A smart home control method and system based on IoT perception

Abstract

The present invention relates to a smart home control method and system based on IoT perception, belonging to the field of smart home control technology. The method includes the following steps: determining the constituent elements according to the target smart home control system, analyzing the influence between the constituent elements; collecting IoT perception data to establish an IoT perception database; selecting execution elements, and fusing the contribution factors with evidence; performing conflict detection on the decision rules, and revising the conflict rules based on the scene and the user; generating control commands according to the revised decision rules to control the execution element devices; matching the IoT perception data and the execution log to generate control rules, and updating the user rule list. The present invention analyzes the constituent elements of the target smart home control system, constructs a user rule library based on the distribution of the constituent elements, and generates rules according to the order of influence, so that the final decision is more in line with the user's habits, and at the same time improves the availability and security of the system through the conflict detection and avoidance mechanism.

Description

Technical Field

[0001] The present invention belongs to the technical field of smart home control, and in particular relates to a smart home control method and system based on Internet of Things perception. Background Art

[0002] Smart home uses residence as a platform and utilizes integrated wiring technology, network communication technology, security technology, automatic control technology, and audio and video technology to integrate facilities related to home life, build an efficient management system for residential facilities and family schedule affairs, improve home safety, convenience, comfort, and artistry, and achieve an environmentally friendly and energy-saving living environment.

[0003] However, the current smart home control system is designed to meet the needs of most users rather than the specific needs of individual users. This "one-size-fits-all" approach may result in the system not being able to adapt well to the unique preferences of individual users, resulting in poor user experience. At the same time, the traditional single sensor data-based control system has poor reliability and stability. In addition, as the system is increasingly widely used, the number of decision rules set by users continues to increase. Without decision rule conflict detection, it is impossible to detect conflicting rules and effectively avoid them. Summary of the invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a smart home control method and system based on Internet of Things perception, which analyzes the constituent elements of the target smart home control system, builds a user rule library based on the distribution of the constituent elements, and generates rules according to the order of influence, so that the final decision is more in line with user habits. At the same time, the conflict detection and avoidance mechanism is used to improve the availability and security of the system.

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

[0006] A first aspect of the present disclosure provides a smart home control method based on IoT perception, comprising the following steps:

[0007] Extract system elements: determine the constituent elements according to the target smart home control system, analyze the influence between the elements, and generate element distribution relationships;

[0008] Collect perception data: Collect IoT perception data through corresponding devices based on status elements and establish IoT perception database;

[0009] Generate decision rules: Select execution factors based on the user rule library through the collected IoT perception data, and then fuse evidence based on the contribution factors of the execution factors to generate decision rules for controlling devices;

[0010] Modify decision rules: Detect conflicts in the generated decision rules and modify conflicting rules based on scenarios and users;

[0011] Execution decision rules: Generate control commands based on the revised decision rules, control the execution element equipment to complete the corresponding actions, and feedback the execution log;

[0012] Update the user rule base: Match the IoT perception data and execution logs to generate control rules. The rule engine adds the generated control rules to the user rule base as the first rule of the regulation factor, re-sorts the historical rules and updates the user rule list.

[0013] Furthermore, the extraction of system elements comprises the following steps:

[0014] Elements and data collection: Determine the elements of the target smart home control system, including various sensors and smart devices, prepare a questionnaire, and have the expert team and users score the direct impact between the elements;

[0015] Construct the direct impact matrix: round off the collected data and take the mean to obtain the direct impact matrix Z;

[0016] Construct a comprehensive impact matrix: Use the row and maximum value method to normalize the direct impact matrix A to obtain the normalized impact matrix X, and calculate the comprehensive impact matrix T;

[0017] Generate factor influence relationship: List the influence matrix of each state factor on each execution factor through the comprehensive influence matrix T, and classify the state factors based on each execution factor, including contribution factor and failure factor;

[0018] Generate factor distribution relationship: calculate the sum of rows and columns of a single factor in the comprehensive influence matrix T, represent them with vectors R and Q respectively, and generate a factor distribution relationship table according to the calculation results of vectors R and Q.

