Smart home equipment control system and control method based on window opening and closing

Abstract

The invention discloses a smart home equipment control system and control method based on window opening and closing, and the system employs a parameter receiving module to receive indoor and outdoor environment natural dynamic parameters, improves the environment collection capability, employs an XGBoost algorithm to carry out the deep learning of the natural dynamic parameters through a parameter processing module, builds a corresponding model, and achieves the control of the window opening and closing. The driving module drives the window to be opened and closed according to the model content of the parameter processing module output by the terminal control module, the result detection module is used for judging the window opening and closing degree of the driving module and making adjustment, and therefore adjustment is made in time under the condition that the window ventilation effect is poor and dust exists in the microscopic quantitative change state; and the window use experience and disaster prevention are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart home control, and in particular to a smart home device control system and a control method based on a window switch. Background Art

[0002] Most of the windows in the home still rely on manual opening and closing, and are relatively backward in the use of automatic rainproof, lightproof and pollution-proof functions. However, with the development of intelligent technology, users have begun to have higher requirements in the use of windows, so home intelligent technology has become an inevitable trend.

[0003] At the current stage, the smart window system uses a single-chip microcomputer to automatically control the opening and closing of windows based on environmental information collected by multiple sensors. When there is rain, strong light or polluted gas in the external environment, the single-chip microcomputer can open and close the windows after receiving the signal from the sensor, and has a strong intelligent response. However, since the smart window system is in a passive learning state at the beginning of its design, it is limited to controlling actions according to the set program. If the window ventilation effect is not good and there is a microscopic quantitative change of dust, the smart window system cannot make timely adjustments, which may threaten life and property safety in the event of a fire. Summary of the invention

[0004] The purpose of the present invention is to provide a smart home device control system and control method based on window switches to solve the problems described in the background technology.

[0005] The technical solution of the present invention is achieved in this way:

[0006] On the one hand, the present invention provides a smart home device control system based on window switches, including a parameter receiving module, a parameter processing module, a driving module, a result detection module and a terminal control module;

[0007] The parameter receiving module is used to receive natural dynamic parameters of outdoor and indoor environments;

[0008] The parameter processing module uses the XGBoost algorithm to perform deep learning on the natural dynamic parameters;

[0009] The driving module drives the window switch according to the output content of the parameter processing module;

[0010] The result detection module is used to determine the degree of window opening and closing of the driving module;

[0011] The parameter processing module, the driving module and the result detection module are all connected to the terminal control module.

[0012] A further technical solution is that the parameter receiving module includes a dust sensor, a raindrop detector, a smoke detector, a strong light detector, a ventilation sensor, an infrared human body induction sensor, a temperature sensor and a sound sensor;

[0013] The dust sensor is used to detect dust from outside that floats toward the window, the raindrop detector is used to detect raindrops, the smoke detector is used to detect smoke generated indoors and inside the room, the strong light detector is used to detect sunlight outdoors, the ventilation sensor is used to detect the ventilation status of the window switch, the infrared human body induction sensor is used to detect the manual opening and closing of the window by a human body, the temperature sensor is used to detect the temperature difference between the outdoor and indoor temperature, and the sound sensor is used to detect outdoor noise;

[0014] The dust sensor, raindrop detector, smoke detector, strong light detector, ventilation sensor, infrared human body induction sensor, temperature sensor and sound sensor detect the natural dynamic parameters and transmit them to the parameter processing module through the terminal control module.

[0015] A further technical solution is that the parameter processing module includes a signal analysis unit, a signal training unit and a signal output unit. The signal analysis unit is used to receive the natural dynamic parameters and perform analysis based on parameters of different categories in the natural dynamic parameters. After the analysis is completed, the parameters of the same category are sent one by one to the signal training unit. The signal training unit uses the XGBoost algorithm for training, and sends the natural dynamic parameters after training to the signal output unit. The signal output unit sends the results to the driving module.

[0016] A further technical solution is that the signal analysis unit includes a first subroutine, a second subroutine and a third subroutine, the first subroutine obtains the natural dynamic parameter signal, the second subroutine screens the natural dynamic parameter into multiple subsets, and the third subroutine outputs the multiple subset signals to the signal training unit one by one.

[0017] A further technical solution is that the signal training unit includes an XGBoost algorithm program, and the XGBoost algorithm program receives a plurality of the subset signals to complete simulation training, and sends the signals to the driving module through the signal output unit.

