Indoor unit room identification method for multi-connected air conditioner based on indoor pressure and multi-connected air conditioner

Abstract

This invention relates to the field of multi-split air conditioning, and provides a method for room identification of indoor units in multi-split air conditioning systems based on indoor pressure, as well as a multi-split air conditioning system. The method includes: acquiring pressure time-series data collected by micro-differential pressure sensors of each indoor unit within a first preset time window, and calculating a pressure fluctuation feature vector for each indoor unit based on the pressure time-series data; calculating a similarity index between the pressure fluctuation feature vectors of any two indoor units using a weighted similarity algorithm; performing a clustering operation on all indoor units based on the similarity comparison result (comparing the similarity index with a preset similarity threshold), grouping indoor units with similarity indices greater than or equal to the preset similarity threshold into the same group, and assigning a unique room identifier to each group to obtain the indoor unit room grouping result. This invention can automatically complete indoor unit room grouping based on indoor pressure fluctuation characteristics without manual preset, significantly improving grouping efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to the field of multi-split air conditioning technology, and in particular to a method for room identification of indoor units of multi-split air conditioning based on indoor pressure, and a multi-split air conditioning system. Background Technology

[0002] In multi-split air conditioning centralized control systems, it is usually necessary to group indoor units installed in different rooms so that the central control host can coordinate and manage the indoor units on a room-by-room basis, such as synchronously starting and stopping multiple indoor units in the same room or uniformly adjusting the temperature. Currently, indoor unit room grouping in multi-split centralized control systems is generally completed manually: after the system is installed and debugged, the operator records the number of each indoor unit and its actual room location, then logs into the central control host interface through a manual operation terminal, manually selects all indoor unit numbers in the same room, issues a grouping command, and names the group. After the central control host stores the manual grouping result, it can then perform coordinated control on a room-by-room basis.

[0003] However, the aforementioned manual grouping method has significant drawbacks. First, in large buildings such as office buildings, shopping malls, and hotels, the number of indoor units can reach hundreds. Operators would need to spend hours or even days recording, verifying, and configuring each unit individually, resulting in extremely low grouping efficiency and high labor costs. Second, manual operation inevitably leads to recording errors and incorrect number selection. Once grouping is incorrect, it will directly affect the collaborative control effect of indoor units in the same room, and may even damage the equipment due to incorrect control commands. Furthermore, when the installation location of indoor units is adjusted, the room function is changed, or the room's sealing condition changes, the existing solution cannot automatically calibrate the grouping results. Operators must re-execute the entire manual grouping process, which cannot adapt to dynamically changing scenarios.

[0004] Therefore, how to achieve automated, high-precision identification and grouping of multi-split indoor unit rooms to avoid manual intervention is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] This invention provides a method for room identification of indoor units in a multi-split air conditioner based on indoor pressure, and a multi-split air conditioner that can automatically group indoor units based on indoor pressure fluctuation characteristics without manual preset, significantly improving grouping efficiency and accuracy.

[0006] In a first aspect, the present invention provides a method for room identification of multi-split air conditioning indoor units based on indoor pressure, wherein each of the multi-split air conditioning indoor units is equipped with a micro differential pressure sensor; The method includes: Within the first preset time window, pressure time-series data collected by the micro differential pressure sensors of each indoor unit are acquired, and pressure fluctuation feature vector of each indoor unit is calculated based on the pressure time-series data. A weighted similarity algorithm is used to calculate the similarity index between the pressure fluctuation feature vectors of any two indoor units; wherein, the similarity index is used to characterize the degree of consistency of the pressure fluctuation features of the two indoor units. Based on the similarity comparison results of the similarity index and the preset similarity threshold, a clustering operation is performed on all indoor units. Indoor units with a similarity index greater than or equal to the preset similarity threshold are grouped into the same group, and a unique room identifier is assigned to each group to obtain the indoor unit room grouping results.

[0007] Preferably, according to the multi-split air conditioning indoor unit room identification method based on indoor pressure provided by the present invention, the pressure fluctuation feature vector includes the following parameters: pressure fluctuation amplitude feature value, pressure change slope feature value, and pressure stabilization duration feature value; The calculation of the pressure fluctuation feature vector for each indoor unit based on the pressure time-series data includes: Calculate the difference between the maximum and minimum values ​​of the pressure time series data within the first preset time window to obtain the pressure difference value, and use the pressure difference value as the pressure fluctuation amplitude characteristic value; The first preset time window is divided into multiple consecutive calculation cycles. The pressure time series data in each calculation cycle is linearly fitted to obtain the pressure change rate in the corresponding calculation cycle. The statistical average of the pressure change rates in multiple calculation cycles is used as the pressure change slope feature value. Set a pressure fluctuation threshold range, and count the cumulative duration of pressure time series data that are continuously within the pressure fluctuation threshold range within the first preset time window. Use the cumulative duration as a pressure stability duration feature value.

[0008] Preferably, according to the multi-split air conditioning indoor unit room identification method based on indoor pressure provided by the present invention, the step of employing a weighted similarity algorithm to calculate the similarity index between the pressure fluctuation feature vectors of any two indoor units includes: Obtain the pressure fluctuation feature vector of the first indoor unit and the pressure fluctuation feature vector of the second indoor unit; The differences between the first indoor unit and the second indoor unit at three characteristic values ​​are calculated respectively to obtain the three characteristic differences; Obtain three preset weight coefficients, each of which corresponds one-to-one with one of the three feature values; Calculate the product of the square of each feature difference and the corresponding weight coefficient, add the three products together and take the square root to obtain the weighted Euclidean distance; The weighted Euclidean distance is calculated to obtain the similarity index between the first indoor unit and the second indoor unit.

[0009] Preferably, according to the multi-split air conditioning indoor unit room identification method based on indoor pressure provided by the present invention, the similarity comparison result based on the similarity index and the preset similarity threshold is used to perform clustering operation on all indoor units, and indoor units with the similarity index greater than or equal to the preset similarity threshold are grouped into the same group, including: Select the indoor unit to be assigned as the current cluster center, and group all indoor units with a similarity index greater than or equal to the preset similarity threshold into the same temporary group; Repeat the steps of selecting the current cluster center and assigning to the same temporary group until all indoor units are assigned to the temporary group.

[0010] Preferably, according to the multi-split air conditioning indoor unit room identification method based on indoor pressure provided by the present invention, after the step of assigning all indoor units to a temporary group, the method includes: Traverse all temporary groups and detect whether there are at least two pairs of indoor units between any two temporary groups, and the similarity index between the two indoor units in the pair is greater than or equal to the preset similarity threshold. If they exist, the two temporary groups are merged into one group, and a unique room identifier is assigned to each group after the merge process.

[0011] Preferably, according to the method for room identification of multi-split air conditioning indoor units based on indoor pressure provided by the present invention, after the step of classifying indoor units with similarity indices greater than or equal to the preset similarity threshold into the same group, the method includes: Traverse each group and verify whether the similarity index of any two indoor units in the group is greater than or equal to the preset similarity threshold. If there is a pair of indoor units with a similarity index less than the preset similarity threshold, remove the abnormal indoor unit in the group whose similarity index with other indoor units in the group is less than the preset similarity threshold from the corresponding group, and re-perform clustering operation on the abnormal indoor units after they are removed from the group to group the abnormal indoor units. Traverse any two different groups and check whether there are indoor unit pairs between the two groups with a similarity index greater than or equal to the preset similarity threshold. If so, merge the two groups.

