. The heat exchange control unit is configured to acquire the temperature value of the air supply area, analyze the temperature difference between the temperature value and the set temperature value, the heat exchange influence value and the heat exchange space value in a historical database, and if there are similar suspected data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on a clustering algorithm. If there are identical data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on the historical database. If there are no identical data in the historical database, a heat exchange model is determined based on the historical database, and the heat exchange operating power of the multi-split air conditioner is determined based on the heat exchange model.
[0016] It is understandable that the above-mentioned control method and system for multi-split air conditioners have the same beneficial effects, and will not be elaborated further here. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 A flowchart of a control method for a multi-split air conditioner provided in an embodiment of the present invention; Figure 2 This is a functional block diagram of a control system for a multi-split air conditioner provided in an embodiment of the present invention. Detailed Implementation
[0019] 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 embodiments of the present invention, and not all embodiments. 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.
[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0021] See Figure 1 As shown in some embodiments of this application, a control method for a multi-split air conditioner includes: S100: Determine the heat exchange area and air supply area of the multi-split air conditioner, set up several heat exchange data areas in the heat exchange area, obtain the heat exchange attribute data corresponding to each heat exchange data area, and construct a heat exchange attribute set by the heat exchange attribute data of the same type. S200: Based on the association rule algorithm, merge the heat exchange attribute data that are related in each heat exchange attribute set to determine the heat exchange association set, and construct the heat exchange disturbance set from the unmerged heat exchange attribute data. Determine the heat exchange influence value based on the heat exchange association set and the heat exchange disturbance set. S300: Obtain the heat exchange area and corresponding space area of all space heat exchange areas within the air supply area, determine the heat exchange ratio based on the heat exchange area and space area, determine the heat exchange type of each space heat exchange area based on the heat exchange ratio, and determine the heat exchange space value based on the heat exchange type. S400: Obtain the temperature value of the air supply area, analyze the temperature difference between the temperature value and the set temperature value, the heat exchange influence value, and the heat exchange space value in the historical database. If there are similar suspected data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on the clustering algorithm. If there are identical data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on the historical database. If there are no identical data in the historical database, the heat exchange model is determined based on the historical database, and the heat exchange operating power of the multi-split air conditioner is determined based on the heat exchange model.
[0022] Specifically, the heat exchange area refers to the functional area where the multi-split air conditioner achieves heat exchange, which is the area where heat exchange occurs in the outdoor environment. The air supply area refers to the space covered by the air supplied and regulated by the multi-split air conditioner, which is the area where air actually circulates indoors. Several heat exchange data zones are set up within the heat exchange area. Each heat exchange data zone is a subdivided area within the heat exchange area, used to collect heat exchange-related information independently. The number of heat exchange data zones can be dynamically adjusted according to the actual size of the heat exchange area. Particle sensors, wind speed sensors, solar radiation sensors, precipitation sensors, and air pressure sensors are deployed in each heat exchange data zone to accurately acquire the heat exchange attribute data corresponding to each heat exchange data zone. The heat exchange attribute data refers to the external factors that affect the heat exchange process in each heat exchange data zone, including dust concentration, wind speed, solar radiation intensity, and atmospheric pressure. All heat exchange attribute data of the same type are classified and integrated to construct heat exchange attribute sets. Each heat exchange attribute set is a dataset for the same environmental factors, avoiding the limitations of single-point sampling. Furthermore, by subdividing the areas and comprehensively collecting multi-dimensional environmental information of the heat exchange area, the one-sidedness caused by single data feedback is avoided, laying a comprehensive data foundation for subsequent precise control. Association rule algorithms are used to mine the relationships between data. By systematically analyzing the data relationships of each heat exchange attribute set, related heat exchange attribute data are merged to form a heat exchange association set. Simultaneously, unmerged and unrelated heat exchange attribute data are separately aggregated to form a heat exchange disturbance set. By analyzing the coupling relationships between various environmental factors, environmentally related data is distinguished from relatively irregular external environmental data, thus eliminating the reverse interference from multidimensional environmental factors. Based on the heat exchange association set and the heat exchange disturbance set, the heat exchange influence value is determined. The heat exchange influence value comprehensively reflects the multidimensional