[0018] A further technical solution is that the driving module includes a steering gear, and the steering gear outputs a signal to the terminal control module according to the parameter processing module, and the terminal control module starts the steering gear output end to adjust the window switch.

[0019] A further technical solution is that the result detection module includes a gap detection unit and a temperature difference detection unit, the gap detection unit is used to detect the closing distance between the window and the window frame, and the temperature difference detection unit is used to detect the indoor and outdoor temperature difference after the window is closed, and the result detection module sends the detection result to the terminal control module.

[0020] A further technical solution is that the gap detection unit is a distance detector, which detects the closing distance between the window and the window frame, collects the distance parameters, and sends them to the terminal control module, and the terminal control module sends a return instruction to start the drive module for regulation.

[0021] On the other hand, the present invention provides a control method of a smart home device control system based on a window switch, comprising the following steps:

[0022] Step S1: The parameter receiving module receives natural dynamic parameters inside and outside the window;

[0023] Step S2: The terminal receiving module receives the natural dynamic parameters and performs A / D signal conversion;

[0024] Step S3: the parameter processing module completes parameter processing according to the natural dynamic parameters received by the terminal receiving module;

[0025] Step S4: the driving module receives the output signal of the parameter processing module and drives the window switch;

[0026] Step S5: The result detection module receives and determines whether the window switch driven by the driving module meets the switch threshold, wherein the detection threshold is 0.1-0.5mm.

[0027] A further technical solution is that the parameter processing module completes parameter processing according to the natural dynamic parameters received by the terminal receiving module, including:

[0028] Step S301: The signal analysis unit receives the natural dynamic parameters and performs analysis according to parameters of different categories in the natural dynamic parameters, wherein the natural dynamic parameter signal is obtained by a first subroutine, the natural dynamic parameter is screened into a plurality of subsets according to the second subroutine, and the plurality of subset signals are output one by one by a third subroutine;

[0029] Step S302: The signal training unit uses the XGBoost algorithm program to train the model according to the signals output one by one by the plurality of subset signals according to the third subprogram;

[0030] Step S303: embedding the model into the driving module according to the XGBoost algorithm program training to adjust the window opening and closing;

[0031] Step S304: the result detection module completes the distance detection according to the window switch controlled by the driving module. If the judgment result is within the threshold, no feedback instruction is given. If it is not within the threshold, a feedback instruction is issued. The distance detection adopts the distance formula detection.

[0032] The beneficial effects of the present invention are:

[0033] The parameter receiving module is used to receive the natural dynamic parameters of the indoor and outdoor environments to increase the environmental collection capability. The parameter processing module uses the XGBoost algorithm to perform deep learning on the natural dynamic parameters to establish a corresponding model. The window switch is driven by the driving module according to the model content of the parameter processing module output by the terminal control module. The result detection module is used to determine the degree of window opening and closing of the driving module and make adjustments. In this way, when the window ventilation effect is poor and there is a microscopic quantitative change state of dust, timely adjustments can be made to improve the window usage experience and disaster prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a system block diagram of the present invention;

[0035] Figure 2 is a flow chart of the method of the present invention;

[0036] Figure 3 for Figure 2 Flowchart of sub-methods of S3. DETAILED DESCRIPTION

[0037] In order to better understand the technical content of the present invention, specific embodiments are provided below, and the present invention is further described in conjunction with the accompanying drawings.

[0038] See also Figures 1 to 3 On the one hand, the present invention provides a smart home device control system based on window switches, including a parameter receiving module, a parameter processing module, a driving module, a result detection module and a terminal control module; the parameter receiving module is used to receive the natural dynamic parameters of outdoor and indoor environments; the parameter processing module uses the XGBoost algorithm to perform deep learning on the natural dynamic parameters; the driving module drives the window switch according to the output content of the parameter processing module; the result detection module is used to judge the degree of window opening and closing of the driving module; the parameter processing module, the driving module and the result detection module are all connected to the terminal control module.

[0039] It should be noted that the terminal control module adopts the STC89C52 microcontroller.

[0040] Specifically, the parameter receiving module is used to receive the natural dynamic parameters of the indoor and outdoor environments to increase the environmental collection capability. The parameter processing module uses the XGBoost algorithm to perform deep learning on the natural dynamic parameters to establish a corresponding model. The window switch is driven by the driving module according to the model content of the parameter processing module output by the terminal control module. The result detection module is used to determine the degree of window opening and closing by the driving module and make adjustments. In this way, when the window ventilation effect is poor and there is microscopic quantitative change of dust, timely adjustments can be made to improve the window usage experience and disaster prevention.

