A smart home control system based on the Internet of Things

Abstract

The present invention provides a smart home control system based on the Internet of Things, which relates to the field of electrical signal processing technology, including: an acquisition module for acquiring the temperature signal monitored by each indoor temperature sensor; a first determination module for determining the optimal smoothing result of the temperature signal curve of each temperature sensor based on trend consistency according to each temperature signal, and the temperature signal curve includes a temperature signal composition curve; a second determination module for subtracting the optimal smoothing result from the temperature signal to determine the residual signal of each temperature sensor; a processing module for forming all amplitudes of the residual signal into a set, using the covariance matrix converted from the set as the noise covariance matrix of the Kalman filter system, and performing denoising on the real-time temperature signal. By considering the actual indoor electrical environment and the different noise interference problems of sensors at different locations, the convergence speed of the Kalman filter is greatly improved, thereby achieving accurate control of the indoor temperature.

Description

Technical Field

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

[0002] IoT smart home refers to a smart home system that connects home appliances to the Internet to achieve remote automated control and intelligent management. The smart home system can monitor, control and optimize the home environment through various sensors, communication modules and intelligent control centers to improve the quality of life and comfort.

[0003] In smart homes, in order to better monitor the indoor temperature and environment, multiple sensors are usually installed in the room to form an IoT monitoring system. For example, the smart air conditioner in the living room can be intelligently adjusted according to the indoor temperature to ensure that the indoor temperature is maintained at a constant temperature. Due to the location of the air conditioner, the actual temperature at different locations in the room may vary greatly. The indoor temperature cannot be accurately adjusted by a sensor in only one location, so multiple sensors are often set up to monitor the indoor temperature in all directions.

[0004] However, since the temperature in different parts of the room is not exactly the same, there are certain differences in the monitoring results of sensors installed in different locations. In addition, due to the electromagnetic interference between indoor electrical appliances and the different degrees of interference, the existing smart home control system cannot accurately control the indoor temperature. Summary of the invention

[0005] In view of the problems existing in the prior art, the present invention provides a smart home control system based on the Internet of Things.

[0006] The present invention is achieved through the following technical solutions:

[0007] The present invention provides a smart home control system based on the Internet of Things, comprising:

[0008] An acquisition module, used for acquiring a temperature signal monitored by each indoor temperature sensor, wherein a plurality of indoor temperature sensors are arranged;

[0009] A first determination module, configured to determine, according to each of the temperature signals, an optimal smoothing result of a temperature signal curve of each temperature sensor based on trend consistency, wherein the temperature signal curve includes the temperature signal component curve;

[0010] A second determination module, configured to subtract the optimal smoothing result from the temperature signal to determine a residual signal of each temperature sensor;

[0011] The processing module is used to group all amplitudes of the residual signal into a set, use the covariance matrix converted from the set as the noise covariance matrix of the Kalman filter system, and perform denoising on the real-time temperature signal.

[0012] Furthermore, the interpolation module is used to perform temperature interpolation processing on the target position according to the correlation coefficient between the temperature change and the distance, and adjust the air conditioner according to the processing result so that the target position obtains the expected temperature value, the temperature change includes the difference between the air outlet temperature of the air conditioner and the temperature monitored by each of the temperature sensors, and the distance includes the distance between the air outlet position of the air conditioner and each of the temperature sensors.

[0013] Furthermore, the first determination module is specifically used to perform mean filtering iterations on the temperature signals of each of the sensors within a period of time, and output the smoothed signals of each iteration; perform first-order derivative on the curve composed of the smoothed signals, delete the sections with first-order derivatives of 0 from the curve of the smoothed signals, and reorganize the remaining signals in sequence to obtain reorganized auxiliary signals; use the dynamic time normalization DTW algorithm to align the reorganized auxiliary signals, and perform similarity processing on the processing results to determine the similarity between the aligned and reorganized auxiliary signals; and determine the optimal smoothing result of the temperature signal curve of each temperature sensor based on the similarity.

