Intelligent Control Methods and Systems for Electronic Devices Based on the Internet of Things

Abstract

This invention discloses an intelligent control method and system for electronic devices based on the Internet of Things (IoT), relating to the field of IoT data analysis technology. It includes analyzing the contribution of each data category to the control objective, considering the contribution or correlation of different data categories to different control objectives, providing a reliable foundation for subsequent data fusion and intelligent device control. Kalman filter parameters are set according to the contribution of each data category to the control objective, and further set based on the attributes and contribution of each data category. Fuzzy sets and membership functions are defined based on the fused features, and the type of membership function is selected according to the characteristics of the fused features to improve the targeting of fuzzy inference. Fuzzy inference is performed according to fuzzy control rules to generate control commands, improving the adaptability and accuracy of intelligent device control, ensuring the normal, automated, and safe operation of the intelligent device, and improving equipment efficiency.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) data analysis technology, and in particular to intelligent control methods and systems for electronic devices based on the Internet of Things. Background Technology

[0002] The Internet of Things (IoT), through technologies such as sensors, RFID, and network communication, enables ubiquitous connectivity between things and between things and people, laying the foundation for intelligent control of electronic devices. This solution utilizes IoT technology to connect electronic devices to the internet, achieving remote monitoring, intelligent adjustment, and automated control. By integrating multiple sensors and intelligent algorithms, the system can perceive environmental changes in real time and automatically adjust the operating status of equipment based on preset rules or user commands, thereby improving equipment efficiency and user experience. This solution not only promotes the intelligent upgrade of electronic devices but also provides strong support for building smart living and smart factory application scenarios.

[0003] In existing technologies, the intelligent control of electronic devices in the Internet of Things involves a wide variety of data categories, and the relationship between the data and the control target is quite complex. This results in poor adaptability and accuracy of intelligent control of devices, and cannot guarantee the degree of automation of intelligent control of electronic devices.

[0004] Therefore, how to improve the adaptability and accuracy of intelligent control of equipment is a technical problem that needs to be solved. Summary of the Invention

[0005] The purpose of this invention is to address the problem of poor adaptability and accuracy in existing intelligent control technologies for equipment, and to propose an intelligent control method for electronic devices based on the Internet of Things, which includes:

[0006] Collect all data categories and control objectives of the target electronic devices in the Internet of Things, and analyze the contribution of each data category to the control objectives;

[0007] The Kalman filter parameters are set according to the contribution of data categories to the control objective, and all data categories are fused to obtain fused features;

[0008] Based on the fusion features, fuzzy sets and membership functions are defined to establish fuzzy control rules for the target electronic device;

[0009] The system collects all data from the target electronic device in the Internet of Things in real time, performs fuzzy inference according to fuzzy control rules, generates control commands, and performs intelligent control of the target electronic device according to the control commands.

[0010] In some embodiments of this application, after collecting all data categories and control objectives of the target electronic device in the Internet of Things, the method further includes,

[0011] The control objectives of the target electronic equipment are classified, and the matching relationship between the partial data categories of all data categories and each control objective is determined, thus establishing the data-control objective category matching relationship;

[0012] Collect effect parameters that describe the effectiveness of each type of control target. Based on the data-control target category matching relationship, add the corresponding effect parameters to form a data-control target-effect parameter category matching relationship.

[0013] In some embodiments of this application, the contribution of each data category to the control objective is analyzed, including,

[0014] Based on the category matching relationship between data, control objectives, and effect parameters, a scatter plot is drawn between each data category and each effect parameter category. The points on the scatter plot are connected and smoothed to determine the shape of the line on the scatter plot, thereby determining the linear and non-linear relationships.

[0015] For linear relationships, the correlation coefficient is used to describe the contribution of each data category to each effect parameter category;

[0016] For non-linear relationships, the SHAP value is used to describe the contribution of each data category to each effect parameter category;

[0017] The contribution values ​​are integrated according to the category of effect parameters to obtain the contribution value of each data category;

[0018] The contribution of each data category to each effect parameter category is the contribution of the data category to the effect parameter.