[0019] Furthermore, the normalized influence matrix X and the comprehensive influence matrix T are calculated as follows:

[0020] ;

[0021] ;

[0022] Among them, Z is the direct influence matrix and Z=[a ij ], a ij It represents the influence degree of factor i on factor j, n is the number of components of the target smart home control system, and I is the unit matrix.

[0023] Furthermore, the calculation formulas of the vectors R and Q are:

[0024] ;

[0025] Among them, R represents the total influence of factor i on other factors, Q represents the total influence of factor i by other factors, and t ij is the element in the comprehensive influence matrix T.

[0026] Furthermore, the generation of decision rules comprises the following steps:

[0027] Establish a user rule library: set trigger conditions according to the user's control requirements and control factors, define the execution factor rules of each control factor in descending order according to the influence degree in the factor distribution relationship table, and generate a rule group by combining the trigger conditions and execution factor rules;

[0028] The control factors are set based on the factors that can be sensed and controlled by the components of the target smart home control system;

[0029] Decision rule fusion: Obtain control factors based on the collected IoT perception data, select the first corresponding rule in the user rule library based on the control factors, filter out execution factors through the first rule, perform evidence fusion on the contribution factors of the execution factors, and obtain the fusion results;

[0030] Generate decision rules: Make a decision by integrating the failure factors and decision rules to obtain the final decision rules for controlling smart devices.

[0031] Furthermore, the decision rule fusion comprises the following steps:

[0032] Define the framework: Define the decision framework U based on the control factors. The decision framework includes all the states of the control factors.

[0033] Collect evidence: Select sensor data to collect evidence based on the contribution factors corresponding to the execution elements;

[0034] Assign a basic trust assignment function: assign a basic probability value to each subset in the decision framework based on the evidence:

[0035] The basic trust allocation function m is a function from the set 2 U A represents any subset of the decision framework U, denoted by A U, and satisfies

[0036] ;

[0037] In the formula, m(A) is the basic trust distribution function of event A, which indicates the degree of trust in A by the evidence;

[0038] Multi-source data fusion: For multi-source data, two pieces of evidence are fused first, and then the fusion result is fused with the third piece of evidence, and so on to complete the fusion of multi-source data evidence. The fusion rules are as follows:

[0039] ;

[0040] Where m1 and m2 are the basic trust allocation functions of the two pieces of evidence respectively, and the focal elements are A i and B j ,in Represents the basic trust allocation function of the conflict focus of event A,

[0041] For the synthesis of n pieces of evidence, the focal elements involved in the conflict are C1, C2, ..., C l And C i ∈P, P is the set of conflicting focal elements, The calculation formula is:

[0042] ;

[0043] In the formula, is the weighting factor, and the calculation formula is:

[0044] ;

[0045] Among them, m j (C i ) represents the trust value of the conflict focal element.

[0046] Furthermore, the modified decision rule comprises the following steps:

[0047] Establish conflict matrix: Generate conflict matrix by clustering the functions of execution elements and judging the same or opposite relationship between each function cluster;

[0048] Conflict judgment: Obtain the device name and action of the execution element through the decision rule, and match the obtained device name and action with the device name and function cluster in the conflict matrix;

[0049] If the function clusters corresponding to the actions between the two rules in the matching result are in opposite relationships, then the judgment decision rules conflict;

[0050] Conflict avoidance mechanism: First, select the decision rules by setting the scene priority, and set the security decision as the highest priority. If the conflicting rules do not involve security, select the corresponding second rule in the user rule library based on the control factors, regenerate the decision rules, and recursively.

[0051] Furthermore, the establishment of the conflict matrix specifically includes the following steps:

[0052] Obtain a function list based on the execution elements of the smart home control system, calculate the phrase vector of each functional knowledge phrase based on the function list and combine the word vector tool to cluster the execution element functions;

[0053] The semantic dictionary is introduced to split the functional knowledge items with opposite semantics in the same cluster of the preliminary clustering results into subcategories, ensuring that each subcategory does not contain functional phrases with opposite semantics, thereby obtaining the list of actuator functional clusters.

[0054] Combined with the semantic dictionary, the function cluster lists of different execution elements are compared pairwise to obtain the correlation between the function clusters, and it is judged whether the effects of each function cluster between the execution elements are the same or opposite relationship. The judgment results of the same or opposite relationship are stored in the form of a conflict matrix.

