Multi-dimensional linkage control method, air conditioner, server and storage medium
By deploying a pre-trained classification model in the air conditioner, the difference combination and benchmark value are determined based on the target temperature value, which solves the problems of intelligence and storage space in the multi-dimensional linkage control of the air conditioner's electronic control terminal and realizes efficient multi-dimensional linkage control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GD MIDEA AIR CONDITIONING EQUIP CO LTD
- Filing Date
- 2022-07-27
- Publication Date
- 2026-07-24
AI Technical Summary
Existing multi-dimensional linkage control methods in air conditioning electronic control terminals lack intelligence, have high storage space requirements, low model prediction accuracy, and cannot take into account regional differences and real-time performance.
By using a pre-trained classification model, the difference combination and multi-dimensional benchmark values are determined based on the target temperature value, compressing storage space and transforming the multi-index classification problem into a single-index classification problem, thereby improving the model's prediction accuracy and intelligence.
It enables multi-dimensional linkage control of air conditioners, improves intelligence and model prediction accuracy, and reduces storage space requirements.
Smart Images
Figure CN117515799B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of air conditioning technology, and in particular to a multi-dimensional linkage control method, an air conditioner, a server, and a storage medium. Background Technology
[0002] Currently, the main multi-dimensional linkage control methods for air conditioning electronic control terminals are as follows: Frequency statistics are used to obtain the most frequent values of other dimensions under a given dimension, and then a multi-dimensional universal mapping table is constructed. During air conditioning operation, the values of other dimensions are retrieved from the mapping table based on the given dimension value, thereby achieving linkage control. However, this method uses national data for high-frequency values, lacking regional and city-specific differences; the mapping table is based on statistical data over a period of time and is updated semi-annually or annually, resulting in a lack of real-time linkage; and the dimension statistics only consider the frequency of past data occurrences, ignoring the factors influencing the dimensions, thus lacking intelligence.
[0003] Currently, there is also a method of linkage control based on separate modeling of each dimension. However, this method requires the separate modeling and storage of multiple indicators, which necessitates the deployment of multiple models, resulting in high storage space requirements. Furthermore, it does not consider the relationship between multiple indicators, leading to low model prediction accuracy.
[0004] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention
[0005] The main objective of this invention is to provide a multi-dimensional linkage control method, an air conditioner, a server, and a storage medium, aiming to solve the technical problems of low intelligence, high storage space requirements, and low model prediction accuracy in current multi-dimensional linkage control methods for air conditioner electronic control terminals.
[0006] To achieve the above objectives, the present invention provides a multi-dimensional linkage control method applied to an air conditioner, the method comprising the following steps:
[0007] Obtain the target temperature value;
[0008] The difference combination corresponding to the target temperature value is determined by a pre-trained classification model;
[0009] Determine multi-dimensional benchmark values based on the target temperature value;
[0010] The multidimensional control parameters are determined based on the combination of differences and the multidimensional benchmark values; and the corresponding control operations are executed based on the multidimensional control parameters.
[0011] Optionally, determining the difference combination corresponding to the target temperature value using a pre-trained classification model includes:
[0012] The target temperature value is input into a pre-trained classification model to obtain the output category label;
[0013] Based on the category labels, query the interval combination dictionary to obtain multi-dimensional interval identifiers; and
[0014] Based on the multi-dimensional interval identifier, query the interval list corresponding to each dimension to obtain the difference combination corresponding to the target temperature value. The interval list includes multiple adjustment intervals and corresponding standard deviations.
[0015] Furthermore, to achieve the above objectives, this invention also proposes a multi-dimensional linkage control method applied to a server, the method comprising the following steps:
[0016] Obtain the operating data corresponding to the air conditioner;
[0017] The running data is preprocessed to generate training samples;
[0018] The initial classification model is trained based on the training samples to obtain a trained classification model; and
[0019] The trained classification model is sent to the air conditioner so that when the air conditioner obtains a target temperature value, it determines the difference combination corresponding to the target temperature value through the trained classification model, determines a multi-dimensional benchmark value based on the target temperature value, determines multi-dimensional control parameters based on the difference combination and the multi-dimensional benchmark value, and executes the corresponding control operation based on the multi-dimensional control parameters.