[0019] In some embodiments of this application, Kalman filter parameters are set according to the contribution of data category to the control objective, including:

[0020] Calculate the measurement noise variance and standard deviation for each data category, and determine the process noise variance for each data category based on the standard deviation.

[0021] Construct the measurement noise covariance matrix and the process noise covariance matrix based on the measurement noise variance and the process noise variance, respectively;

[0022] Based on the contribution of each data category to the control objective, element values ​​are assigned to each data category to construct an observation matrix;

[0023] The Kalman gain is determined based on the measurement noise covariance matrix, the process noise covariance matrix, and the observation matrix.

[0024] The Kalman filter parameters include the preset initial error covariance matrix, measurement noise covariance matrix, process noise covariance matrix, observation matrix, and Kalman gain.

[0025] In some embodiments of this application, all data categories are fused to obtain fused features, including:

[0026] Under the category matching relationship between data and control targets, prediction, update and feature fusion steps are performed according to the Kalman filter parameters to obtain the fused features under the category matching relationship between data and control targets for each group.

[0027] In some embodiments of this application, fuzzy sets and membership functions are defined based on the fused features, including:

[0028] Define the fuzzy set of fused features according to the category of the control target;

[0029] Construct a distribution map for each fusion feature, connect the points on the distribution map, and perform smoothing to obtain the fusion feature curve. Perform a normality test on the fusion feature curve according to the normality test method to obtain the test value.

[0030] The skewness and kurtosis on the statistical fusion characteristic curve are used to obtain the normality index based on the test statistic, skewness, and kurtosis;

[0031] The membership function type is selected based on the normality index, thereby constructing membership functions for different fuzzy sets. The membership function types include Gaussian membership functions and other membership functions.

[0032] Other membership functions include the triangular membership function and the trapezoidal membership function.

[0033] In some embodiments of this application, fuzzy control rules for the target electronic device are established, including:

[0034] Fuzzy control rules for each control objective of the target electronic device are established based on multiple fuzzy sets and multiple membership functions.

[0035] In some embodiments of this application, fuzzy inference is performed according to fuzzy control rules to generate control commands, including...

[0036] Based on the facts, collect all data over a period of time, plot the change curves for each type of data, analyze the change curves to determine the confidence level of each type of data, and add the confidence level to fuzzy inference to assist in generating control instructions.

[0037] Correspondingly, this application also provides an intelligent control system for electronic devices based on the Internet of Things, including,

[0038] The first module is used to collect all data categories and control objectives of the target electronic devices in the Internet of Things, and analyze the contribution of each data category to the control objectives;

[0039] The second module is used to set the Kalman filter parameters according to the contribution of data categories to the control objective, and to fuse all data categories to obtain fused features;

[0040] The third module is used to define fuzzy sets and membership functions based on the fused features, and to establish fuzzy control rules for the target electronic device.

[0041] The fourth module is used to collect all data from the target electronic device in the Internet of Things in real time, perform fuzzy inference according to fuzzy control rules, generate control commands, and perform intelligent control of the target electronic device according to the control commands.

[0042] Compared with the prior art, the beneficial effects of this invention are as follows:

[0043] 1. Analyze the contribution of each data category to the control objective, considering the contribution or correlation between different data categories and different control objectives, to provide a reliable foundation for subsequent data fusion and intelligent equipment control. Set Kalman filter parameters based on the contribution of each data category to the control objective, and further set Kalman filter parameters based on both the attributes and contribution of the data categories to reasonably reduce noise and improve the real-time performance, robustness, and scalability of data analysis and processing.

[0044] 2. Based on the fused features, fuzzy sets and membership functions are defined. The type of membership function is selected according to the characteristics of the fused features to improve the targeting of fuzzy inference. Fuzzy inference is performed according to fuzzy control rules to generate control commands, and the confidence level of the data is considered to assist in determining the control commands. This improves the adaptability and accuracy of intelligent equipment control, ensures the normal, automated, and safe operation of intelligent equipment, and enhances equipment working efficiency. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating the intelligent control method for electronic devices based on the Internet of Things proposed in this invention.

