Intelligent door and window control method and system

Abstract

The invention provides an intelligent door and window control method and system, and relates to the technical field of intelligent door and window regulation, and the method comprises the steps: adding at least one probability adjustment label for each control strategy; setting an initial executable probability for the control strategy of each intelligent door and window, collecting real-time data including real-time environment data, real-time user data and the real-time state of each intelligent door and window, and updating the initial executable probability of the control strategy of each intelligent door and window based on the real-time data, and recording the updated initial executable probability as a real-time executable probability, and executing the control strategy of which the real-time executable probability is greater than a preset probability threshold value. According to the method, the executable probability of each control strategy can be evaluated according to the real-time environment data and the real-time user data, and intelligent management and optimal control of the intelligent door and window system are realized.

Description

Technical Field

[0001] The present application relates to the technical field of regulating smart doors and windows, and in particular to a smart door and window control method and system. Background Art

[0002] With the rapid advancement of technology and the continuous improvement of people's living standards, smart home systems have gradually become an important part of modern families. Smart homes aim to provide users with a more convenient, comfortable, safe, and energy-efficient living environment by integrating various intelligent devices, such as smart doors and windows, smart lighting, and smart appliances. Smart doors and windows, as a key component of smart home systems, not only provide basic opening and closing functions but also intelligently adjust to different environments and user needs, such as automatic opening and closing and adjustable opening, thereby improving users' quality of life.

[0003] A Chinese invention patent with an application publication date of February 13, 2024, provides a door and window control method and system for a smart home. The patent obtains door and window control instructions by calculating the time difference eigenvector and temperature difference eigenvector of indoor and outdoor light intensity.

[0004] However, the real environment is constantly changing, and user needs can shift at any time. Originally set door and window control instructions may not be suitable for execution when a user has a special need (such as a quiet environment for rest). Due to the lack of dynamic evaluation and adjustment of control strategy execution probabilities, the aforementioned patent cannot flexibly select a more appropriate control strategy based on real-time conditions, thus affecting the control effectiveness and user experience of smart doors and windows. Summary of the Invention

[0005] In order to evaluate the executable probability of each control strategy based on real-time environmental data and real-time user data and improve user satisfaction, the present application provides an intelligent door and window control method and system.

[0006] In the first aspect, the present application provides an intelligent door and window control method, which adopts the following technical solutions: An intelligent door and window control method comprises the following steps: Establish a database storing the control strategy of each smart door and window, and add at least one probability adjustment tag to each control strategy, wherein the control strategy includes the opening mode, closing mode, and opening degree of each smart door and window, and the probability adjustment tag includes an environmental tag and a user tag; An initial executable probability is set for the control strategy of each smart door and window, and real-time data is collected. The real-time data includes real-time environmental data, real-time user data, and the real-time status of each smart door and window. The initial executable probability of the control strategy of each smart door and window is updated based on the real-time data, and the updated initial executable probability is recorded as the real-time executable probability. The control strategy with a real-time executable probability greater than a preset probability threshold is executed.

[0007] This application establishes a database storing control policies, enabling centralized storage and management of all control policies, facilitating rapid retrieval and recall. Subsequently, this application adds environmental and user tags to each control policy, enabling the initial executable probability of each control policy to be updated based on different environmental and user data. This application then sets an initial executable probability for each control policy, providing a benchmark for subsequent real-time probability updates, enabling the control system to dynamically adjust the execution probability of the control policy based on actual conditions. Subsequently, this application collects real-time environmental data, real-time user data, and the real-time status of each smart door and window to understand changes in the current environment and user needs in real time. Based on this real-time data, the control system can update the initial executable probability of each control policy to obtain a real-time executable probability. The real-time data collection and probability updates enable the control system to adapt to changing environments and user needs, improving its adaptability and robustness. Subsequently, this application only executes control policies with a real-time executable probability greater than a preset probability threshold, ensuring that only those control policies with high execution value given the current environment and user needs are executed, thereby reducing resource waste caused by executing inefficient or ineffective control policies. At the same time, this application improves user satisfaction by using real-time data to update the initial executable probability of the control strategy and screening the control strategy based on the real-time executable probability.