[0014] Furthermore, the first determination module is also used to use the DTW algorithm to align the trend terms of the curve composed of the recombinant auxiliary signals to determine each aligned recombinant auxiliary signal; and determine the similarity between each aligned recombinant auxiliary signal based on the mean square error of the relative trend change between the aligned recombinant auxiliary signals.

[0015] Furthermore, the first determination module is specifically based on the formula Determine the similarity MSE (a, b) between the auxiliary signal a after alignment and reorganization and the auxiliary signal b after alignment and reorganization, is the temperature signal of sampling point number i of auxiliary signal a after alignment and reorganization, is the temperature signal of sampling point number i-1 of auxiliary signal a after alignment and reorganization, is the temperature signal of sampling point number i of auxiliary signal b after alignment and reorganization, is the temperature signal of the sampling point number i-1 of the auxiliary signal b after alignment and reorganization, is the maximum value of the temperature signal among all sampling point numbers of the auxiliary signal a after alignment and reorganization, is the most corrected value of the temperature signal among all sampling point numbers of the auxiliary signal a after alignment and reorganization, is the maximum value of the temperature signal among all sampling point numbers of the auxiliary signal b after alignment and reorganization, is the most correct value of the temperature signal among all sampling point numbers of the auxiliary signal b after alignment and reorganization. There are n sampling points on the auxiliary signal a after alignment and reorganization, and there are n sampling points on the auxiliary signal b after alignment and reorganization.

[0016] Furthermore, the first determination module is also used to determine the optimal smoothing result according to the objective function E=min[σ(MSE)+μ(MSE)], wherein σ(MSE) is the mean square error of the mean square errors between all the auxiliary signals after alignment and reorganization, and μ(MSE) is the average value of the mean square errors between all the auxiliary signals after alignment and reorganization.

[0017] Furthermore, the interpolation module is specifically used to determine the temperature difference ΔI between the air conditioner display temperature and the temperature monitored by the temperature sensor P through the air conditioner display temperature and the temperature monitored by the temperature sensor P. p According to the formula Determine the correlation coefficient ρ, where o represents the air conditioner, p represents any temperature sensor, and I p -I o Represents the temperature difference ΔI between the air conditioner and the pth temperature sensor during the adjustment cycle p , Represents the average value of the temperature difference between all temperature sensors and the air conditioner display. represents the distance between the air conditioner and the pth temperature sensor, Represents the average distance between the air conditioner and the temperature sensor.

[0018] Furthermore, the interpolation module is used to calculate the value according to the formula Determine the temperature interpolation of the target position R, where the number of temperature sensors is N, and ρ represents the correlation coefficient between the temperature change and the distance from the air-conditioning temperature to different temperature sensor positions. represents the distance between the target position and the pth temperature sensor, is the distance normalization result.

[0019] Compared with the prior art, the present invention has the following beneficial technical effects:

[0020] The smart home control system based on the Internet of Things provided by the present invention includes: an acquisition module for acquiring the temperature signal monitored by each indoor temperature sensor, wherein a plurality of the temperature sensors are arranged indoors; a first determination module for determining the optimal smoothing result of the temperature signal curve of each temperature sensor based on trend consistency according to each temperature signal, wherein the temperature signal curve includes the temperature signal composition curve; a second determination module for subtracting the optimal smoothing result from the temperature signal to determine the residual signal of each temperature sensor; a processing module for forming a set of all amplitudes of the residual signal, using the covariance matrix converted from the set as the noise covariance matrix of the Kalman filter system, and performing denoising on the real-time temperature signal. By obtaining the optimal smoothing result of each temperature sensor on the basis of ensuring trend consistency, the residual signal of each temperature sensor data is obtained, which is used as the system noise covariance matrix of the Kalman filter. Then, the real-time temperature signal of the adjustment cycle is denoised and smoothed to obtain accurate temperature monitoring data. Then, by calculating the correlation coefficient between the temperature change and the distance between the air-conditioning outlet position and each temperature sensor, the monitoring data error that may be caused by external factors such as the living room window orientation, the sunlight exposure area, and the air-conditioning outlet direction is eliminated. Finally, the convolution method is used to interpolate the temperature at the location of the personnel and automatically adjust to the expected temperature value. Compared with the traditional method of denoising multiple sensors separately using default parameters, the present invention takes into account the actual indoor electrical environment and the different noise interference of sensors at different locations. Therefore, it has better adaptive ability, optimizes the system noise setting, and can also greatly improve the convergence speed of the Kalman filter, thereby improving the real-time performance of the denoising algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a structural schematic diagram of a smart home control system based on the Internet of Things according to an embodiment of the present invention;