[0046] Figure 2 This is a schematic diagram of the structure of the intelligent control system for electronic devices based on the Internet of Things proposed in this invention. Detailed Implementation

[0047] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0048] Reference Figure 1A method for intelligent control of electronic devices based on the Internet of Things includes the following steps:

[0049] Step S101: Collect all data categories and control objectives of the target electronic devices in the Internet of Things, and analyze the contribution of each data category to the control objectives.

[0050] In this embodiment, the target electronic device in the Internet of Things (IoT) can be a smart home control device, industrial automation equipment, or other IoT-related electronic devices. Different target electronic devices may involve numerous data categories, which can be collected by sensors. The control objectives can include start / stop conditions for control commands, alarm monitoring, device protection, and specific effect adjustments (energy consumption optimization, comfort adjustment), etc. The data includes environmental parameters, device operating parameters, user parameters, etc. Different control objectives correspond to different data categories. For example, in a smart home, the control objective is the start / stop of certain control commands, which requires environmental data to trigger thresholds (such as temperature > 30℃, humidity meeting the threshold, automatically turning on the air conditioner), and user location information to help determine the usage scenario. The control objective is energy consumption optimization, which requires a comprehensive judgment based on environmental data and device operating data. Therefore, different control objectives correspond to different data categories.

[0051] In some embodiments of this application, after collecting all data categories and control objectives of the target electronic device in the Internet of Things, the method further includes,

[0052] The control objectives of the target electronic equipment are classified, and the matching relationship between the partial data categories of all data categories and each control objective is determined, thus establishing the data-control objective category matching relationship;

[0053] Collect effect parameters that describe the effectiveness of each type of control target. Based on the data-control target category matching relationship, add the corresponding effect parameters to form a data-control target-effect parameter category matching relationship.

[0054] In this embodiment, the effect parameters describing the effect of each type of control target include the accuracy of the start command, the accuracy of the warning, and the change in energy consumption. The effect parameters of the start command include the start accuracy rate, response delay, and false trigger rate. The effect parameters of energy efficiency optimization include the energy efficiency ratio improvement rate and unit energy consumption cost. The effect parameters of equipment protection include the failure rate reduction rate, overheat protection success rate, maintenance cost, and downtime.

[0055] In some embodiments of this application, the contribution of each data category to the control objective is analyzed, including,

[0056] Based on the category matching relationship between data, control objectives, and effect parameters, a scatter plot is drawn between each data category and each effect parameter category. The points on the scatter plot are connected and smoothed to determine the shape of the line on the scatter plot, thereby determining the linear and non-linear relationships.

[0057] For linear relationships, the correlation coefficient is used to describe the contribution of each data category to each effect parameter category;

[0058] For non-linear relationships, the SHAP value is used to describe the contribution of each data category to each effect parameter category;

[0059] The contribution values ​​are integrated according to the category of effect parameters to obtain the contribution value of each data category;

[0060] The contribution of each data category to each effect parameter category is the contribution of the data category to the effect parameter.

[0061] In this embodiment, to analyze the contribution or correlation of each data category to the control objective, each control objective is refined, and all effect parameters that can describe the control objective are collected. The strength of the correlation between data and effect parameters is calculated to indirectly reflect the contribution of each data category to the control objective. Linear and non-linear relationships are identified based on the shape of the scatter plot between each data category and each effect parameter category. For linear relationships, the correlation coefficient (Pearson correlation coefficient) is used to describe the linear relationship; for non-linear relationships, the SHAP value is used to describe the non-linear relationship. A suitable machine learning model (such as random forest, gradient boosting tree, neural network, etc.) is selected for training. To ensure the model can capture non-linear relationships in the data, the SHAP (SHapley Additive exPlanations) method is used to calculate the contribution of each data category to each effect parameter category. The SHAP value is based on the Shapley value in game theory, which can fairly allocate the contribution of each feature to the prediction result. Specific calculation steps include: a. For each data sample, calculate the prediction difference when a certain data category is included and not included. b. Calculate the weighted sum of all possible feature combinations to obtain the SHAP value for the prediction result for that data category. c. Repeat the above steps to calculate the SHAP value for each data category for each effect parameter category.