[0041] In this embodiment, the parameter receiving module includes a dust sensor, a raindrop detector, a smoke detector, a strong light detector, a ventilation sensor, an infrared human body sensing sensor, a temperature sensor and a sound sensor; the dust sensor is used to detect dust from outdoors floating toward the window, the raindrop detector is used to detect raindrops, the smoke detector is used to detect smoke occurring indoors and inside the room, the strong light detector is used to detect outdoor sunlight, the ventilation sensor is used to detect the switch ventilation status of the window, the infrared human body sensing sensor is used to detect manual opening and closing of the window by a human body, the temperature sensor is used to detect the temperature difference between outdoors and indoors, and the sound sensor is used to detect outdoor noise; after the dust sensor, raindrop detector, smoke detector, strong light detector, ventilation sensor, infrared human body sensing sensor, temperature sensor and sound sensor detect the natural dynamic parameters, they are transmitted to the parameter processing module through the terminal control module.

[0042] It should be noted that dust sensors, raindrop detectors, smoke detectors, strong light detectors, ventilation sensors, infrared human body induction sensors, temperature sensors and sound sensors. Dust sensors, raindrop detectors, smoke detectors, strong light detectors, ventilation sensors and temperature sensors are used to collect changes in the environment. When in use, the signals detected by dust sensors, raindrop detectors, smoke detectors, strong light detectors, ventilation sensors and temperature sensors are converted into dust content, rainfall and smoke content, light intensity, wind speed change and temperature values ​​by A / D converters and transmitted to parameter processing modules for processing. Among them, infrared human body induction sensors are used to detect human body temperature and to distinguish manual opening and closing of windows, and sound sensors are used to detect environmental decibel values. By detecting the natural parameters of the environment through multiple sensors, the integrity and intelligent development of safety detection can be improved.

[0043] In addition, when human body temperature is detected, there is no need to start the control terminal module to open or close the windows, thus reducing the energy consumption of the drive module.

[0044] In this embodiment, the parameter processing module includes a signal analysis unit, a signal training unit and a signal output unit. The signal analysis unit is used to receive the natural dynamic parameters and perform analysis based on parameters of different categories in the natural dynamic parameters. After the analysis is completed, the parameters of the same category are sent one by one to the signal training unit. The signal training unit uses the XGBoost algorithm for training, and sends the natural dynamic parameters after training to the signal output unit. The signal output unit sends the results to the driving module.

[0045] It should be noted that natural dynamic parameters include dust content, rainfall and smoke content, light intensity, wind speed change, temperature value, and human body temperature and decibel value. In this process, the signal analysis unit analyzes various types of parameters to facilitate subsequent training.

[0046] Specifically, the signal analysis unit includes a first subroutine, a second subroutine and a third subroutine. The first subroutine obtains the natural dynamic parameter signal, the second subroutine screens the natural dynamic parameter into multiple subsets, and the third subroutine outputs the multiple subset signals to the signal training unit one by one.

[0047] Exemplarily, the first subroutine is an API. After collecting data through dust sensors, raindrop detectors, smoke detectors, strong light detectors, ventilation sensors, infrared human body sensors, temperature sensors and sound sensors, the collected data is read using the API to improve data accuracy and stability. The second subroutine uses a technology stack to preset an existing natural dynamic parameter model. When the first subroutine obtains natural dynamic parameters, it is compared with the existing natural dynamic parameter model to obtain different types of subsets. The third subroutine uses ZigBee to achieve low-power transmission.

[0048] Preferably, the signal training unit includes an XGBoost algorithm program, and the XGBoost algorithm program receives a plurality of the subset signals to complete simulation training, and sends the signals to the driving module through the signal output unit.

[0049] Exemplarily, the XGBoost algorithm uses multiple subsets and iterative operations of weak classifiers to realize strong classifiers, thereby improving the prediction speed of multi-type subsets. Among them, the user APP connects to the system, sets the preset value of the window switch in the initialization state, and when the natural dynamic parameters change, the system determines whether the window state needs to be changed based on whether the time and environmental data exceed the set threshold. At the same time, the multi-type subset historical data trained by the XGBoost algorithm is used to predict the current window state based on the probability of natural dynamic parameters in the current time, environment and future time period, and make a switch action. The weight of the subset model changes with the increase of data in the subset or the user modifies the preset value. When the data in the subset increases, the weight of the preset value decreases, and when the user modifies the preset value, the weight of the preset value control increases. By integrating the preset value control and the XGBoost algorithm and adjusting the weight, the advantages of the preset value control and the XGBoost algorithm can be combined to improve the prediction accuracy of the system.