environmental impact during the heat exchange process. By determining the heat exchange influence value, control can be tailored to the actual conditions of the heat exchange area. By considering the environment and heat exchange conditions, and avoiding adjustment errors caused by environmental interference, the reliability of heat exchange control for multi-split air conditioners is improved. After analyzing the external environment, the heat exchange area and corresponding space area of all space heat exchange areas within the air supply area are obtained. The space heat exchange area refers to the room in the air supply area where air heat exchange occurs independently. The heat exchange area refers to the effective area of the space heat exchange area that participates in air heat exchange, which is the actual area of the room's doors and windows. The space area refers to the actual area corresponding to the space heat exchange area. Based on the heat exchange ratio, the heat exchange type of each space heat exchange area is further determined. The heat exchange type is a result that reflects the degree of heat exchange openness of the space heat exchange area, based on the heat exchange ratio.The spatial structure of different rooms within the air supply area varies, and the actual door and window area of each room, which serves as a separate heat exchange area, differs significantly from the actual room area. This directly determines the speed of heat exchange and the degree of external interference. By determining the heat exchange type, it is possible to accurately adapt to complex application scenarios such as multi-space and long, narrow functional areas, avoiding problems such as excessive or insufficient heat exchange in local areas due to differences in spatial heat exchange characteristics. At the same time, it avoids the repeated adjustments and response lag during the operation of multi-split air conditioners. This allows multi-split air conditioners to better meet the actual space heat exchange needs while avoiding adjustment errors caused by environmental interference, ensuring the reliability of control.
[0023] Understandably, based on the determined heat exchange type, comprehensive statistical calculations are performed to determine the heat exchange space value. The heat exchange space value reflects the heat exchange characteristics of the overall space of the air supply area. Combined with the actual spatial structure and heat exchange layout of the air supply area, the heat exchange characteristics of different spaces are quantified, thus avoiding the risk of excessive or insufficient heat exchange in local areas. The temperature value of the air supply area is acquired, reflecting the real-time monitored air temperature within that area. The difference between this temperature value and the user-preset temperature value is calculated. The set temperature value is the target value that the air supply area needs to achieve, as preset by the user. The temperature difference represents the difference between the actual temperature and the set temperature value. These three parameters—temperature difference, heat exchange impact value, and heat exchange space value—are simultaneously imported into a historical database for matching analysis. The historical database stores various relevant data and corresponding heat exchange operating power from past VRF air conditioner operations. The heat exchange operating power is determined based on the matching results. Heat exchange operating power represents the operating power parameter required for the VRF air conditioner to achieve its heat exchange function. If there is similar data in the historical database, meaning there is no completely identical historical data in the database, but some high-value data exists, the analysis will be performed. For historical data with similarity, clustering algorithms are used to classify and integrate similar data to determine the heat exchange operating power of the multi-split air conditioner. If there is data in the historical database that is exactly the same as the three types of data, the corresponding parameters are directly retrieved from the historical database to determine the heat exchange operating power. If there is no data in the historical database that is the same as the three types of data, a mathematical model for calculating the heat exchange operating power, i.e., the heat exchange model, is trained and constructed based on the past data in the historical database. Then, the heat exchange operating power of the multi-split air conditioner is determined based on the heat exchange model. By using the historical database, the efficiency of operating condition matching and control accuracy are taken into account. It can quickly respond to similar historical operating conditions to achieve real-time adjustment, and can also adapt to newly emerging operating conditions. This ensures that the heat exchange operating power of the multi-split air conditioner is always compatible with the actual heat exchange demand, thereby comprehensively improving the reliability of the closed-loop control of air heat exchange.
[0024] In some embodiments of this application, determining the heat exchange association set and the heat exchange disturbance set includes: dividing the heat exchange attribute data of each heat exchange attribute set into several data items, determining the support of each data item, constructing a frequent data item header table based on the support, determining the starting point based on the frequent data item header table, constructing the corresponding conditional pattern base from bottom to top to determine the frequent data item set, determining the association results and non-association results among all heat exchange attribute data based on the frequent data item set, merging all heat exchange attribute data with association results into a heat exchange association set, and merging all heat exchange attribute data with non-association results into a heat exchange disturbance set.