[0055] A second aspect of the present disclosure provides a smart home control system based on IoT perception, which is applied to a smart home control method based on IoT perception as described above, including a decision rule generation module, a decision rule fusion module and a decision rule correction module;

[0056] The decision rule generation module is used to select execution factors based on the user rule library through the collected IoT perception data, and then perform evidence fusion based on the contribution factors of the execution factors to generate decision rules for controlling the device;

[0057] The decision rule fusion module is used to obtain the control factor according to the collected IoT perception data, select the first corresponding rule in the user rule library according to the control factor, screen out the execution element through the first rule, and perform evidence fusion on the contribution factor of the execution element;

[0058] The decision rule correction module is used to perform conflict detection on the generated decision rules and correct the conflicting rules based on the scenario and the user.

[0059] As a preferred technical solution of the present invention, it also includes a system component analysis module, an IoT perception data acquisition module and a decision rule control execution module;

[0060] The system component analysis module is used to determine the component elements according to the target smart home control system, analyze the influence degree between the elements, and generate the element distribution relationship;

[0061] The IoT perception data collection module is used to collect IoT perception data through corresponding devices based on state elements and establish an IoT perception database. The steps of performing the collection include:

[0062] Collect real-time and periodic information of the home environment through existing equipment, where real-time information includes switch quantity, analog quantity, internal quantity, and parameter information; periodic information includes data automatically received from the equipment through the set collection cycle;

[0063] Formulate a time-series-based status element information collection format, including time, device number, device name, device data, device status, and device category, and build an IoT perception database;

[0064] The decision rule control execution module is used to generate a control command according to the decision rule, control the execution element device to complete the corresponding action, and feed back the execution log.

[0065] The beneficial effects of the present invention are:

[0066] The present invention first analyzes the constituent elements of the target smart home control system to obtain the distribution relationship of the constituent elements, then builds a user rule library based on the distribution relationship of the constituent elements, generates rules according to the order of influence, and gives priority to the execution devices preferred by the user, so as to better meet the user's control intention. At the same time, the contribution factors screened out by the execution elements are fused, which reduces error accumulation, improves the computational efficiency of evidence fusion, and makes the decision rule fusion more accurate.

[0067] The present invention adopts evidence theory to fuse contribution factors, thereby improving the accuracy and stability of the Internet of Things perception data control system and reducing decision-making risks. At the same time, the final decision generation is based on the decision rule fusion result, and the failure factor is also considered. The user's prerequisites are added to the measurement of decision rule generation, making the decision rules more practical and effective. Finally, conflict detection and conflict avoidance are performed on the generated decision rules, which improves the availability and security of the system. Secondly, the execution elements are reselected to meet the needs of user control factors, thereby improving the system flexibility and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0069] Figure 1 A schematic diagram of the steps of a smart home control method based on IoT perception provided by an embodiment of the present invention;

[0070] Figure 2 A schematic diagram of the steps of extracting system elements provided in an embodiment of the present invention;

[0071] Figure 3 A schematic diagram of the steps for generating decision rules provided in an embodiment of the present invention;

[0072] Figure 4A schematic diagram of the steps of modifying a decision rule provided in an embodiment of the present invention;

[0073] Figure 5 A schematic diagram of the structure of a smart home control system based on IoT perception provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0074] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0075] At least one embodiment of the present disclosure provides a smart home control method based on IoT perception, such as Figure 1 As shown, the following steps are included:

[0076] S1. Extract system elements: Determine the constituent elements according to the target smart home control system, analyze the influence degree between the elements, and generate element distribution relationships, such as Figure 2 As shown, the specific steps include:

[0077] S11. Elements and data collection: Determine the constituent elements according to the target smart home control system, including various sensors and smart devices, prepare a questionnaire, and have the expert team and users score the direct impact between the elements. The judgment standard is 0 to 4, which respectively represents no impact, low impact, medium impact, high impact and very high impact.

