An energy-saving control method for central air conditioning based on Internet of Things

Abstract

The present invention discloses an energy-saving control method based on an Internet of Things central air conditioner, which relates to the field of energy-saving control technology, including collecting environmental data in real time through an Internet of Things sensor network, and using a mobile application to collect user preference information and historical behavior data, outputting a user behavior feature matrix and a personal preference profile; based on the user behavior feature matrix and the personal preference profile, using a machine learning model to analyze the user's behavior pattern, and generating a personalized energy-saving strategy set; based on real-time feedback information, evaluating the effectiveness of the personalized energy-saving strategy, and updating the user behavior feature matrix and the personal preference profile according to the evaluation results. The present invention achieves the dual improvement of efficient energy saving and user comfort of the central air-conditioning system through personalized energy-saving strategies and multi-objective optimization, and solves the problem of inaccurate energy-saving control in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of energy-saving control, and in particular to an energy-saving control method based on a central air conditioner of the Internet of Things. Background Art

[0002] Some existing IoT-based central air-conditioning energy-saving control systems can achieve energy-saving effects to a certain extent, but there are still defects in the formulation of personalized energy-saving strategies. Most systems rely only on environmental data (such as temperature, humidity, etc.), while ignoring the importance of user behavior data. Due to the large differences in user preferences, simple environmental parameters are not enough to fully reflect the real needs of users. Therefore, energy-saving strategies formulated without fully considering user behavior characteristics often fail to achieve the optimal energy-saving effect and user experience. In addition, the existing technology usually adopts a single optimization goal (such as minimizing energy consumption) in the process of formulating energy-saving strategies, but fails to take into account both energy saving and user comfort, which makes it difficult to achieve a balance in practical applications. Summary of the invention

[0003] In view of the above existing problems, the present invention is proposed.

[0004] Therefore, the present invention provides an energy-saving control method based on the Internet of Things central air conditioner to solve the problem of inaccurate formulation of personalized energy-saving strategies in the prior art.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides an energy-saving control method based on an Internet of Things central air conditioner, which includes collecting environmental data in real time through an Internet of Things sensor network, and using a mobile application to collect user preference information and historical behavior data, and outputting a user behavior feature matrix and a personal preference profile; based on the user behavior feature matrix and the personal preference profile, using a machine learning model to analyze the user's behavior pattern and generate a personalized energy-saving strategy set; combining the personalized energy-saving strategy set with the actual needs of the operation of the Internet of Things central air conditioner, using a multi-objective optimization algorithm to find the optimal operating parameter configuration; according to the optimal operating parameter configuration, controlling the actual operation of the Internet of Things central air conditioner, and continuously collecting real-time feedback information during the actual operation process; based on the real-time feedback information, evaluating the effectiveness of the personalized energy-saving strategy, and updating the user behavior feature matrix and the personal preference profile according to the evaluation results.

[0007] As a preferred solution of the energy-saving control method based on the Internet of Things central air conditioner of the present invention, wherein: the output of the user behavior feature matrix and the personal preference profile comprises the following specific steps:

[0008] Install IoT sensors and connect to the Alibaba Cloud IoT platform via ZigBee;

[0009] Upload data in a standardized format to the cloud database every minute, use Z-score anomaly detection, and linear interpolation to fill in missing values;

[0010] Record user setting preferences and operation behaviors through mobile applications;

[0011] Integrate environmental data with user preference and behavior data into the same dataset;

[0012] Extract key features from the integrated data and use DataFrame to construct a user behavior feature matrix;

[0013] Generates a personal preference profile in JSON format based on the feature matrix.

[0014] As a preferred solution of the energy-saving control method based on the Internet of Things central air conditioner described in the present invention, wherein: extracting key features from the integrated data and using DataFrame to construct a user behavior feature matrix, the specific steps are:

[0015] Based on the integrated data, the Pearson correlation coefficient is used to quantify the correlation between features and user behavior;

[0016] Using the feature importance scores in random forests, we extract key features based on the correlation between quantified features and user behaviors, and use principal component analysis to reduce the number of features.

[0017] Use Python's Pandas library to create a DataFrame with the selected key features as columns and user IDs as row indices to form a user behavior feature matrix.