[0025] Specifically, all heat exchange attribute data within each heat exchange attribute set is broken down into several data items, laying the foundation for data correlation mining. For each data item, its support is determined individually. Support represents the frequency of a single data item's occurrence within the entire heat exchange attribute set. By calculating support, the frequency of data item occurrences can be intuitively distinguished, accurately filtering out high-frequency data related to the heat exchange environment, thus improving the targeting of correlation analysis. A frequent data item header table is constructed based on the support levels of each data item. This frequent data item header table is an index table sorted by data item support and used to organize correlation relationships. It can organize the hierarchy and arrangement of high-frequency data, thereby improving the efficiency of data correlation analysis. Based on the constructed frequent data item header table, the corresponding correlation mining starting point is determined. The starting point is the core correlation mining starting point selected in the frequent data item header table, thus ensuring the accuracy of correlation mining. To ensure coherence and unity, the system constructs conditional pattern bases corresponding to a selected starting point from the bottom up. These conditional pattern bases represent subsets of associated data obtained by tracing back from the starting point, fully tracing the implicit and explicit associations between various heat exchange attribute data. Frequent data itemsets are datasets formed by combinations of data items with support and stable associations. Based on the FP-Growth association rule algorithm, all heat exchange attribute data are judged one by one, clearly distinguishing between associated and non-associated results. Ultimately, all heat exchange attribute data judged to have associated results are unified, as are all heat exchange attribute data judged to have non-associated results. This clarifies the coupling relationship between multi-dimensional heat exchange attribute data such as dust concentration, wind speed, solar radiation intensity, and atmospheric pressure, thereby reducing the adjustment deviation of multi-split air conditioners and comprehensively improving the stability and reliability of air heat exchange closed-loop control.
[0026] In some embodiments of this application, determining the heat exchange impact value includes: preprocessing the heat exchange association set and the heat exchange disturbance set, the preprocessing including data noise reduction and data normalization, determining the initial weight of each heat exchange attribute data in the preprocessed heat exchange disturbance set based on the type of the corresponding heat exchange attribute data, and using the improved initial weight as the target weight of each heat exchange attribute data in the preprocessed heat exchange association set, and performing a weighted calculation on all preprocessed heat exchange attribute data and the corresponding initial weight or target weight to determine the heat exchange impact value.
[0027] Specifically, the heat exchange correlation set and heat exchange disturbance set are preprocessed. Data denoising removes invalid stray information and abnormal fluctuation signals from the heat exchange attribute data. Data normalization adjusts heat exchange attribute data of different types and dimensions to a fixed range, eliminating differences in dimensions and magnitudes between various data types and avoiding calculation errors caused by different data magnitudes and dimensions. Based on the corresponding type of each heat exchange attribute data in the heat exchange disturbance set, the initial weight corresponding to that data is determined one by one. The types of heat exchange attribute data include dust concentration, wind speed, solar radiation intensity, atmospheric pressure, etc. Solar radiation intensity directly affects the air conditioning heat exchange load; the level of solar radiation intensity directly changes the degree of heat exchange in the indoor space and directly affects the heat exchange environment of the outdoor heat exchanger. Wind speed reflects the airflow speed around the outdoor unit heat exchanger, and the wind speed directly determines the heat exchange rate between the air and the heat exchanger, directly affecting... The efficiency of heat exchange is also a key factor affecting the heat exchange effect. Therefore, solar radiation intensity and wind speed are given a greater initial weight, set to 2. Atmospheric pressure fluctuations have only an indirect impact on the air heat exchange process and are considered minor environmental disturbances. Dust concentration is an occasional and irregular interference factor, so dust concentration and atmospheric pressure are given a lower initial weight, set to 1. The initial weight is determined according to the type of heat exchange attribute data, so that the calculation results can fit the actual heat exchange conditions and further improve the accuracy of air conditioning control. For the coupled heat exchange association set, assuming that the initial weight of solar radiation intensity in the heat exchange disturbance set is 2, and solar radiation intensity heat exchange attribute data also appears in the heat exchange association set, the target weight of the heat exchange attribute data for solar radiation intensity in the heat exchange association set is set to 3, thus increasing the initial weight and using it as the target weight. Since solar radiation intensity directly determines the air conditioning heat exchange load and affects outdoor heat exchange, and since it is interrelated with other environmental factors, a target weight higher than the initial weight is necessary to match its true influence and highlight the strong coupling relationship between this factor and the heat exchange process. All pre-processed heat exchange attribute data and their corresponding initial or target weights are weighted to determine the heat exchange influence value. This value reflects the comprehensive impact of environmental factors on the heat exchange area of the multi-split air conditioner. A larger influence value indicates a higher level of comprehensive environmental interference in the heat exchange area, while a smaller influence value indicates a more stable impact. The heat exchange influence value lays the data foundation for subsequent accurate matching of heat exchange operating power.