[0012] Preferably, according to the multi-split air conditioning indoor unit room identification method based on indoor pressure provided by the present invention, within a second preset time window, the method includes: A new indoor unit room grouping result is generated, and the new indoor unit room grouping result is compared with the currently effective indoor unit room grouping result. When the grouping difference is greater than or equal to a preset difference threshold, the new indoor unit room grouping result is updated to the currently effective indoor unit room grouping result.

[0013] Preferably, according to the multi-split air conditioning indoor unit room identification method based on indoor pressure provided by the present invention, the step of comparing the new indoor unit room grouping result with the currently effective indoor unit room grouping result, and updating the new indoor unit room grouping result to the currently effective indoor unit room grouping result when the grouping difference is greater than or equal to a preset difference threshold, includes: The new indoor unit room grouping results are compared with the currently effective indoor unit room grouping results for each indoor unit, and the number of indoor units whose room labels have been changed is counted. The group change rate is obtained by dividing the number of indoor units whose room identification has changed by the total number of indoor units in the multi-split central control system. Compare the group change rate with a preset change threshold; If the group change rate is greater than or equal to the preset change threshold, the currently effective indoor unit room grouping result is updated to the new indoor unit room grouping result, and the new indoor unit room grouping result is sent to each indoor unit; If the group change rate is less than the preset change threshold, the currently effective indoor unit room grouping results are maintained.

[0014] Preferably, according to the multi-split air conditioning indoor unit room identification method based on indoor pressure provided by the present invention, after the step of updating the new indoor unit room grouping result to the currently effective indoor unit room grouping result, the method includes: The updated and currently effective indoor unit room grouping results are stored in groups. Each group records the corresponding room identifier and a list of unique indoor unit identifiers belonging to the group. In response to a collaborative control command for a target room, the system searches for a list of unique indoor unit identifiers belonging to the group based on the room identifier corresponding to the target room, and issues unified control commands to all indoor units in the list.

[0015] In a second aspect, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the room identification method for multi-split indoor units based on indoor pressure as described above.

[0016] Thirdly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the room identification method for multi-split indoor units based on indoor pressure as described above.

[0017] Fourthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the multi-split indoor unit room identification method based on indoor pressure as described above.

[0018] Fifthly, the present invention also provides a multi-split air conditioner, which includes at least two indoor units, each indoor unit being equipped with a micro differential pressure sensor. The multi-split air conditioner stores a computer program, which, when executed by a processor, implements any of the above-described methods for room identification of multi-split indoor units based on indoor pressure. This invention provides a method for room identification of indoor units in a multi-split air conditioning system based on indoor pressure, and a multi-split air conditioner. The method involves acquiring time-series pressure data collected by micro-differential pressure sensors of each indoor unit within a first preset time window, and calculating a pressure fluctuation feature vector for each indoor unit based on the pressure time-series data. A weighted similarity algorithm is then used to calculate a similarity index between the pressure fluctuation feature vectors of any two indoor units. This similarity index characterizes the degree of consistency in the pressure fluctuation characteristics of the two indoor units. Based on the similarity comparison result between the similarity index and a preset similarity threshold, clustering is performed on all indoor units. Indoor units with a similarity index greater than or equal to the preset similarity threshold are grouped into the same group, and each group is assigned a unique room identifier, resulting in the indoor unit room grouping result. This method can automatically group indoor units based on indoor pressure fluctuation characteristics without manual preset, significantly improving grouping efficiency and accuracy. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the method for room identification of multi-split air conditioning indoor units based on indoor pressure provided by the present invention. Figure 2 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0022] First, let's analyze some of the terms used in this invention: Micro differential pressure sensor: A sensor capable of accurately measuring minute pressure differences, typically covering a measurement range from -500 Pascals (Pa) to 500 Pascals, characterized by its small size, high accuracy, and modular integration. In this invention, the micro differential pressure sensor is configured on each indoor unit to collect real-time air pressure fluctuation data at the location of the indoor unit.

[0023] Pressure time series data: refers to the sequence of pressure values ​​collected continuously by a micro differential pressure sensor at a preset sampling frequency (e.g., once per second) and arranged in chronological order. This sequence reflects the dynamic change of indoor pressure over time.

[0024] Pressure fluctuation feature vector: A multi-dimensional vector composed of multiple feature parameters extracted from pressure time-series data. In this invention, this vector includes three dimensions: pressure fluctuation amplitude feature value, pressure change slope feature value, and pressure stabilization duration feature value, used to quantitatively describe the pressure fluctuation characteristics of an indoor unit's location.

[0025] Weighted similarity algorithm: A method for measuring the degree of similarity between two multidimensional vectors. This algorithm assigns different weight coefficients to different dimensions of the vectors to reflect the difference in importance of each dimension in the similarity determination, and then calculates the weighted distance and converts the distance into a similarity index.

[0026] Weighted Euclidean distance: Euclidean distance is the straight-line distance between two points in Euclidean space. Weighted Euclidean distance is based on standard Euclidean distance, with different weight coefficients assigned to each dimension.

[0027] Clustering is an unsupervised learning method used to divide samples in a dataset into several subsets (clusters) based on a certain similarity metric, such that the similarity of samples within the same cluster is as high as possible, and the similarity of samples between different clusters is as low as possible. In this invention, a clustering algorithm based on a similarity threshold is used to group indoor units.

[0028] K-means clustering algorithm: A commonly used clustering algorithm. Its basic idea is to randomly select K initial cluster centers, calculate the distance from each sample to each cluster center, assign the sample to the cluster containing the nearest cluster center, and then recalculate the center of each cluster. This process iterates until the cluster centers no longer change or a preset number of iterations is reached. In the embodiments of this invention, the K-means algorithm can be used as the specific implementation of the clustering operation.

[0029] All actions involving the acquisition of signal information or data in this invention are carried out in compliance with the relevant data protection laws and policies of the country where the device is located, and with the authorization granted by the owner of the device.

[0030] The following is combined Figures 1-2 This invention describes a method for identifying indoor unit rooms in a multi-split air conditioner based on indoor pressure, and a multi-split air conditioner that can automatically group indoor unit rooms based on indoor pressure fluctuation characteristics without manual preset, significantly improving grouping efficiency and accuracy.

[0031] Figure 1 This is one of the flowcharts illustrating a method for room identification of multi-split air conditioning indoor units based on indoor pressure provided by the present invention, such as... Figure 1 As shown, the method may include, but is not limited to, steps S100 to S300: S100, within the first preset time window, acquire the pressure time-series data collected by the micro differential pressure sensor of each indoor unit, and calculate the pressure fluctuation feature vector of each indoor unit based on the pressure time-series data; S200 uses a weighted similarity algorithm to calculate the similarity index between the pressure fluctuation feature vectors of any two indoor units; wherein, the similarity index is used to characterize the degree of consistency of the pressure fluctuation characteristics of the two indoor units. S300: Based on the similarity comparison result of the similarity index and the preset similarity threshold, perform clustering operation on all indoor units, divide indoor units with the similarity index greater than or equal to the preset similarity threshold into the same group, and assign a unique room identifier to each group to obtain the indoor unit room grouping result.