[0020] Optionally, the preprocessing of the running data to generate training samples includes:
[0021] The multidimensional mapping value is obtained by querying the multidimensional mapping table based on the operating temperature in the operating data.
[0022] Multiple sets of mapping differences are determined based on the operational data and the multi-dimensional mapping values;
[0023] The target category corresponding to the running data is determined based on the multiple sets of mapping differences; and
[0024] Training samples are generated based on the running data and the target category.
[0025] Optionally, determining the target category corresponding to the running data based on the multiple sets of mapping differences includes:
[0026] Based on the multiple sets of mapping differences, query the interval list corresponding to each dimension to determine the multiple interval identifiers corresponding to the multiple sets of mapping differences; and
[0027] The target category corresponding to the running data is determined by querying the interval combination dictionary based on the multiple sets of interval identifiers.
[0028] Optionally, before obtaining the operating data corresponding to the air conditioner, the method further includes:
[0029] Real-time reported data is obtained based on the model of the air conditioner;
[0030] Based on the real-time reported data, statistically analyze reference values for multiple dimensions corresponding to different set temperatures; and
[0031] A multidimensional mapping table is constructed based on the set temperature and the reference values corresponding to the multiple dimensions.
[0032] Optionally, after calculating the reference values corresponding to multiple dimensions at different set temperatures based on the real-time reported data, the method further includes:
[0033] Determine the distribution information of the difference between the real-time reported data and the reference value under each dimension;
[0034] The maximum variation range for each dimension is determined based on the difference distribution information; and
[0035] Based on the maximum range of variation, construct a list of intervals and a dictionary of interval combinations for each dimension.
[0036] Furthermore, to achieve the above objectives, the present invention also proposes an air conditioner, the air conditioner comprising: a memory, a processor, and a multi-dimensional linkage control program stored in the memory and executable on the processor, the multi-dimensional linkage control program being configured to implement the multi-dimensional linkage control method as described above.
[0037] Furthermore, to achieve the above objectives, the present invention also proposes a server, the server comprising: a memory, a processor, and a multi-dimensional linkage control program stored in the memory and executable on the processor, the multi-dimensional linkage control program being configured to implement the multi-dimensional linkage control method as described above.
[0038] In addition, to achieve the above objectives, the present invention also proposes a storage medium storing a multi-dimensional linkage control program, which, when executed by a processor, implements the multi-dimensional linkage control method as described above.
[0039] In this invention, a pre-trained classification model predicts multi-dimensional difference combinations based on a single temperature value index, transforming a multi-index classification problem into a single-index classification problem. This reduces the storage space for predicted indicators, considers the linkage between multi-dimensional control parameters and temperature values, and improves the model's prediction accuracy. Multi-dimensional benchmark values are determined based on the target temperature value, and multi-dimensional control parameters for air conditioner adjustment are determined based on the difference combinations and multi-dimensional benchmark values, thereby enhancing the intelligence of air conditioner control. Attached Figure Description
[0040] Figure 1 This is a schematic diagram of the structure of a multi-dimensional linkage control device for the hardware operating environment involved in the embodiments of the present invention;
[0041] Figure 2 This is a flowchart illustrating the first embodiment of the multi-dimensional linkage control method of the present invention;
[0042] Figure 3 This is a flowchart illustrating the second embodiment of the multi-dimensional linkage control method of the present invention;
[0043] Figure 4 This is a flowchart illustrating the third embodiment of the multi-dimensional linkage control method of the present invention;
[0044] Figure 5 This is a data processing flowchart for an example of the multi-dimensional linkage control method of the present invention.
[0045] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation
[0046] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.
[0047] Reference Figure 1 , Figure 1 This is a schematic diagram of the structure of a multi-dimensional linkage control device for the hardware operating environment involved in an embodiment of the present invention. The multi-dimensional linkage control device in this embodiment is an air conditioner or a server.
[0048] like Figure 1As shown, the multi-dimensional linkage control device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen, input units such as buttons or a keyboard, and optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be high-speed random access memory (RAM) or stable non-volatile memory (NVM), such as a disk storage device. The memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001.
