Intelligent control method and system of temperature controller and intelligent electrical appliance

Abstract

The present invention relates to an intelligent control method, system and intelligent appliance for a thermostat, and belongs to the field of thermostat control technology. Through the collected behavior data, the relationship between input variables and output variables is analyzed by using data mining technology to obtain user behavior rules; the similarity of user behavior rules is calculated through multiple dimensions, and a control mode similar to the current user is matched, and a control suggestion is generated based on the matching result; each generated control suggestion is verified according to the user's geographic location information, and when the verification passes, the control suggestion is used to extract the start time point and setting parameters. The present invention mines the user's control behavior rules, matches the user's control behavior in the thermostat control mode library, generates control suggestions, and verifies the pre-start control parameters in real time through the user's activity mode, thereby improving the accuracy and intelligence level of pre-start.

Description

Technical Field

[0001] The present invention belongs to the technical field of thermostat control, and in particular relates to an intelligent control method and system for a thermostat and an intelligent electrical appliance. Background Art

[0002] Thermostats are control devices that need to control temperature in temperature-adjusting places. For common thermostats, they can not only complete temperature sampling, but also have built-in relays to directly control the working state of the fan disk corresponding to the cooling condition or the water valve corresponding to the heating condition. Therefore, users can control the temperature in the temperature-adjusting place through the thermostat.

[0003] However, traditional thermostats have a large time lag in pre-start, making it difficult to heat or cool in advance according to specific needs, thus affecting the user's comfort experience and energy efficiency. Or the accuracy of pre-start is poor, resulting in untimely opening or closing of control devices, resulting in energy waste. Therefore, it is urgent to propose an intelligent control method, system and intelligent appliance for thermostats, which can improve the intelligence level of thermostats, optimize the startup timing and improve user satisfaction by obtaining pre-start parameters. Summary of the invention

[0004] In order to solve the above-mentioned problems existing in the prior art, the present invention provides an intelligent control method, system and intelligent appliance of a thermostat, which mines the control behavior rules of users, matches them with the user control behaviors in the thermostat control pattern library to generate control suggestions, and obtains pre-start control parameters through real-time verification through the user's activity pattern, thereby improving the accuracy and intelligence level of pre-start.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A first aspect of the present disclosure provides an intelligent control method for a thermostat, comprising the following steps:

[0007] Mining behavioral rules: Through the collected behavioral data, data mining technology is used to analyze the relationship between input variables and output variables to obtain user behavior rules;

[0008] Associated control mode: Calculate the similarity of user behavior rules through multiple dimensions, match the control mode similar to the current user, and generate control suggestions based on the matching results;

[0009] Obtain pre-start parameters: Verify each generated control suggestion based on the user's geographic location information. If the verification passes, use the control suggestion to extract the start time point and set parameters.

[0010] The mining behavior rules include the following steps:

[0011] Select mining variables: select input variables and output variables through the collected behavioral data, where the input variables include set temperature, actual temperature, thermostat operation time, operation time period, date type and area external temperature, and the output variables include thermostat usage status;

[0012] Variable relationship mining: Discretize the input variables, convert the data into transaction format, group the data by date type and build a transaction database; scan the transaction database in groups using the association rule mining algorithm to identify frequent item sets of different date types;

[0013] Obtaining behavioral rules: Generate all rules from frequent item sets of different date types. During the continuous collection and update of behavioral data, select grid point parameters at different stages through coverage to screen association rules.

[0014] Furthermore, the establishment of the grid point parameters includes:

[0015] By setting the minimum support and minimum confidence, the initial parameter range is set, a series of minimum support and minimum confidence are selected, and all combinations of minimum support and minimum confidence are used to construct a two-dimensional grid, where each grid point represents a pair of parameter combinations.

[0016] Furthermore, the association control mode comprises the following steps:

[0017] Establish a pattern library: store all users' behavior rules in the thermostat control pattern library and update and maintain it regularly;

[0018] Matching control mode: Calculate the similarity of user behavior patterns through multiple dimensions, and select several users with the highest similarity to the current user from the thermostat control mode library based on the similarity calculation results;

[0019] The dimensions are set according to the mining variables in the behavior rule mining, including the temperature setting dimension, the time mode dimension, the external environment dimension, and the operation mode dimension;

[0020] Generate control suggestions: Extract the behavior rules of the matched users, including the thermostat startup conditions, and generate thermostat control suggestions for the current user;

[0021] The thermostat start-up conditions include time point and environmental value.