[0078] In the above process, the scores of target users are added on the basis of the expert team's scores, that is, the actual needs and preferences of users themselves are integrated, so that the later system decisions are more in line with the user's behavior habits, thereby improving the practicality of system control and user satisfaction. For example, in temperature adjustment, when paying attention to whether there is a human body in the home environment, if the user prefers the temperature adjustment effect of the environment, the human infrared sensor has a lower degree of influence on the start and stop of the air conditioner. If the user prefers energy-saving control, the human infrared sensor has a higher degree of influence on the start and stop of the air conditioner.

[0079] S12. Construct the direct impact matrix: round off the collected data and take the average to get the direct impact matrix Z=[a ij ], a ij Indicates the degree of influence of factor i on factor j.

[0080] S13, constructing a comprehensive influence matrix: using the row and maximum value method, normalize the direct influence matrix A to obtain the normalized influence matrix X, and calculate the comprehensive influence matrix T. That is, select the maximum value of all the row sums in Z so that each element a in Z ijDivide them by the row sum maximum value to get the normalized influence matrix X, and then calculate the comprehensive influence matrix T through the normalized influence matrix X. The calculation formula is as follows:

[0081] ;

[0082] ;

[0083] Among them, n is the number of components of the target smart home control system, I is the unit matrix, and T is the comprehensive superposition matrix of the direct and indirect influence relationships of each component, which is used to determine the degree of influence of each component on other elements.

[0084] S14. Generate factor influence relationships: List the influence matrix of each state factor on each execution factor through the comprehensive influence matrix T, and classify the state factors based on each execution factor, including contribution factors and failure factors.

[0085] It is understandable that the contribution factor and failure factor obtained through the comprehensive influence matrix T are used to judge the influence of a certain state element on a certain execution element. For example, when the start and stop of the execution element air conditioner does not consider whether there is a human body, the state element human infrared sensor has no contribution to the start and stop of the air conditioner. At this time, the state element human infrared sensor is the failure factor; the state elements include various sensors and measuring instruments such as smart electricity meters and smart water meters, and the execution elements include smart devices that execute command actions.

[0086] S15. Generate factor distribution relationship: Calculate the sum of the rows and columns of a single factor in the comprehensive influence matrix T, represented by vectors R and Q respectively. The calculation formula is:

[0087] ;

[0088] Among them, R represents the total influence of factor i on other factors, Q represents the total influence of factor i by other factors, and t ij is the element in the comprehensive influence matrix T.

[0089] Generate an element distribution relationship table based on the calculation results of vectors R and Q.

[0090] The Decision Analysis Laboratory (DEMATEL) is a methodology that can analyze complex problems in the real world. It analyzes the causal relationship between the components of the system through matrix or structural modeling that visualizes complex causal relationships. The DEMATEL method can determine the logical relationship between the elements in the system, calculate the degree of influence of a single element on other elements in the system with the help of various influence matrices, and transform the interdependence between elements into causal groups, and finally discover the key factors of complex structural systems through finite relationship diagrams.

[0091] In this embodiment, in the factor distribution relationship table, the weights of different household products in the target user experience can be analyzed through the values ​​of influence R and influence Q. By analyzing the influence and influence of different sensors, the behavioral habit bias characteristics of different users can be obtained for later decision-making and execution factor selection. For example, for energy-saving users, smart meters and human infrared sensors have a greater influence, that is, the execution factor devices that users prefer to enable for adjusting a certain control factor.

[0092] S2. Collecting perception data: Based on the state elements, IoT perception data is collected through corresponding devices to establish an IoT perception database, which specifically includes the following steps:

[0093] S21. Collect real-time and periodic information of the home environment through existing equipment, where real-time information includes various types of information such as switch quantity, analog quantity, internal quantity, parameters, etc.; periodic information includes data automatically received from the device through a set collection cycle.

[0094] S22. Develop a time series-based status element information collection format, including time, device number, device name, device data, device status, and device category, and build an IoT perception database.

[0095] S3, Generate decision rules: Select execution factors based on the user rule library through the collected IoT perception data, and then fuse evidence based on the contribution factors of the execution factors to generate decision rules for controlling devices, such as Figure 3 As shown, the specific steps include:

[0096] S31, establish a user rule library: set trigger conditions according to the user's control requirements and control factors, define the execution factor rules of each control factor in descending order according to the influence degree in the factor distribution relationship table, and generate a rule group by combining the trigger conditions and the execution factor rules;

[0097] The control factors are set based on the factors that can be sensed and controlled by the components of the target smart home control system, such as temperature, humidity, etc.