[0028] In some embodiments of this application, determining the heat exchange type includes: obtaining the heat exchange area of each spatial heat exchange region and determining the corresponding spatial area; determining the ratio of the heat exchange area to the corresponding spatial area as the heat exchange ratio; if the heat exchange ratio is less than or equal to a heat exchange ratio threshold, then the heat exchange type of the spatial heat exchange region is determined to be a closed exchange region; if the heat exchange ratio is greater than the heat exchange ratio threshold, then the heat exchange type of the spatial heat exchange region is determined to be an open exchange region.
[0029] In some embodiments of this application, determining the heat exchange space value includes: counting the number of closed exchanges in the closed exchange area, counting the number of open exchanges in the open exchange area, determining the exchange quantity and value of the closed exchange quantity and the open exchange quantity, and determining the ratio of the open exchange quantity and the exchange quantity and value as the heat exchange space value.
[0030] Specifically, the heat exchange area corresponding to each space heat exchange zone within the air supply area is obtained, and the corresponding space area is determined. The heat exchange area and the corresponding space area are proportionally calculated, and the resulting ratio is determined as the heat exchange ratio. The heat exchange ratio is compared with a preset heat exchange ratio threshold. If the heat exchange ratio is less than or equal to the threshold, it indicates that the actual room area is large, but the actual door and window area is small, and the heat exchange type of this space heat exchange zone is a closed heat exchange zone. If the heat exchange ratio is greater than the threshold, it indicates that the actual room area is small, but the actual door and window area is large, and the heat exchange type of this space heat exchange zone is an open heat exchange zone. The structure of the heat exchange area precisely distinguishes the degree of heat exchange openness in different spaces, quantifying and classifying the heat exchange characteristics of the spaces. This adapts to diverse spatial scenarios such as large office buildings, multi-room residences, and long, narrow functional areas. It also avoids the risk of excessive or insufficient heat exchange in localized areas at the spatial structure level. The number of heat exchange areas in all spaces identified as closed exchange areas is statistically analyzed to determine the number of closed exchange areas, and the number of heat exchange areas in all spaces identified as open exchange areas is also statistically analyzed to determine the number of open exchange areas. Because the air supply area encompasses diverse spatial scenarios such as large office buildings, multi-room residences, and long, narrow functional areas, there are numerous heat exchange areas, and cases where the number of open exchange areas is zero are almost nonexistent. The sum of the closed and open exchange areas is calculated to determine the total exchange quantity, and the ratio between the open exchange quantity and the total exchange quantity is calculated to determine the heat exchange space value. The quantitative statistics intuitively reflect the proportion of open heat exchange in the overall space within the air supply area, enabling the control of multi-split air conditioners to match the overall heat exchange characteristics of the air supply area. This further improves the reliability and accuracy of the closed-loop control of air heat exchange, ensuring the adaptability of the air conditioner's heat exchange operating power to the actual heat exchange requirements.
[0031] In some embodiments of this application, when analyzing the temperature difference between the temperature value and the set temperature value, the heat exchange influence value, and the heat exchange space value in a historical database, the following steps are taken: the historical database includes several historical data groups and several historical heat exchange operating powers, and each historical data group corresponds to a historical heat exchange operating power. Each historical data group includes historical temperature difference, historical heat exchange influence value, and historical heat exchange space value. The temperature difference, heat exchange influence value, and heat exchange space value are compared with each historical data group to determine the data similarity. If there is a data similarity of 1, it is determined that there is a historical data group in the historical database that is the same as each data. If there is a data similarity greater than or equal to the data similarity threshold, and there is no data similarity of 1, it is determined that there is a suspected historical data group in the historical database that is similar to each data. If the data similarity is less than the data similarity threshold, it is determined that there is no historical data group in the historical database that is the same as each data.