[0022] Further, the obtaining of pre-start parameters comprises the following steps:

[0023] Obtaining distance judgment value: Mining the user's home route through the user's daily activity trajectory, and then searching for the distance judgment value based on route features;

[0024] Verify control suggestions: Calculate the average time for users to return home through the first route based on the selected decision value. Use the average time as the time difference compensation start time point based on the thermostat start time point of a control suggestion as the decision time point. Obtain the user's location information at the decision time point, set a distance threshold, and enable the control suggestion when the difference between the user's location and the home value and the decision value is within the distance threshold.

[0025] Generate pre-start parameters: Based on the enabled control suggestions, extract the start time point and set parameters, generate control commands and send them to the thermostat to execute actions.

[0026] Further, the obtaining of the distance judgment value comprises the following steps:

[0027] Data collection: The geographic location information of each user is obtained through the IoT devices of user members, and each acquired location information is bound to a timestamp to construct a time series trajectory;

[0028] At the same time, the user's home address coordinates are pre-set;

[0029] Select the decision value: Based on the collected data, draw the user trajectory during the peak time period of going home, find the route with the largest proportion among all trajectories as the first route, and the remaining trajectories as the second route. Search the reverse turning trajectory point closest to home from the second route as the decision point, and calculate the distance between the decision point and the home address as the decision value.

[0030] Furthermore, the intelligent control method further comprises the steps of:

[0031] Collecting behavioral data: Collecting behavioral data on a timeline based on user control requirements, including the temperature parameters and operating time set by the user for the controlled object and the real-time temperature of the spatial area, while obtaining the geographic location information of all user members.

[0032] Furthermore, the intelligent control method further comprises the steps of:

[0033] Mode feedback correction: Collect user's control behavior data at home to update the thermostat control mode library, regularly associate control modes to supplement the current control mode, and gradually screen out high-frequency control suggestions for optimization.

[0034] A second aspect of the present disclosure provides an intelligent control system for a thermostat, applying an intelligent control method for a thermostat as described above, including an IoT sensing module, an intelligent control module, and a pattern library cloud platform, wherein the IoT sensing module, the intelligent control module, and the pattern library cloud platform are sequentially connected in communication;

[0035] The IoT sensing module is used to collect user behavior data, including sensors set in the control environment and IoT devices worn by users, to obtain temperature parameters, thermostat operating status and user member location information in real time;

[0036] The data collection strategy of the IoT sensing module includes:

[0037] Thermostat intelligent control learning stage: Set the first collection frequency to collect user behavior data around the clock;

[0038] Thermostat intelligent control optimization stage: Set the second collection frequency to collect user behavior data in stages.

[0039] As a preferred technical solution of the present invention, the intelligent control module is used to mine the user's behavior rules through the user behavior data, and to search for the control mode similar to the current user through the pattern library cloud platform to generate control suggestions;

[0040] At the same time, the user's activity mode is obtained according to the user's location information, so as to obtain the user's location based on the start time of the control suggestion to verify the feasibility of the control suggestion;

[0041] The intelligent control module is also used to modify the control suggestions for high-frequency activation through the behavior data collected during the intelligent control optimization phase of the thermostat, and adjust the parameters and duration of the thermostat operation;

[0042] The pattern library cloud platform is used to establish an information-sharing thermostat control pattern library, store the mined user behavior rules in the thermostat control pattern library, and update and maintain the thermostat control pattern library by collecting the behavior rules of different types of users.

[0043] A third aspect of the present disclosure provides an intelligent appliance, comprising a thermostat, a memory, a processor, and a control program of the intelligent appliance stored in the memory and executable on the processor, wherein:

[0044] The thermostat is used to adjust the on / off of the cooling or heating circuit of the smart appliance;

[0045] When the control program of the intelligent electrical appliance is executed by the processor, the steps of the intelligent control method of the thermostat as described above are implemented.