[0098] It should be noted that the execution element rules of each control factor are generated in order of influence. For example, when the control factor is temperature, the following rule group is generated:

[0099] Rule 1: When the temperature is too high, turn on the air conditioning cooling mode;

[0100] Rule 2: When the temperature is too high, turn on the fan;

[0101] In this embodiment, according to the rules generated in order of influence, the execution device preferred by the user is selected first, so as to better meet the control intention of the user.

[0102] S32, decision rule fusion: obtain the control factors according to the collected IoT perception data, select the first corresponding rule in the user rule library according to the control factors, filter out the execution elements through the first rule, perform evidence fusion on the contribution factors of the execution elements, and obtain the fusion results. Specifically, it includes the following steps:

[0103] S321, define framework: define a decision framework U according to the control factors, the decision framework includes all states of the control factors, such as in the temperature control system, the decision framework includes {temperature is too high, temperature is appropriate, temperature is too low};

[0104] S322, Collect evidence: Select sensor data to collect evidence based on the contribution factors corresponding to the execution elements to provide support for the elements in the decision framework;

[0105] S323. Assign a basic trust assignment function: assign a basic probability value to each subset in the decision framework based on the evidence:

[0106] The basic trust allocation function m is a function from the set 2 U A represents any subset of the decision framework U, denoted by A U, and satisfies

[0107] ;

[0108] Where m(A) is the basic trust distribution function of event A, which indicates the degree of trust in A from the evidence.

[0109] S324, Multi-source data fusion: For multi-source data, two pieces of evidence are fused first, and then the fusion result is fused with the third piece of evidence, and so on to complete the fusion of multi-source data evidence. The fusion rules are as follows:

[0110] ;

[0111] Where m1 and m2 are the basic trust allocation functions of the two pieces of evidence respectively, and the focal elements are A i and B j ,in Represents the basic trust allocation function of the conflict focus of event A,

[0112] For the synthesis of n pieces of evidence, the focal elements involved in the conflict are C1, C2, ..., C l And C i ∈P, P is the set of conflicting focal elements, The calculation formula is:

[0113] ;

[0114] In the formula, is the weighting factor, and the calculation formula is:

[0115] ;

[0116] Among them, m j (C i ) represents the trust value of the conflict focal element.

[0117] DS (Dempster-Shafer) evidence theory originates from the upper probability and lower probability derived from multi-value mapping. Later, by establishing a corresponding relationship between propositions and sets, the uncertainty problem of propositions is transformed into the uncertainty problem of sets, which satisfies the situation weaker than probability theory and forms a set of mathematical theories about evidence reasoning. In this embodiment, DS evidence theory is used to fuse multiple data, and the fused evidence is screened by contribution factors instead of fusing all sensor data in the smart home control system, which reduces error accumulation, improves the computational efficiency of evidence fusion, and makes the decision rule fusion more accurate.

[0118] S33, generating decision rules: making a decision by integrating the failure factors and the decision rules, and finally obtaining the decision rules for controlling the intelligent device.

[0119] It should be noted that the contribution value of the failure factor to the control factor is 0, but it cannot be directly discarded in the final decision of the control device. For example, in the temperature control system, the contribution value of the human infrared sensor to the temperature judgment is 0, and the final decision rule fusion result is "when the temperature is too high, turn on the air conditioning cooling mode". In the actual execution process, based on the user's preference, it is only necessary to turn on the corresponding equipment in the environmental range where the human body exists. At this time, the failure factor-human infrared sensor needs to be considered for the decision rule. Therefore, in this embodiment, the failure factor is considered in the process of generating the decision rule, so that the final decision is more in line with the user's habits and the action execution is more reasonable.