[0018] As a preferred solution of the energy-saving control method based on the Internet of Things central air conditioner described in the present invention, wherein: the specific steps of generating a personalized energy-saving strategy set are as follows:

[0019] The energy-saving strategy set S contains a series of energy-saving measures, among which the energy-saving measures ;

[0020] Calculates a personalized energy-saving strategy score for each user , the expression is:

[0021] ;

[0022] in, Indicates The personalized score of the energy-saving strategy, N is the number of features, For the The weight of the feature, Indicates The feature for The contribution function of each energy-saving strategy is: represents the index variable identifying each energy-saving strategy in the energy-saving strategy set, represents the index variable identifying the feature set extracted from the user behavior feature matrix and personal preference profile;

[0023] The contribution function of the energy-saving strategy is expressed as:

[0024] ;

[0025] in, Indicates The feature for The contribution value of each energy saving strategy is Indicates environmental factors For The contribution function of the feature, represents the contribution function of user preference y to the jth feature, is the sensitivity parameter, E is the value range of the environmental data, and dz represents the integration operation of a variable in the environmental data;

[0026] Environmental factor contribution function , the expression is:

[0027] ;

[0028] in, represents the average value of the environmental factor corresponding to the jth feature, represents the standard deviation of the environmental factor corresponding to the jth feature;

[0029] User preference contribution function , the expression is:

[0030] ;

[0031] in, represents the sensitivity parameter of the jth feature, Indicates the threshold of user preference corresponding to the jth feature;

[0032] For all energy saving strategies, Sort in descending order and select the top K energy-saving strategies to form a personalized energy-saving strategy set , the expression is: .

[0033] As a preferred solution of the energy-saving control method based on the Internet of Things central air conditioner described in the present invention, the multi-objective optimization algorithm is used to find the optimal operating parameter configuration, and the specific steps are:

[0034] Set the operating parameter range of central air conditioners based on personalized energy-saving strategies and equipment capabilities, environmental conditions, and user preferences;

[0035] For the set parameter range, a non-dominated sorting genetic algorithm is used to adjust energy consumption and comfort;

[0036] A set of operating parameters of the central air conditioner in actual operation is used as the initial population to calculate the energy consumption and comfort index of each individual;

[0037] Generate new individuals through crossover and mutation, and use non-dominated sorting for iterative optimization until the predetermined number of iterations is reached to form the final population;

[0038] The individuals on the Pareto frontier are selected from the final population as the optimal operating parameter configuration.

[0039] As a preferred solution of the energy-saving control method based on the Internet of Things central air conditioner described in the present invention, wherein: the non-dominated sorting is iteratively optimized, and the specific steps are:

[0040] Sort all individuals in the current population according to the non-dominated hierarchy;

[0041] The first layer is the set with the most non-dominated individuals, the second layer is the set of individuals that do not dominate the individuals in the first layer, and so on, until all individuals are assigned to the bottom layer;

[0042] For each layer, the crowding distance of each individual is calculated;

[0043] Select individuals from the current population to produce the next generation, pair the selected individuals, and use a crossover operator to produce offspring;

[0044] Apply mutation operators to offspring individuals and merge the parent population and offspring population into a new temporary population;

[0045] Perform non-dominated sorting on the merged population again and calculate the crowding distance;

[0046] Select a specified number of individuals from the merged population to form the next generation population;

[0047] Repeat the steps of sorting all individuals in the current population according to the non-dominated hierarchy to the step of forming the next generation population until a predetermined number of iterations is reached.

[0048] As a preferred solution of the energy-saving control method based on the Internet of Things central air conditioner of the present invention, wherein: the actual operation of the Internet of Things central air conditioner is controlled according to the optimal operating parameter configuration, and real-time feedback information during the actual operation is continuously collected. The specific steps are:

[0049] Use MQTT protocol to send optimal operating parameter configuration to the central air conditioning control unit;

[0050] After receiving the configuration, the central air conditioning control unit parses the message in JSON format and extracts the required operating parameters;

[0051] The control unit adjusts the operating state of the central air conditioner according to the received parameters and performs real-time operation;

[0052] Continuously monitor environmental changes through the IoT sensor network and send monitoring data to the cloud using the ZigBee protocol;

[0053] The Z-score method is used to detect anomalies on the cloud data, identify and remove outliers, and linear interpolation is used to fill in missing values;

[0054] Store the processed data in a cloud database.

[0055] As a preferred solution of the energy-saving control method based on the Internet of Things central air conditioner described in the present invention, wherein: the user behavior feature matrix and the personal preference profile are updated according to the evaluation results, and the specific steps are as follows:

[0056] Calculate energy saving efficiency by comparing energy consumption changes before and after implementing energy saving strategies;

[0057] Using user rating data, the overall satisfaction score is calculated by weighted average;

[0058] Calculate the comfort index by combining real-time environmental data;

[0059] Using online learning algorithms, the feature matrix is ​​updated based on user satisfaction and comfort index;

[0060] Recalculate the score of the personalized energy-saving strategy based on the new user behavior feature matrix and personal preference profile, and update the profile;

[0061] According to the updated personalized energy-saving strategy set and user behavior feature matrix, the multi-objective optimization algorithm is used again to find the optimal operating parameter configuration.

[0062] In a second aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the energy-saving control method for central air conditioning based on the Internet of Things as described in the first aspect of the present invention is implemented.

[0063] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the energy-saving control method for central air conditioning based on the Internet of Things as described in the first aspect of the present invention is implemented.

