Air conditioner control system and control method

Abstract

The embodiment of the invention provides an air conditioner control system and method, and relates to the technical field of intelligent electric appliances. The air conditioner control system predicts the setting condition of a user through a machine learning model. The air conditioner control system comprises an air conditioner system and an algorithm server. The algorithm server is configured to obtain historical data of the air conditioner, construct a first data set, perform feature analysis on the first data set to obtain a second data set, and perform model training on the second data set through a machine learning algorithm to generate a final model; the historical data comprises user habit historical data; the algorithm server is further configured to obtain real-time data of the air conditioner, construct a third data set and input the third data set into the final model to obtain prediction data; the real-time data comprises real-time data set by a user; the algorithm server is further configured to control the air conditioner according to the prediction data.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent electrical appliances, and in particular to an air-conditioning control system and a control method. Background Art

[0002] With the continuous development of daily household appliance technology, more and more household appliances have entered people's daily life and work. At the same time, due to the improvement of scientific and technological levels in real life, people also have higher demands for the intelligence of household appliances. Therefore, household appliances should also be optimized and improved accordingly.

[0003] Taking air conditioning as an example, currently air conditioning can only be turned on and off at different time periods through time-based switches. However, due to the different ambient temperatures at different time periods every day, users have different requirements for the operating mode and target temperature of the air conditioner. The operating mode and target temperature are set at different time periods every day, while the setting parameters for the corresponding time periods of a certain day and the days before and after are the same. This makes it inconvenient and unintelligent to manually repeat the settings every day. Therefore, how to make the air conditioner intelligently learn the user's air conditioning setting habits, improve the intelligence of air conditioning control, and meet user needs has become an existing technical problem. Summary of the invention

[0004] An embodiment of the present application provides an air-conditioning control system and a control method, wherein the air-conditioning control system predicts a user's settings through a machine learning model.

[0005] In a first aspect, an embodiment of the present application provides an air-conditioning control system, which includes: an air-conditioning system and an algorithm server; the air-conditioning system includes an air conditioner.

[0006] The algorithm server is configured to: obtain historical data of the air conditioner and construct a first data set, perform feature analysis on the first data set to obtain a second data set, perform model training on the second data set through a machine learning algorithm, and generate a final model; the historical data includes user habit historical data.

[0007] The algorithm server is also configured to: obtain real-time data of the air conditioner, construct a third data set, input the third data set into the final model, and obtain prediction data; the real-time data includes user-set real-time data.

[0008] The algorithm server is further configured to control the air conditioner according to the prediction data.

[0009] Based on the above technical solution, some embodiments of the present application provide an air conditioning control system, which constructs the collected historical data of the air conditioner as a first data set based on the Internet of Things technology and cloud platform big data technology, and performs feature analysis on the first data set to obtain a second data set, simulates and trains the second data set through a machine learning algorithm, and finally generates a final model; constructs a third data set based on the Internet of Things technology and cloud platform big data technology with the collected real-time data of the air conditioner, and inputs the third data set into the final model to obtain predicted data; the cloud platform receives the predicted data and controls the air conditioner according to the predicted data. The user's settings are predicted by the machine learning model, which greatly increases the intelligent use experience of the smart central air conditioner users.

[0010] In some embodiments, the algorithm server is further configured to: obtain the initial historical data of the air conditioner; the initial historical data of the air conditioner includes: initial historical data of the outdoor environment, initial historical data of the indoor unit operation, initial historical data of the user habits and initial historical data of the user settings.

[0011] The initial historical data are merged and screened to obtain the indoor unit operation historical data, the user habit historical data and the user setting historical data.

[0012] The initial historical data are merged and screened to obtain the outdoor environment historical data.

[0013] The obtained outdoor environment historical data, the indoor unit operation historical data, the user habit historical data and the user setting historical data are constructed into the first data set.

[0014] The first data set is subjected to feature analysis to screen out data whose feature distribution meets set requirements, thereby obtaining a second data set.