[0046] The beneficial effects of the present invention are:

[0047] The present invention mines the control behavior rules of the current user, matches them with the big data in the thermostat control mode library to obtain more control behaviors similar to the user, then generates control suggestions based on the matched control behaviors, and verifies them in real time through the current user's activity mode. Under the joint action of these two parts of operation, the pre-start control parameters are obtained, which improves the accuracy of the pre-start parameters and solves the hysteresis problem of the thermostat startup; at the same time, through the correlation mining of user behavior data, the user behavior analysis is gradually refined, and the behavior rules of each stage are improved, which provides support for matching similar control behaviors and enriching control suggestions; and the user's activity trajectory is analyzed to search for the best decision point to verify the feasibility of the control suggestion pre-start, thereby improving the automation level of the thermostat and the accuracy of intelligent control.

[0048] The present invention realizes accurate pre-start control, so that the thermostat can realize the control of the operating state on demand and in a timely manner, thereby reducing energy consumption, improving energy utilization efficiency, and reducing operating costs. In addition, the startup control of the thermostat is carried out in a predictive manner, which improves the response speed and dynamic adjustment ability of the temperature control system, adapts to changes in different environmental conditions and user needs, and optimizes the control mode through analysis and learning of behavioral data, thereby improving the long-term performance and stability of the thermostat. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0050] Figure 1 A schematic diagram of the steps of an intelligent control method for a temperature controller provided in Embodiment 1 of the present invention;

[0051] Figure 2 A schematic diagram of the steps of mining behavior rules provided in the first embodiment of the present invention;

[0052] Figure 3 A schematic diagram of the steps of the association control mode provided in the first embodiment of the present invention;

[0053] Figure 4 A schematic diagram of the steps for obtaining pre-start parameters provided in the first embodiment of the present invention;

[0054] Figure 5 This is a structural schematic diagram of an intelligent control system of a temperature controller provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0055] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0056] The purpose of the present invention is to achieve temperature adjustment of the controlled object through intelligent control of the thermostat. This embodiment takes the current smart home as the scene, sets thermostats for smart devices in different spatial areas, and achieves temperature adjustment of the spatial area through intelligent control of the thermostat.

[0057] Specifically, the intelligent control method, system and intelligent appliance of a thermostat of the present invention are described in detail through the following contents:

[0058] Embodiment 1

[0059] This embodiment provides an intelligent control method for a temperature controller. Figure 1 As shown, the following steps are included:

[0060] S1. Collecting behavior data: Collecting behavior data on a time axis according to user control requirements, including the temperature parameters and operating time set by the user for the controlled object and the real-time temperature of the spatial area, and obtaining the geographic location information of all user members.

[0061] It should be noted that the collection of user behavior data is used to construct user behavior rules for temperature control, thereby associating the control mode. The geographic location information of all user members refers to the location information obtained through the perception of the Internet of Things devices used by the members, thereby providing a trigger condition for the user to pre-start the thermostat. The intention of collecting the location information of all members is to establish an electronic fence centered on the residential address location information, so that the location of the members entering the fence meets the trigger condition.

[0062] S2. Mining behavioral rules: Through the collected behavioral data, data mining technology is used to analyze the relationship between input variables and output variables to obtain user behavior rules, such as Figure 2 As shown, the specific steps include:

[0063] S21. Select mining variables: select input variables and output variables through the collected behavioral data, where the input variables include the set temperature, actual temperature, thermostat operation time, operation time period, date type and area external temperature, and the output variables include the thermostat usage status.

[0064] It can be understood that the date type includes the season and whether it is a weekday, which is used to associate different usage characteristics and improve the accuracy of behavior rule mining; the regional external temperature refers to the area outside the spatial area where the thermostat is set, such as the outdoor area or the indoor area outside the room where the thermostat is located.

[0065] S22. Variable relationship mining: Discretize the input variables, convert the data into transaction format, group the data by date type and establish a transaction database; scan the transaction database in groups through the association rule mining algorithm to identify frequent item sets of different date types.

[0066] It is understandable that continuous variables (such as temperature, operating time) are converted into discrete values ​​to facilitate association rule analysis. For example, the temperature is divided into below 25℃, 25~27℃, 27~29℃, etc. The data is converted into transaction format, and each transaction contains the value of the input variable and the corresponding thermostat usage status (on or off).

[0067] In this embodiment, the input variables are discretized, such as the date types are divided into early July, mid-July, late July, etc. In order to improve the behavior rule library and use it to associate with the control mode, the date type is used as the group, and the association rules of each group are screened out as the behavior rules of the date type. Therefore, in the process of establishing the above rules, the data is grouped according to the date type and then the variable relationship mining is performed.