[0120] S4, modify decision rules: perform conflict detection on the generated decision rules, and modify the conflicting rules based on the scenario and user, such as Figure 4 As shown, the following steps are included:

[0121] S41, establishing a conflict matrix: generating a conflict matrix by clustering the functions of the execution elements and determining the same or opposite relationship between the function clusters, specifically including the following steps:

[0122] Obtain a function list based on the execution elements of the smart home control system, calculate the phrase vector of each functional knowledge phrase based on the function list and combine the word vector tool to cluster the execution element functions;

[0123] The semantic dictionary is introduced to split the functional knowledge items with opposite semantics in the same cluster of the preliminary clustering results into subcategories, ensuring that each subcategory does not contain functional phrases with opposite semantics, thereby obtaining the list of actuator functional clusters.

[0124] Combined with the semantic dictionary, the function cluster lists of different execution elements are compared pairwise to obtain the correlation between the function clusters, and it is judged whether the effects of each function cluster between the execution elements are the same or opposite relationship. The judgment results of the same or opposite relationship are stored in the form of a conflict matrix.

[0125] S42, conflict judgment: obtaining the device name and action of the execution element through the decision rule, and matching the obtained device name and action with the device name and function cluster in the conflict matrix;

[0126] If the function clusters corresponding to the actions between the two rules in the matching result are in opposite relationships, then the judgment decision rules conflict;

[0127] S43, conflict avoidance mechanism: First, select the decision rules by setting the scene priority, and set the security decision as the highest priority. If the conflicting rules do not involve security, select the corresponding second rule in the user rule library based on the control factors, regenerate the decision rules, and recursively.

[0128] In this embodiment, the conflict avoidance mechanism first avoids conflicts between user demand scenarios based on priority, thereby improving the availability and security of the system. Secondly, it meets the needs of user control factors by reselecting execution elements, thereby improving system flexibility and user experience.

[0129] S5. Execute decision rules: Generate control commands according to the revised decision rules, control the execution element devices to complete corresponding actions, and feedback the execution log;

[0130] S6. Update the user rule base: Match the IoT perception data and execution logs to generate control rules. The rule engine adds the generated control rules to the user rule base as the first rule of the regulation factor, re-sorts the historical rules and updates the user rule list.

[0131] It is understandable that by taking the executed control rule as the first rule of the regulation factor, the complexity of subsequent decision generation is reduced, and the response time of the system is reduced. At the same time, the system continuously updates the rules during operation, making the control decision more accurate and efficient.

[0132] At least one embodiment of the present disclosure further provides a smart home control system based on IoT perception, such as Figure 5As shown, it includes a system component analysis module, an IoT perception data acquisition module, a decision rule generation module, a decision rule fusion module, a decision rule correction module and a decision rule control execution module.

[0133] The system component analysis module is used to determine the component elements according to the target smart home control system, analyze the influence degree between the elements, and generate the element distribution relationship.

[0134] The IoT perception data collection module is used to collect IoT perception data through corresponding devices based on state elements and establish an IoT perception database. The steps of performing the collection include:

[0135] Real-time and periodic information of the home environment is collected through existing equipment. Real-time information includes various types of information such as switching quantity, analog quantity, internal quantity, parameters, etc.; periodic information includes data automatically received from the equipment through the set collection cycle (such as temperature, humidity, power consumption, etc.).

[0136] Formulate a time series-based status element information collection format, including time, device number, device name, device data, device status and device category, and build an IoT perception database.

[0137] The decision rule generation module is used to select execution factors based on the user rule library through the collected IoT perception data, and then perform evidence fusion based on the contribution factors of the execution factors to generate decision rules for controlling the device.

[0138] The decision rule fusion module is used to obtain control factors based on the collected IoT perception data, select the corresponding first rule in the user rule library based on the control factors, screen out execution elements through the first rule, and perform evidence fusion on the contribution factors of the execution elements.

[0139] The decision rule correction module is used to perform conflict detection on the generated decision rules and correct the conflicting rules based on the scenario and the user.

[0140] The decision rule control execution module is used to generate a control command according to the decision rule, control the execution element device to complete the corresponding action, and feed back the execution log.

[0141] The present invention first analyzes the constituent elements of the target smart home control system to obtain the distribution relationship of the constituent elements, then builds a user rule library based on the distribution relationship of the constituent elements, generates rules according to the order of influence, and gives priority to the execution devices preferred by the user, so as to better meet the user's control intention. At the same time, the contribution factors screened out by the execution elements are fused, which reduces error accumulation, improves the computational efficiency of evidence fusion, and makes the decision rule fusion more accurate.