[0008] Optionally, the method further includes: Based on the probability adjustment label, a profit function is constructed, and the control strategy with a real-time executable probability greater than the preset probability threshold is recorded as the target strategy. It is determined whether there are conflicting target strategies. If so, calculate the payoff value of each target strategy according to the payoff function, integrate the payoff values ​​of all target strategies into a payoff matrix, use the Nash equilibrium algorithm to obtain the target strategy combination in the payoff matrix when the payoff reaches an equilibrium state, retain the target strategies included in the target strategy combination, and delete the remaining target strategies; If not, no action will be taken.

[0009] The present application constructs a profit function based on probability adjustment labels, so that the execution effect of each control strategy can be quantitatively evaluated. The present application also records the control strategy whose real-time executable probability is greater than the preset probability threshold as the target strategy, and determines whether there are conflicting target strategies, which helps to timely discover and deal with potential control conflicts, and try to avoid waste of resources or poor control effects caused by the execution of conflicting strategies. In the case of conflicting target strategies, the present application calculates the profit value of each target strategy according to the profit function, integrates the profit values ​​of all target strategies into a profit matrix, and uses the Nash equilibrium algorithm to obtain the target strategy combination in the profit matrix when the profit reaches an equilibrium state, retains the target strategy combination that maximizes the overall profit, and deletes the remaining conflicting strategies. By applying the Nash equilibrium algorithm, the present application can effectively resolve conflicts between conflicting target strategies, so that the executed control strategy can maximize the overall profit.

[0010] Optionally, the method further includes: Determine whether the number of conflicting target strategies is greater than two, If so, record the conflicting target strategies as conflicting strategies, map all conflicting strategies to quantum bits, construct a superposition state based on the quantum bits, construct a quantum state based on the superposition state, calculate the expected return of the quantum state using a quantum algorithm, collapse the quantum state corresponding to the maximum expected return into a conflicting strategy combination, and update the conflicting strategies contained in the conflicting strategy combination as the target strategy; If not, no action will be taken.

[0011] If the number of conflicting target strategies is greater than two, it proves that there are multiple conflicting control strategies for a certain smart door and window. At this time, the conflicting strategies are mapped to quantum bits, and a superposition state is constructed based on these quantum bits, realizing the representation of the conflicting strategies under the quantum computing framework. The conflicting strategy processing under the quantum computing framework can take advantage of the parallelism and advantages of quantum computing to improve processing efficiency, especially when dealing with a large number of conflicting strategies. By adopting the above scheme, the present application can take advantage of the parallelism and advantages of quantum computing to deal with complex conflict problems. Subsequently, the present application uses a quantum algorithm to calculate the expected return of the quantum state, providing a quantitative basis for the combination optimization of the conflicting strategies, and then evaluating the execution effect of various conflicting strategy combinations. Subsequently, the present application collapses the quantum state corresponding to the maximum expected return into a conflicting strategy combination, realizing the conversion from the quantum state to the specific conflicting strategy combination, and then updates the conflicting strategies contained in the conflicting strategy combination to the target strategy. During execution, the optimal conflicting strategy combination can be selected, thereby improving the overall control effect. The present application improves the efficiency and accuracy of handling complex conflicting strategies by judging the number of conflicting strategies, quantum mapping and superposition state construction, quantum state expected return calculation, and quantum state collapse and strategy update.

[0012] Optionally, the method further includes: The conflicting policies not included in the deleted target policy and the conflicting policy combination are recorded as deleted policies, and the average value of the real-time executable probabilities of the deleted policies is calculated and recorded as the first data; the conflicting policies included in the retained target policy and the conflicting policy combination are recorded as retained policies, and the average value of the real-time executable probabilities of the retained policies is calculated and recorded as the second data; Determine whether the first data is greater than the second data. If so, issue an alarm signal; if not, do nothing.