[0049] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on multi-dimensional linkage control devices, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.
[0050] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and a multi-dimensional linkage control program.
[0051] exist Figure 1 In the multi-dimensional linkage control device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the multi-dimensional linkage control device of the present invention can be set in the multi-dimensional linkage control device, and the multi-dimensional linkage control device calls the multi-dimensional linkage control program stored in the memory 1005 through the processor 1001 and executes the multi-dimensional linkage control method provided in the embodiment of the present invention.
[0052] This invention provides a multi-dimensional linkage control method applied to air conditioners, as described in the following embodiments. Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the multi-dimensional linkage control method of the present invention.
[0053] In this embodiment, the multi-dimensional linkage control method includes the following steps:
[0054] Step S10: Obtain the target temperature value.
[0055] It is understood that the executing entity in this embodiment can be an electronic control unit installed on the air conditioner. The target temperature value is the temperature value set by the user via remote control or the temperature adjustment button on the air conditioner after the air conditioner is turned on.
[0056] Step S20: Determine the difference combination corresponding to the target temperature value using a pre-trained classification model.
[0057] It should be noted that the pre-trained classification model is obtained by the server acquiring the operating data corresponding to the air conditioner, preprocessing the operating data to generate training samples, training the initial classification model based on the training samples to obtain the trained classification model, and then sending the trained classification model to the air conditioner for deployment.
[0058] In practical implementation, the server statistically analyzes usage data from multiple users of a specific air conditioner model, constructs a multi-dimensional mapping table, analyzes the current air conditioner's operating data, determines the mapping values and actual values of other dimensions at various given temperatures, calculates the combined difference based on the mapping values and actual values of multiple dimensions, and trains an initial classification model constructed by a shallow machine learning algorithm based on the corresponding temperature and combined difference of each group. This model is then shared with the current air conditioner. The current air conditioner rewrites the model framework of the classification model trained on the server using C language and deploys it on the control unit. The classification model in this embodiment is based on a shallow machine learning algorithm, which refers to traditional machine learning algorithms other than neural networks, such as logistic regression, perceptrons, and decision trees, which have relatively low complexity. By reconstructing the shallow machine learning algorithm at its underlying level and deploying it on the air conditioner's control unit, the algorithm control logic on the control unit is simplified, the model deployment difficulty on the control unit is reduced, and the model can be deployed to the control unit where resources are relatively insufficient, thereby achieving multi-dimensional linkage intelligent recommendation and intelligent control, and improving the intelligence of the air conditioner.
[0059] It should be understood that the difference combination refers to the combination formed by the difference between the multi-dimensional control parameters predicted based on the user-defined temperature value and the mapped benchmark value. In this embodiment, the multi-dimensional control indicators are compressed into a multi-indicator combination, saving the model calculation and storage resources of the air conditioner.
[0060] Specifically, step S20 includes: inputting the target temperature value into a pre-trained classification model to obtain the output category label; querying an interval combination dictionary based on the category label to obtain a multi-dimensional interval identifier; querying the interval list corresponding to each dimension based on the multi-dimensional interval identifier to obtain the difference combination corresponding to the target temperature value, wherein the interval list includes multiple adjustment intervals and corresponding standard deviations.
[0061] It should be noted that, optionally, in addition to temperature control, the multi-dimensional control in this embodiment also includes any two or more of the following: airflow control, humidity control, cleanliness control, and freshness control. This embodiment uses multi-dimensional control including airflow control, humidity control, cleanliness control, and freshness control as an example for explanation. The classification model in this embodiment outputs category labels. Based on the category labels, a pre-built interval combination dictionary is queried to determine the multi-dimensional interval identifier corresponding to the category label. Based on the multi-dimensional interval labels, a pre-built interval list under each dimension is queried to determine the difference combination predicted based on the target temperature value. In this embodiment, by combining the indicators of each dimension into corresponding categories, the number of categories is limited, thereby reducing the storage space of the prediction indicators.
[0062] The difference combination determination process in this embodiment is illustrated by the following example: the category labels include 1, 2, ..., n; the interval combination dictionary includes {1:[W1,H1,C1,F1], 2:[W2,H2,C2,F2], ..., n:[Wn,Hn,Cn,Fn]}; the interval list includes the air volume variation interval list {W1:x1,W2:x2, ...,Wn:xn} and the humidity variation interval list.