[0015] In some embodiments, the algorithm server is further configured to: perform model training on the second data set through multiple machine learning methods to generate multiple prediction models; and compare the performance of the multiple prediction models to select a prediction model that meets the set conditions as the final model.

[0016] In some embodiments, the algorithm server is further configured to: obtain the indoor unit operation initial historical data and the user-set initial historical data according to the equipment code and indoor unit number of the air conditioner; the algorithm server is further configured to: filter the indoor unit operation initial historical data; and use the data in the indoor unit operation initial historical data that meets a first set condition as the indoor unit operation historical data; the first set condition is that the user-set initial historical data has a stable duration greater than a first time.

[0017] In some embodiments, the algorithm server is further configured to: obtain initial historical data of the outdoor environment in the same area as the user according to the area code; the algorithm server is further configured to: filter the initial historical data of the outdoor environment; and use the data in the initial historical data of the outdoor environment that meets a second set condition as the outdoor environment historical data; the second set condition is that the interval between the set time of the initial historical data set by the user and the collection time of the initial historical data of the outdoor environment is less than or equal to a second time.

[0018] In some embodiments, the plurality of machine learning algorithms include: a decision tree-based distributed gradient algorithm and a random forest algorithm.

[0019] In some embodiments, the algorithm server is further configured to: obtain initial real-time data of the air conditioner; merge and filter the initial real-time data to obtain the real-time operation data of the indoor unit and the real-time user habit data; construct the obtained real-time operation data of the indoor unit and the real-time user habit data into the third data set; input the third data set into the final model to obtain the predicted data; the predicted data includes: user-set temperature data, user-set wind speed data and user-set mode data.

[0020] In some embodiments, the air conditioning control system also includes: a cloud platform; the algorithm server is also configured to send the prediction data to the cloud platform; the cloud platform is configured to detect whether the prediction data needs to be sent to the air conditioner; if so, the cloud platform sends the prediction data to the air conditioner.

[0021] In a second aspect, some embodiments of the present application provide a control method for an air-conditioning control system.

[0022] The control method includes: acquiring historical data of the air conditioner and constructing a first data set, processing the first data set to obtain a second data set, performing model training on the second data set through a machine learning algorithm to generate a final model; the historical data includes user habit historical data; acquiring real-time data of the air conditioner and constructing a third data set, inputting the third data set into the final model to obtain predicted data; the real-time data includes user-set real-time data; and controlling the air conditioner according to the predicted data.

[0023] The beneficial effects of the control method of the air-conditioning control system are the same as those of the above-mentioned air-conditioning control system, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0025] Figure 1 A schematic diagram of the composition of an air conditioning control system provided by an embodiment of the present invention;

[0026] Figure 2 A flow chart of a control method for an air conditioning control system provided by an embodiment of the present invention;

[0027] Figure 3 A flow chart of a control method of an air conditioning control system in a model training phase provided by an embodiment of the present invention;

[0028] Figure 4 A flow chart of another control method of an air conditioning control system in a model training phase provided by an embodiment of the present invention;

[0029] Figure 5 A flow chart of a control method of an air conditioning control system in a model training phase provided by an embodiment of the present invention;

[0030] Figure 6 A flow chart of a control method of an air conditioning control system in a model training phase provided by an embodiment of the present invention;

[0031] Figure 7 A flow chart of a control method of an air conditioning control system in a model training phase provided by an embodiment of the present invention;

[0032] Figure 8 A schematic diagram of outdoor current temperature characteristic analysis provided by an embodiment of the present invention;

[0033] Fig. 9 A schematic diagram of outdoor current humidity characteristic analysis provided by an embodiment of the present invention;

[0034] Fig.10 A schematic diagram of outdoor current wind force characteristic analysis provided by an embodiment of the present invention;

[0035] Fig.11 A schematic diagram of indoor comfort characteristics analysis provided by an embodiment of the present invention;

[0036] Fig.12 A schematic diagram of outdoor comfort characteristics analysis provided by an embodiment of the present invention;

[0037] Fig.13 A schematic diagram of indoor and outdoor temperature difference characteristic analysis provided by an embodiment of the present invention;