[0032] It should be noted that this embodiment provides a method for room identification of indoor units in a multi-split air conditioning system based on indoor pressure. This method is applied to a multi-split air conditioning system, which includes at least two indoor units, a single outdoor unit, and a central control unit. Each indoor unit is independently equipped with a micro-differential pressure sensor. The micro-differential pressure sensor is electrically connected to the indoor unit's main control board and is used to collect pressure data at the location of the indoor unit in real time. The pressure data is then bound to the unique identifier of the indoor unit via the main control board and uploaded to the core control unit.

[0033] In this embodiment of the invention, the core control unit is the centralized control device. Pressure timing data collected by the micro-differential pressure sensors of each indoor unit is transmitted from the indoor unit to the outdoor unit, and then periodically uploaded by the outdoor unit to the centralized control device via a transmission module.

[0034] In step S100 of some embodiments, within a first preset time window, pressure time-series data collected by the micro differential pressure sensors of each indoor unit are acquired, and pressure fluctuation feature vector of each indoor unit is calculated based on the pressure time-series data.

[0035] Understandably, within the first preset time window, the core control unit acquires the pressure time-series data collected by the micro-differential pressure sensors of each indoor unit. The first preset time window can be configured according to the actual application scenario; for example, it is set to 60 seconds by default, but can also be adjusted to 30 seconds, 90 seconds, or 120 seconds based on the periodicity of indoor pressure changes. The core control unit uses this time window as the unit of analysis, extracting features from the pressure time-series data of each indoor unit within this window, and calculating the pressure fluctuation feature vector for each indoor unit.

[0036] The pressure fluctuation feature vector is a multidimensional mathematical expression used to describe the pressure fluctuation characteristics at the location of the indoor unit. In this embodiment, the vector includes at least feature parameters such as pressure fluctuation amplitude, pressure change slope, and pressure stabilization time. Subsequent embodiments will describe in detail the specific calculation methods for these feature parameters.

[0037] The technical effect achieved by this step is that by configuring a micro differential pressure sensor and extracting pressure fluctuation characteristics, the continuous analog signal of indoor pressure changes in the physical world is transformed into discrete numerical characteristics that can be processed by a computer, providing a standardized data foundation for subsequent intelligent group calculations.

[0038] In step S200 of some embodiments, a weighted similarity algorithm is used to calculate the similarity index between the pressure fluctuation feature vectors of any two indoor units; wherein, the similarity index is used to characterize the degree of consistency of the pressure fluctuation features of the two indoor units.

[0039] Understandably, the core control unit employs a weighted similarity algorithm to calculate the similarity index between the pressure fluctuation feature vectors of any two indoor units. This similarity index is a numerical value, typically ranging from [0, 1], used to quantitatively characterize the degree of consistency in the pressure fluctuation characteristics of the two indoor units. The closer the similarity index is to 1, the more similar the pressure fluctuation characteristics of the two indoor units are, meaning that the pressure changes at their locations are highly synchronized; the closer the similarity index is to 0, the greater the difference in the pressure fluctuation characteristics of the two indoor units, meaning that the pressure changes at their locations are independent of each other.

[0040] The core idea of ​​the weighted similarity algorithm is to assign different weight coefficients to different feature parameters in the pressure fluctuation feature vector to reflect the differences in the importance of each feature parameter in the similarity determination. For example, the pressure fluctuation amplitude can most intuitively reflect the intensity of indoor disturbance, so it can be given a higher weight; while the pressure change slope and pressure stabilization duration, as auxiliary features, can be given relatively lower weights. Subsequent embodiments will explain the specific calculation process of the weighted similarity algorithm in detail.

[0041] The technical effect achieved by this step is that, through a weighted similarity algorithm, the consistency of pressure fluctuation characteristics between any two indoor units can be quantitatively evaluated, providing an accurate quantitative basis for subsequent clustering and grouping, and avoiding the subjectivity and inaccuracy of human experience judgment.

[0042] In step S300 of some embodiments, based on the similarity comparison result of the similarity index and the preset similarity threshold, a clustering operation is performed on all indoor units, and indoor units with the similarity index greater than or equal to the preset similarity threshold are grouped into the same group, and a unique room identifier is assigned to each group to obtain the indoor unit room grouping result.

[0043] Understandably, the core control unit compares the similarity index calculated in step S200 with a preset similarity threshold, performing clustering operations on all indoor units. The preset similarity threshold is a critical value used to determine whether two indoor units belong to the same room; for example, it can be preset to 85% (i.e., 0.85). When the similarity index of two indoor units is greater than or equal to this threshold, they are determined to be highly similar in features and meet the matching conditions of being in the same room; when the similarity index is less than this threshold, they are determined to be significantly different in features, excluding the possibility of matching them in the same room.

[0044] The specific process of clustering is as follows: The core control unit uses a similarity index as a distance metric to divide all indoor units into several clusters. Within each cluster, the similarity index between any two indoor units is greater than or equal to a preset similarity threshold. After the clustering operation is completed, the core control unit assigns a unique room identifier to each cluster (i.e., each group), such as "Room_001", "Room_002", or "Meeting Room 101", "Office 102", etc., forming a mapping relationship of "room identifier - indoor unit number", thus obtaining the indoor unit room grouping results.

[0045] The technical effect achieved by this step is that it realizes the automatic identification and grouping of indoor unit rooms through clustering operations, completely replacing the traditional manual preset method and fundamentally solving the problems of low efficiency and easy error in manual grouping.

[0046] Taking a 10-story office building as an example, each floor has 8 indoor units, totaling 80 indoor units. The method used in this embodiment is to group the rooms: the core control unit acquires the pressure time-series data of all 80 indoor units within a 60-second time window, calculates the pressure fluctuation feature vector for each indoor unit, then calculates the similarity index between any two indoor units, and finally performs clustering operations based on an 85% similarity threshold. Assuming each floor of the office building is an open-plan office area, and the 8 indoor units on each floor are actually located in the same room, the clustering operation will automatically group the 8 indoor units on each floor into the same group, forming 10 groups, each corresponding to one floor. The entire calculation process takes approximately 3-5 seconds, requires no manual intervention, and the grouping accuracy can reach over 95%.

[0047] In some embodiments of the present invention, the pressure fluctuation feature vector includes the following parameters: pressure fluctuation amplitude feature value, pressure change slope feature value, and pressure stabilization duration feature value; The calculation of the pressure fluctuation feature vector for each indoor unit based on the pressure time-series data includes: Calculate the difference between the maximum and minimum values ​​of the pressure time series data within the first preset time window to obtain the pressure difference value, and use the pressure difference value as the pressure fluctuation amplitude characteristic value; The first preset time window is divided into multiple consecutive calculation cycles. The pressure time series data in each calculation cycle is linearly fitted to obtain the pressure change rate in the corresponding calculation cycle. The statistical average of the pressure change rates in multiple calculation cycles is used as the pressure change slope feature value. Set a pressure fluctuation threshold range, and count the cumulative duration of pressure time series data that are continuously within the pressure fluctuation threshold range within the first preset time window. Use the cumulative duration as a pressure stability duration feature value.

[0048] It is understood that this embodiment further defines the specific composition of the pressure fluctuation feature vector and the calculation method of each feature parameter.