[0064] The beneficial effects of the present invention are as follows: the present invention realizes the intelligent optimization configuration of the operating parameters of the central air-conditioning system by integrating the Internet of Things sensor network, mobile application preference information collection, machine learning model analysis of user behavior patterns, multi-objective optimization algorithm and dynamic feedback mechanism; the user experience and comfort are greatly improved through the customized implementation of personalized energy-saving strategies; by collecting environmental data and user preference information in real time, a user behavior feature matrix and a personal preference profile are constructed, and then a personalized energy-saving strategy set is analyzed using a machine learning model, and the optimal operating parameter configuration is determined using a multi-objective optimization algorithm in combination with equipment operation requirements; the dynamic adaptability and continuous optimization capability of the energy-saving strategy are ensured by evaluating the effectiveness of the energy-saving strategy and timely updating the user behavior feature matrix and the personal preference profile. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0066] Figure 1 This is a flow chart of the energy-saving control method of the central air conditioner based on the Internet of Things in Example 1.

[0067] Figure 2 This is a flow chart for finding the optimal operating parameter configuration in Example 1. DETAILED DESCRIPTION

[0068] In order to make the above-mentioned objects, features and advantages of the present invention more understandable, the specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings.

[0069] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0070] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0071] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides an energy-saving control method for central air conditioning based on the Internet of Things, comprising the following steps:

[0072] S1. Collect environmental data in real time through the IoT sensor network, and use mobile applications to collect user preference information and historical behavior data, and output user behavior feature matrix and personal preference profile.

[0073] Going a step further, IoT sensors (such as temperature and humidity sensors) are installed and connected to the Alibaba Cloud IoT platform via ZigBee;

[0074] Upload data in a standardized format (JSON) to the cloud database every minute, use Z-score anomaly detection, and linear interpolation to fill in missing values;

[0075] Through mobile applications, record user setting preferences and operation behaviors, such as device usage frequency, preference settings for specific functions, etc.;

[0076] It should be noted that the user's setting preferences and operating behaviors are recorded through mobile applications. Specific examples are as follows:

[0077] Example 1: Suppose a user controls a central air conditioner through a smart home management application (mobile application), and the application interface provides the following functions: Temperature setting: users can select their favorite room temperature range (for example: 22°C-26°C); Humidity setting: users can select a suitable humidity range (for example: 40%-60%); Wind speed selection: users can select high, medium, and low wind speeds (for example: low speed, medium speed, high speed); Mode selection: users can select different air conditioner operating modes (for example: cooling, heating, dehumidification, automatic mode); Time setting: users can set the timer switch of the air conditioner (for example: automatically turn off at 10 pm and automatically turn on at 7 am).

[0078] Operation process: The user opens the mobile app and enters the air conditioning control interface. In the app interface, the user adjusts and saves his or her air conditioning setting preferences through UI elements such as sliders, buttons, or selection boxes. For example, the user sets the temperature to 22°C, the humidity to 50%, the wind speed to medium, and selects "automatic mode".

[0079] Example 2: In addition to recording the preferences set by users, mobile applications can also track users' operating behaviors in real time to understand their usage habits. For example: Manual temperature adjustment: users frequently manually adjust the air conditioner temperature during a specific time period (for example, users lower the temperature to 20°C every day at noon); mode switching: users switch to dehumidification mode in certain special weather conditions (for example, very humid weather); frequency of use: the time period during which users use the air conditioner every day (for example, users usually use the air conditioner from 7pm to 11pm).

[0080] Operation process: Users manually adjust the temperature or mode of the air conditioner through the mobile app at different times of the day. For example, the user adjusts the temperature to 20°C at 12 noon and switches the mode to "dehumidification" at 7 pm.

[0081] Integrate contextual data with user preference and behavior data into the same dataset for unified analysis;

[0082] Extract key features from the integrated data, such as temperature change trends, user usage habits, etc., and use DataFrame to construct a user behavior feature matrix, where each column represents a feature;

[0083] Generate a personal preference profile in JSON format based on the feature matrix to facilitate subsequent application and service customization.

[0084] Furthermore, the specific steps of “extracting key features from the integrated data and using DataFrame to construct a user behavior feature matrix” are to use the Pearson correlation coefficient to quantify the correlation between features and user behavior based on the integrated data;

[0085] Using the feature importance scores in random forests, we extract key features based on the correlation between quantified features and user behaviors. We also use principal component analysis to reduce the number of features, thereby reducing the data dimension while retaining the main information.

[0086] Preferably, principal component analysis (PCA), a dimensionality reduction technique, can map multidimensional data to a lower dimensional space while retaining the variation information of the original data as much as possible. PCA can reduce the dimension of the data and simplify the analysis process.

[0087] Use Python's Pandas library to create a DataFrame with the selected key features as columns and user IDs as row indices to form a user behavior feature matrix.

[0088] It should be noted that DataFrame is a two-dimensional data structure in the Pandas library, similar to a table in a spreadsheet or database. The main purpose of using DataFrame is to structure user behavior feature data to facilitate subsequent data processing and analysis.