[0038] Fig.14A schematic diagram of a machine learning algorithm model provided by an embodiment of the present invention;

[0039] Fig.15 A schematic diagram of another machine learning algorithm model provided by an embodiment of the present invention;

[0040] Fig.16 A schematic diagram of four seasons prediction accuracy provided by an embodiment of the present invention;

[0041] Fig.17 A flow chart of a control method of an air conditioning control system in a setting prediction stage provided by an embodiment of the present invention;

[0042] Fig.18 A schematic diagram of a fully connected neural network model provided by an embodiment of the present invention;

[0043] Fig.19 A flow chart of a control method of an air-conditioning control system in a control phase provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0046] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0047] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. In addition, when describing the pipeline, the "connected" and "connection" used in the present invention have the meaning of conduction. The specific meaning needs to be understood in conjunction with the context.

[0048] In the embodiments of the present invention, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0049] As described in the background technology, with the continuous development of daily household appliance technology, more and more household appliances have entered people's daily life and work. At the same time, due to the improvement of scientific and technological levels in real life, people also have corresponding higher demands for the intelligence of household appliances. For this reason, household appliances should also be optimized and improved accordingly.

[0050] Taking air conditioning as an example, currently air conditioning can only be turned on and off at different time periods through time-based switching. However, due to the different ambient temperatures at different time periods every day, users have different requirements for the operating mode and target temperature of the air conditioner. The operating mode and target temperature are set at different time periods every day, while the setting parameters for the corresponding time periods of a certain day and the days before and after are the same. Currently, users mainly control the air conditioner by operating a wire controller or a mobile phone APP. This manual repeated setting every day is very inconvenient and unintelligent. Therefore, how to make the air conditioner intelligently learn the user's air conditioning setting habits, improve the intelligence of air conditioning control, and meet user needs has become an existing technical problem.

[0051] In order to solve this technical problem, it is necessary to propose an intelligent air-conditioning control method to realize the automatic and intelligent regulation process of air-conditioning to meet the user's requirements for the operating mode and target temperature of the air-conditioning at different time periods every day.

[0052] Based on this, the present application provides an air conditioning control system. Figure 1 As shown, the air conditioning control system 1000 includes: an algorithm server 100 and an air conditioning system 200, the air conditioning system 200 includes an air conditioner 50; the air conditioner 50 includes: a plurality of indoor units 20 and a plurality of outdoor units 30 (only one indoor unit and one outdoor unit are shown in the figure).

[0053] Reference Figure 2 The algorithm server 100 is configured as follows:

[0054] S1. Obtain historical data of the air conditioner and construct a first data set, perform feature analysis on the first data set to obtain a second data set, perform model training on the second data set through a machine learning algorithm, and generate a final model.

[0055] Among them, historical data includes user habit historical data.

[0056] S2. Acquire real-time data of the air conditioner, construct a third data set, and input the third data set into the final model to obtain prediction data.

[0057] The real-time data includes real-time data set by the user.

[0058] S3. Control the air conditioner according to the predicted data.

[0059] It should be noted that the above-mentioned air conditioning control is suitable for home multi-connected network users.

[0060] Some embodiments of the present application provide an air conditioning control system, which constructs the collected historical data of the air conditioner as a first data set, performs feature analysis on the first data set to obtain a second data set, performs simulation training on the second data set through a machine learning algorithm, and finally generates a final model; constructs a third data set based on the Internet of Things technology and cloud platform big data technology with the collected real-time data of the air conditioner, and inputs the third data set into the final model to obtain predicted data; the cloud platform receives the predicted data and controls the air conditioner according to the predicted data. Predicting the user's settings through a machine learning model greatly increases the intelligent use experience of the smart central air conditioner user.

[0061] like Figure 3 As shown, the algorithm server 100 is also configured as:

[0062] S11. Obtaining initial historical data of the air conditioner.