[0022] Figure 2 This is a schematic diagram of temperature data sampling of a smart home control system based on the Internet of Things according to an embodiment of the present invention;

[0023] Figure 3 It is a structural schematic diagram of a smart home control system based on the Internet of Things according to another embodiment of the present invention;

[0024] Figure 4 It is a structural schematic diagram of an air conditioner and a temperature sensor of a smart home control system based on the Internet of Things according to an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The present invention is further described in detail below in conjunction with specific embodiments, which are intended to explain the present invention rather than to limit it.

[0026] An embodiment of the present invention provides an Internet of Things-based smart home control system, which is applicable to the following scenarios: the indoor temperature is intelligently adjusted by the smart air conditioner in the living room through the temperature monitored by sensors at different locations in the room.

[0027] The applicable scenarios of the embodiments of the present invention are: the indoor air conditioner can be intelligently adjusted according to the indoor temperature to ensure that the indoor temperature is maintained at a constant temperature. However, due to the location restrictions of the air conditioner, the actual temperature at different locations in the room may be quite different. Multiple temperature sensors are set at different locations in the room to adjust the indoor temperature. Since the number and type of electrical appliances in each household are different, the electromagnetic compatibility is relatively poor. For example, abnormal data often appears in temperature sensors, humidity sensors, wind speed sensors, etc., which has a great impact on the all-round monitoring and intelligent adjustment of indoor temperature. The present invention proposes a smart home control system based on the Internet of Things, in order to suppress the interference of multi-distance noise on sensor data, improve the accuracy of the interpolation of indoor air conditioner multi-temperature distribution data, and then optimize the air conditioner temperature intelligent adjustment system.

[0028] Figure 1 FIG. 1 is a schematic diagram of a smart home control system based on the Internet of Things according to an embodiment of the present invention; Figure 1 As shown, the smart home control system based on the Internet of Things includes: an acquisition module 11, a first determination module 12, a second determination module 13 and a processing module 14, wherein:

[0029] An acquisition module 11 is used to acquire a temperature signal monitored by each indoor temperature sensor, where a plurality of indoor temperature sensors are arranged;

[0030] A first determination module 12, for determining, according to each of the temperature signals, an optimal smoothing result of a temperature signal curve of each temperature sensor based on trend consistency, the temperature signal curve including the temperature signal component curve;

[0031] A second determination module 13, configured to subtract the optimal smoothing result from the temperature signal to determine a residual signal of each temperature sensor;

[0032] The processing module 14 is used to group all the amplitudes of the residual signal into a set, use the covariance matrix converted from the set as the noise covariance matrix of the Kalman filter system, and perform denoising on the real-time temperature signal.

[0033] Figure 2 FIG. 1 is a schematic diagram of temperature data sampling of a smart home control system based on the Internet of Things according to an embodiment of the present invention; Figure 2 As shown in the figure, the monitoring data of each temperature sensor is obtained. The sampling frequency of household appliances is about 100 Hz, that is, 100 sampling points per second.