[0068] It should be noted that the Apriori algorithm in association analysis is used in this embodiment to describe the association between the variables. The Apriori algorithm is a breadth-first algorithm, which uses an iterative method of layer-by-layer search, that is, k-item sets are used to explore (k+1)-item sets, and based on the a priori properties of frequent item sets (that is, non-empty subsets of frequent item sets are all frequent item sets), all frequent item sets are found by multiple traversals of the database. The first step of the algorithm scans the entire database, counts the number of each item, and collects items that meet the minimum support, and finds a set of frequent 1-item sets, which is recorded as L1. The second step uses L1 to find the set L2 of frequent 2 items, and uses L2 to find L3, and this step is repeated until no frequent k-item sets can be found. After the frequent item sets are obtained from the data in the database, strong association rules that meet the minimum support and minimum confidence can be directly generated from them.

[0069] S23. Obtaining behavioral rules: Generate all rules from frequent item sets of different date types. During the continuous collection and update of behavioral data, select grid point parameters at different stages through coverage to screen association rules.

[0070] The establishment of grid point parameters includes:

[0071] By setting the minimum support and minimum confidence, the initial parameter range is set, a series of minimum support and minimum confidence are selected, and all combinations of minimum support and minimum confidence are used to construct a two-dimensional grid. Each grid point represents a pair of parameter combinations. For example, the minimum support is set to 0.1, 0.2, and 0.3; the minimum confidence is set to 0.6, 0.7, and 0.8; then the grid points are: (0.1, 0.6), (0.1, 0.7), (0.1, 0.8), (0.2, 0.6), (0.2, 0.7), (0.2, 0.8), (0.3, 0.6), (0.3, 0.7), and (0.3, 0.8).

[0072] It should be noted that in the mining of the above-mentioned user behavior rules, the features of different levels are mined in groups, the user behavior analysis is gradually refined, and the behavior rules of each stage are improved; at the same time, in the process of screening association rules, the rules are screened by selecting the most suitable parameter combinations for different stages. The different stages refer to the stages of application of thermostats in home scene temperature control systems, such as the thermostat intelligent control learning stage and the thermostat intelligent control optimization stage. Based on the amount of behavioral data collected, in the initial stage, lower minimum support and minimum confidence are set to obtain sufficient behavioral rules, so as to associate more accurate control modes in the subsequent process. This process is equivalent to improving the matching accuracy of the control mode by enhancing the data in the early stage.

[0073] S3, associated control mode: similarity calculation is performed on user behavior rules through multiple dimensions, and control modes similar to the current user are matched, and control suggestions are generated based on the matching results, such as Figure 3 As shown, the specific steps include:

[0074] S31. Establish a pattern library: store all users' behavior rules in the thermostat control pattern library, and update and maintain it regularly.

[0075] It is understandable that by collecting user behavior data from multiple sources, the diversity of the pattern library is expanded, including: behavior data of different types of users (such as homes, offices, and commercial places); user behavior data in different regions and climatic conditions.

[0076] S32, matching control mode: performing similarity calculation on user behavior patterns through multiple dimensions, and selecting several users with the highest similarity to the current user from the thermostat control mode library according to the similarity calculation results.

[0077] The dimensions are set according to the mining variables in the behavior rule mining, including the temperature setting dimension, the time mode dimension, the external environment dimension and the operation mode dimension.

[0078] Specifically, a similarity threshold is set to ensure that the behavior pattern of the selected users is close enough to the current user's pattern; according to different behavioral rule dimensions, users with high similarity in certain specific dimensions can be prioritized, such as users with similar temperature setting behaviors. In the calculation process, cosine similarity calculation is selected to evaluate the similarity between multi-dimensional features, such as users' multi-scenario behavior patterns.

[0079] S33, generate control suggestions: extract their behavior rules from the matched users, including the thermostat start conditions, and generate thermostat control suggestions for the current user.

[0080] It should be noted that the thermostat startup conditions include time points and environmental values. Through similarity matching, the control modes of similar users in all time periods are obtained, and the startup conditions of each time period are generated for the current user. Then, combined with the potential association between the subsequent geographic location and the startup, the accurate startup trigger mechanism for the current user is customized. The control suggestions finally generated are such as turning on the air conditioner at 18:00 and setting the temperature to 25°C, and setting the temperature to 26°C at 18:30, etc.