[0062] In this embodiment, the linear and non-linear relationships (multiple effect parameters) of each data category are converted into contribution degrees, and the contribution degree of each data category is determined after integration. The calculation formula is as follows:

[0063] ;

[0064] in, For the first The contribution of each data category , These are the conversion coefficients for linear and nonlinear relationships, respectively. , The first The number of linear and non-linear relationships for each data category, i.e., the number of effect parameters for each category. , The first The first linear relationship and the second The combined weights of the contributions of each nonlinear relationship. , The first The first data category The first linear relationship and the second The contribution of each nonlinear relationship , They are respectively and The maximum value in, , The first The first and second constants for each data category , They represent and The maximum value in the formula is used to correct the average value. The first and second constants exist to balance the magnitude of the correction function.

[0065] Step S102: Set the Kalman filter parameters according to the contribution of data categories to the control objective, and fuse all data categories to obtain fused features.

[0066] In this embodiment, Kalman filtering is based on a state-space model, which consists of state equations and observation equations. The state equations describe the dynamic changes in the system state, while the observation equations describe the relationship between sensor measurements and the system state. In multi-class data fusion, the state vector can contain data from different sensors, such as environmental data (temperature, humidity, etc.) and equipment status data (energy consumption, operating mode, etc.). Kalman filtering is a recursive estimation algorithm that uses the estimated value from the previous time step and the observed value from the current time step to update the estimate of the system state. Through this recursive approach, Kalman filtering can dynamically combine multiple types of data to achieve real-time estimation of the system state. Kalman filtering handles noise in the system model and sensor measurements through the process noise covariance matrix and the measurement noise covariance matrix. The process noise covariance matrix describes the uncertainty in the dynamic changes of the system state, while the measurement noise covariance matrix describes the uncertainty in sensor measurements.

[0067] As is understandable, Kalman filtering is a recursive algorithm capable of processing newly arriving observation data in real time and updating the system state estimate. This makes it extremely useful in scenarios requiring real-time decision-making and control. Kalman filtering can handle noise and uncertainty in system models and sensor measurements. By adjusting the weights of the noise covariance matrix, Kalman filtering can adaptively handle different noise levels. Kalman filtering can be easily extended to handle multi-class data. This is achieved simply by incorporating different classes of data into the state vector and observation matrix, and adjusting the noise covariance matrix accordingly.

[0068] In some embodiments of this application, Kalman filter parameters are set according to the contribution of data category to the control objective, including:

[0069] Calculate the measurement noise variance and standard deviation for each data category, and determine the process noise variance for each data category based on the standard deviation.

[0070] Construct the measurement noise covariance matrix and the process noise covariance matrix based on the measurement noise variance and the process noise variance, respectively;

[0071] Based on the contribution of each data category to the control objective, element values ​​are assigned to each data category to construct an observation matrix;

[0072] The Kalman gain is determined based on the measurement noise covariance matrix, the process noise covariance matrix, and the observation matrix.

[0073] The Kalman filter parameters include the preset initial error covariance matrix, measurement noise covariance matrix, process noise covariance matrix, observation matrix, and Kalman gain.

[0074] In this embodiment, the process noise covariance matrix (Q) describes the noise caused by external factors (such as random disturbances) during the system state transition process. It reflects the uncertainty of the system model and is an important basis for updating the covariance matrix in the Kalman filter prediction step.

[0075] The measurement noise covariance matrix (R) describes the noise caused by factors such as sensor errors during the observation process. It reflects the uncertainty of the observed values ​​and is a key parameter for calculating the Kalman gain in the Kalman filter update step. The observation matrix (H) describes the relationship between the observed values ​​and the state variables. It maps the system state to the observation space, allowing the Kalman filter to update the state estimate using the observed values. The noise variance (measurement noise variance) of each data category can be calculated by analyzing historical observation data, and the diagonal elements of the R matrix can be set accordingly. The stability of each data category can be evaluated by calculating indicators such as the time series variance, rate of change, and standard deviation. A process noise variance is assigned to each data category based on stability. These process noise variances are combined into a Q matrix, which is similar in form to the R matrix and is usually also a diagonal matrix. The elements of the H matrix can be set according to the contribution of the data category to the control objective. Data categories with higher contributions can be assigned larger observation matrix element values, thus giving them higher weights in the Kalman filter.