[0050] In this embodiment, the driving module includes a steering gear, and the steering gear outputs a signal to the terminal control module according to the parameter processing module, and the terminal control module starts the output end of the steering gear to adjust the window switch.

[0051] It should be noted that the control terminal module is used to control the servo, so that the servo can control the opening and closing of the window according to the output signal of the parameter processing module, thereby improving the automation of window use and enhancing the user experience of using smart windows.

[0052] In this embodiment, the result detection module includes a gap detection unit and a temperature difference detection unit. The gap detection unit is used to detect the closing distance between the window and the window frame, and the temperature difference detection unit is used to detect the indoor and outdoor temperature difference after the window is closed. The result detection module sends the detection result to the terminal control module.

[0053] Specifically, the gap detection unit is a distance detector, which detects the closing distance between the window and the window frame, collects the distance parameters, and sends them to the terminal control module. The terminal control module sends a return instruction to start the driving module for regulation.

[0054] The gap detection unit and the temperature difference detection unit use laser distance sensors and thermocouples for distance measurement and temperature difference measurement respectively. The gap detection unit makes a judgment based on the measured distance and the detection threshold of 0.1-0.5mm. If it is greater than 0.5mm, the window is prone to leaking when closed. If it is less than 0.1mm, the ventilation effect is poor, and the data needs to be uploaded to the terminal control module for correction to within 0.1-0.5mm according to the preset detection threshold. At the same time, under thermocouple detection, if the distance between the window and the window frame is not greater than 0.5mm when the window is closed, in winter, the temperature difference is too small, indicating that the room is cold, which is not conducive to the user's rest. Therefore, it is necessary to start the drive module to adjust the window opening and closing distance to achieve a suitable window closing effect.

[0055] Another aspect of the present invention provides a control method for a smart home device control system based on a window switch, comprising the following steps:

[0056] Step S1: The parameter receiving module receives natural dynamic parameters inside and outside the window;

[0057] Step S2: The terminal receiving module receives the natural dynamic parameters and performs A / D signal conversion;

[0058] Step S3: the parameter processing module completes parameter processing according to the natural dynamic parameters received by the terminal receiving module;

[0059] Step S4: the driving module receives the output signal of the parameter processing module and drives the window switch;

[0060] Step S5: The result detection module receives and determines whether the window switch driven by the driving module meets the switch threshold, wherein the detection threshold is 0.1-0.5mm.

[0061] Preferably, the parameter processing module completes parameter processing according to the natural dynamic parameters received by the terminal receiving module, including:

[0062] Step S301: The signal analysis unit receives the natural dynamic parameters and performs analysis according to parameters of different categories in the natural dynamic parameters, wherein the natural dynamic parameter signal is obtained by a first subroutine, the natural dynamic parameter is screened into a plurality of subsets according to the second subroutine, and the plurality of subset signals are output one by one by a third subroutine;

[0063] Step S302: The signal training unit uses the XGBoost algorithm program to train the model according to the signals output one by one by the plurality of subset signals according to the third subprogram;

[0064] Step S303: embedding the model into the driving module according to the XGBoost algorithm program training to adjust the window opening and closing;

[0065] Step S304: the result detection module completes the distance detection according to the window switch controlled by the driving module. If the judgment result is within the threshold, no feedback instruction is given. If it is not within the threshold, a feedback instruction is issued. The distance detection adopts the distance formula detection.

[0066] It should be noted that the working principle of this method is consistent with the principle of the system and will not be repeated here.

[0067] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A smart home device control system based on window switches, characterized in that: It includes a parameter receiving module, a parameter processing module, a driving module, a result detection module and a terminal control module; The parameter receiving module is used to receive natural dynamic parameters of outdoor and indoor environments; The parameter processing module uses the XGBoost algorithm to perform deep learning on the natural dynamic parameters; The driving module drives the window switch according to the output content of the parameter processing module; The result detection module is used to determine the degree of window opening and closing of the driving module; The parameter processing module, the driving module and the result detection module are all connected to the terminal control module.