[0142] The present invention adopts evidence theory to fuse contribution factors, thereby improving the accuracy and stability of the Internet of Things perception data control system and reducing decision-making risks. At the same time, the final decision generation is based on the decision rule fusion result, and the failure factor is also considered. The user's prerequisites are added to the measurement of decision rule generation, making the decision rules more practical and effective. Finally, conflict detection and conflict avoidance are performed on the generated decision rules, which improves the availability and security of the system. Secondly, the execution elements are reselected to meet the needs of user control factors, thereby improving the system flexibility and user experience.

[0143] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A smart home control method based on IoT perception, characterized in that: The following steps are involved: Extract system elements: determine the constituent elements according to the target smart home control system, analyze the influence between the elements, and generate element distribution relationships; Collect perception data: Collect IoT perception data through corresponding devices based on status elements and establish IoT perception database; Generate decision rules: Select execution factors based on the user rule library through the collected IoT perception data, and then fuse evidence based on the contribution factors of the execution factors to generate decision rules for controlling devices; Modify decision rules: Detect conflicts in the generated decision rules and modify conflicting rules based on scenarios and users; Execution decision rules: Generate control commands based on the revised decision rules, control the execution element equipment to complete the corresponding actions, and feedback the execution log; Update the user rule base: Match IoT sensing data and execution logs to generate control rules. The rule engine adds the generated control rules to the user rule base as the first rule of the regulation factor, reorders the historical rules and updates the user rule list. The generation decision rule comprises the following steps: Establish a user rule library: set trigger conditions according to the user's control requirements and control factors, define the execution factor rules of each control factor in descending order according to the influence degree in the factor distribution relationship table, and generate a rule group by combining the trigger conditions and execution factor rules; The control factors are set based on the factors that can be sensed and controlled by the components of the target smart home control system; Decision rule fusion: Obtain control factors based on the collected IoT perception data, select the first corresponding rule in the user rule library based on the control factors, filter out execution factors through the first rule, perform evidence fusion on the contribution factors of the execution factors, and obtain the fusion results; Generate decision rules: Make a decision by integrating the failure factors and decision rules, and finally obtain the decision rules used to control the smart device; The modified decision rule comprises the following steps: Establish conflict matrix: Generate conflict matrix by clustering the functions of execution elements and judging the same or opposite relationship between each function cluster; Conflict judgment: Obtain the device name and action of the execution element through the decision rule, and match the obtained device name and action with the device name and function cluster in the conflict matrix; If the function clusters corresponding to the actions between the two rules in the matching result are in opposite relationships, then the judgment decision rules conflict; Conflict avoidance mechanism: First, select the decision rules by setting the scene priority, and set the security decision as the highest priority. If the conflicting rules do not involve security, select the corresponding second rule in the user rule library based on the control factors, regenerate the decision rules, and recursively.

2. The smart home control method based on IoT perception according to claim 1, characterized in that: The extraction system elements include the following steps: Elements and data collection: Determine the elements of the target smart home control system, including various sensors and smart devices, prepare a questionnaire, and have the expert team and users score the direct impact between the elements; Construct the direct impact matrix: round off the collected data and take the mean to obtain the direct impact matrix Z; Construct a comprehensive impact matrix: Use the row and maximum value method to normalize the direct impact matrix Z to obtain the normalized impact matrix X, and calculate the comprehensive impact matrix T; Generate factor influence relationship: List the influence matrix of each state factor on each execution factor through the comprehensive influence matrix T, and classify the state factors based on each execution factor, including contribution factor and failure factor; Generate factor distribution relationship: calculate the sum of rows and columns of a single factor in the comprehensive influence matrix T, represent them with vectors R and Q respectively, and generate a factor distribution relationship table according to the calculation results of vectors R and Q.

3. The smart home control method based on IoT perception according to claim 2 is characterized in that: The normalized influence matrix X and the comprehensive influence matrix T are calculated as follows: ； ； Among them, Z is the direct influence matrix and Z=[a ij ], a ij It represents the influence degree of factor i on factor j, n is the number of components of the target smart home control system, and I is the unit matrix.