[0076] In this embodiment, the Kalman gain is a key parameter that determines the weight of the observations in the state estimate update. The calculation of the Kalman gain depends on the measurement noise covariance matrix (R), the process noise covariance matrix (Q), and the observation matrix (H). The Kalman gain (K), calculated in the update step, is used to weigh the weights between the predicted and observed values. It is one of the core parameters of the Kalman filter and determines the degree of state estimate update.

[0077] In some embodiments of this application, all data categories are fused to obtain fused features, including:

[0078] Under the category matching relationship between data and control targets, prediction, update and feature fusion steps are performed according to the Kalman filter parameters to obtain the fused features under the category matching relationship between data and control targets for each group.

[0079] This embodiment provides the specific steps for Kalman filtering to achieve multi-type data fusion:

[0080] initialization:

[0081] Set initial values ​​for the state vector and the covariance matrix (initial error covariance matrix). These initial values ​​are typically determined based on prior knowledge or historical data.

[0082] Prediction steps:

[0083] The system model (state equation) is used to predict the state vector and covariance matrix at the next time step. This step does not consider the observations at the current time step, but only makes predictions based on the estimates from the previous time step and the system model.

[0084] Update steps:

[0085] The Kalman gain is calculated by combining the current observations and the observation matrix. The Kalman gain is used to weigh the weights between the predicted and observed values, resulting in a more accurate estimate. The state vector and covariance matrix are then updated using the Kalman gain to obtain the optimal estimate for the current time step.

[0086] Output fusion features:

[0087] Key features are extracted from the fused data, such as the overall energy efficiency index and user comfort index. These features can be calculated based on the fused state vector for subsequent decision-making and control.

[0088] It should be noted that in the prediction step, the update of the prediction error covariance matrix (P) considers the process noise covariance matrix (Q), which is the noise caused by external factors (such as random disturbances) during the system state transition. In the update step, the Kalman gain is calculated by combining the current observation value and the observation matrix, and the state vector and covariance matrix are updated using the Kalman gain. The updated covariance matrix reflects the uncertainty of the state estimate after fusing the observation values. This update process also considers the measurement noise covariance matrix (R), which is the noise caused by factors such as sensor errors during the observation process.

[0089] Step S103: Based on the fused features, define fuzzy sets and membership functions to establish fuzzy control rules for the target electronic device.

[0090] In this embodiment, the control objectives of control commands can be categorized into fuzzy sets such as performance, energy saving, and comfort. Control commands for parameter optimization can be categorized into fuzzy sets such as environmental comfort (high, medium, low), equipment load (high, medium, low), and user activity (high, medium, low). Fuzzy sets are an extension of classical sets, allowing elements to belong to a set to a certain degree (membership degree), rather than simply "belonging" or "not belonging." Fuzzy sets are the foundation of fuzzy inference. In fuzzy inference, rule matching and reasoning are performed based on fuzzy sets and membership functions to arrive at fuzzy conclusions, making it more suitable for intelligent control of electronic devices in the Internet of Things (IoT) environment. The membership function is the core of fuzzy sets; it defines the degree to which an element belongs to the fuzzy set. The range of the membership function is typically [0,1], where 0 indicates that the element does not belong to the set at all, 1 indicates that the element belongs to the set completely, and values ​​between 0 and 1 represent the degree to which the element belongs to the set.

[0091] In some embodiments of this application, fuzzy sets and membership functions are defined based on the fused features, including:

[0092] Define the fuzzy set of fused features according to the category of the control target;

[0093] Construct a distribution map for each fusion feature, connect the points on the distribution map, and perform smoothing to obtain the fusion feature curve. Perform a normality test on the fusion feature curve according to the normality test method to obtain the test value.