[0089] It should be noted that by installing IoT sensors (such as temperature and humidity sensors) and connecting to the Alibaba Cloud IoT platform via ZigBee, the function of uploading data in a standardized format (JSON) to the cloud database once a minute is realized. In order to ensure data quality, Z-score anomaly detection is used to identify and exclude outliers, and linear interpolation is used to fill missing values ​​to ensure data integrity and accuracy. In addition, mobile applications are used to record user setting preferences and operation behaviors, such as device usage frequency and preference settings for specific functions. These data are integrated into the same data set with environmental data for unified analysis. By extracting key features such as temperature change trends and user usage habits, and using DataFrame to construct a user behavior feature matrix, in which each column represents a feature, user needs can be understood and predicted more accurately. Finally, a personal preference profile in JSON format is generated based on the feature matrix to provide users with personalized application and service customization, greatly improving the user experience and the intelligence level of the system.

[0090] S2. Based on the user behavior feature matrix and personal preference profile, the machine learning model is used to analyze the user's behavior pattern and generate a personalized energy-saving strategy set.

[0091] Furthermore, the energy-saving strategy set S includes a series of energy-saving measures, among which ;

[0092] Calculates a personalized energy-saving strategy score for each user , the expression is:

[0093] ;

[0094] in, Indicates The personalized score of the energy-saving strategy, N is the number of features, For the The weight of the feature, Indicates The feature for The contribution function of each energy-saving strategy is: represents the index variable identifying each energy-saving strategy in the energy-saving strategy set, represents the index variable identifying the feature set extracted from the user behavior feature matrix and personal preference profile;

[0095] It should be noted that the personalized energy saving strategy score Describes each energy saving strategy The suitability and benefits for a particular user are calculated based on the user's behavioral characteristics and personal preference profile.

[0096] Among them, the higher the score, the more suitable the energy-saving strategy is for the user and the more effective it may be in saving energy.

[0097] The contribution function of the energy-saving strategy is expressed as:

[0098] ;

[0099] in, Indicates The feature for The contribution value of each energy saving strategy is Indicates environmental factors For The contribution function of the feature, represents the contribution function of user preference y to the jth feature, is the sensitivity parameter, E is the value range of the environmental data, and dz represents the integration operation of a variable in the environmental data;

[0100] It should be noted that the contribution function of the energy-saving strategy Describes the The feature for Energy-saving strategies The contribution function is used to quantify the relationship between energy-saving strategies and user behavior characteristics and personal preference profiles.

[0101] It should also be noted that the contribution function of the energy-saving strategy Score for personalized energy saving strategies Provides the necessary inputs to personalize energy saving strategy scores It is calculated based on these contribution functions, which reflects the applicability and benefits of energy-saving strategies for specific users. These two parts are closely linked and together constitute the core of personalized energy-saving strategy evaluation.

[0102] Environmental factor contribution function , the expression is:

[0103] ;

[0104] in, represents the average value of the environmental factor corresponding to the jth feature, represents the standard deviation of the environmental factor corresponding to the jth feature;

[0105] It should be noted that the specific steps for obtaining the environmental factor z are as follows: install various types of sensors in the environment that needs to be monitored, such as temperature sensors, humidity sensors, light sensors, etc. The sensors should cover all important environmental areas to ensure the comprehensiveness and accuracy of the data; the sensors should be configured to collect data regularly (for example, every minute or every half hour) and send these data to the central data processing server, and use wireless communication technology (such as Wi-Fi, Bluetooth, Zigbee, etc.) to transmit the data to the receiving device or gateway, and then upload it to the cloud platform or local server by the gateway; store the received raw data in the database, clean and preprocess the data, remove outliers, missing values ​​or noise, and ensure the quality of the data; use data analysis tools (such as Python's Pandas library, R language, etc.) to perform statistical analysis on the data, extract useful features, and use data visualization tools (such as Matplotlib, Tableau, etc.) to create charts to intuitively display the trends and changes of environmental data.

[0106] User preference contribution function , the expression is:

[0107] ;

[0108] in, represents the sensitivity parameter of the jth feature, Indicates the threshold of user preference corresponding to the jth feature;

[0109] It should be noted that the specific steps for obtaining user preference y are as follows: design a user-friendly user interface (UI) so that users can easily input their preference information. The user interface can be implemented through mobile applications, web applications or other interactive methods; collect user preference information through questionnaires, such as their favorite indoor temperature range, humidity range, etc., and set some options for users to choose from, or allow users to enter custom preference values; implement user behavior tracking functions to record users' actual operations at different times and scenarios, such as adjusting the temperature of the air conditioner, opening and closing curtains, etc., and collect these data through mobile applications or smart devices; combine the preference information collected from the user interface with the information obtained from behavioral data tracking to form a complete user preference profile, which can automatically identify user preference patterns through machine learning algorithms (such as collaborative filtering, cluster analysis, etc.); regularly update the user's preference profile to reflect the user's latest preferences, which can be achieved through active updates by the user or automatic learning by the system.