[0032] Specifically, the historical database contains several historical data sets and several historical heat exchange operating powers, with each historical data set and historical heat exchange operating power maintaining a one-to-one correspondence. Each historical data set includes historical temperature differences, historical heat exchange impact values, and historical heat exchange space values. The historical database can be constructed based on experimental simulations and experience from corresponding human controls. The temperature difference values, heat exchange impact values, and heat exchange space values are compared one by one with each historical data set within the historical database to determine the data similarity between each historical data set. Data similarity can be determined using methods such as cosine similarity and Euclidean distance. If any data in the comparison results has a similarity of 1, it indicates that there is a historical data group in the historical database that is the same as the current data. This provides a basis for directly calling the corresponding historical heat exchange operating power, thereby improving the response efficiency of air conditioning control. If any data in the comparison results has a similarity greater than or equal to the data similarity threshold, and there is no data similarity of 1, it indicates that there is a suspected historical data group in the historical database that is similar to the current data. This provides data support for accurately screening historical data that is close to the current operating conditions, thereby determining the heat exchange operating power. If the data similarity in all comparison results is less than the data similarity threshold, it indicates that there is no historical data group in the historical database that is the same as the current data. Accurate division based on the degree of matching with historical data ensures both rapid response under similar or identical operating conditions and control accuracy under new operating conditions, thus ensuring the reliability of closed-loop control of air heat exchange.
[0033] In some embodiments of this application, when determining the heat exchange operating power of a multi-split air conditioner based on a clustering algorithm, the process includes: extracting suspected historical data groups with a similarity greater than or equal to a data similarity threshold and their corresponding historical heat exchange operating power, and using them, along with temperature difference, heat exchange influence value, and heat exchange space value, as the heat exchange set to be clustered; determining the desired number of clusters and initializing the parameters of the Gaussian distribution; determining the suspected heat exchange set corresponding to the temperature difference, heat exchange influence value, and heat exchange space value based on the probability that each data in the heat exchange set to be clustered belongs to each Gaussian distribution; and using the average historical heat exchange operating power in the suspected heat exchange set as the heat exchange operating power of the multi-split air conditioner.
[0034] Specifically, all suspected historical data groups with a data similarity greater than or equal to the data similarity threshold are extracted from the historical database, along with the historical heat exchange operating power corresponding to each of these suspected historical data groups. These extracted suspected historical data groups and their corresponding historical heat exchange operating power are then integrated with the currently calculated temperature difference, heat exchange influence value, and heat exchange space value, forming a unified heat exchange set to be clustered. A clustering algorithm is then used to analyze this heat exchange set, combining the operating conditions of multiple similar historical operating conditions to find the suspected heat exchange set closest to the temperature difference, heat exchange influence value, and heat exchange space value. This improves the accuracy of determining the heat exchange operating power, ensuring that the heat exchange operating power aligns with the operating experience of similar historical operating conditions while also adapting to the actual heat exchange requirements of the current operating conditions. This balances control response efficiency and adjustment accuracy, further enhancing the reliability of the closed-loop control of air heat exchange in multi-split air conditioning systems.
[0035] In some embodiments of this application, when determining the heat exchange operating power of a multi-split air conditioner based on a historical database, the method includes: extracting historical data groups with a data similarity of 1; if there is only one historical data group, the historical heat exchange operating power corresponding to that historical data group is used as the heat exchange operating power of the multi-split air conditioner; if there are multiple historical data groups, the average of the historical heat exchange operating power corresponding to each historical data group is used as the heat exchange operating power of the multi-split air conditioner.