[0063] {H1:y1,H2:y2,…,Hn:yn}, List of Clarity Variation Ranges {C1:z1,C2:z2,…,Cn:zn}, Freshness Variation
[0064] The dynamic range list {F1:s1,F2:s2,…,Fn:sn} is used to predict the target temperature value using a classification model.
[0065] When the category label is 2, the interval combination dictionary is queried to determine the multi-dimensional interval identifier as [W2,H2,C2,F2]. Based on [W2,H2,C2,F2], the interval list under each dimension is queried to obtain the standard deviation values corresponding to multiple dimensions: x2,y2,z2,s2. The difference combination [x2,y2,z2,s2] is constructed, where each difference value can be a positive number, a negative number, or 0.
[0066] Step S30: Determine the multi-dimensional benchmark value based on the target temperature value.
[0067] It should be understood that, based on real-time statistical data reported by each air conditioner model, the most frequent airflow, humidity, cleanliness, and freshness at a given temperature are determined in advance, and a multi-dimensional mapping table is constructed. The multi-dimensional mapping table is then queried based on the target temperature value to determine the multi-dimensional baseline values.
[0068] Step S40: Determine the multidimensional control parameters based on the difference combination and the multidimensional benchmark value.
[0069] It should be noted that the control parameters for that dimension are obtained by adding each difference in the difference combination to the corresponding benchmark value in the multi-dimensional benchmark. For example, if the difference combination is [x2, y2, z2, s2], and the multi-dimensional benchmark values obtained by querying the multi-dimensional mapping table are w, h, c, and f, then the multi-dimensional control parameters determined by the difference combination and the multi-dimensional mapping table are x2+w, y2+h, z2+c, and s2+f.
[0070] Step S50: Execute the corresponding control operation according to the multi-dimensional control parameters.
[0071] In practice, the air conditioner's electronic control unit automatically recommends suitable multi-dimensional control parameters based on the user-set target temperature value. Based on these multi-dimensional control parameters, the air conditioner controls airflow, humidity, cleanliness, and freshness, thus achieving multi-dimensional linkage control of the air conditioner.
[0072] In this embodiment, a pre-trained classification model predicts multi-dimensional difference combinations based on a single temperature value index, transforming the multi-index classification problem into a single-index classification problem. This reduces the storage space for the predicted indexes, considers the linkage between multi-dimensional control parameters and temperature values, and improves the model's prediction accuracy. Multi-dimensional benchmark values are determined based on the target temperature value, and multi-dimensional control parameters for air conditioner adjustment are determined based on the difference combinations and multi-dimensional benchmark values, thereby enhancing the intelligence of air conditioner control.
[0073] This invention also provides a multi-dimensional linkage control method, applied to a server, see reference. Figure 3 , Figure 3 This is a flowchart illustrating the second embodiment of the multi-dimensional linkage control method of the present invention.
[0074] The multi-dimensional linkage control method in this embodiment includes:
[0075] Step S01: Obtain the operating data corresponding to the air conditioner.
[0076] It should be understood that users adjust the temperature, air volume, humidity, cleanliness, and freshness of the air conditioner based on their usage habits. When the air conditioner receives the adjustment signal, it collects operating data, or collects multiple pieces of operating data based on a preset collection frequency, and sends them to the server.
[0077] Step S02: Preprocess the running data to generate training samples.
[0078] It should be noted that the server obtains actual multi-dimensional data (temperature, air volume, humidity, cleanliness, and freshness values) from multiple running data respectively, and looks up the mapped values of other dimensions at this temperature based on a pre-constructed multi-dimensional mapping table. Determine the differences between the actual data of each dimension at a given temperature and the looked-up mapped values to obtain a difference combination, and generate training samples based on different temperatures and the difference combinations corresponding to each temperature.
[0079] Step S03: Train the initial classification model based on the training samples to obtain a trained classification model.