[0013] This application records the conflicting strategies not included in the deleted target strategy and conflict strategy combination as deleted strategies, and records the conflicting strategies included in the retained target strategy and conflict strategy combination as retained strategies. The application then calculates the average value (first data) of the real-time executable probabilities of the deleted strategies and the average value (second data) of the real-time executable probabilities of the retained strategies. It then determines whether the first data is greater than the second data to detect whether an anomaly exists. If the first data (the average execution probability of the deleted strategies) is greater than the second data (the average execution probability of the retained strategies), this means that the deleted control strategy has a higher execution value, or that the retained control strategy has a poor execution effect. In this case, an alarm signal needs to be issued to the relevant personnel to inform them of the anomaly.

[0014] Optionally, after the step of issuing an alarm signal, the method further comprises: gradually reducing the initial executable probability of the deleted policy according to a preset step size until the first data is no greater than the second data.

[0015] Optionally, the method further includes: Calculate the partial derivative of the profit function with respect to each probability adjustment label, input the real-time environment data and real-time user data into the partial derivative, obtain the partial derivative value, and integrate the partial derivative value into the Jacobian matrix; Based on the elements in the Jacobian matrix, the sensitivity of the benefit function with respect to each probability adjustment label is obtained, and each sensitivity is normalized respectively to obtain the normalized result. The normalized result is used as the probability compensation value of the corresponding real-time environmental data or the corresponding real-time user data, and the sum of the real-time executable probability and the probability compensation value is used as the new real-time executable probability.

[0016] This application calculates the partial derivative of the reward function with respect to each probability-adjusted label to determine its sensitivity. For example, if the partial derivative of a particular environmental label (such as temperature) is large, it indicates that temperature changes significantly impact the reward function, and the application should prioritize that label. Subsequently, the application integrates the partial derivative values ​​into a Jacobian matrix, visually displaying the sensitivity of the reward function to all probability-adjusted labels. The application then normalizes each sensitivity individually and uses the normalized result as a probability compensation value for the corresponding real-time environmental data or real-time user data. This allows the application to adjust the real-time executable probability based on the sensitivity. By dynamically adjusting the real-time executable probability, the application can better align with users' actual usage habits and needs, improving user satisfaction and comfort. For example, if a user habitually opens doors and windows during specific time periods, the application can automatically increase the real-time executable probability of the opening policy during those time periods. Using the normalized sensitivity value directly as the probability compensation value reflects the extent to which the real-time environmental data or real-time user data adjusts the real-time executable probability, improving the rationality and effectiveness of the probability compensation value.

[0017] Optionally, the method further includes: Determine whether the sensitivity is a non-negative number, If so, no action will be taken; If not, the absolute value of the sensitivity that is less than zero is recorded as the new sensitivity.

[0018] Optionally, if the number of environment class labels and user class labels is different, the number zero is used to fill the empty positions in the Jacobian matrix.

[0019] Optionally, the method further includes: Obtain historical user data, obtain the user's adjustment habits and adjustment timestamps for each smart door and window based on the historical user data, and increase the new real-time executable probability of the control strategy corresponding to the adjustment habit within a preset time before the adjustment timestamp.

[0020] In the second aspect, the present application provides an intelligent door and window control system, which adopts the following technical solutions: An intelligent door and window control system, comprising: a memory and a processor, The memory stores a computer-readable storage medium and training data; When the processor calls the program code in the memory, the steps of the method described in the first aspect are executed.