[0094] The skewness and kurtosis on the statistical fusion characteristic curve are used to obtain the normality index based on the test statistic, skewness, and kurtosis;

[0095] The membership function type is selected based on the normality index, thereby constructing membership functions for different fuzzy sets. The membership function types include Gaussian membership functions and other membership functions.

[0096] Other membership functions include the triangular membership function and the trapezoidal membership function.

[0097] In this embodiment, multiple data sets are fused to obtain a parameter for a fused feature. The distribution plot is analyzed to determine whether it conforms to a normal distribution. Statistical tests, such as the Shapiro-Wilk test, the Kolmogorov-Smirnov test (KS test), or the Anderson-Darling test, are used to formally test whether the data follows a normal distribution. These tests provide a p-value (test scalar value). If the p-value is greater than a certain significance level (e.g., 0.05), the hypothesis that the data follows a normal distribution cannot be rejected. Skewness measures the symmetry of the data distribution; the skewness of a normal distribution is 0. Kurtosis measures the kurtosis of the data distribution; the kurtosis of a normal distribution is 3 (or 0 under some definitions, depending on whether 3 is subtracted for standardization). Combining the test scalar value, skewness, and kurtosis generates a normality index to determine whether the data follows a normal distribution.

[0098] In this embodiment, an appropriate membership function type is selected based on the distribution of the fused features. For example, if the data exhibits a normal distribution, a Gaussian membership function can be selected. Otherwise, triangular or trapezoidal membership functions are chosen.

[0099] In some embodiments of this application, fuzzy control rules for the target electronic device are established, including:

[0100] Fuzzy control rules for each control objective of the target electronic device are established based on multiple fuzzy sets and multiple membership functions.

[0101] In this embodiment, for example, for smart homes, the fuzzy set may include environmental comfort (low, medium, high), user activity (low, medium, high), operating mode (comfort, energy saving), and wind speed. The fuzzy rule is: IF environmental comfort is "high" AND user activity is "low" THEN operating mode is "energy saving". Implementation: When the membership function value of environmental comfort is in the "high" range and the membership function value of user activity is in the "low" range, the membership function value of the operating mode is output to be in the "energy saving" range.

[0102] Step S104: Collect all data of the target electronic device in the Internet of Things in real time, perform fuzzy inference according to fuzzy control rules, generate control commands, and perform intelligent control of the target electronic device according to the control commands.

[0103] In this embodiment, the fuzzy reasoning process is as follows:

[0104] Fuzzification: Mapping fused features to a fuzzy set.

[0105] Rule matching: Activate fuzzy rules that meet the conditions.

[0106] Inference synthesis: Calculate the activation strength of each rule.

[0107] Defuzzification: Obtain precise control commands through the centroid method or the maximum membership method.

[0108] Control command execution

[0109] Send instructions to the device actuators (such as adjusting the air conditioner compressor frequency or changing the fan speed).

[0110] Monitor the execution results and report them to the system logs for future optimization.

[0111] In some embodiments of this application, fuzzy inference is performed according to fuzzy control rules to generate control commands, including...

[0112] Based on the facts, collect all data over a period of time, plot the change curves for each type of data, analyze the change curves to determine the confidence level of each type of data, and add the confidence level to fuzzy inference to assist in generating control instructions.

[0113] In this embodiment, to ensure the reliability of fuzzy inference, the confidence level is determined by analyzing the credibility of real-time collected data. Plotting tools are selected: plotting libraries such as Matplotlib and Seaborn, or spreadsheet software such as Excel. Curves are plotted: for each type of data, a curve is plotted with its time series as the horizontal axis and the data values ​​as the vertical axis.

[0114] Confidence assessment:

[0115] Stability: Data with small curve fluctuations and stable trends have higher confidence levels; data with large fluctuations and unstable trends have lower confidence levels.

[0116] Consistency: Data that is consistent with other relevant data or expected trends has a higher confidence level; data that shows significant deviations has a lower confidence level.