[0080] It can be understood that the initial classification model is trained based on the training data and corresponding labels in the training samples to learn the relationship between the data and the labels, and adjust the model parameters until the prediction result of the initial classification model is consistent with the label, obtaining a trained classification model. The classification model in this embodiment is a model composed of shallow machine learning algorithms, such as a model constructed based on algorithms with relatively low complexity such as logistic regression, perceptron, decision tree, etc.
[0081] Step S04: Send the trained classification model to the air conditioner, so that when the air conditioner obtains the target temperature value, determine the difference combination corresponding to the target temperature value through the trained classification model, determine the multi-dimensional reference value according to the target temperature value, determine the multi-dimensional control parameters according to the difference combination and the multi-dimensional reference value, and perform corresponding control operations according to the multi-dimensional control parameters.
[0082] It should be noted that after the server trains the model based on a large number of training samples, it sends the trained classification model to the air conditioner. The air conditioner rewrites the trained model framework in C language and then deploys it. When it obtains the target temperature value set by the user, it automatically predicts the difference combination, determines the multi-dimensional control parameters, and realizes multi-dimensional linkage control.
[0083] In this embodiment, the server trains the initial classification model according to the training samples obtained by preprocessing the running data of the air conditioner, and then sends the trained classification model to be deployed on the corresponding air conditioner, so that the air conditioner determines the multi-dimensional control parameters according to a single temperature value index, converts the multi-index classification problem into a single-index classification problem, compresses the storage space of the prediction index, realizes the multi-dimensional linkage control of the air conditioner, and improves the intelligence of the air conditioner control. The model training process in this embodiment is set outside the air conditioner, saving the storage space and computing resources of the air conditioner.
[0084] Reference Figure 4 , Figure 4 is a schematic flowchart of the third embodiment of the multi-dimensional linkage control method of the present invention.
[0085] Based on the second embodiment described above, step S02 of the multi-dimensional linkage control method in this embodiment includes:
[0086] Step S021: Query the multidimensional mapping table based on the operating temperature in the operating data to obtain the multidimensional mapping value.
[0087] Understandably, the multidimensional mapping table includes different temperatures and parameters such as air volume, humidity, cleanliness, and freshness. By querying the multidimensional mapping table based on the operating temperature, the corresponding mapping values for temperature, air volume, humidity, cleanliness, and freshness can be determined.
[0088] Step S022: Determine multiple sets of mapping differences based on the running data and the multi-dimensional mapping values.
[0089] It should be noted that the actual values of temperature, air volume, humidity, cleanliness, and freshness adjusted by the user when using the air conditioner are determined based on the operating data. The mapping difference between the actual value and the mapping value under each dimension is determined based on the actual operating data and multi-dimensional mapping values, resulting in multiple sets of mapping difference values.
[0090] Step S023: Determine the target category corresponding to the running data based on the multiple sets of mapping differences.
[0091] Specifically, step S023 includes: querying the interval list corresponding to each dimension according to the multiple sets of mapping differences to determine the multiple sets of interval identifiers corresponding to the multiple sets of mapping differences; querying the interval combination dictionary according to the multiple sets of interval identifiers to determine the target category corresponding to the running data.
[0092] It should be understood that by querying the corresponding interval list under each dimension, the interval identifier corresponding to the calculated mapping difference under each dimension is determined. For example, if the list of air volume variation intervals is {W1(-inf,-15):-20,W2(-15,-5):-10,W3(-5,5):0,W4(5,15):10,W5(15,+inf):20}, the mapping difference of the air volume dimension is matched with multiple intervals in the interval list. Assuming the mapping difference of the air volume dimension is -12, then this mapping difference matches the interval (-15,-5), and the corresponding interval identifier is determined to be W2.
[0093] It should be noted that, assuming that the interval lists obtained by querying each dimension are multiple interval identifiers [W2,H2,C2,F2], and the interval combination dictionary includes 2: [W2,H2,C2,F2], the target category corresponding to the current running data can be determined by querying the interval combination dictionary.
[0094] Step S024: Generate training samples based on the running data and the target category.
[0095] Understandably, multiple training samples are generated by using the operating temperatures corresponding to multiple operating data points as model input data and the target category as the model training label.