[0036] Specifically, all historical data groups with a similarity of 1 are extracted from the historical database, and all invalid historical data that are not completely matched are eliminated. If only one historical data group is extracted, the historical heat exchange operating power corresponding to that unique historical data group is directly determined as the current heat exchange operating power of the multi-split air conditioner. This fully utilizes the single historical operating experience of a completely matched data group, enabling rapid retrieval and determination of the heat exchange operating power, thereby shortening the response time of air conditioner control and adapting to regular operating scenarios with recurring operating conditions. If multiple historical data groups are extracted, the historical heat exchange operating power corresponding to each historical data group is obtained separately, and the average value of all corresponding historical heat exchange operating powers is calculated to determine the current heat exchange operating power of the multi-split air conditioner. This ensures both the control response efficiency under completely matched operating conditions and the accuracy of determining the heat exchange operating power under multiple sets of the same historical data, thus ensuring the reliability of the closed-loop control of air heat exchange.
[0037] In some embodiments of this application, when determining the heat exchange operating power of a multi-split air conditioner based on a heat exchange model, the process includes: acquiring a machine learning model, training the machine learning model based on a historical database, determining the heat exchange model based on the training results, substituting the temperature difference, heat exchange influence value, and heat exchange space value into the heat exchange model, and using the output of the heat exchange model as the heat exchange operating power of the multi-split air conditioner.
[0038] Specifically, machine learning models, including random forest models and convolutional neural networks, are used. All historical data sets stored in the historical database, along with their corresponding historical heat exchange operating power, are used as training samples to train the machine learning model. Through iterative learning and feature extraction on massive amounts of historical data, the machine learning model can grasp the intrinsic correlation between three parameters—temperature difference, heat exchange influence, and heat exchange space—and historical heat exchange operating power, thereby determining the heat exchange model. Using the current actual operating conditions' temperature difference, heat exchange influence, and heat exchange space as input, the heat exchange model outputs the heat exchange operating power of the multi-split air conditioner, thus comprehensively covering all possible heat exchange operating scenarios and improving the accuracy and reliability of multi-split air conditioner control in various scenarios such as large office buildings, narrow functional areas, and multi-room residences.
[0039] It should be noted that the machine learning model in this application can be implemented using existing technologies in the field, and it is not the focus of the improvement claimed in this application. The focus of this application is on the data connection relationship and processing logic of temperature difference, heat exchange influence value, heat exchange space value and historical heat exchange operating power. Furthermore, it can be adapted, replaced or equivalently implemented by combining conventional engineering methods based on the input-output relationship and parameter configuration rules disclosed in this application, without affecting the implementation of the technical solution of this application.
[0040] In summary, the beneficial effects of this invention are as follows: By determining the heat exchange area and air supply area of the multi-split air conditioner, and setting several heat exchange data areas in the heat exchange area, the heat exchange attribute data corresponding to each heat exchange data area is obtained and a heat exchange attribute set is constructed. This reflects the heat exchange degree of each spatial area and the heat exchange relationship of primary air in different areas, laying a data foundation for precise control. By dividing the heat exchange association set and the heat exchange disturbance set through the association rule algorithm, the heat exchange influence value is determined, and a processing mechanism for the coupling relationship between environments is constructed. This distinguishes the influence of the multi-dimensional external environment on heat exchange, reduces the deviation of the adjustment process, and determines the heat exchange operating power by adapting the heat exchange ratio, heat exchange type, and heat exchange space value to the heat exchange characteristics of different spaces. This ensures the reliability of air heat exchange in complex scenarios of multi-zone heat exchange and environmental disturbances, and improves the reliability of multi-split air conditioner control.
[0041] In another preferred embodiment based on the above embodiments, see [reference] Figure 2 As shown, this embodiment provides a control system for a multi-split air conditioner, for applying a control method for a multi-split air conditioner, including: The data acquisition and analysis unit is configured to determine the heat exchange area and air supply area of the multi-split air conditioner, set up several heat exchange data areas in the heat exchange area, acquire the heat exchange attribute data corresponding to each heat exchange data area, and construct a heat exchange attribute set by the heat exchange attribute data of the same type. The first heat exchange unit is configured to merge the heat exchange attribute data that are related in each heat exchange attribute set based on the association rule algorithm to determine the heat exchange association set, and construct the heat exchange disturbance set from the unmerged heat exchange attribute data. The heat exchange influence value is determined based on the heat exchange association set and the heat exchange disturbance set. The second heat exchange unit is configured to acquire the heat exchange area and corresponding space area of all space heat exchange areas within the air supply area, determine the heat exchange ratio based on the heat exchange area and space area, determine the heat exchange type of each space heat exchange area based on the heat exchange ratio, and determine the heat exchange space value based on the heat exchange type. The heat exchange control unit is configured to acquire the temperature value of the air supply area, analyze the temperature difference between the temperature value and the set temperature value, the heat exchange influence value, and the heat exchange space value in a historical database. If there are similar suspected data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on a clustering algorithm. If there are identical data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on the historical database. If there are no identical data in the historical database, a heat exchange model is determined based on the historical database, and the heat exchange operating power of the multi-split air conditioner is determined based on the heat exchange model.