[0021] In summary, the beneficial technical effects of this application are as follows: This application dynamically updates the executable probability of each control strategy through real-time data, making the execution of each control strategy more consistent with the actual situation, improving environmental adaptability and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flow chart of Example 1 of the present application; Figure 2 This is a flow chart of Example 2 of the present application; Figure 3 This is a flowchart of Example 3 of the present application. DETAILED DESCRIPTION

[0023] The following combination Figures 1 to 3 This application is described in further detail.

[0024] Example 1: This example discloses a smart door and window control method, referring to Figure 1 The method includes: S11 processing strategy, S12 executing strategy. This embodiment first establishes a database storing each smart door and window control strategy and adds at least one probability adjustment tag (covering environmental and user tags) to each control strategy. Subsequently, this embodiment sets an initial executable probability for each control strategy and collects real-time data including real-time environmental data, user data, and door and window status to dynamically update the initial executable probability of the control strategy, obtain a real-time executable probability, and execute the control strategy whose real-time executable probability exceeds a preset threshold. This embodiment includes the following steps: S11 processing strategy, establishing a database that stores the control strategy of each smart door and window, each smart door and window contains at least one control strategy, and the database records the control strategy of each smart door and window in detail, including the opening method of each smart door and window (such as inward opening / outward opening of casement windows, sliding opening of sliding windows, rotation opening of top-hung windows, rotation opening of bottom-hung windows, folding opening of folding windows, sliding up and down opening of lifting windows, center rotation opening of rotating windows, etc.), closing method (such as inward closing / outward closing of casement windows, sliding closing of sliding windows, rotation closing of top-hung windows, rotation closing of bottom-hung windows, folding closing of folding windows, sliding up and down closing of lifting windows, center rotation closing of rotating windows, etc.) and opening degree (such as fully open, half open, slightly open, etc. specific angles or proportions).

[0025] At least one probability adjustment tag is added to each control strategy to dynamically adjust the strategy execution probability according to different environments and user needs. The probability adjustment tags include environment tags and user tags.

[0026] Environmental tags include environmental factors such as temperature, humidity, and air quality. Environmental tags will affect the opening or closing decision of smart doors and windows, as well as the opening degree of each smart door and window.

[0027] User-class tags include user preferences, daily behavior patterns, family member composition, etc. User-class tags reflect the user's personalized needs for the use of smart doors and windows.

[0028] By adding a probability adjustment tag to the control strategy of each smart door and window, this embodiment can dynamically adjust the execution probability of each smart door and window control strategy based on real-time environmental data, real-time user data and the real-time status of each smart door and window, thereby achieving more intelligent and personalized control.

[0029] S12 executes the strategy, setting an initial executable probability for each smart door and window control strategy. The initial executable probability reflects the possibility of executing the control strategy under the real-time state of the smart door and window. The initial executable probability can be set based on historical data, expert experience or system default values.

[0030] The solution for setting the initial executable probability of a control strategy based on historical data is to calculate the initial executable probability of the strategy by analyzing the usage frequency and time distribution of each control strategy over a period of time (which can be a month or a week), as well as the frequency and time of user calls to the control strategy.

[0031] For example, the probability that a user calls a control strategy to open the bedroom window to 40° at 7:00 every morning in the past 30 days is 0.8, and the probability that a non-user calls this control strategy is 0. Then the initial executable probability of the control strategy can be set to (0.8+0) / 2=0.4.

[0032] The solution for setting the initial executable probability of a control strategy based on expert experience is: combining expert knowledge in the fields of architecture, environmental science, or smart home, and setting the initial executable probability of the control strategy based on building type, climate conditions, and user needs.

[0033] For example, if experts suggest that the probability of closing outward-opening windows in high-rise buildings on typhoon days is 1, the initial executable probability of this strategy can be set to 1.

[0034] The solution for setting the initial executable probability of a control strategy according to the system default value is as follows: the system developer presets the initial executable probability of the control strategy according to common scenarios or industry standards.

[0035] For example, the system sets the initial executable probability of the control strategy "close all windows at night" to 0.7 by default.