[0096] Furthermore, prior to step S01, the method further includes: obtaining real-time reported data according to the model of the air conditioner; calculating reference values corresponding to multiple dimensions under different set temperatures based on the real-time reported data; and constructing a multi-dimensional mapping table based on the set temperature and the reference values corresponding to the multiple dimensions.
[0097] It should be noted that the server receives real-time operational data uploaded by multiple air conditioners of various models. Each air conditioner connected to the server, after starting operation, returns real-time operational data to the server at fixed intervals, including at least the air conditioner's status values under different control dimensions, such as temperature, airflow, humidity, cleanliness, and freshness. Specifically, the server stores the operational data reported by each air conditioner and distinguishes the managed operational data according to the different model identifiers of the air conditioners. For the large amount of real-time reported data corresponding to a specific model of air conditioner, other dimension data are statistically analyzed under different set temperatures to determine reference data for multiple dimensions. Optionally, the average values corresponding to airflow, humidity, cleanliness, and freshness at the set temperature are calculated to obtain reference values for each dimension. Optionally, the most frequently used airflow, humidity, cleanliness, and freshness values at the set temperature are statistically analyzed to obtain reference values for each dimension.
[0098] Furthermore, after calculating the reference values corresponding to multiple dimensions under different set temperatures based on the real-time reported data, the method further includes: determining the difference distribution information between the real-time reported data and the reference values under each dimension; determining the maximum variation range of each dimension based on the difference distribution information; and constructing an interval list and an interval combination dictionary corresponding to each dimension based on the maximum variation range.
[0099] Understandably, using airflow as an example, the difference between the airflow in each real-time reported data point and the reference value is determined. The maximum variation range is then analyzed to identify this range. For instance, if the difference between the reported airflow value and the reference value is concentrated in the range [-20, 20], the maximum variation range is determined to be [-20, 20]. This maximum variation range is then divided into intervals according to a preset width. Each interval is labeled with an interval identifier, and the median value of each interval is determined as its standard deviation. Based on the interval identifier, the divided intervals, and the standard deviation, a list of intervals corresponding to the airflow dimension is constructed. For example, the airflow interval list could be {W1(-inf,-15):-20,W2(-15,-5):-10,W3(-5,5):0,W4(5,15):10,W5(15,+inf):20}, and so on, constructing a list of intervals for each dimension.
[0100] It should be noted that an interval identifier is randomly selected from multiple interval identifiers across various dimensions and combined, for example, [W1,H2,C3,F4], [W2,H3,C4,F1], etc., to generate multiple interval combinations. These combinations are then labeled with category tags to generate an interval combination dictionary. In this embodiment, by setting multi-dimensional numerical variation ranges, the number of combined categories across multiple dimensions is reduced, decreasing the classification difficulty of the classification model. Furthermore, this ensures a relatively balanced data distribution across each dimension, improving the robustness of the algorithm.
[0101] In the specific implementation, refer to Figure 5 , Figure 5 This is a data processing flowchart for an example of the multi-dimensional linkage control method of the present invention. It involves selecting real-time reported data from a specific air conditioner model, statistically analyzing the most frequent values of wind, humidity, cleanliness, and freshness at different temperatures to form a five-dimensional universal mapping table; acquiring air conditioner operating data to obtain the temperature, humidity, wind, cleanliness, and freshness values, and determining the actual humidity, wind, and freshness values at the current temperature; querying the five-dimensional universal mapping table to determine the universally mapped humidity, wind, and freshness values at the current temperature; calculating the difference between the actual value and the mapped value of the humidity, wind, and freshness value dimension at the current temperature; querying the interval list to determine the interval identifier for each dimension; and querying the interval combination dictionary to determine the interval category combination, thereby determining the category label.
[0102] In this embodiment, the data of other dimensions at different temperatures are classified by combining the difference values, which limits the total number of predicted categories, saves the storage space of training data, reduces the classification difficulty of the classification model, and improves the training efficiency of the classification model.
[0103] Furthermore, this embodiment of the invention also proposes a storage medium storing a multi-dimensional linkage control program, which, when executed by a processor, implements the multi-dimensional linkage control method as described above.
[0104] Since this storage medium adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought about by the technical solutions of the above embodiments, which will not be repeated here.