[0110] For all energy saving strategies, Sort in descending order and select the top K energy-saving strategies to form a personalized energy-saving strategy set , the expression is: .

[0111] S3, combined with the personalized energy-saving strategy set and the actual needs of IoT central air-conditioning operation, use the multi-objective optimization algorithm to find the best operating parameter configuration, refer to Figure 2 .

[0112] Furthermore, the operating parameter range of the central air conditioner can be set based on the personalized energy-saving strategy and equipment capabilities, environmental conditions and user preferences;

[0113] For the set parameter range, a non-dominated sorting genetic algorithm is used to adjust energy consumption and comfort;

[0114] A set of operating parameters of the central air conditioner in actual operation is used as the initial population to calculate the energy consumption and comfort index of each individual;

[0115] Generate new individuals through crossover and mutation, and use non-dominated sorting for iterative optimization until the predetermined number of iterations is reached to form the final population;

[0116] The individuals located on the Pareto frontier are selected from the final population as the optimal operating parameter configuration, where the Pareto frontier is translated into Chinese as Pareto frontier.

[0117] It should be noted that the Pareto front represents a set of solutions that find the best compromise between multiple objectives (such as energy consumption and comfort), that is, these individuals cannot further optimize another objective without sacrificing one objective.

[0118] Going further, the specific steps of "Iterative Optimization Using Non-Dominated Sorting" are as follows:

[0119] Sort all individuals in the current population according to the non-dominated hierarchy;

[0120] The first layer is the set with the most non-dominated individuals, the second layer is the set of individuals that do not dominate the individuals in the first layer, and so on, until all individuals are assigned to the bottom layer;

[0121] It should be noted that the “bottom layer” refers to the last non-empty layer, which contains all individuals that have been dominated by individuals in the previous layers. In short, this is the last filled level in the sorting process, which contains all the remaining individuals.

[0122] For each layer, the crowding distance of each individual is calculated. The larger the crowding distance, the more "sparse" the individual is in the target space, that is, there are fewer individuals around it.

[0123] Select individuals from the current population to produce the next generation, pair the selected individuals, and use a crossover operator to produce offspring;

[0124] Apply mutation operators to offspring individuals and merge the parent population and offspring population into a new temporary population;

[0125] Perform non-dominated sorting on the merged population again and calculate the crowding distance;

[0126] Select a specified number of individuals from the merged population to form the next generation population;

[0127] It should be noted that when selecting individuals from the current population to produce the next generation, the selection usually starts from the individuals in the lowest non-dominated level. If the individuals in this level are not enough to constitute the next generation, individuals are selected from the next level. When selecting individuals from the same level, individuals with a larger crowding distance are given priority to maintain the diversity of the population.

[0128] Repeat the steps of sorting all individuals in the current population according to the non-dominated hierarchy to the step of forming the next generation population until a predetermined number of iterations is reached.

[0129] It should be noted that in multi-objective optimization problems, we usually face multiple objective functions, which are often conflicting with each other. Therefore, non-dominated sorting is used to evaluate the relationship between the advantages and disadvantages of individuals in the population. Specifically, the "non-dominated" relationship between individuals is as follows:

[0130] Suppose there are two individuals A and B. If for all objective functions , both , and there exists at least one objective function j such that , then individual A is said to be non-dominated by individual B; if there is no other individual C that does not dominate individual A, then individual A is said to be non-dominated.

[0131] Preferably, through the above non-dominated sorting, a series of optimal solutions can be found among multiple objectives. These solutions are usually called Pareto optimal solutions, which help to maintain the diversity of the population and find the optimal solution set when solving multi-objective optimization problems.

[0132] S4. According to the optimal operating parameter configuration, control the actual operation of the IoT central air conditioner and continuously collect real-time feedback information during the actual operation process.

[0133] Furthermore, the MQTT protocol is used to send the optimal operating parameter configuration to the central air conditioning control unit;

[0134] After receiving the configuration, the central air conditioning control unit parses the message in JSON format and extracts the required operating parameters;

[0135] The control unit adjusts the operating state of the central air conditioner according to the received parameters and performs real-time operation;

[0136] Continuously monitor environmental changes through the IoT sensor network and send monitoring data to the cloud using the ZigBee protocol;

[0137] It should be noted that the specific implementation process of "using ZigBee protocol to send monitoring data to the cloud" is as follows:

[0138] First, deploy various types of sensors in the environment that needs to be monitored, such as temperature sensors, humidity sensors, light sensors, etc. After deploying various types of sensors, collect environmental data through sensors regularly, such as once every minute or every five minutes, depending on the application requirements and the capabilities of the sensors.