[0042] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0043] 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 scope of protection of the claims of the present invention.
Claims
1. A control method for a multi-split air conditioner, characterized in that, include: Determine the heat exchange area and air supply area of the multi-split air conditioner, and set up several heat exchange data areas in the heat exchange area. Obtain the heat exchange attribute data corresponding to each heat exchange data area, and construct a heat exchange attribute set by heat exchange attribute data of the same type. Based on the association rule algorithm, the heat exchange attribute data that are associated in each heat exchange attribute set are merged to determine the heat exchange association set, and the unmerged heat exchange attribute data are constructed into the heat exchange disturbance set. The heat exchange influence value is determined based on the heat exchange association set and the heat exchange disturbance set. Obtain the heat exchange area and corresponding space area of all space heat exchange areas within the air supply area; determine the heat exchange ratio based on the heat exchange area and space area; determine the heat exchange type of each space heat exchange area based on the heat exchange ratio; and determine the heat exchange space value based on the heat exchange type. The temperature value of the air supply area is obtained, and the temperature difference between the temperature value and the set temperature value, the heat exchange influence value, and the heat exchange space value are analyzed in a historical database. If there are similar suspected data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on a clustering algorithm. If there are identical data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on the historical database. If there are no identical data in the historical database, a heat exchange model is determined based on the historical database, and the heat exchange operating power of the multi-split air conditioner is determined based on the heat exchange model.
2. The control method for multi-split air conditioners according to claim 1, characterized in that, The process of determining the heat exchange association set and the heat exchange disturbance set includes: dividing the heat exchange attribute data of each heat exchange attribute set into several data items, determining the support of each data item, constructing a frequent data item header table based on the support, determining a starting point based on the frequent data item header table, constructing a corresponding conditional pattern base from bottom to top to determine the frequent data item set, determining the association results and non-association results among all heat exchange attribute data based on the frequent data item set, merging all heat exchange attribute data with association results into the heat exchange association set, and merging all heat exchange attribute data with non-association results into the heat exchange disturbance set.
3. The control method for multi-split air conditioners according to claim 2, characterized in that, Determining the heat exchange impact value includes: preprocessing the heat exchange association set and the heat exchange disturbance set, the preprocessing including data noise reduction and data normalization; determining the initial weight of each heat exchange attribute data in the preprocessed heat exchange disturbance set based on the type of the corresponding heat exchange attribute data; using the improved initial weight as the target weight of each heat exchange attribute data in the preprocessed heat exchange association set; and performing a weighted calculation on all preprocessed heat exchange attribute data and the corresponding initial weight or target weight to determine the heat exchange impact value.
4. The control method for multi-split air conditioners according to claim 3, characterized in that, When determining the heat exchange type, the process includes: obtaining the heat exchange area of each spatial heat exchange region and determining the corresponding spatial area; determining the ratio of the heat exchange area to the corresponding spatial area as the heat exchange ratio; if the heat exchange ratio is less than or equal to a heat exchange ratio threshold, then the heat exchange type of the spatial heat exchange region is determined to be a closed exchange region; if the heat exchange ratio is greater than the heat exchange ratio threshold, then the heat exchange type of the spatial heat exchange region is determined to be an open exchange region.
5. The control method for multi-split air conditioners according to claim 4, characterized in that, Determining the heat exchange space value includes: counting the number of closed exchanges in the closed exchange area, counting the number of open exchanges in the open exchange area, determining the sum of the number of closed exchanges and the number of open exchanges, and determining the ratio of the number of open exchanges to the sum of the number of exchanges as the heat exchange space value.