[0081] S4, obtaining pre-start parameters: verifying the generated control suggestion according to the user's geographic location information, and when the verification passes, using the control suggestion to extract the start time point and setting parameters, such as Figure 4 As shown, the following steps are included:

[0082] S41. Obtaining a distance judgment value: mining the user's home route through the user's daily activity trajectory, and then obtaining a distance judgment value by searching with route features.

[0083] S411, data collection: obtaining the geographic location information of each user through the IoT devices of user members, and binding each acquired location information with a timestamp to construct a time series trajectory;

[0084] At the same time, the system also needs to pre-set the user's home address information.

[0085] S412, select the decision value: based on the collected data, draw the user trajectory during the peak time period of going home, find the route with the largest proportion among all the trajectories as the first route, and the remaining trajectories as the second route, search for the reverse turning trajectory point closest to the home from the second route as the decision point, and calculate the distance between the decision point and the home address as the decision value.

[0086] S42. Verify control suggestions: Calculate the average time it takes for the user to return home through the first route based on the selected judgment value, use the average time as the time difference compensation turning-on time point as the judgment time point based on the thermostat opening time point of a certain control suggestion, obtain the user's location information at the judgment time point, set a distance threshold, and enable the control suggestion when the difference between the user's location distance from home and the judgment value is within the distance threshold.

[0087] S43, generating pre-start parameters: according to the enabled control suggestions, extracting the start time point and setting parameters, generating a control command and sending it to the thermostat to execute the action.

[0088] It should be noted that the above-mentioned pre-start control process includes two parts. First, by mining user behavior rules and matching with big data, more control behaviors similar to the user are obtained, and then the matched control behaviors are verified according to the user's activity pattern. With the combination of these two parts, the pre-start control parameters are obtained, which improves the accuracy and fit of intelligent control.

[0089] S5. Mode feedback correction: Collect the user's control behavior data at home to update the thermostat control mode library, regularly associate the control mode to supplement the current control mode, and gradually screen out the control suggestions that are frequently activated and optimize them.

[0090] The optimization of control suggestions for high-frequency activation refers to the user's adjustment behavior during actual operation, including parameter adjustment and device shutdown, so as to improve the perfection of high-frequency control suggestions.

[0091] In this embodiment, the mode feedback correction improves the thermostat control mode library through the collected behavioral data to provide more control suggestions for the user. On the other hand, it optimizes the frequently used control suggestions through the user's behavioral data, such as the operating time of the thermostat, parameter settings, etc., so as to improve the flexibility of intelligent control and make its automated control more accurate and practical.

[0092] Embodiment 2

[0093] This embodiment also provides an intelligent control system for a temperature controller, such as Figure 5 As shown, it includes an Internet of Things perception module, an intelligent control module and a pattern library cloud platform, and the Internet of Things perception module, the intelligent control module and the pattern library cloud platform are communicatively connected in sequence.

[0094] The IoT sensing module is used to collect user behavior data, including sensors set in the control environment and IoT devices worn by users, to obtain temperature parameters, thermostat operating status and user member location information in real time.

[0095] It should be noted that the data collection strategy of the IoT perception module includes:

[0096] Thermostat intelligent control learning stage: Set the first collection frequency to collect user behavior data around the clock;

[0097] Thermostat intelligent control optimization stage: Set the second collection frequency to collect user behavior data in stages.

[0098] In order to collect more comprehensive user behavior data in the thermostat intelligent control learning stage, a higher first collection frequency is set. In the thermostat intelligent control optimization stage, behavior data is collected according to the concentrated time period of thermostat startup, so a lower second collection frequency is set. This achieves data collection for the system at different stages, improves system efficiency, and reduces storage and computing costs.

[0099] The intelligent control module is used to mine the user's behavior rules through the user behavior data, and to search for similar control modes of the current user through the pattern library cloud platform to generate control suggestions;

[0100] At the same time, the user's activity mode is obtained according to the user's location information, so as to obtain the user's location based on the start time of the control suggestion to verify the feasibility of the control suggestion.