[0139] In order to achieve low-power and long-distance communication, a wireless sensor network is established between sensor nodes through the ZigBee protocol. There is usually a coordinator node in the network as a gateway, which is responsible for receiving data from each sensor and forwarding the data to the cloud; the sensor node packages the collected data into the format specified by the ZigBee protocol, usually including information such as sensor ID, timestamp and measurement value; the sensor node sends the data packet to the coordinator node in the network.

[0140] Preferably, the coordinator node can be a dedicated device or a computer / server with a ZigBee module; if the network is large, there may be multiple hops to reach the coordinator. In this case, the data packet will be forwarded to the coordinator according to a predefined routing table or multi-hop mechanism.

[0141] After receiving the data from all sensor nodes, the coordinator node repackages the data into a format suitable for cloud reception, such as JSON format.

[0142] In order to protect the security of the data, the coordinator node may encrypt the data to prevent it from being intercepted during transmission.

[0143] Send data to the cloud using the appropriate upload protocol.

[0144] Among them, common protocols include HTTP / HTTPS, MQTT, CoAP, etc. Here, you can choose to use HTTP / HTTPS or MQTT protocols because they support data transmission in JSON format, and MQTT is particularly suitable for low-bandwidth, high-latency or unreliable network connections.

[0145] The cloud server receives the data packet from the coordinator node and depacketizes it into original data.

[0146] Use the Z-score method to detect anomalies in the data, identify and remove outliers; for missing values, linear interpolation and other methods can be used to fill them.

[0147] Store the processed data in a cloud database for subsequent analysis and access.

[0148] The Z-score method is used to detect anomalies on the cloud data, identify and remove outliers, and linear interpolation is used to fill in missing values;

[0149] Preferably, linear interpolation is a commonly used data processing method for estimating unknown values ​​between two known data points. It is particularly useful for handling missing values, especially in time series data.

[0150] Store the processed data in a cloud database;

[0151] Preferably, when storing the processed data in a cloud database, Amazon DynamoDB should be selected as the cloud database because it supports high-concurrency read and write operations and can automatically scale to cope with the growing amount of data. At the same time, DynamoDB also provides a powerful API interface for easy integration with other AWS services, such as Lambda functions, S3 storage buckets, etc., to facilitate data processing and analysis tasks.

[0152] S5. Based on real-time feedback information, evaluate the effectiveness of personalized energy-saving strategies, and update the user behavior feature matrix and personal preference profile according to the evaluation results.

[0153] Furthermore, the energy saving efficiency is calculated by comparing the energy consumption changes before and after the implementation of the energy saving strategy. The specific steps are as follows:

[0154] Before implementing energy-saving strategies, collect energy consumption data within a certain period as baseline data (such as one week, one month, etc.), which can be automatically collected through the IoT sensor network.

[0155] After implementing the energy-saving strategy, energy consumption data in the same period are also collected as experimental data.

[0156] Remove outliers and missing values ​​to ensure data quality; if necessary, standardize the data to facilitate subsequent comparison and analysis.

[0157] Calculate the total energy consumption during the baseline period and calculate the total energy consumption during the period when the energy saving strategy is implemented.

[0158] The energy saved is the baseline energy consumption minus the experimental energy consumption. The energy saving efficiency can be calculated by the following formula: ;

[0159] For the results found, statistical methods can be used to verify the energy-saving efficiency, such as t-test; analyze the trend of energy-saving efficiency over time to understand the long-term effects of energy-saving strategies;

[0160] The examples for the above content are as follows:

[0161] Suppose we have the following data:

[0162] Baseline energy consumption: The total energy consumption before implementing energy-saving strategies is 1,000 kilowatt-hours (kWh).

[0163] Experimental energy consumption: The total energy consumption after implementing the energy-saving strategy is 800 kilowatt-hours (kWh).

[0164] Calculate the energy saving efficiency according to the above formula:

[0165] ;

[0166] Among other things, this meant a 20% reduction in energy consumption through the implementation of energy-saving strategies.

[0167] It is also important to note that external factors affect energy consumption: it is necessary to consider the impact of external factors (such as weather changes, equipment aging, etc.) on energy consumption and ensure that the comparison is made under similar conditions; time window selection: select a suitable time window to collect energy consumption data to ensure that the data is representative; repeat the experiment: in order to improve the reliability of the results, the above experiment can be repeated in different time periods to verify the effectiveness of the energy-saving strategy.

[0168] Use user rating data to calculate the overall satisfaction score by weighted average. The specific steps are as follows:

[0169] First, it is necessary to collect rating data about services or products from users. These ratings can be collected within a specific period of time or in real time. The ratings can use different scales, such as star ratings from 1 to 5, numerical ratings from 1 to 10, etc.

[0170] Secondly, determine the weight of each rating. The weight can be determined based on a variety of factors. For example, higher weights may be given to time factors: recent ratings may be considered more important; user activity: users who frequently provide feedback may have a greater impact on ratings; specific rating values: some systems may consider extreme ratings (very satisfied or very dissatisfied) to be more important than medium ratings.