[0049] In this embodiment, the pressure fluctuation feature vector includes the following three parameters: pressure fluctuation amplitude feature value, pressure change slope feature value, and pressure stabilization duration feature value. These three feature parameters characterize the indoor pressure fluctuation from different dimensions, and together constitute a complete mathematical expression describing the pressure fluctuation characteristics of an indoor unit.

[0050] The steps for calculating the characteristic value of pressure fluctuation amplitude are as follows: The core control unit calculates the difference between the maximum and minimum values ​​of the pressure time sequence data within the first preset time window to obtain the pressure difference value, and uses this pressure difference value as the characteristic value of the pressure fluctuation amplitude.

[0051] Specifically, let the duration of the first preset time window be T (unit: seconds), and the pressure time series data within this window be: , ,....., Where n is the number of sampling points. The formula for calculating the characteristic value A of the pressure fluctuation amplitude is: The formula represents the difference between the maximum and minimum pressure values ​​within a given time window, where A is the characteristic value of the pressure fluctuation amplitude. This represents pressure time series data, where n is the number of sampling points.

[0052] The pressure fluctuation amplitude characteristic value reflects the intensity of pressure disturbance experienced by the indoor unit's location within a preset time window. Indoor units located in different positions within the same room should have basically the same pressure fluctuation amplitude due to spatial connectivity; however, the pressure fluctuation amplitude may differ significantly between different rooms due to wall isolation.

[0053] The steps for calculating the characteristic value of the slope of pressure change are as follows: The core control unit divides the first preset time window into multiple consecutive calculation cycles, performs linear fitting on the pressure time series data within each calculation cycle to obtain the pressure change rate within the corresponding calculation cycle, and then uses the statistical average of the pressure change rates of multiple calculation cycles as the pressure change slope feature value.

[0054] Specifically, the calculation cycle is designed to last for Δt (e.g., 10 seconds), and the first preset time window T is divided into m = T / Δt consecutive calculation cycles. For the j-th calculation cycle, the corresponding pressure time series data is as follows: , ,....., The least squares method was used to perform a linear fit on the pressure data within this period, and the fitted line equation was: ,in This represents the rate of pressure change within the calculation period. A positive value indicates that the pressure is rising, a negative value indicates that the pressure is falling, and the larger the absolute value, the more drastic the pressure change.

[0055] Then, the statistical average of the pressure change rate over all calculation cycles is calculated to obtain the characteristic value S of the pressure change slope: in, This represents the characteristic value of the slope of the pressure change, and m represents the number of calculation periods. This indicates the rate of pressure change within the calculation period.

[0056] The characteristic value of the pressure change slope reflects the rate of pressure change at the location of the indoor unit. Indoor units in different locations within the same room should have a basically consistent pressure change slope because pressure disturbances are transmitted synchronously within the space; however, the pressure change slope of different rooms may differ due to differences in room volume, sealing, and other factors.

[0057] The calculation steps for the characteristic value of pressure stabilization time are as follows: The core control unit sets a pressure fluctuation threshold range, and counts the cumulative duration of pressure time series data that are continuously within the pressure fluctuation threshold range within the first preset time window, and uses the cumulative duration as the pressure stability duration feature value.

[0058] Specifically, let the pressure fluctuation threshold be δ (e.g., ±5 Pascals), with the reference pressure... (This could be the average pressure within that time window) as the center, define the steady-state interval as [ , The core control unit traverses the pressure time-series data, detects time periods where the pressure values ​​of multiple consecutive sampling points fall within the steady-state range, and calculates the cumulative duration of these time periods. , which serves as a characteristic value for the duration of pressure stabilization.

[0059] The pressure stabilization time characteristic value reflects the speed at which the pressure at the location of the indoor unit recovers to stability. Indoor units in different locations within the same room should have a basically consistent pressure stabilization time due to spatial connectivity; however, the pressure stabilization time may differ between different rooms due to differences in sealing.

[0060] This embodiment extracts feature parameters from three dimensions: pressure fluctuation amplitude, pressure change slope, and pressure stabilization duration. It can comprehensively characterize the physical properties of indoor pressure fluctuations from three perspectives: disturbance intensity, change rate, and recovery characteristics. This provides rich and accurate quantitative basis for subsequent similarity calculations, improving the accuracy and robustness of room identification.

[0061] In some embodiments of the present invention, the step of employing a weighted similarity algorithm to calculate the similarity index between the pressure fluctuation feature vectors of any two indoor units includes: Obtain the pressure fluctuation feature vector of the first indoor unit and the pressure fluctuation feature vector of the second indoor unit; The differences between the first indoor unit and the second indoor unit at three characteristic values ​​are calculated respectively to obtain the three characteristic differences; Obtain three preset weight coefficients, each of which corresponds one-to-one with one of the three feature values; Calculate the product of the square of each feature difference and the corresponding weight coefficient, add the three products together and take the square root to obtain the weighted Euclidean distance; The weighted Euclidean distance is calculated to obtain the similarity index between the first indoor unit and the second indoor unit.

[0062] It is understood that in this embodiment, weighted Euclidean distance is used as the basis for similarity measurement, and the weighted Euclidean distance is converted into a similarity index. Specifically, it includes the following sub-steps: the core control unit obtains the pressure fluctuation feature vector of the first indoor unit and the pressure fluctuation feature vector of the second indoor unit. Let the pressure fluctuation feature vector of the first indoor unit A be VA=(A1, A2, A3), where A1 is the pressure fluctuation amplitude feature value, A2 is the pressure change slope feature value, and A3 is the pressure stabilization time feature value; the pressure fluctuation feature vector of the second indoor unit B is VB=(B1, B2, B3), where the meanings of B1, B2, and B3 are the same as those of unit A.

[0063] Furthermore, the differences between the first indoor unit and the second indoor unit on three characteristic values ​​are calculated separately to obtain the three characteristic differences. Specifically: Δ1=A1-B1, Δ2=A2-B2, Δ3=A3-B3.

[0064] Wherein, Δ1 represents the difference in pressure fluctuation amplitude characteristic value between the first indoor unit A and the second indoor unit B, Δ2 represents the pressure change slope characteristic value between the first indoor unit A and the second indoor unit B, and Δ3 represents the difference in pressure stabilization time characteristic value between the first indoor unit A and the second indoor unit B.

[0065] Furthermore, the core control unit acquires three preset weighting coefficients, each corresponding to one of the three feature values. Let w1 be the weighting coefficient corresponding to the pressure fluctuation amplitude, w2 be the weighting coefficient corresponding to the pressure change slope, and w3 be the weighting coefficient corresponding to the pressure stabilization time, satisfying w1+w2+w3=1.

[0066] In a preferred embodiment, w1=0.4, w2=0.3, and w3=0.3. This is because the pressure fluctuation amplitude is the most critical feature, most directly reflecting the intensity of indoor disturbance. The pressure fluctuation amplitude differences between different indoor units within the same room are minimal, therefore, they are assigned the highest weight. The pressure change slope and pressure stabilization duration, as auxiliary matching features, provide additional discriminative information and are assigned relatively lower weights. This weight allocation highlights the role of the core features while also considering the value of auxiliary features, thus improving matching accuracy.