[0036] Obtain current environmental data, user data, and real-time status of smart doors and windows to dynamically adjust the execution probability of the strategy.

[0037] Real-time environmental data includes current temperature, humidity, air quality, etc.

[0038] Real-time user data includes the user's current location, activity status, etc. (which can be obtained through sensors in the smart home system or an IoT terminal worn by the user, such as a smart watch).

[0039] The real-time status of smart doors and windows includes whether they are currently open, the opening degree, etc.

[0040] The initial executable probability of each control strategy is dynamically adjusted based on real-time data to reflect changes in the current environment and user needs.

[0041] The existing control strategies and the probability adjustment labels and initial executable probabilities of each control strategy are shown in Table 1. The real-time data are as follows: temperature 24°C, humidity 30%, good air quality, the user is in a resting state, smart door and window 1 is in an open state of 80°, smart door and window 2 is in a closed state, smart door and window 3 is in an open state of 20°, and smart door and window 4 is in a closed state.

[0042] Table 1 Control strategies and their probability adjustment labels, initial executable probabilities

[0043] This embodiment uses a voting method to update the initial executable probability, setting the update step size to 0.1, and the method is as follows: Taking the control strategy of smart door and window 1, "casement window, open 20°", as an example, the real-time temperature, real-time humidity, and air quality all meet the probability adjustment label of smart door and window 1. Therefore, the initial executable probability of the control strategy is adjusted from 0.2 to 0.5, that is, the real-time executable probability of the control strategy is 0.5.

[0044] Similarly, the real-time executable probability of the control strategy "casement window, closed 10°" for smart door and window 1 is 0.7.

[0045] The real-time executable probability of the control strategy "top-hung window, open 10°" for smart door and window 2 is 0.5.

[0046] The real-time executable probability of the control strategy "upper-hung window, closed 5°" for smart door and window 2 is 0.1.

[0047] The real-time executable probability of the control strategy "casement window, closed 5°" for smart door and window 2 is 0.1.

[0048] The real-time executable probability of the control strategy "casement window, open 20°" for smart door and window 2 is 0.7.

[0049] The real-time executable probability of the control strategy "top-hung window, open 5°" for smart door and window 3 is 0.4.

[0050] The real-time executable probability of the control strategy "upper-hung window, closed 5°" for smart door and window 3 is 0.

[0051] The real-time executable probability of the control strategy "casement window, open 10°" for smart door and window 4 is 0.5.

[0052] In other embodiments, the real-time executable probability can also be predicted using a machine learning model, and the process is as follows: The control strategy, initial executable probability, and historical data are used as training data, and the historical executable probability is used as a training label to train the machine learning model to obtain a trained machine learning model. The historical data includes historical environmental data, historical user data, and the historical status of each smart door and window. The historical executable probability refers to the probability value obtained by updating the initial executable probability using the historical data.

[0053] The control strategy, initial executable probability and real-time data are input into the trained machine learning model, and the predicted real-time executable probability is output.

[0054] Execute the control strategy whose real-time executable probability exceeds the preset threshold (such as 0.5) to achieve automatic control of smart doors and windows.

[0055] By adopting the above solution, this embodiment realizes refined management, dynamic adjustment and automated execution of the control strategy, thereby improving the flexibility, adaptability and user experience of the control system.

[0056] Example 2: Reference Figure 2 The difference between this embodiment and embodiment 1 is that, before executing the control strategy whose real-time executable probability exceeds a preset threshold, the method further includes: S21 strategy judgment, based on the probability adjustment label to build a profit function, the profit function is used to quantify the execution effect of each control strategy under different environments and user needs. The calculation model is as follows: ; ; Where n is the number of probability adjustment labels of the sth control strategy; Adjust the label for the i-th probability The satisfaction function of .

[0057] In other embodiments, other types of functions may be selected as the profit function according to requirements.