[0171] For each rating , the weighted score expression is calculated as:

[0172] ;

[0173] in, is the weight corresponding to the score, and Q represents the weight value associated with the score;

[0174] Adding up all weighted scores gives the weighted sum expression: ;

[0175] Adding all weights together gives the total weight expression: ;

[0176] Finally, the expression for calculating the weighted average score A is: ;

[0177] The specific examples are as follows:

[0178] Assume the following rating data and corresponding weights:

[0179] User A gave a 4-star rating, with a weight of 1 (standard weight).

[0180] User B gave a 5-star rating, which has a weight of 1.2 (because user B gave a rating most recently).

[0181] User C gave a 3-star rating with a weight of 0.8 (because user C is an occasional feedback provider).

[0182] Then, follow these steps to calculate the weighted average score:

[0183] Calculate the weighted score:

[0184] User A: 4×1=44×1=4

[0185] User B: 5×1.2=65×1.2=6

[0186] User C: 3×0.8=2.43×0.8=2.4

[0187] Weighted sum:

[0188] ;

[0189] Sum of weights:

[0190] ;

[0191] Weighted average score:

[0192] ;

[0193] Therefore, the overall satisfaction score is 4.13 stars.

[0194] Calculate the comfort index by combining real-time environmental data (such as temperature, humidity and other environmental factors data);

[0195] For example, a comfort model can be used, such as the ASHRAE (American Society of Heating, Refrigerating and Air-Conditioning Engineers) Standard 55-2017, which provides a method for assessing thermal comfort;

[0196] Use online learning algorithms (such as online gradient descent (OGD), stochastic gradient descent (SGD), Adagrad, etc.) to update the feature matrix based on user satisfaction and comfort index;

[0197] Recalculate the score of the personalized energy-saving strategy based on the new user behavior feature matrix and personal preference profile, and update the profile;

[0198] According to the updated personalized energy-saving strategy set and user behavior feature matrix, the multi-objective optimization algorithm is used again to find the optimal operating parameter configuration.

[0199] This embodiment also provides a computer device, which is suitable for the energy-saving control method of central air conditioning based on the Internet of Things, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the energy-saving control method of central air conditioning based on the Internet of Things as proposed in the above embodiment.

[0200] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0201] The present embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the energy-saving control method for central air conditioning based on the Internet of Things is implemented as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0202] In summary, the present invention realizes the intelligent optimization configuration of the operating parameters of the central air-conditioning system by integrating the Internet of Things sensor network, mobile application preference information collection, machine learning model analysis of user behavior patterns, multi-objective optimization algorithm and dynamic feedback mechanism; the user experience and comfort are greatly improved through the customized implementation of personalized energy-saving strategies; by collecting environmental data and user preference information in real time, a user behavior feature matrix and a personal preference profile are constructed, and then a personalized energy-saving strategy set is analyzed using a machine learning model, and the optimal operating parameter configuration is determined using a multi-objective optimization algorithm in combination with equipment operation requirements; by evaluating the effectiveness of the energy-saving strategy and timely updating the user behavior feature matrix and personal preference profile, the dynamic adaptability and continuous optimization capability of the energy-saving strategy are ensured.