[0105] Furthermore, it should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.
[0106] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.
[0107] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory (ROM) / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.
[0108] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.
Claims
1. A multi-dimensional linkage control method, applied to an air conditioner, characterized in that, The multi-dimensional linkage control method includes: Obtain the target temperature value; The target temperature value is input into a pre-trained classification model to obtain the output category label; Based on the category labels, query the interval combination dictionary to obtain multi-dimensional interval identifiers; and Based on the multi-dimensional interval identifier, query the interval list corresponding to each dimension to obtain the difference combination corresponding to the target temperature value, wherein the interval list includes multiple adjustment intervals and corresponding standard deviations; Determine multi-dimensional benchmark values based on the target temperature value; Determine multidimensional control parameters based on the difference combination and the multidimensional benchmark value; and Execute the corresponding control operation based on the multi-dimensional control parameters.
2. A multi-dimensional linkage control method, applied to a server, characterized in that, The multi-dimensional linkage control method includes: Obtain the operating data corresponding to the air conditioner; The running data is preprocessed to generate training samples; The initial classification model is trained based on the training samples to obtain a trained classification model; and The trained classification model is sent to the air conditioner so that when the air conditioner obtains a target temperature value, it inputs the target temperature value into the trained classification model to obtain the output category label. The category label is then used to query an interval combination dictionary to obtain a multi-dimensional interval identifier. The multi-dimensional interval identifier is then used to query the interval list corresponding to each dimension to obtain the difference combination corresponding to the target temperature value. The interval list includes multiple adjustment intervals and their corresponding standard deviations. A multi-dimensional benchmark value is determined based on the target temperature value. Multi-dimensional control parameters are determined based on the difference combination and the multi-dimensional benchmark value. The corresponding control operation is then executed based on the multi-dimensional control parameters.
3. The multi-dimensional linkage control method as described in claim 2, characterized in that, The preprocessing of the running data to generate training samples includes: The multidimensional mapping value is obtained by querying the multidimensional mapping table based on the operating temperature in the operating data. Multiple sets of mapping differences are determined based on the operational data and the multi-dimensional mapping values; The target category corresponding to the running data is determined based on the multiple sets of mapping differences; and Training samples are generated based on the running data and the target category.
4. The multi-dimensional linkage control method as described in claim 3, characterized in that, The step of determining the target category corresponding to the running data based on the multiple sets of mapping differences includes: Based on the multiple sets of mapping differences, query the interval list corresponding to each dimension to determine the multiple interval identifiers corresponding to the multiple sets of mapping differences; and The target category corresponding to the running data is determined by querying the interval combination dictionary based on the multiple sets of interval identifiers.
5. The multi-dimensional linkage control method as described in claim 4, characterized in that, Before obtaining the operating data corresponding to the air conditioner, the method further includes: Real-time reported data is obtained based on the model of the air conditioner; Based on the real-time reported data, statistically analyze reference values for multiple dimensions corresponding to different set temperatures; and A multidimensional mapping table is constructed based on the set temperature and the reference values corresponding to the multiple dimensions.
6. The multi-dimensional linkage control method as described in claim 5, characterized in that, After calculating the reference values for multiple dimensions corresponding to different set temperatures based on the real-time reported data, the method further includes: Determine the distribution information of the difference between the real-time reported data and the reference value under each dimension; The maximum variation range for each dimension is determined based on the difference distribution information; and Based on the maximum range of variation, construct a list of intervals and a dictionary of interval combinations for each dimension.
7. An air conditioner, characterized in that, The air conditioner includes: a memory, a processor, and a multi-dimensional linkage control program stored in the memory and executable on the processor, the multi-dimensional linkage control program being configured to implement the multi-dimensional linkage control method as described in any one of claims 1.
8. A server, characterized in that, The server includes: a memory, a processor, and a multi-dimensional linkage control program stored in the memory and executable on the processor, the multi-dimensional linkage control program being configured to implement the multi-dimensional linkage control method as described in any one of claims 2 to 6.
9. A storage medium, characterized in that, The storage medium stores a multi-dimensional linkage control program, which, when executed by a processor, implements the multi-dimensional linkage control method as described in any one of claims 1 to 6.