4. The smart home control method based on IoT perception according to claim 3 is characterized by: The calculation formulas of the vectors R and Q are: ； Among them, R represents the total influence of factor i on other factors, Q represents the total influence of factor i by other factors, and t ij is the element in the comprehensive influence matrix T.

5. The smart home control method based on IoT perception according to claim 1, characterized in that: The decision rule fusion comprises the following steps: Define the framework: Define the decision framework U based on the control factors. The decision framework includes all the states of the control factors. Collect evidence: Select sensor data to collect evidence based on the contribution factors corresponding to the execution elements; Assign a basic trust assignment function: assign a basic probability value to each subset in the decision framework based on the evidence: The basic trust allocation function m is a function from the set 2 U A represents any subset of the decision framework U, denoted by A. U, and satisfies ； In the formula, m(A) is the basic trust distribution function of event A, which indicates the degree of trust in A by the evidence; Multi-source data fusion: For multi-source data, two pieces of evidence are fused first, and then the fusion result is fused with the third piece of evidence, and so on to complete the fusion of multi-source data evidence. The fusion rules are as follows: ； Where m1 and m2 are the basic trust allocation functions of the two pieces of evidence respectively, and the focal elements are A i and B j ,in Represents the basic trust allocation function of the conflict focus of event A, For the synthesis of n pieces of evidence, the focal elements involved in the conflict are C1, C2, ..., C l And C i ∈P, P is the set of conflicting focal elements, The calculation formula is: ； In the formula, is the weighting factor, and the calculation formula is: ； Among them, m j (C i ) represents the trust value of the conflict focal element.

6. The method for controlling a smart home based on IoT perception according to claim 1, characterized in that: The establishment of the conflict matrix specifically includes the following steps: Obtain a function list based on the execution elements of the smart home control system, calculate the phrase vector of each functional knowledge phrase based on the function list and combine the word vector tool to cluster the execution element functions; The semantic dictionary is introduced to split the functional knowledge items with opposite semantics in the same cluster of the preliminary clustering results into subcategories, ensuring that each subcategory does not contain functional phrases with opposite semantics, thereby obtaining the list of actuator functional clusters. Combined with the semantic dictionary, the function cluster lists of different execution elements are compared pairwise to obtain the correlation between the function clusters, and it is judged whether the effects of each function cluster between the execution elements are the same or opposite relationship. The judgment results of the same or opposite relationship are stored in the form of a conflict matrix.

7. A smart home control system based on Internet of Things perception, applied to a smart home control method based on Internet of Things perception as claimed in any one of claims 1 to 6, characterized in that: It includes a decision rule generation module, a decision rule fusion module and a decision rule correction module; The decision rule generation module is used to select execution factors based on the user rule library through the collected IoT perception data, and then perform evidence fusion based on the contribution factors of the execution factors to generate decision rules for controlling the device; The decision rule fusion module is used to obtain the control factor according to the collected IoT perception data, select the first corresponding rule in the user rule library according to the control factor, screen out the execution element through the first rule, and perform evidence fusion on the contribution factor of the execution element; The decision rule correction module is used to perform conflict detection on the generated decision rules and correct the conflicting rules based on the scenario and the user.

8. The smart home control system based on IoT perception according to claim 7 is characterized in that: It also includes a system component analysis module, an IoT perception data acquisition module, and a decision rule control execution module; The system component analysis module is used to determine the component elements according to the target smart home control system, analyze the influence degree between the elements, and generate the element distribution relationship; The IoT perception data collection module is used to collect IoT perception data through corresponding devices based on state elements and establish an IoT perception database. The steps of performing the collection include: Collect real-time and periodic information of the home environment through existing equipment, where real-time information includes switch quantity, analog quantity, internal quantity, and parameter information; The periodic information includes data automatically received from the device and uploaded through the set collection cycle; Formulate a time-series-based status element information collection format, including time, device number, device name, device data, device status, and device category, and build an IoT perception database; The decision rule control execution module is used to generate a control command according to the decision rule, control the execution element device to complete the corresponding action, and feed back the execution log.

Images