[0063] The initial historical data of the air conditioner includes: initial historical data of the outdoor environment, initial historical data of the indoor unit operation, initial historical data of user habits, and initial historical data of user settings. The initial historical data of the air conditioner is collected based on the Internet of Things technology and cloud platform big data technology. The algorithm server can store the initial historical data of the air conditioner.

[0064] S12: Merge and filter the initial historical data to obtain indoor unit operation historical data, user habit historical data, and user setting historical data.

[0065] S13: Merge and filter the initial historical data to obtain outdoor environment historical data.

[0066] S14: construct the obtained outdoor environment historical data, indoor unit operation historical data, user habit historical data and user setting historical data into a first data set.

[0067] S15. Perform feature analysis on the first data set, filter out data whose feature distribution meets set requirements, and obtain a second data set.

[0068] In some embodiments of the present application, the above-mentioned air-conditioning control system is merged and screened to make the model application more accurate.

[0069] Reference Figure 3 , the algorithm server 100 is further configured as:

[0070] S16. Perform model training on the second data set using multiple machine learning methods to generate multiple prediction models.

[0071] S17. By comparing the performance of multiple prediction models, a prediction model that meets the set conditions is selected as the final model.

[0072] In some embodiments of the present application, the air-conditioning control system analyzes the user's usage habits through big data, and obtains a prediction model based on the user's usage habits that is closer to the user, thereby making the model application more accurate.

[0073] The initial historical data types of the above air conditioners can be found in Table 1 below.

[0074] Table 1 Parameters used by the user to set the prediction function

[0075]

[0076]

[0077] The outdoor environment initial historical data, the indoor unit operation initial historical data, the user habit initial historical data and the user setting initial historical data are equivalent to the outdoor environment parameters, indoor unit operation parameters, user habit parameters and user setting parameters mentioned above.

[0078] The parameters in Table 1 include direct parameters and construction parameters, and the construction parameters are constructed from direct parameters.

[0079] In some embodiments, the variable outdoor comfort outDI in the outdoor environmental parameters in Table 1 includes two direct parameters: the current outdoor temperature l1 and the current outdoor humidity l2, and the calculation formula of the comfort index DI is: DI = T-(0.55-0.55f)*(T-58), wherein T refers to the conversion of the current outdoor temperature l1 from Celsius to Fahrenheit; f refers to the current outdoor humidity l2.

[0080] In some embodiments, the variable user temperature tendency rec_temp_range_mean in the user habit parameters of Table 1 includes: set temperature iu31Temp, season number season, user ID userId and timestamp roundTime; user temperature tendency rec_temp_range_mean refers to the value of the user's set temperature in the past week, the past month and the current season weighted according to a certain weighting method, that is, rec_temp_range_mean = Average('iu31Temp')by('time_period','weeks','iuId').

[0081] It should be noted that Average('iu31Temp')by('time_period','weeks','iuId') is the average value of the user-set temperature within a week; if the above 'weeks' is changed to 'months', it is the average value of the user-set temperature within a month; if the above 'weeks' is changed to 'season', it is the average value of the user-set temperature in the current season.

[0082] Similarly, Average('iu30Wind')by('time_period','weeks','iuId') is the average wind speed set by the user within a week; if the above 'weeks' is changed to 'months', it is the average wind speed set by the user within a month; if the above 'weeks' is changed to 'season', it is the average wind speed set by the user in the current season.

[0083] mostly_select('iu29Mode')by('time_period','weeks','iuId') is the mode that the user has set in the past week. For example, if the user has set the cooling mode and dehumidification mode in the past week, and the proportion of the cooling mode is higher than that of the dehumidification mode, the user mode tends to be the cooling mode.

[0084] Among them, the timestamp roundTime is the Unix timestamp, which is defined as the total number of seconds from 00:00:00 Greenwich Mean Time on January 1, 1970 to the present. Then the timestamp will not change with the change of time zone, so no matter which time zone you are in, the timestamp is consistent. In this way, we can avoid the errors caused by repeated modification of data. In order to achieve cross-platform, we can also use the method of recording timestamps to achieve cross-platform when recording time in the application system.