[0117] Data completeness: Data with few missing data points and complete coverage of time periods has a higher confidence level; data with many missing data points and incomplete coverage of time periods has a lower confidence level.

[0118] The formula for calculating the confidence level for each data category is as follows:

[0119] ;

[0120] in, For the first Confidence of class data , , These are the conversion coefficients for data stability, consistency, and integrity, respectively. , , The first The confidence levels determined by the stability, consistency, and integrity of the data type are as follows: , They are respectively The maximum and minimum values ​​in the range. For the first The third constant of class data, This represents the correction of the average of the sum of the three values ​​to the average of the average determined by the maximum and minimum values.

[0121] Confidence fusion: In the process of fuzzy inference, the confidence level of each type of data is used as an additional input variable or weighting factor to affect the result of fuzzy inference.

[0122] Fuzzy inference result processing: Based on the output of the fuzzy inference system, defuzzification processing is performed to obtain specific control command parameters.

[0123] Command generation and transmission: Converting control command parameters into actual control commands and sending them to the actuators (such as intelligent devices, automation systems, etc.).

[0124] Correspondingly, this application also provides an intelligent control system for electronic devices based on the Internet of Things, such as... Figure 2 As shown, including,

[0125] The first module is used to collect all data categories and control objectives of the target electronic devices in the Internet of Things, and analyze the contribution of each data category to the control objectives;

[0126] The second module is used to set the Kalman filter parameters according to the contribution of data categories to the control objective, and to fuse all data categories to obtain fused features;

[0127] The third module is used to define fuzzy sets and membership functions based on the fused features, and to establish fuzzy control rules for the target electronic device.

[0128] The fourth module is used to collect all data from the target electronic device in the Internet of Things in real time, perform fuzzy inference according to fuzzy control rules, generate control commands, and perform intelligent control of the target electronic device according to the control commands.

[0129] Compared with the prior art, the beneficial effects of this invention are as follows:

[0130] 1. Analyze the contribution of each data category to the control objective, considering the contribution or correlation between different data categories and different control objectives, to provide a reliable foundation for subsequent data fusion and intelligent equipment control. Set Kalman filter parameters based on the contribution of each data category to the control objective, and further set Kalman filter parameters based on both the attributes and contribution of the data categories to reasonably reduce noise and improve the real-time performance, robustness, and scalability of data analysis and processing.

[0131] 2. Based on the fused features, fuzzy sets and membership functions are defined. The type of membership function is selected according to the characteristics of the fused features to improve the targeting of fuzzy inference. Fuzzy inference is performed according to fuzzy control rules to generate control commands, and the confidence level of the data is considered to assist in determining the control commands. This improves the adaptability and accuracy of intelligent equipment control, ensures the normal, automated, and safe operation of intelligent equipment, and enhances equipment working efficiency.

[0132] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0133] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0134] Those skilled in the art will understand that the modules in the system of the implementation scenario can be distributed throughout the system of the implementation scenario as described, or they can be modified to reside in one or more systems different from this implementation scenario. The modules of the above-mentioned implementation scenario can be merged into one module, or they can be further divided into multiple sub-modules.

[0135] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An intelligent control method for electronic devices based on the Internet of Things, characterized in that, include, Collect all data categories and control objectives of the target electronic devices in the Internet of Things, and analyze the contribution of each data category to the control objectives; The Kalman filter parameters are set according to the contribution of data categories to the control objective, and all data categories are fused to obtain fused features; Based on the fusion features, fuzzy sets and membership functions are defined to establish fuzzy control rules for the target electronic device; Real-time acquisition of all data from target electronic devices in the Internet of Things, fuzzy inference according to fuzzy control rules, generation of control commands, and intelligent control of target electronic devices according to control commands; in, After collecting all data categories and control objectives of the target electronic devices in the Internet of Things, the method further includes, The control objectives of the target electronic equipment are classified, and the matching relationship between the partial data categories of all data categories and each control objective is determined, thus establishing the data-control objective category matching relationship; Collect effect parameters that describe the effectiveness of each type of control target. Based on the data-control target category matching relationship, add the corresponding effect parameters to form a data-control target-effect parameter category matching relationship. Analyze the contribution of each data category to the control objective, including: Based on the category matching relationship between data, control objectives, and effect parameters, a scatter plot is drawn between each data category and each effect parameter category. The points on the scatter plot are connected and smoothed to determine the shape of the line on the scatter plot, thereby determining the linear and non-linear relationships. For linear relationships, the correlation coefficient is used to describe the contribution of each data category to each effect parameter category; For non-linear relationships, the SHAP value is used to describe the contribution of each data category to each effect parameter category; The contribution values ​​are integrated according to the category of effect parameters to obtain the contribution value of each data category; The contribution of each data category to each effect parameter category is the contribution of the data category to the effect parameter.