6. The control method for multi-split air conditioners according to claim 5, characterized in that, When analyzing the temperature difference between the temperature value and the set temperature value, the heat exchange influence value, and the heat exchange space value in a historical database, the process includes: the historical database includes several historical data groups and several historical heat exchange operating powers, with each historical data group corresponding to a historical heat exchange operating power; each historical data group includes historical temperature difference, historical heat exchange influence value, and historical heat exchange space value; the temperature difference, heat exchange influence value, and heat exchange space value are compared with each historical data group to determine data similarity; if there is a data similarity of 1, it is determined that there is a historical data group in the historical database that is the same as each data group; if there is a data similarity greater than or equal to the data similarity threshold, and there is no data similarity of 1, it is determined that there is a suspected historical data group in the historical database that is similar to each data group; if the data similarity is less than the data similarity threshold, it is determined that there is no historical data group in the historical database that is the same as each data group.
7. The control method for multi-split air conditioners according to claim 6, characterized in that, When determining the heat exchange operating power of the multi-split air conditioner based on a clustering algorithm, the process includes: extracting suspected historical data groups with a similarity greater than or equal to a data similarity threshold and their corresponding historical heat exchange operating power, and using these groups along with the temperature difference, heat exchange influence value, and heat exchange space value as the heat exchange set to be clustered; determining the desired number of clusters and initializing the parameters of the Gaussian distribution; determining the suspected heat exchange set corresponding to the temperature difference, heat exchange influence value, and heat exchange space value based on the probability that each data point in the heat exchange set belongs to each Gaussian distribution; and using the mean of the historical heat exchange operating power in the suspected heat exchange set as the heat exchange operating power of the multi-split air conditioner.
8. The control method for multi-split air conditioners according to claim 7, characterized in that, When determining the heat exchange operating power of the multi-split air conditioner based on the historical database, the method includes: extracting historical data groups with a data similarity of 1; if there is only one historical data group, the historical heat exchange operating power corresponding to the historical data group is taken as the heat exchange operating power of the multi-split air conditioner; if there are multiple historical data groups, the average of the historical heat exchange operating power corresponding to each historical data group is taken as the heat exchange operating power of the multi-split air conditioner.
9. The control method for multi-split air conditioners according to claim 8, characterized in that, When determining the heat exchange operating power of the multi-split air conditioner based on the heat exchange model, the process includes: acquiring a machine learning model, training the machine learning model based on the historical database, determining the heat exchange model based on the training results, substituting the temperature difference, heat exchange influence value, and heat exchange space value into the heat exchange model, and using the output of the heat exchange model as the heat exchange operating power of the multi-split air conditioner.
10. A control system for a multi-split air conditioner, used for applying the control method for a multi-split air conditioner as described in any one of claims 1-9, characterized in that, include: The data acquisition and analysis unit is configured to determine the heat exchange area and air supply area of the multi-split air conditioner, and set up several heat exchange data areas in the heat exchange area, acquire the heat exchange attribute data corresponding to each heat exchange data area, and construct a heat exchange attribute set by the heat exchange attribute data of the same type. The first heat exchange unit is configured to merge the associated heat exchange attribute data in each heat exchange attribute set based on the association rule algorithm to determine the heat exchange association set, and construct the unmerged heat exchange attribute data into a heat exchange disturbance set, and determine the heat exchange influence value based on the heat exchange association set and the heat exchange disturbance set; The second heat exchange unit is configured to acquire the heat exchange area and corresponding space area of all space heat exchange areas within the air supply area, determine the heat exchange ratio based on the heat exchange area and space area, determine the heat exchange type of each space heat exchange area based on the heat exchange ratio, and determine the heat exchange space value based on the heat exchange type. The heat exchange control unit is configured to acquire the temperature value of the air supply area, analyze the temperature difference between the temperature value and the set temperature value, the heat exchange influence value and the heat exchange space value in a historical database, and if there are similar suspected data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on a clustering algorithm. If there are identical data in the historical database, the heat exchange operating power of the multi-split air conditioner is determined based on the historical database. If there are no identical data in the historical database, a heat exchange model is determined based on the historical database, and the heat exchange operating power of the multi-split air conditioner is determined based on the heat exchange model.