[0067] Furthermore, the product of the square of each feature difference and its corresponding weight coefficient is calculated. The sum of these three products is then taken as the square root to obtain the weighted Euclidean distance D. Where D represents the weighted Euclidean distance, w1 is the weighting coefficient corresponding to the pressure fluctuation amplitude, w2 is the weighting coefficient corresponding to the pressure change slope, and w3 is the weighting coefficient corresponding to the pressure stabilization time. Δ1 represents the difference in the characteristic value of the pressure fluctuation amplitude between the first indoor unit A and the second indoor unit B, Δ2 represents the characteristic value of the pressure change slope between the first indoor unit A and the second indoor unit B, and Δ3 represents the difference in the characteristic value of the pressure stabilization time between the first indoor unit A and the second indoor unit B.

[0068] The range of values ​​for the weighted Euclidean distance D depends on the magnitude of the eigenvalues. After standardization, the eigenvalues ​​are mapped to the interval [0,1], and the range of values ​​for D is [0,1].

[0069] Furthermore, the weighted Euclidean distance D is transformed and calculated to obtain the similarity index S between the first indoor unit and the second indoor unit. In a preferred embodiment, the formula for calculating the similarity index S is: S=1-D Where S represents the similarity index and D represents the weighted Euclidean distance.

[0070] At this point, the similarity index S also ranges from [0, 1]. When the pressure fluctuation characteristics of the two indoor units are completely identical, Δ1=Δ2=Δ3=0, then D=0, S=1; when the difference in pressure fluctuation characteristics between the two indoor units reaches its maximum, D=1, S=0.

[0071] In another implementation, other similarity calculation methods such as cosine similarity and Pearson correlation coefficient can also be used, but weighted Euclidean distance is the preferred option because it is simple to calculate, has clear physical meaning, and is easy to implement in engineering.

[0072] This embodiment quantifies multidimensional feature differences into a single similarity index using weighted Euclidean distance. It considers both the degree of difference in each dimension and the importance differences of different features through weight allocation, making similarity determination more scientific and accurate. The preset weight coefficients (0.4, 0.3, 0.3) have been experimentally verified to achieve optimal grouping accuracy in various indoor scenarios.

[0073] In some embodiments of the present invention, the similarity comparison result based on the similarity index and the preset similarity threshold is used to perform clustering operations on all indoor units, and indoor units with the similarity index greater than or equal to the preset similarity threshold are grouped into the same group, including: Select the indoor unit to be assigned as the current cluster center, and group all indoor units with a similarity index greater than or equal to the preset similarity threshold into the same temporary group; Repeat the steps of selecting the current cluster center and assigning to the same temporary group until all indoor units are assigned to the temporary group.

[0074] It is understood that in this embodiment, clustering operations are performed on all indoor units based on similarity metrics and preset similarity thresholds, specifically including the following sub-steps: The specific steps for selecting cluster centers and forming temporary groups are as follows: The core control unit is selected from the set of indoor units that have not been assigned to any temporary group. One indoor unit to be assigned is chosen as the current cluster center (or seed indoor unit). Then, all unassigned indoor units are traversed, and the similarity index between each indoor unit and the current cluster center is calculated. All indoor units with a similarity index greater than or equal to a preset similarity threshold (e.g., 0.85) are grouped into the same temporary group.

[0075] In one specific implementation, assuming a preset similarity threshold of Th = 0.85, and the current cluster center is indoor unit i, then the members of the temporary group G are: G = {j | Sij ≥ 0.85, and j is not assigned} Where Sij represents the similarity index between indoor unit i and indoor unit j.

[0076] Furthermore, the process is iterated until all indoor units are assigned.

[0077] Repeat the steps of selecting cluster centers and forming temporary groups, that is, sequentially select indoor units that have not been assigned to any temporary group as new current cluster centers, and perform the temporary group formation operation until all indoor units are assigned to a certain temporary group.

[0078] It should be noted that the current cluster centers can be selected randomly, in the order of the indoor unit numbers, or using other preset selection strategies. Different selection strategies may affect the order in which temporary groups are formed, but will not affect the final grouping results.

[0079] This embodiment uses an iterative method to select cluster centers, quickly dividing all indoor units into multiple temporary groups, thus achieving preliminary clustering of the indoor units. The computational complexity of the clustering grouping algorithm provided in the above embodiment of the present invention is O(n^2). 2 ), where n is the number of indoor units. In typical multi-split system scenarios (n≤200), it has good computational efficiency and can complete the initial grouping of all indoor units within a few seconds.

[0080] In some embodiments of the present invention, after the step of assigning all indoor units to a temporary group, the method includes: Traverse all temporary groups and detect whether there are at least two pairs of indoor units between any two temporary groups, and the similarity index between the two indoor units in the pair is greater than or equal to the preset similarity threshold. If they exist, the two temporary groups are merged into one group, and a unique room identifier is assigned to each group after the merge process.

[0081] It is understood that in this embodiment, after all indoor units are assigned to temporary groups, the following group merging step is also included: The core control unit traverses all temporary groups and detects whether there are at least two pairs of indoor units between any two temporary groups, and the similarity index between the two indoor units in the pair is greater than or equal to a preset similarity threshold.

[0082] Let Gp and Gq be two temporary groups, where Gp represents the first temporary group and Gq represents the second temporary group. An indoor unit pair is an ordered pair consisting of an indoor unit taken from the first temporary group Gp and an indoor unit taken from the second temporary group Gq. For any indoor unit u∈Gp and v∈Gq (∈ indicates belonging), if Suv≥Th, then (u,v) is called a pair of indoor units that satisfy the condition. This step requires that there exist at least two such pairs of indoor units. Th is a preset similarity threshold, and Suv represents the similarity index between indoor unit u belonging to the first temporary group and indoor unit v belonging to the second temporary group.

[0083] If two temporary groups that meet the conditions are detected, then these two temporary groups are determined to be missing groups in the same room, and they are merged into one group.

[0084] Assign a unique room identifier to each group after the merge process.

[0085] For example, suppose a large conference room actually has four indoor units installed: M1, M2, M3, and M4. After initial clustering, M1 and M2 have extremely high similarity (92%), so they are assigned to temporary group G1; M3 and M4 have extremely high similarity (90%), so they are assigned to temporary group G2. Meanwhile, M2 and M3 have a similarity of 87% (≥85%), and M1 and M4 have a similarity of 86% (≥85%), meaning there are two cross-group indoor unit pairs that meet the similarity threshold condition. Therefore, this step merges G1 and G2 into one group, correctly identifying M1, M2, M3, and M4 as all indoor units in the same room.

[0086] This embodiment effectively solves the problem of indoor units in the same room being incorrectly split into multiple temporary groups due to minor differences in feature data by introducing a grouping and merging rule based on "at least two pairs of indoor units". This rule utilizes the idea of ​​"transitive clustering," which states that if multiple strongly correlated bridge pairs of indoor units exist between two groups, it proves that they are physically connected and must be merged. This significantly improves the completeness and accuracy of the grouping results, avoiding the defect of the same room being incorrectly split into multiple groups.

[0087] In some embodiments of the present invention, after the step of grouping indoor units with similarity indices greater than or equal to the preset similarity threshold into the same group, the method includes: Traverse each group and verify whether the similarity index of any two indoor units in the group is greater than or equal to the preset similarity threshold. If there is a pair of indoor units with a similarity index less than the preset similarity threshold, remove the abnormal indoor unit in the group whose similarity index with other indoor units in the group is less than the preset similarity threshold from the corresponding group, and re-perform clustering operation on the abnormal indoor units after they are removed from the group to group the abnormal indoor units. Traverse any two different groups and check whether there are indoor unit pairs between the two groups with a similarity index greater than or equal to the preset similarity threshold. If so, merge the two groups.