[0085] Among them, the variable outdoor comfort outDI in the above outdoor environmental parameters is constructed by the outdoor environmental parameters themselves, while the variable user temperature tendency rec_temp_range_mean in the user habit parameters will use the variable season number season in the outdoor environmental parameters and the variable set temperature iu31Temp in the user setting parameters.

[0086] For the above step S11, the initial historical data of the air conditioner is obtained; the initial historical data of the air conditioner has a large amount of raw data and many categories, but little effective information. At the same time, it is necessary to mine the user's usage habits to achieve the prediction of the user's set parameters. The libraries used in processing data can be referred to in Table 2 below:

[0087] Table 2 Libraries used to process data and their descriptions

[0088]

[0089] Among them, the acquisition and screening of the initial historical data of the above air conditioner can be referred to Figure 4 , Figure 5 and Figure 6 .

[0090] The algorithm server 100 is also configured as (refer to Figure 4 and Figure 5 ):

[0091] S111. Acquire the indoor unit operation initial historical data and the user set initial historical data according to the equipment code and indoor unit number of the air conditioner.

[0092] S121. Filter the initial historical data of the indoor unit operation.

[0093] S122: taking the data satisfying the first setting condition in the initial historical data of the indoor unit operation as the historical data of the indoor unit operation.

[0094] The first setting condition is that the user sets the initial historical data stability duration T W1 Greater than the first time T1, that is, T W1 >T1.

[0095] In some embodiments, the first time T1 is 10 minutes. For example, referring to Table 1, the set temperature iu31Temp in the user setting data is 25°C. If the set temperature iu31Temp changes to another temperature within 10 minutes, the set temperature iu31Temp is invalid data and will be filtered out; if the set temperature iu31Temp is still 25°C after 10 minutes, the set temperature iu31Temp is valid data and will be retained.

[0096] The algorithm server 100 is also configured as (refer to Figure 4 and Figure 6 ):

[0097] S112. Acquire initial historical data of the outdoor environment in the same area as the user according to the area code.

[0098] S131. Screening initial historical data of the outdoor environment.

[0099] S132: Use the data in the outdoor environment initial historical data that meets the second set condition as the outdoor environment historical data.

[0100] The second setting condition is the interval T between the setting time of the initial historical data set by the user and the collection time of the initial historical data of the outdoor environment. W2 Less than or equal to the second time T2, that is, T W2 ≤T2.

[0101] In some embodiments, the second time is 75 minutes. For example, if the user adjusts the air conditioner at 19:00, if the outdoor environment data of the user in the same area obtained by the air conditioning control system before 17:45 is invalid data, the data will be filtered out; the outdoor environment data of the user in the same area obtained by the air conditioning control system after 17:45 is valid data, that is, the data in the time period of 17:45 to 19:00 will be retained.

[0102] The air-conditioning control system provided in some embodiments of the present application merges and filters the initial historical data of the air-conditioning, thereby eliminating invalid operation data of some users and unstable data generated by blind operation of users, thereby ensuring the validity of the obtained indoor unit operation history data, user habit history data and outdoor environment history data.

[0103] For the real-time data of indoor unit operation, real-time user setting data and real-time data of outdoor environment obtained above, some data have poor feature distribution and low discrimination, which makes it difficult to analyze the effectiveness and is not conducive to algorithm learning. Figure 7 As shown, S15, feature analysis is performed on the first data set, and data with feature distribution that meets the set requirements is screened out to obtain a second data set; this step also includes the following steps:

[0104] S151. Visually analyze the feature distribution in the first data set and delete features with poor distribution.

[0105] The visual analysis of the feature distribution of the second data set can be found in Figure 8 to Figure 13 .in, Figure 8 It is a visual analysis chart of the current outdoor temperature l1; Fig. 9 It is a visual analysis chart of the current outdoor humidity l2; Fig.10 It is a visual analysis diagram of the current outdoor wind force l3; Fig.11 It is a visual analysis chart of indoor comfort inDI; Fig.12 It is a visual analysis chart of outdoor comfort outDI; Fig.13 It is a visual analysis chart of the indoor and outdoor temperature difference temp_range.