[0058] The control strategy whose real-time executable probability is greater than a preset probability threshold is marked as a target strategy, and the target strategy is a strategy that is considered to have a higher execution value in the current environment.

[0059] Conflict detection is performed on the target strategies described above to determine whether there are conflicting target strategies. Conflicts arise from different strategies requiring different operations for the same smart door or window. For example, one strategy requires opening the door or window for ventilation, while another requires closing it for heat preservation.

[0060] If there is a conflict, S22 quantity determination is performed to further process the conflict strategy.

[0061] If there is no conflict, no further processing is done and the target policy that meets the conditions is directly executed.

[0062] S22: Quantity judgment, judging whether the number of conflicting target strategies is greater than two. If the number of conflicting strategies is greater than two, S23 quantum coding is executed.

[0063] S23 quantum coding marks conflicting target strategies as conflicting strategies and maps all conflicting strategies to quantum bits. A quantum bit is the fundamental unit in quantum computing, possessing properties such as superposition and entanglement, capable of representing multiple states simultaneously. For example, if there are four conflicting strategies, each corresponds to a quantum bit, with the strategy states represented in binary, where |0> indicates that the control strategy cannot be executed and |1> indicates that the control strategy can be executed. Therefore, four conflicting strategies require four quantum bits, represented as |s1s2s3s4>.

[0064] Based on the above quantum bits, a superposition state is constructed, which can simultaneously consider all possible conflicting strategy combinations. Then, a quantum state is constructed based on the superposition state. The quantum state is the collection of all possible states in the quantum system.

[0065] The expected return of the quantum state is calculated using quantum algorithms (such as quantum annealing algorithm, variational quantum eigensolver, etc.). The expected return reflects the average return of executing conflicting strategy combinations under different quantum states. The calculation method is as follows, ; in, is the weight of the energy-saving benefit; A is the amount of electricity saved, for example, by closing smart doors and windows when the temperature is low, the electricity used for indoor air conditioning heating can be saved; is the weight of user satisfaction; B is the difference between the user satisfaction value and the satisfaction threshold.

[0066] The quantum state corresponding to the maximum expected return is collapsed into the conflicting strategy combination. The collapse process is the transition from the quantum state to the classical state. The optimal conflicting strategy combination under the current environment is determined, and the conflicting strategies contained in the conflicting strategy combination are updated to the target strategy. Then, the S24 strategy screening is performed.

[0067] If the number of conflicting strategies is not greater than two, no further processing is performed.

[0068] S24 strategy screening calculates the payoff value of each new target strategy based on the payoff function. These payoff values ​​reflect the execution performance of each new target strategy in the current environment. The payoff values ​​of all new target strategies are integrated into a payoff matrix, where each element in the payoff matrix represents the payoff value of a new target strategy.

[0069] The Nash equilibrium algorithm is used to find a new target strategy combination in the payoff matrix that achieves equilibrium payoff. Nash equilibrium is a concept in game theory, referring to a state in which every player in a game has chosen the optimal strategy, and no player has the incentive to individually change their strategy to achieve higher payoffs.

[0070] The new target strategy contained in the new target strategy combination that retains the returns reaching the equilibrium state.

[0071] Delete the remaining new target strategies that are not in the equilibrium strategy portfolio because these strategies are not optimal choices in the current environment.

[0072] S25 generates an alarm. The new target policies deleted in S24, as well as any conflicting policies not included in the conflicting policy combination, are collectively marked as deleted policies. These control policies are deemed suboptimal in the current environment and are therefore excluded. The average real-time executable probability of the deleted policies is calculated and recorded as the first data point.

[0073] The new target policies retained in the policy screening in S24 and the conflicting policies included in the conflicting policy combination are uniformly marked as retained policies, and the average value of the real-time executable probabilities of the retained policies is calculated and recorded as the second data.