[0088] It is understood that in this embodiment, after classifying indoor units with similarity indices greater than or equal to a preset similarity threshold into the same group, two sub-steps are included: intra-group verification and inter-group verification, to ensure the accuracy of the grouping results.

[0089] First, an intra-group verification is performed. The core control unit traverses each group and verifies whether the similarity index of any two indoor units in the group is greater than or equal to the preset similarity threshold.

[0090] Specifically, for a given group G={i1,i2,…,ik}, the verification condition is: u, v∈G, u≠v, Suv≥Th Th is the preset similarity threshold, and Suv represents the similarity index between indoor unit u belonging to the first temporary group and indoor unit v belonging to the second temporary group.

[0091] If there are indoor unit pairs with a similarity index lower than a preset similarity threshold, it indicates that there is an abnormal indoor unit in that group. In this case, the core control unit removes the abnormal indoor unit from the corresponding group that has a similarity index lower than the preset similarity threshold with other indoor units in the group, and re-performs the clustering operation on the abnormal indoor units after they are removed from the group, so as to regroup the abnormal indoor units.

[0092] In some embodiments, inter-group verification is also included. After completing intra-group verification, the core control unit traverses any two different groups to verify whether there are indoor unit pairs between the two groups whose similarity index is greater than or equal to a preset similarity threshold.

[0093] Specifically, for two different temporary groups, the first temporary group Gp and the second temporary group Gq, it is checked whether there exists an indoor unit u∈Gp (first temporary group) and an indoor unit v∈Gq (second temporary group) such that Suv≥Th (preset similarity threshold). If they exist, it means that these two groups actually belong to the same room, and the core control unit merges these two groups.

[0094] It should be noted that intra-group and inter-group verification may need to be performed iteratively. For example, after merging two groups in the inter-group verification, the newly merged group may need to undergo intra-group verification again. The core control unit will continue to perform verification until all groups pass both intra-group and inter-group verification.

[0095] This embodiment employs a dual verification mechanism to rigorously verify the accuracy of the clustering grouping results. Intra-group verification ensures that indoor units within each group exhibit high characteristic consistency, while inter-group verification ensures that indoor units across different groups exhibit significant characteristic differences. This dual verification mechanism can detect and correct errors that may occur during the clustering process, such as misgrouping caused by temporary sensor malfunctions or data anomalies, significantly improving the reliability and robustness of the grouping results.

[0096] In some embodiments of the present invention, within a second preset time window, the method includes: A new indoor unit room grouping result is generated, and the new indoor unit room grouping result is compared with the currently effective indoor unit room grouping result. When the grouping difference is greater than or equal to a preset difference threshold, the new indoor unit room grouping result is updated to the currently effective indoor unit room grouping result.

[0097] Understandably, this embodiment incorporates a dynamic calibration mechanism to adapt to dynamic changes in the room environment.

[0098] In this embodiment, within the second preset time window, the core control unit performs the following dynamic calibration steps: First, within the second preset time window, the core control unit reacquires the pressure timing data collected by the micro differential pressure sensors of each indoor unit, and re-executes the steps described in the above embodiment to generate new indoor unit room grouping results.

[0099] The second preset time window can be the same as or different from the first preset time window. In one implementation, the second preset time window can be set to 24 hours, meaning that regrouping is performed automatically once a day; in another implementation, the second preset time window can be set to 1 hour, meaning that regrouping is performed automatically once an hour. The length of the second preset time window can be configured according to the real-time requirements of the actual application scenario.

[0100] The core control unit compares the newly generated indoor unit room grouping results with the currently effective indoor unit room grouping results and calculates the degree of grouping difference.

[0101] When the grouping difference is greater than or equal to the preset difference threshold, the core control unit updates the newly generated indoor unit room grouping result to the currently effective indoor unit room grouping result; when the grouping difference is less than the preset difference threshold, the currently effective indoor unit room grouping result remains unchanged.

[0102] The preset difference threshold is used to measure whether the difference between the old and new grouping results is significant enough to require an update.

[0103] This embodiment uses a dynamic calibration mechanism to enable the system to automatically recalculate the room grouping results of the indoor units periodically or as needed, thereby adaptively responding to changes in the indoor environment (such as room partition modifications, changes in door and window sealing, and adjustments to the position of indoor units). It can maintain the long-term accuracy and stability of the grouping results without manual intervention, significantly reducing the operation and maintenance costs of the multi-split air conditioning system.

[0104] In some embodiments of the present invention, the step of comparing the new indoor unit room grouping result with the currently effective indoor unit room grouping result, and updating the new indoor unit room grouping result to the currently effective indoor unit room grouping result when the grouping difference is greater than or equal to a preset difference threshold, includes: The new indoor unit room grouping results are compared with the currently effective indoor unit room grouping results for each indoor unit, and the number of indoor units whose room labels have been changed is counted. The group change rate is obtained by dividing the number of indoor units whose room identification has changed by the total number of indoor units in the multi-split central control system. Compare the group change rate with a preset change threshold; If the group change rate is greater than or equal to the preset change threshold, the currently effective indoor unit room grouping result is updated to the new indoor unit room grouping result, and the new indoor unit room grouping result is sent to each indoor unit; If the group change rate is less than the preset change threshold, the currently effective indoor unit room grouping results are maintained.

[0105] It is understood that this embodiment further defines the specific calculation method and update judgment logic for the degree of group difference.

[0106] In this embodiment, the newly generated indoor unit room grouping results are compared with the currently effective indoor unit room grouping results. When the grouping difference is greater than or equal to a preset difference threshold, an update is performed. Specifically, this includes the following sub-steps: First, the number of indoor units that have changed is counted. The core control unit compares the newly generated indoor unit room grouping results with the currently effective indoor unit room grouping results for each indoor unit to count the number of indoor units whose room labels have changed.

[0107] Specifically, let the total number of indoor units in the multi-split system be N. For each indoor unit i, let its room identifier in the currently effective grouping result be Ro(i), and its room identifier in the newly generated grouping result be Rn(i). If Rn(i) ≠ Ro(i), then the room identifier of that indoor unit has been changed. Count the number M of all indoor units that have changed: Where M represents the number of indoor units that have been modified. This is an indicator function that takes the value 1 when the condition is true, i.e., Rn(i) ≠ Ro(i), and takes the value 0 otherwise.

[0108] Furthermore, the core control unit divides the number of indoor units with changed room identification by the total number of indoor units in the multi-split system to obtain the group change rate. : in, This indicates the group change rate, where M represents the number of indoor units that have changed, and N represents the total number of indoor units.

[0109] Group change rate The value range is [0,1]. When the room identifiers of all indoor units remain unchanged... =0; When the room identifier of all indoor units changes. =1.

[0110] Furthermore, the core control unit will group change rates. Compared with the preset change threshold Comparisons can be made. The preset change threshold can be configured according to the stability requirements of the actual application scenario; for example, it can be set to 0.1 (i.e., 10%).

[0111] If the group change rate Greater than or equal to the preset change threshold This indicates that the grouping results have changed significantly. The core control unit updates the currently effective indoor unit room grouping results to the newly generated indoor unit room grouping results, and sends the newly generated indoor unit room grouping results to the main control board of each indoor unit for local storage.