[0106] S152: Construct the remaining features of the first data set into a second data set.

[0107] The remaining features of the first data set can be found in Table 3 below.

[0108] Table 3. Features constructed in the prediction stage

[0109]

[0110]

[0111] By constructing features that are more distinctive and can better reflect user personality preferences, we can avoid directly using user ID information, improve the learning effect of the machine learning algorithm, and solve the cold start problem of new users joining.

[0112] In some embodiments, the multiple machine learning algorithms include: a distributed gradient algorithm based on a decision tree and a random forest algorithm.

[0113] like Fig.14 As shown in the figure, the distributed gradient algorithm based on decision tree (Light Gradient Boosting Machine) uses multiple weak classifiers (decision trees) for iterative training to obtain the optimal model. This model integrating multiple weak classifiers has the advantages of good training effect and not easy to overfit. It is fast and efficient, low memory usage, high accuracy, supports large-scale data parallel processing, and is suitable for large-scale data.

[0114] The model uses a depth-limited leaf-wise growth algorithm, that is, each time from all the current leaves, find a leaf with the largest split gain to split (such as Fig.14 ) until the set depth limit is reached. Compared with the GBDT algorithm, this algorithm can reduce the error caused by traversing nodes and selecting features by layer to obtain better accuracy. And because LightGBM has a maximum depth limit, it can prevent the growth of deeper decision trees and overfitting while ensuring high efficiency.

[0115] like Fig.15As shown in the figure, Random Forest is an algorithm that integrates multiple trees through the Bagging idea of ​​ensemble learning. It builds multiple decision trees and combines them to get a more accurate and stable model. It is a combination of the Bagging idea and random selection features. Random Forest constructs multiple decision trees. When a sample needs to be predicted, the prediction results of each tree in the forest for the sample are counted, and then the final result is selected from these prediction results through voting.

[0116] Randomness is reflected in two aspects: one is the random selection of features, and the other is the random selection of samples, so that each tree in the forest has both similarities and differences.

[0117] Among them, each tree in the random forest grows as follows:

[0118] (1) N represents the number of training cases (samples), and M represents the number of features.

[0119] (2) Input the number of features m, which is used to determine the decision result of a node on the decision tree; m should be much smaller than M.

[0120] (3) Sample N times with replacement from N training cases (samples) to form a training set (i.e. bootstrap sampling), and use the unsampled cases (samples) to make predictions and evaluate their errors.

[0121] (4) For each node, m features are randomly selected. The decision of each node in the decision tree is determined based on these m features. Based on these m features, the best splitting method is calculated.

[0122] (5) Each tree will grow completely without pruning, which may be used after building a normal tree classifier.

[0123] It should be noted that the larger the value of m, the higher the correlation in (3) and the stronger the splitting ability in (4), so m is an important parameter in random forests.

[0124] Reference Fig.15 According to the above steps, the first decision tree, the second decision tree, and the Nth decision tree are constructed. All decision trees constitute a random forest, and each decision tree corresponds to a different category. Finally, the random forest integrates all classification voting results and designates the category with the most votes as the final result.

[0125] Random forests have powerful learning capabilities, are not prone to overfitting, have excellent performance, can handle very high-dimensional data, and do not require feature selection.

[0126] like Fig.17As shown, the algorithm server 100 is also configured as:

[0127] S21. Obtain initial real-time data of the air conditioner.

[0128] S22: Merge and filter the initial real-time data to obtain indoor unit operation real-time data and real-time user habit data.

[0129] S23: construct the obtained indoor unit operation real-time data and real-time user habit data into a third data set.

[0130] S24. Input the third data set into the final model to obtain predicted data.

[0131] In some embodiments, the rules for merging and filtering the initial real-time data are the same as the rules of step S12 above, and are not described in detail here.

[0132] The prediction data includes: user-set temperature data, user-set wind speed data and user-set mode data.