[0074] Determine whether the first data is greater than the second data. If the first data is greater than the second data, it means that the average execution probability of the deleted policy in the current environment is higher than that of the retained policy, indicating that the initial executable probability of the deleted policy is set too high. At this time, the executable probability of the deleted policy needs to be lowered.

[0075] If the first data is not greater than the second data, it indicates that the average execution probability of the deleted policy is not higher than that of the retained policy, the current policy selection is reasonable, and no alarm processing is required.

[0076] When an alarm signal is issued, the control system gradually reduces the initial probability of execution of the deleted policy according to a preset step size. By adjusting the initial probability of execution of the policy, the likelihood of it being incorrectly selected in the future is reduced. After adjusting the initial probability of execution of the deleted policy, the steps from S11 policy processing to S25 alarm are repeated, and a new round of policy screening, evaluation, and adjustment is performed until the first value is no longer greater than the second value.

[0077] Example 3: Reference Figure 3 The difference between this embodiment and embodiment 2 is that the method further includes: S31 calculates the sensitivity, uses the partial derivative formula in mathematics to derive the benefit function, obtains the partial derivative expression for each probability adjustment label, inputs the real-time environmental data and real-time user data into the partial derivative expression, and calculates the partial derivative value of each probability adjustment label in the current environment. By calculating the partial derivative of the benefit function with respect to each probability adjustment label, this embodiment can understand the degree of influence of each label on the benefit function.

[0078] The partial derivatives of the reward function for the same control strategy with respect to each probability adjustment label are used as elements in the same row or column of the Jacobian matrix, and the elements are arranged in the order of the labels to construct a complete Jacobian matrix. The Jacobian matrix can more intuitively demonstrate the sensitivity of the reward function to all probability adjustment labels. This example calculates the sensitivity of the reward function with respect to each probability adjustment label. Sensitivity reflects the responsiveness of the reward function to changes in the probability adjustment labels.

[0079] S32 value determination, traverse each element in the Jacobian matrix to determine whether the sensitivity is a non-negative number, If so, no action will be taken; If not, the absolute value of the sensitivity that is less than zero is recorded as the new sensitivity, and S33 is executed to update the real-time executable probability.

[0080] S33 updates the real-time executable probability and uses methods such as linear normalization or minimum-maximum normalization to normalize each sensitivity, map the sensitivity value to [0,1], and use the normalization result as the probability compensation value of the corresponding real-time environmental data or the corresponding real-time user data to reflect the impact of these data on the real-time executable probability. Subsequently, the real-time executable probability of each control strategy is added to its corresponding probability compensation value to obtain a new real-time executable probability to achieve dynamic adjustment.

[0081] In other embodiments, the method further comprises, The user's adjustment habits and adjustment timestamps for each smart door and window are obtained from historical user data. Within a preset time period before the adjustment timestamp, the new real-time executable probability of the control strategy corresponding to the adjustment habit is increased.

[0082] In other embodiments, the step S31 of calculating the sensitivity further includes: if the number of environment class labels and the number of user class labels are different, filling the vacant positions in the Jacobian matrix with the number zero.

[0083] By adopting the above scheme, this embodiment can more accurately evaluate the impact of each probability adjustment label on the benefit function, and dynamically adjust the real-time executable probability of each smart door and window control strategy based on the sensitivity value, which helps to improve the intelligence level and user experience of the control system, so that it can better adapt to the ever-changing environment and user needs.

[0084] Embodiment 4: This embodiment discloses an intelligent door and window control system, the system comprising: a memory and a processor, The memory stores a computer-readable storage medium; When the processor processes the computer program stored on the computer-readable storage medium, the method described above is implemented.

[0085] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A smart door and window control method, characterized in that: include: Establish a database storing the control strategy of each smart door and window, and add at least one probability adjustment tag to each control strategy, wherein the control strategy includes the opening mode, closing mode, and opening degree of each smart door and window, and the probability adjustment tag includes an environmental tag and a user tag; An initial executable probability is set for the control strategy of each smart door and window, and real-time data is collected. The real-time data includes real-time environmental data, real-time user data, and the real-time status of each smart door and window. The initial executable probability of the control strategy of each smart door and window is updated based on the real-time data, and the updated initial executable probability is recorded as the real-time executable probability. The control strategy with a real-time executable probability greater than a preset probability threshold is executed.