[0112] If the group change rate Less than the preset change threshold This indicates that the change in the grouping results is not significant (it may be due to accidental changes caused by minor disturbances such as sensor noise). The core control unit maintains the currently effective indoor unit room grouping results unchanged and does not perform any update operations.

[0113] This embodiment achieves quantitative evaluation and intelligent decision-making regarding changes in grouping results by calculating the grouping change rate and comparing it with a preset change threshold. When environmental changes are significant, the grouping results are automatically updated; when environmental changes are insignificant, frequent updates are avoided. This ensures the real-time accuracy of the grouping while preventing invalid updates caused by minor disturbances such as sensor noise, thus improving the system's stability and efficiency.

[0114] In some embodiments of the present invention, after the step of updating the new indoor unit room grouping result to the currently effective indoor unit room grouping result, the method includes: The updated and currently effective indoor unit room grouping results are stored in groups. Each group records the corresponding room identifier and a list of unique indoor unit identifiers belonging to the group. In response to a collaborative control command for a target room, the system searches for a list of unique indoor unit identifiers belonging to the group based on the room identifier corresponding to the target room, and issues unified control commands to all indoor units in the list.

[0115] It is understood that this embodiment further limits the application of the grouping results.

[0116] In this embodiment, after updating the newly generated indoor unit room grouping results to the currently effective indoor unit room grouping results, the following steps are also included: The core control unit stores the updated, currently effective room grouping results for indoor units in groups. Each group records the corresponding room identifier and a list of unique identifiers for the indoor units belonging to that group.

[0117] The storage structure can be in the form of key-value pairs, for example: Room label "101 Meeting Room": Indoor unit number list [M01, M02, M03].

[0118] Room label "Office 102": Indoor unit number list [M04, M05].

[0119] Room label "103 Storage Room": Indoor unit number list [M06].

[0120] The storage result can be saved to the output storage unit (such as non-volatile memory) so that the group information can still be recovered after the multi-unit system is powered off and restarted.

[0121] Furthermore, when the multi-split central control host needs to issue a collaborative control command to a target room (such as "adjust the temperature of meeting room 101 to 24℃" or "turn off all indoor units in office 102"), the core control unit or central control host responds to the collaborative control command for the target room and, based on the room identifier corresponding to the target room, searches the stored grouping results for a list of unique identifiers of indoor units belonging to that group.

[0122] The core control unit or centralized control host issues unified control commands to all indoor units in the list of unique indoor unit identifiers. For example, for the "101 Conference Room" group, the temperature adjustment command is sent simultaneously to three indoor units, M01, M02, and M03, to achieve synchronous control of all indoor units in the same room.

[0123] This embodiment stores the grouping results as a mapping relationship between room identifiers and indoor unit lists, and responds to collaborative management and control commands by uniformly issuing control commands to all indoor units within the group. This achieves room-level collaborative management and control based on automatic identification results, fully leveraging the technical value of automated grouping and improving the practicality and user experience of the multi-unit centralized control system.

[0124] In some embodiments of the present invention, the method further includes: in the data acquisition step, after receiving the pressure data collected by the micro differential pressure sensor, the indoor unit main control board can also perform an abnormal data filtering step to improve the quality and reliability of the data. Specifically, this includes: Determine whether the current pressure data exceeds the preset sensor range (e.g., -500Pa to 500Pa). If it does, mark it as invalid data and discard it. Determine whether the rate of change of the current pressure data compared to the pressure data at the previous sampling time exceeds a preset mutation threshold. If it does, mark it as abnormal fluctuation data and remove it. The remaining pressure data after removing invalid and abnormal fluctuation data will be used as valid data for subsequent feature extraction steps.

[0125] After feature extraction and before similarity calculation, a feature standardization step can be included. Due to slight differences in precision among different differential pressure sensors and minor deviations in the installation location of the indoor unit, the magnitudes of the three types of feature parameters will differ, leading to errors in direct calculation. Therefore, a normalization calculation method can be used to uniformly transform the three types of feature parameters of each indoor unit into the [0,1] interval. The standardization formula is: Where X represents the original value of a certain characteristic of a single indoor unit. This represents the maximum value of this type of characteristic for all indoor units. This is the minimum value for this type of characteristic across all indoor units. These are the standardized feature values.

[0126] The embodiments provided by this invention can completely replace the traditional manual pre-setting grouping method, realizing automatic identification and grouping of indoor unit rooms without any manual intervention, significantly improving grouping efficiency and reducing labor costs. Furthermore, through multi-dimensional feature extraction, weighted similarity calculation, clustering grouping, and a dual verification mechanism, the accuracy of the grouping results is ensured, avoiding errors that may be caused by manual operation. Through a dynamic calibration mechanism, it can automatically respond to changes in the room environment and update the grouping results in real time without manual reconfiguration, demonstrating strong adaptability. Moreover, by introducing pressure sensing, data analysis, and machine learning into the field of multi-split air conditioning system centralized control, it realizes intelligent decision-making based on environmental physical characteristics, providing technical support for the intelligent upgrade of multi-split air conditioning systems.

[0127] Figure 2 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 2As shown, the electronic device may include: a processor 210, a communication interface 220, a memory 230, and a communication bus 240, wherein the processor 210, the communication interface 220, and the memory 230 communicate with each other through the communication bus 240. The processor 210 can call logic instructions in the memory 230 to execute a multi-split air conditioning indoor unit room identification method based on indoor pressure. The method includes: acquiring pressure time-series data collected by the micro-differential pressure sensors of each indoor unit within a first preset time window, and calculating the pressure fluctuation feature vector of each indoor unit based on the pressure time-series data; using a weighted similarity algorithm to calculate a similarity index between the pressure fluctuation feature vectors of any two indoor units; wherein the similarity index is used to characterize the degree of consistency of the pressure fluctuation characteristics of the two indoor units; performing clustering operation on all indoor units based on the similarity comparison result of the similarity index and a preset similarity threshold, classifying indoor units with similarity indices greater than or equal to the preset similarity threshold into the same group, and assigning a unique room identifier to each group to obtain the indoor unit room grouping result.

[0128] Furthermore, the logical instructions in the aforementioned memory 230 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0129] On the other hand, the present invention also provides 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, when the program instructions are executed by the computer, the computer can execute the multi-split air conditioning indoor unit room identification method based on indoor pressure provided by the above methods, the method including: within a first preset time window, acquiring pressure time-series data collected by the micro differential pressure sensors of each indoor unit, and calculating the pressure fluctuation feature vector of each indoor unit based on the pressure time-series data; using a weighted similarity algorithm to calculate the similarity index between the pressure fluctuation feature vectors of any two indoor units; wherein, the similarity index is used to characterize the degree of consistency of the pressure fluctuation characteristics of the location of the two indoor units; based on the similarity comparison result of the similarity index and the preset similarity threshold, performing a clustering operation on all indoor units, classifying indoor units with the similarity index greater than or equal to the preset similarity threshold into the same group, and assigning a unique room identifier to each group to obtain the indoor unit room grouping result.