[0203] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An energy-saving control method for central air conditioning based on the Internet of Things, characterized in that: include, Collect environmental data in real time through the IoT sensor network, and use mobile applications to collect user preference information and historical behavior data, and output user behavior feature matrix and personal preference profile; Based on the user behavior feature matrix and personal preference profile, a machine learning model is used to analyze the user's behavior pattern and generate a personalized energy-saving strategy set; Combining the personalized energy-saving strategy set with the actual needs of IoT central air-conditioning operation, a multi-objective optimization algorithm is used to find the optimal operating parameter configuration; According to the optimal operating parameter configuration, control the actual operation of the IoT central air conditioner and continuously collect real-time feedback information during the actual operation; Based on real-time feedback information, evaluate the effectiveness of personalized energy-saving strategies and update user behavior feature matrix and personal preference profile according to the evaluation results; The output of the user behavior feature matrix and the personal preference profile comprises the following specific steps: Install IoT sensors and connect to the Alibaba Cloud IoT platform via ZigBee; Upload data in a standardized format to the cloud database every minute, use Z-score anomaly detection, and linear interpolation to fill in missing values; Record user setting preferences and operation behaviors through mobile applications; Integrate environmental data with user preference and behavior data into the same dataset; Extract key features from the integrated data and use DataFrame to construct a user behavior feature matrix; Generate a personal preference profile in JSON format based on the feature matrix; The specific steps of extracting key features from the integrated data and using DataFrame to construct a user behavior feature matrix are as follows: Based on the integrated data, the Pearson correlation coefficient is used to quantify the correlation between features and user behavior; Using the feature importance scores in random forests, we extract key features based on the correlation between quantified features and user behaviors, and use principal component analysis to reduce the number of features. Use Python's Pandas library to create a DataFrame with the selected key features as columns and user IDs as row indices to form a user behavior feature matrix; The specific steps of generating a personalized energy-saving strategy set are as follows: The energy-saving strategy set S contains a series of energy-saving measures, among which the energy-saving measures ; Calculates a personalized energy-saving strategy score for each user , the expression is: ； in, Indicates The personalized score of the energy-saving strategy, N is the number of features, For the The weight of the feature, Indicates The feature for The contribution function of each energy-saving strategy is: represents the index variable identifying each energy-saving strategy in the energy-saving strategy set, represents the index variable identifying the feature set extracted from the user behavior feature matrix and personal preference profile; The contribution function of the energy-saving strategy is expressed as: ； in, Indicates The feature for The contribution value of each energy-saving strategy is Indicates environmental factors For The contribution function of the feature, represents the contribution function of user preference y to the jth feature, is the sensitivity parameter, E is the value range of the environmental data, and dz represents the integration operation of a variable in the environmental data; Environmental factor contribution function , the expression is: ； in, represents the average value of the environmental factor corresponding to the jth feature, represents the standard deviation of the environmental factor corresponding to the jth feature; User preference contribution function , the expression is: ； in, represents the sensitivity parameter of the jth feature, Indicates the threshold of user preference corresponding to the jth feature; For all energy saving strategies, Sort in descending order and select the top K energy-saving strategies to form a personalized energy-saving strategy set , the expression is: ; The multi-objective optimization algorithm is used to find the best operating parameter configuration, and the specific steps are: Set the operating parameter range of central air conditioners based on personalized energy-saving strategies and equipment capabilities, environmental conditions, and user preferences; For the set parameter range, a non-dominated sorting genetic algorithm is used to adjust energy consumption and comfort; A set of operating parameters of the central air conditioner in actual operation is used as the initial population to calculate the energy consumption and comfort index of each individual; Generate new individuals through crossover and mutation, and use non-dominated sorting for iterative optimization until the predetermined number of iterations is reached to form the final population; Select the individuals on the Pareto frontier from the final population as the optimal operating parameter configuration; The non-dominated sorting is iteratively optimized, and the specific steps are: Sort all individuals in the current population according to the non-dominated hierarchy; The first layer is the set with the most non-dominated individuals, the second layer is the set of individuals that do not dominate the individuals in the first layer, and so on, until all individuals are assigned to the bottom layer; For each layer, the crowding distance of each individual is calculated; Select individuals from the current population to produce the next generation, pair the selected individuals, and use a crossover operator to produce offspring; Apply mutation operators to offspring individuals and merge the parent population and offspring population into a new temporary population; Perform non-dominated sorting on the merged population again and calculate the crowding distance; Select a specified number of individuals from the merged population to form the next generation population; Repeat the steps of sorting all individuals in the current population according to the non-dominated hierarchy to the step of forming the next generation population until a predetermined number of iterations is reached.

2. The energy-saving control method based on the central air conditioner of the Internet of Things as claimed in claim 1, characterized in that: The actual operation of the IoT central air conditioner is controlled according to the optimal operating parameter configuration, and real-time feedback information during the actual operation is continuously collected. The specific steps are: Use MQTT protocol to send optimal operating parameter configuration to the central air conditioning control unit; After receiving the configuration, the central air conditioning control unit parses the message in JSON format and extracts the required operating parameters; The control unit adjusts the operating state of the central air conditioner according to the received parameters and performs real-time operation; Continuously monitor environmental changes through the IoT sensor network and send monitoring data to the cloud using the ZigBee protocol; The Z-score method is used to detect anomalies on the cloud data, identify and remove outliers, and linear interpolation is used to fill in missing values; Store the processed data in a cloud database.

3. The energy-saving control method based on the central air conditioner of the Internet of Things as claimed in claim 2 is characterized in that: The specific steps of updating the user behavior feature matrix and the personal preference profile according to the evaluation results are as follows: Calculate energy saving efficiency by comparing energy consumption changes before and after implementing energy saving strategies; Using user rating data, the overall satisfaction score is calculated by weighted average; Calculate the comfort index by combining real-time environmental data; Using online learning algorithms, the feature matrix is ​​updated based on user satisfaction and comfort index; Recalculate the score of the personalized energy-saving strategy based on the new user behavior feature matrix and personal preference profile, and update the profile; According to the updated personalized energy-saving strategy set and user behavior feature matrix, the multi-objective optimization algorithm is used again to find the optimal operating parameter configuration.

4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the energy-saving control method based on the Internet of Things central air conditioner according to any one of claims 1 to 3 are implemented.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the energy-saving control method based on the Internet of Things central air conditioner according to any one of claims 1 to 3 are implemented.

Images