2. The intelligent control method for electronic devices based on the Internet of Things according to claim 1, characterized in that, Set the Kalman filter parameters according to the contribution of the data category to the control objective. include, Calculate the measurement noise variance and standard deviation for each data category, and determine the process noise variance for each data category based on the standard deviation. Construct the measurement noise covariance matrix and the process noise covariance matrix based on the measurement noise variance and the process noise variance, respectively; Based on the contribution of each data category to the control objective, element values ​​are assigned to each data category to construct an observation matrix; The Kalman gain is determined based on the measurement noise covariance matrix, the process noise covariance matrix, and the observation matrix. The Kalman filter parameters include the preset initial error covariance matrix, measurement noise covariance matrix, process noise covariance matrix, observation matrix, and Kalman gain.

3. The intelligent control method for electronic devices based on the Internet of Things according to claim 2, characterized in that, Then, all data categories are fused to obtain fused features, including: Under the category matching relationship between data and control targets, prediction, update and feature fusion steps are performed according to the Kalman filter parameters to obtain the fused features under the category matching relationship between data and control targets for each group.

4. The intelligent control method for electronic devices based on the Internet of Things according to claim 1, characterized in that, Based on the fusion features, fuzzy sets and membership functions are defined, including: Define the fuzzy set of fused features according to the category of the control target; Construct a distribution map for each fusion feature, connect the points on the distribution map, and perform smoothing to obtain the fusion feature curve. Perform a normality test on the fusion feature curve according to the normality test method to obtain the test value. The skewness and kurtosis on the statistical fusion characteristic curve are used to obtain the normality index based on the test statistic, skewness, and kurtosis; The membership function type is selected based on the normality index, thereby constructing membership functions for different fuzzy sets. The membership function types include Gaussian membership functions and other membership functions. Other membership functions include the triangular membership function and the trapezoidal membership function.

5. The intelligent control method for electronic devices based on the Internet of Things according to claim 1, characterized in that, Establish fuzzy control rules for the target electronic device, including: Fuzzy control rules for each control objective of the target electronic device are established based on multiple fuzzy sets and multiple membership functions.

6. The intelligent control method for electronic devices based on the Internet of Things according to claim 1, characterized in that, Fuzzy inference is performed according to fuzzy control rules to generate control commands, including... Based on the facts, collect all data over a period of time, plot the change curves for each type of data, analyze the change curves to determine the confidence level of each type of data, and add the confidence level to fuzzy inference to assist in generating control instructions.

7. An intelligent control system for electronic devices based on the Internet of Things, characterized in that, The system is used to implement the intelligent control method for IoT-based electronic devices as described in any one of claims 1-6, the system comprising: The first module is used to collect all data categories and control objectives of the target electronic devices in the Internet of Things, and analyze the contribution of each data category to the control objectives; The second module is used to set the Kalman filter parameters according to the contribution of data categories to the control objective, and to fuse all data categories to obtain fused features; The third module is used to define fuzzy sets and membership functions based on the fused features, and to establish fuzzy control rules for the target electronic device. The fourth module is used to collect all data from the target electronic device in the Internet of Things in real time, perform fuzzy inference according to fuzzy control rules, generate control commands, and perform intelligent control of the target electronic device according to the control commands.

Images