[0133] In step S24, the third data set is input into the final model to obtain predicted data. The final model is Fig.14 The LightGBM model in Fig.15 The model prediction accuracy was verified by exporting 1,000 sets of data from the Central China region in one year from the database, and the overall prediction accuracy of the model was shown in Table 4:

[0134] Table 4 Overall accuracy of model verification

[0135] Model Temperature prediction accuracy Wind speed prediction accuracy Model prediction accuracy Random Forest 95.26％ 96.51％ 99.82％ LightGBM 93.27％ 94.76％ 96.89％

[0136] At the same time, considering that the usage of air conditioners in different seasons and the size of the quantity set are different, the prediction accuracy in summer may be much higher than that in spring, autumn and winter, which may lead to an inflated overall prediction accuracy. Therefore, we consider continuing to calculate the accuracy of the user-set prediction for each season. The prediction accuracy for the four seasons is referenced Fig.16 .

[0137] In summary, compared with the LightGBM model, the Random Forest model has higher prediction accuracy.

[0138] In some embodiments, the prediction model may be a fully connected neural network model.

[0139] Reference Fig.18The fully connected neural network is a network with multiple layers of neurons, and there is a nonlinear activation function between the layers. The outdoor environment history data, indoor unit operation history data, user habit history data and user setting history data are sequentially input into the input layer of the neural network; among them, X1, X2, and so on. n Represents the input data, i1, i2, until i n Represents the input layer.

[0140] Then, the predicted value of the target parameter is obtained in the output layer by calculating the activation function in the multi-layer hidden layer, and the loss between the predicted data value and the true value is calculated by the loss function, and then back-propagated to the neural network through the optimizer so that it can correct the weight W between the two adjacent layers of neurons. Finally, through multiple rounds of iterations, the accuracy is improved to the ideal level. Then it is saved to the algorithm server for later use.

[0141] It should be noted that neurons are Fig.18 The circles in the input layer are i1, i2, and so on. n ; The circle of the hidden layer is H1 (1) 、H2 (1) 、H3 (1) Until H m (1) and H1 (s) 、H2 (s) Until H m (s) ; The hidden layer includes multiple hidden layers, such as H1 (1) 、H2 (1) 、H3 (1) Until H m (1) Belongs to the first hidden layer, H1 (s) 、H2 (s) Until H m (s) Belongs to the Sth hidden layer; the circles of the output layer are O1 to O t .

[0142] There is a weight W between two adjacent neurons. The previous neuron uses different weights W to derive the next neuron, and finally obtain the predicted data. The accuracy of the model is judged according to the closeness between the predicted data value and the expected value. The closer the predicted data value is to the expected value, the higher the accuracy of the model. Conversely, the lower the accuracy of the model.

[0143] When the weight W is modified, the next neuron will change, and so on. Through multiple modification iterations, the value of the predicted data becomes closer and closer to the expected value, thereby improving the accuracy of the model.

[0144] like Fig.19 As shown, the air conditioning control system 1000 also includes: a cloud platform 300 (refer to Figure 1 ).

[0145] S31. Transmit the prediction data to the cloud platform.

[0146] S32, the cloud platform detects whether the prediction data needs to be sent to the air conditioner; if so, proceed to the next step S33, if not, go to step S34.

[0147] S33. The cloud platform sends the prediction data to the air conditioner.

[0148] S34. The cloud platform does not take any action.

[0149] S35: The air conditioner receives the prediction data and performs automatic adjustment.

[0150] Some embodiments of the present application provide a control method for an air conditioning control system, and the control method for the air conditioning control system is as follows:

[0151] S1. Obtain historical data of the air conditioner and construct a first data set, process the first data set to obtain a second data set, perform model training on the second data set through a machine learning algorithm, and generate a final model; the historical data includes user habit historical data;

[0152] S2, acquiring real-time data of the air conditioner, and constructing a third data set, and inputting the third data set into the final model to obtain prediction data; the real-time data includes real-time data set by the user;

[0153] S3. Control the air conditioner according to the prediction data.

[0154] The control method of this air-conditioning control system can predict the user's settings in real time through machine learning algorithms, thereby greatly enhancing the intelligent usage experience of smart central air-conditioning users.