2. The smart home device control system based on window switches according to claim 1, characterized in that: The parameter receiving module includes a dust sensor, a raindrop detector, a smoke detector, a strong light detector, a ventilation sensor, an infrared human body induction sensor, a temperature sensor and a sound sensor; The dust sensor is used to detect dust from outside that floats toward the window, the raindrop detector is used to detect raindrops, the smoke detector is used to detect smoke generated indoors and inside the room, the strong light detector is used to detect sunlight outdoors, the ventilation sensor is used to detect the ventilation status of the window switch, the infrared human body induction sensor is used to detect the manual opening and closing of the window by a human body, the temperature sensor is used to detect the temperature difference between the outdoor and indoor temperature, and the sound sensor is used to detect outdoor noise; The dust sensor, raindrop detector, smoke detector, strong light detector, ventilation sensor, infrared human body induction sensor, temperature sensor and sound sensor detect the natural dynamic parameters and transmit them to the parameter processing module through the terminal control module.

3. The smart home device control system based on window switches according to claim 1, characterized in that: The parameter processing module includes a signal analysis unit, a signal training unit and a signal output unit. The signal analysis unit is used to receive the natural dynamic parameters and perform analysis based on parameters of different categories in the natural dynamic parameters. After the analysis is completed, the parameters of the same category are sent one by one to the signal training unit. The signal training unit uses the XGBoost algorithm for training, and sends the natural dynamic parameters after training to the signal output unit. The signal output unit sends the results to the driving module.

4. The smart home device control system based on window switches according to claim 3 is characterized in that: The signal analysis unit includes a first subroutine, a second subroutine and a third subroutine. The first subroutine obtains the natural dynamic parameter signal, the second subroutine screens the natural dynamic parameter into multiple subsets, and the third subroutine outputs the multiple subset signals to the signal training unit one by one.

5. The smart home device control system based on window switches according to claim 3 is characterized in that: The signal training unit includes an XGBoost algorithm program, and the XGBoost algorithm program receives a plurality of the subset signals to complete simulation training, and sends the signals to the driving module through the signal output unit.

6. The smart home device control system based on window switches according to claim 1, characterized in that: The driving module includes a steering gear, and the steering gear outputs a signal to the terminal control module according to the parameter processing module. The terminal control module starts the output end of the steering gear to adjust the window switch.

7. The smart home device control system based on window switches according to claim 1, characterized in that: The result detection module includes a gap detection unit and a temperature difference detection unit. The gap detection unit is used to detect the closing distance between the window and the window frame, and the temperature difference detection unit is used to detect the indoor and outdoor temperature difference after the window is closed. The result detection module sends the detection result to the terminal control module.

8. The smart home device control system based on window switches according to claim 7, characterized in that: The gap detection unit is a distance detector, which detects the closing distance between the window and the window frame, collects the distance parameters, and sends them to the terminal control module. The terminal control module sends a return instruction to start the driving module for regulation.

9. A control method for a smart home device control system based on a window switch according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step S1: The parameter receiving module receives natural dynamic parameters inside and outside the window; Step S2: The terminal receiving module receives the natural dynamic parameters and performs A / D signal conversion; Step S3: the parameter processing module completes parameter processing according to the natural dynamic parameters received by the terminal receiving module; Step S4: the driving module receives the output signal of the parameter processing module and drives the window switch; Step S5: The result detection module receives and determines whether the window switch driven by the driving module meets the switch threshold, wherein the detection threshold is 0.1-0.5mm.

10. The control method of a smart home device control system based on a window switch according to claim 9, characterized in that: The parameter processing module completes parameter processing according to the natural dynamic parameters received by the terminal receiving module, including: Step S301: The signal analysis unit receives the natural dynamic parameters and performs analysis according to parameters of different categories in the natural dynamic parameters, wherein the natural dynamic parameter signal is obtained by a first subroutine, the natural dynamic parameter is screened into a plurality of subsets according to the second subroutine, and the plurality of subset signals are output one by one by a third subroutine; Step S302: The signal training unit uses the XGBoost algorithm program to train the model according to the signals output one by one by the plurality of subset signals according to the third subprogram; Step S303: embedding the model into the driving module according to the XGBoost algorithm program training to adjust the window opening and closing; Step S304: the result detection module completes the distance detection according to the window switch controlled by the driving module. If the judgment result is within the threshold, no feedback instruction is given. If it is not within the threshold, a feedback instruction is issued. The distance detection adopts the distance formula detection.

Images