2. The intelligent door and window control method according to claim 1, characterized in that: The method further comprises: Based on the probability adjustment label, a profit function is constructed, and the control strategy with a real-time executable probability greater than the preset probability threshold is recorded as the target strategy. It is determined whether there are conflicting target strategies. If so, calculate the payoff value of each target strategy according to the payoff function, integrate the payoff values ​​of all target strategies into a payoff matrix, use the Nash equilibrium algorithm to obtain the target strategy combination in the payoff matrix when the payoff reaches an equilibrium state, retain the target strategies included in the target strategy combination, and delete the remaining target strategies; If not, no action will be taken.

3. The intelligent door and window control method according to claim 2, characterized in that: The method further comprises: Determine whether the number of conflicting target strategies is greater than two, If so, record the conflicting target strategies as conflicting strategies, map all conflicting strategies to quantum bits, construct a superposition state based on the quantum bits, construct a quantum state based on the superposition state, calculate the expected return of the quantum state using a quantum algorithm, collapse the quantum state corresponding to the maximum expected return into a conflicting strategy combination, and update the conflicting strategies contained in the conflicting strategy combination as the target strategy; If not, no action will be taken.

4. The intelligent door and window control method according to claim 3, characterized in that: The method further comprises: The conflicting policies not included in the deleted target policy and the conflicting policy combination are recorded as deleted policies, and the average value of the real-time executable probabilities of the deleted policies is calculated and recorded as the first data; the conflicting policies included in the retained target policy and the conflicting policy combination are recorded as retained policies, and the average value of the real-time executable probabilities of the retained policies is calculated and recorded as the second data; Determine whether the first data is greater than the second data. If so, issue an alarm signal; if not, do nothing.

5. The intelligent door and window control method according to claim 4, characterized in that: After the alarm signal is issued, the method further includes: gradually reducing the initial executable probability of the deleted policy according to a preset step size until the first data is no greater than the second data.

6. The intelligent door and window control method according to any one of claims 2 to 5, characterized in that: The method further comprises: Calculate the partial derivative of the profit function with respect to each probability adjustment label, input the real-time environment data and real-time user data into the partial derivative, obtain the partial derivative value, and integrate the partial derivative value into the Jacobian matrix; Based on the elements in the Jacobian matrix, the sensitivity of the benefit function with respect to each probability adjustment label is obtained, and each sensitivity is normalized respectively to obtain the normalized result. The normalized result is used as the probability compensation value of the corresponding real-time environmental data or the corresponding real-time user data, and the sum of the real-time executable probability and the probability compensation value is used as the new real-time executable probability.

7. The intelligent door and window control method according to claim 6, characterized in that: The method further comprises: Determine whether the sensitivity is a non-negative number, If so, no action will be taken; If not, the absolute value of the sensitivity that is less than zero is recorded as the new sensitivity.

8. The intelligent door and window control method according to claim 7, characterized in that: If the number of environment class labels and user class labels is different, the number zero is used to fill the vacant positions in the Jacobian matrix.

9. The intelligent door and window control method according to claim 6, characterized in that: The method further comprises: Obtain historical user data, obtain the user's adjustment habits and adjustment timestamps for each smart door and window based on the historical user data, and increase the new real-time executable probability of the control strategy corresponding to the adjustment habit within a preset time before the adjustment timestamp.

10. An intelligent door and window control system, characterized in that: include: memory and processor, The memory stores a computer-readable storage medium; When the processor processes the computer program stored on the computer-readable storage medium, the method according to any one of claims 1 to 9 is implemented.

Images