[0155] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An air conditioning control system, characterized in that: include: Air conditioning system and algorithm server; the air conditioning system includes air conditioning; The algorithm server is configured to: obtain the historical data of the air conditioner and construct a first data set, perform feature analysis on the first data set to obtain a second data set, perform model training on the second data set through multiple machine learning algorithms, and generate a final model; the historical data includes user habit historical data; The algorithm server is further configured to: obtain real-time data of the air conditioner, construct a third data set, and input the third data set into the final model to obtain prediction data; Wherein, the real-time data includes real-time data set by the user; The algorithm server is further configured to control the air conditioner according to the prediction data.

2. The air conditioning control system according to claim 1, characterized in that: The algorithm server is further configured to: Acquire the initial historical data of the air conditioner; the initial historical data of the air conditioner includes: initial historical data of outdoor environment, initial historical data of indoor unit operation, initial historical data of user habits and initial historical data of user settings; Merging and filtering the initial historical data to obtain the indoor unit operation historical data, the user habit historical data and the user setting historical data; Merging and screening the initial historical data to obtain the outdoor environment historical data; constructing the obtained outdoor environment historical data, the indoor unit operation historical data, the user habit historical data and the user setting historical data into the first data set; The first data set is subjected to feature analysis to screen out data whose feature distribution meets set requirements, thereby obtaining a second data set.

3. The air conditioning control system according to claim 1, characterized in that: The algorithm server is further configured to: Performing model training on the second data set by using a plurality of the machine learning methods to generate a plurality of prediction models; By comparing the performances of the multiple prediction models, a prediction model that meets the set conditions is selected as the final model.

4. The air conditioning control system according to claim 2, characterized in that: The algorithm server is further configured to: According to the equipment code and indoor unit number of the air conditioner, acquiring the indoor unit operation initial historical data and the user set initial historical data; The algorithm server is also configured to: filter the initial historical data of the indoor unit operation; use the data in the initial historical data of the indoor unit operation that meets a first set condition as the historical data of the indoor unit operation; the first set condition is that the user sets the initial historical data to have a stable duration greater than a first time.

5. The air conditioning control system according to claim 2, characterized in that: The algorithm server is further configured to: According to the area code, obtain the initial historical data of the outdoor environment in the same area as the user; The algorithm server is also configured to: filter the initial historical data of the outdoor environment; and use the data in the initial historical data of the outdoor environment that meets a second set condition as the outdoor environment historical data; the second set condition is that the interval between the set time of the initial historical data set by the user and the collection time of the initial historical data of the outdoor environment is less than or equal to a second time.

6. The air conditioning control system according to claim 1, characterized in that: The multiple machine learning algorithms include: a distributed gradient algorithm based on a decision tree and a random forest algorithm.

7. The air conditioning control system according to claim 1, characterized in that: The algorithm server is further configured to: Acquiring initial real-time data of the air conditioner; Merging and filtering the initial real-time data to obtain the indoor unit operation real-time data and the real-time user habit data; constructing the obtained indoor unit operation real-time data and the real-time user habit data into the third data set; Inputting the third data set into the final model to obtain the predicted data; The prediction data includes: user-set temperature data, user-set wind speed data and user-set mode data.

8. The air conditioning control system according to claim 1, characterized in that: The air conditioning control system further includes: a cloud platform; The algorithm server is also configured to transmit the prediction data to the cloud platform; The cloud platform is configured to detect whether the prediction data needs to be sent to the air conditioner; if so, the cloud platform sends the prediction data to the air conditioner.

9. A control method for an air conditioning control system, characterized in that: The control method includes: Acquire historical data of the air conditioner and construct a first data set, process the first data set to obtain a second data set, perform model training on the second data set through a machine learning algorithm, and generate a final model; the historical data includes user habit historical data; Acquire real-time data of the air conditioner, construct a third data set, and input the third data set into the final model to obtain prediction data; the real-time data includes real-time data set by the user; The air conditioner is controlled according to the prediction data.

Images