[0101] The intelligent control module is also used to modify the control suggestions for high-frequency activation through the behavior data collected during the intelligent control optimization phase of the thermostat, and adjust the parameters and duration of the thermostat operation.

[0102] It is understandable that the intelligent control module, as the core of the thermostat control, completes data processing and decision-making. By associating the data of both parties to obtain the user's possible control habits, and then verifying and executing them through the user's activity pattern, it improves the accuracy of the thermostat pre-start, thereby improving the practicality and satisfaction of intelligent control.

[0103] The pattern library cloud platform is used to establish an information-sharing thermostat control pattern library, store the mined user behavior rules in the thermostat control pattern library, and update and maintain the thermostat control pattern library by collecting the behavior rules of different types of users.

[0104] It should be noted that this embodiment only exemplifies the overall control of home environment temperature regulation. The detailed control of different areas is analogous. A thermostat is set in each area. Under the constraint of overall temperature control, the regional thermostat further refines the control according to the temperature control habits of different members, which will not be elaborated here.

[0105] Embodiment 3

[0106] This embodiment further provides a smart appliance, including a thermostat, a memory, a processor, and a control program of the smart appliance stored in the memory and executable on the processor, wherein:

[0107] The thermostat is used to adjust the on / off of the cooling or heating circuit of the smart appliance;

[0108] When the control program of the smart appliance is executed by the processor, the steps of the intelligent control method of the thermostat as described in the first embodiment are implemented.

[0109] The present invention mines the control behavior rules of the current user, matches them with the big data in the thermostat control mode library to obtain more control behaviors similar to the user, then generates control suggestions based on the matched control behaviors, and verifies them in real time through the current user's activity mode. Under the joint action of these two parts of operation, the pre-start control parameters are obtained, which improves the accuracy of the pre-start parameters and solves the hysteresis problem of the thermostat startup; at the same time, through the correlation mining of user behavior data, the user behavior analysis is gradually refined, and the behavior rules of each stage are improved, which provides support for matching similar control behaviors and enriching control suggestions; and the user's activity trajectory is analyzed to search for the best decision point to verify the feasibility of the control suggestion pre-start, thereby improving the automation level of the thermostat and the accuracy of intelligent control.

[0110] The present invention realizes accurate pre-start control, so that the thermostat can realize the control of the operating state on demand and in a timely manner, thereby reducing energy consumption, improving energy utilization efficiency, and reducing operating costs. In addition, the startup control of the thermostat is carried out in a predictive manner, which improves the response speed and dynamic adjustment ability of the temperature control system, adapts to changes in different environmental conditions and user needs, and optimizes the control mode through analysis and learning of behavioral data, thereby improving the long-term performance and stability of the thermostat.

[0111] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. An intelligent control method for a temperature controller, characterized in that: The following steps are involved: Mining behavioral rules: Through the collected behavioral data, data mining technology is used to analyze the relationship between input variables and output variables to obtain user behavior rules; Associated control mode: Calculate the similarity of user behavior rules through multiple dimensions, match the control mode similar to the current user, and generate control suggestions based on the matching results; Obtain pre-start parameters: Verify each generated control suggestion based on the user's geographic location information. If the verification passes, use the control suggestion to extract the start time point and set parameters. The mining behavior rules include the following steps: Select mining variables: select input variables and output variables through the collected behavioral data, where the input variables include set temperature, actual temperature, thermostat operation time, operation time period, date type and area external temperature, and the output variables include thermostat usage status; Variable relationship mining: Discretize the input variables, convert the data into transaction format, group the data by date type and build a transaction database; scan the transaction database in groups using the association rule mining algorithm to identify frequent item sets of different date types; Obtaining behavioral rules: Generate all rules from frequent item sets of different date types. During the continuous collection and update of behavioral data, select grid point parameters at different stages through coverage to screen association rules. The step of obtaining the pre-start parameters comprises the following steps: Obtaining distance judgment value: Mining the user's home route through the user's daily activity trajectory, and then searching for the distance judgment value based on route features; Verify control suggestions: Calculate the average time for users to return home through the first route based on the selected distance judgment value. According to the thermostat opening time point of a control suggestion, use the average time as the time difference compensation opening time point to calculate the judgment time point. Obtain the user's location information at the judgment time point, set the distance threshold, and enable the control suggestion when the difference between the user's location distance from home and the distance judgment value is within the distance threshold. Generate pre-start parameters: Extract the start time and set parameters based on the enabled control suggestions, generate control commands and send them to the thermostat to execute actions; The step of obtaining the distance judgment value comprises the following steps: Data collection: The geographic location information of each user is obtained through the IoT devices of user members, and each acquired location information is bound to a timestamp to construct a time series trajectory; At the same time, the user's home address coordinates are pre-set; Select the distance judgment value: Based on the collected data, draw the user trajectory during the peak time period of going home, find the route with the largest proportion among all trajectories as the first route, and the remaining trajectories as the second route. Search for the reverse turning trajectory point closest to home from the second route as the judgment point, and calculate the distance between the judgment point and the home address as the distance judgment value.