[0130] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program performs the aforementioned multi-split air conditioning indoor unit room identification methods based on indoor pressure. The method includes: acquiring pressure time-series data collected by micro-differential pressure sensors of each indoor unit within a first preset time window, and calculating a pressure fluctuation feature vector for each indoor unit based on the pressure time-series data; calculating a similarity index between the pressure fluctuation feature vectors of any two indoor units using a weighted similarity algorithm; wherein the similarity index is used to characterize the degree of consistency of the pressure fluctuation characteristics of the locations of the two indoor units; performing a clustering operation on all indoor units based on the similarity comparison result of the similarity index and a preset similarity threshold, grouping indoor units with similarity indices greater than or equal to the preset similarity threshold into the same group, and assigning a unique room identifier to each group to obtain the indoor unit room grouping result.

[0131] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0132] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0133] 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 them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for room identification of multi-split air conditioning indoor units based on indoor pressure, characterized in that, Each of the multi-split indoor units is equipped with a micro differential pressure sensor; The method includes: Within the first preset time window, pressure time-series data collected by the micro differential pressure sensors of each indoor unit are acquired, and pressure fluctuation feature vector of each indoor unit is calculated based on the pressure time-series data. A weighted similarity algorithm is used to calculate the similarity index between the pressure fluctuation feature vectors of any two indoor units; wherein, the similarity index is used to characterize the degree of consistency of the pressure fluctuation features of the two indoor units. Based on the similarity comparison results of the similarity index and the preset similarity threshold, a clustering operation is performed on all indoor units. Indoor units with a similarity index greater than or equal to the preset similarity threshold are grouped into the same group, and a unique room identifier is assigned to each group to obtain the indoor unit room grouping results.

2. The method for room identification of multi-split air conditioning indoor units based on indoor pressure according to claim 1, characterized in that, The pressure fluctuation feature vector includes the following parameters: pressure fluctuation amplitude feature value, pressure change slope feature value, and pressure stabilization duration feature value; The calculation of the pressure fluctuation feature vector for each indoor unit based on the pressure time-series data includes: Calculate the difference between the maximum and minimum values ​​of the pressure time series data within the first preset time window to obtain the pressure difference value, and use the pressure difference value as the pressure fluctuation amplitude characteristic value; The first preset time window is divided into multiple consecutive calculation cycles. The pressure time series data in each calculation cycle is linearly fitted to obtain the pressure change rate in the corresponding calculation cycle. The statistical average of the pressure change rates in multiple calculation cycles is used as the pressure change slope feature value. Set a pressure fluctuation threshold range, and count the cumulative duration of pressure time series data that are continuously within the pressure fluctuation threshold range within the first preset time window. Use the cumulative duration as a pressure stability duration feature value.

3. The method for room identification of multi-split air conditioning indoor units based on indoor pressure according to claim 2, characterized in that, The weighted similarity algorithm is used to calculate the similarity index between the pressure fluctuation feature vectors of any two indoor units, including: Obtain the pressure fluctuation feature vector of the first indoor unit and the pressure fluctuation feature vector of the second indoor unit; The differences between the first indoor unit and the second indoor unit at three characteristic values ​​are calculated respectively to obtain the three characteristic differences; Obtain three preset weight coefficients, each of which corresponds one-to-one with one of the three feature values; Calculate the product of the square of each feature difference and the corresponding weight coefficient, add the three products together and take the square root to obtain the weighted Euclidean distance; The weighted Euclidean distance is calculated to obtain the similarity index between the first indoor unit and the second indoor unit.

4. The method for room identification of multi-split air conditioning indoor units based on indoor pressure according to claim 2, characterized in that, The similarity comparison result based on the similarity index and the preset similarity threshold is used to perform clustering operations on all indoor units, grouping indoor units with a similarity index greater than or equal to the preset similarity threshold into the same group, including: Select the indoor unit to be assigned as the current cluster center, and group all indoor units with a similarity index greater than or equal to the preset similarity threshold into the same temporary group; Repeat the steps of selecting the current cluster center and assigning to the same temporary group until all indoor units are assigned to the temporary group.

5. The method for room identification of multi-split air conditioning indoor units based on indoor pressure according to claim 4, characterized in that, After the step of assigning all indoor units to the temporary group, the method includes: Traverse all temporary groups and detect whether there are at least two pairs of indoor units between any two temporary groups, and the similarity index between the two indoor units in the pair is greater than or equal to the preset similarity threshold. If they exist, the two temporary groups are merged into one group, and a unique room identifier is assigned to each group after the merge process.

6. The method for room identification of multi-split air conditioning indoor units based on indoor pressure according to claim 4, characterized in that, After the step of grouping indoor units with a similarity index greater than or equal to the preset similarity threshold into the same group, the method includes: Traverse each group and verify whether the similarity index of any two indoor units in the group is greater than or equal to the preset similarity threshold. If there is a pair of indoor units with a similarity index less than the preset similarity threshold, remove the abnormal indoor unit in the group whose similarity index with other indoor units in the group is less than the preset similarity threshold from the corresponding group, and re-perform clustering operation on the abnormal indoor units after they are removed from the group to group the abnormal indoor units. Traverse any two different groups and check whether there are indoor unit pairs between the two groups with a similarity index greater than or equal to the preset similarity threshold. If so, merge the two groups.

7. The method for room identification of multi-split air conditioning indoor units based on indoor pressure according to any one of claims 1 to 6, characterized in that, Within a second preset time window, the method includes: A new indoor unit room grouping result is generated, and the new indoor unit room grouping result is compared with the currently effective indoor unit room grouping result. When the grouping difference is greater than or equal to a preset difference threshold, the new indoor unit room grouping result is updated to the currently effective indoor unit room grouping result.

8. The method for room identification of multi-split air conditioning indoor units based on indoor pressure according to claim 7, characterized in that, The step of comparing the new indoor unit room grouping result with the currently effective indoor unit room grouping result, and updating the new indoor unit room grouping result to the currently effective indoor unit room grouping result when the grouping difference is greater than or equal to a preset difference threshold, includes: The new indoor unit room grouping results are compared with the currently effective indoor unit room grouping results for each indoor unit, and the number of indoor units whose room labels have been changed is counted. The group change rate is obtained by dividing the number of indoor units whose room identification has changed by the total number of indoor units in the multi-split central control system. Compare the group change rate with a preset change threshold; If the group change rate is greater than or equal to the preset change threshold, the currently effective indoor unit room grouping result is updated to the new indoor unit room grouping result, and the new indoor unit room grouping result is sent to each indoor unit; If the group change rate is less than the preset change threshold, the currently effective indoor unit room grouping results are maintained.

9. The method for room identification of multi-split air conditioning indoor units based on indoor pressure according to claim 7, characterized in that, After the step of updating the new indoor unit room grouping results to the currently effective indoor unit room grouping results, the method includes: The updated and currently effective indoor unit room grouping results are stored in groups. Each group records the corresponding room identifier and a list of unique indoor unit identifiers belonging to the group. In response to a collaborative control command for a target room, the system searches for a list of unique indoor unit identifiers belonging to the group based on the room identifier corresponding to the target room, and issues unified control commands to all indoor units in the list.

10. A multi-split air conditioner, characterized in that, The multi-split air conditioner includes at least two indoor units, each indoor unit is equipped with a micro differential pressure sensor, and the multi-split air conditioner stores a computer program. When the computer program is executed by a processor, it implements the steps of the multi-split indoor unit room identification method based on indoor pressure as described in any one of claims 1 to 9.

Images