2. The intelligent control method of a temperature controller according to claim 1, characterized in that: The establishment of the grid point parameters includes: By setting the minimum support and minimum confidence, the initial parameter range is set, a series of minimum support and minimum confidence are selected, and all combinations of minimum support and minimum confidence are used to construct a two-dimensional grid, where each grid point represents a pair of parameter combinations.

3. The intelligent control method of a temperature controller according to claim 1, characterized in that: The association control mode comprises the following steps: Establish a pattern library: store all users' behavior rules in the thermostat control pattern library and update and maintain it regularly; Matching control mode: Calculate the similarity of user behavior patterns through multiple dimensions, and select several users with the highest similarity to the current user from the thermostat control mode library based on the similarity calculation results; The dimensions are set according to the mining variables in the behavior rule mining, including the temperature setting dimension, the time mode dimension, the external environment dimension, and the operation mode dimension; Generate control suggestions: Extract the behavior rules of the matched users, including the thermostat startup conditions, and generate thermostat control suggestions for the current user; The thermostat start-up conditions include time point and environmental value.

4. The intelligent control method of a temperature controller according to claim 1, characterized in that: The intelligent control method further comprises the steps of: Collecting behavioral data: Collecting behavioral data on a timeline based on user control requirements, including the temperature parameters and operating time set by the user for the controlled object and the real-time temperature of the spatial area, while obtaining the geographic location information of all user members.

5. The intelligent control method of a temperature controller according to claim 1, characterized in that: The intelligent control method further comprises the steps of: Mode feedback correction: Collect user's control behavior data at home to update the thermostat control mode library, regularly associate control modes to supplement the current control mode, and gradually screen out high-frequency control suggestions for optimization.

6. An intelligent control system for a thermostat, using an intelligent control method for a thermostat as claimed in any one of claims 1 to 5, characterized in that: It includes an IoT sensing module, an intelligent control module and a pattern library cloud platform, wherein the IoT sensing module, the intelligent control module and the pattern library cloud platform are sequentially connected in communication; The IoT sensing module is used to collect user behavior data, including sensors set in the control environment and IoT devices worn by users, to obtain temperature parameters, thermostat operating status and user member location information in real time; The data collection strategy of the IoT sensing module includes: Thermostat intelligent control learning stage: Set the first collection frequency to collect user behavior data around the clock; Thermostat intelligent control optimization stage: Set the second collection frequency to collect user behavior data in stages.

7. The intelligent control system of a temperature controller according to claim 6, characterized in that: The intelligent control module is used to mine the user's behavior rules through the user behavior data, and to search for similar control modes of the current user through the pattern library cloud platform to generate control suggestions; At the same time, the user's activity mode is obtained according to the user's location information, so as to obtain the user's location based on the start time of the control suggestion to verify the feasibility of the control suggestion; The intelligent control module is also used to modify the control suggestions for high-frequency activation through the behavior data collected during the intelligent control optimization phase of the thermostat, and adjust the parameters and duration of the thermostat operation; The pattern library cloud platform is used to establish an information-sharing thermostat control pattern library, store the mined user behavior rules in the thermostat control pattern library, and update and maintain the thermostat control pattern library by collecting the behavior rules of different types of users.

8. A smart appliance, characterized in that: The invention comprises a thermostat, a memory, a processor and a control program of an intelligent electrical appliance stored in the memory and operable on the processor, wherein: The thermostat is used to adjust the on / off of the cooling or heating circuit of the smart appliance; When the control program of the smart appliance is executed by the processor, the steps of the smart control method of the thermostat according to any one of claims 1 to 5 are implemented.

Images