[0034] The interferences to the multiple temperature sensors in this embodiment during the temperature data collection process are different, which is due to the following reasons: First, in addition to temperature sensors, the interference degree of various types of household appliances to electronic equipment will be different. For example, video and audio equipment such as televisions and stereos usually have strong electromagnetic radiation; secondly, different frequencies will have different effects on the interference range. Some low-frequency signals with operating frequencies of about 50HZ-60HZ can penetrate walls and obstacles; then the greater the power of household appliances, the wider the electromagnetic interference range. In addition, the degree of electromagnetic interference also depends on the distance and isolation between the electrical appliances and electronic equipment. In short, the range of electromagnetic interference generated by household appliances is mainly affected by factors such as frequency, power, device type, and distance.

[0035] Due to different interference factors, the noise interference of temperature signals collected by different sensors is also different. Therefore, in order to achieve comprehensive monitoring, the IoT monitoring system composed of multiple temperature sensors in the room must adaptively process the sensor data at different locations. Only after obtaining accurate monitoring data can the temperature of the sensorless area be interpolated and a better intelligent temperature control effect be obtained.

[0036] Temperature data is real-time data, and Kalman filtering is often used for noise reduction and smoothing. Kalman filtering is highly dependent on the setting of system noise, which is the core parameter of Kalman filtering prediction, error iteration, and denoising and smoothing. When the noise interference in each sensor is different, the default system noise setting will inevitably have a large error.

[0037] The Kalman filter only needs to smooth the real-time temperature signal when approaching the temperature adjustment cycle. The temperature change feedback is relatively slow. Generally, the intelligent adjustment cycle of the air conditioner is about 10-30 minutes. We use the temperature signal data of the temperature sensor at each location within 30 minutes as the auxiliary signal, that is, Kalman filter prediction is not performed in the non-adjustment cycle, but the auxiliary signal in the non-adjustment cycle is used to estimate the system noise, optimize the denoising accuracy of the Kalman filter in the adjustment cycle, and will not affect its real-time performance.

[0038] When the air-conditioning temperature changes, the indoor temperature will inevitably change slowly. Although the temperature transmission, temperature change, and even temperature change nodes and duration are different at different locations due to factors such as the air-conditioning wind direction and indoor layout, when the air-conditioning temperature changes, the temperature sensors at various locations will inevitably be affected, and their temperature signals will have a certain similarity in the trend of low-frequency changes, which cannot be covered or disguised by noise.

[0039] In this embodiment, the temperature signals of each sensor are first smoothed by mean filtering using mean filtering, and the mean filtering scale is iterated respectively. This is because the smoothing effect of the mean filtering is only related to its filtering scale, so its iterative process is easier to control. Then, the low-frequency signals after smoothing of the sensor signals are interacted, and the objective function is set to obtain the residual noise amount at each position.

[0040] Specifically, the first determination module 12 is specifically used to perform mean filtering iterations on the temperature signals of each of the sensors within a period of time, and output the smoothed signals of each iteration; perform first-order derivative on the curve composed of the smoothed signals, delete the sections with first-order derivatives of 0 from the curve of the smoothed signals, and reorganize the remaining signals in sequence to obtain reorganized auxiliary signals; use the dynamic time normalization DTW algorithm to align the reorganized auxiliary signals, and perform similarity processing on the processing results to determine the similarity between the aligned and reorganized auxiliary signals; and determine the optimal smoothing result of the temperature signal curve of each temperature sensor based on the similarity.

[0041] Furthermore, the first determination module 12 is specifically used to align the trend items of the curve composed of the recombinant auxiliary signals using the DTW algorithm to determine each aligned recombinant auxiliary signal; and determine the similarity between each aligned recombinant auxiliary signal based on the mean square error of the relative trend change between the aligned recombinant auxiliary signals.

[0042] Preferably, based on the above embodiment, the first determining module 12 is specifically based on the formula Determine the similarity MSE (a, b) between the auxiliary signal a after alignment and reorganization and the auxiliary signal b after alignment and reorganization, is the temperature signal of sampling point number i of auxiliary signal a after alignment and reorganization, is the temperature signal of sampling point number i-1 of auxiliary signal a after alignment and reorganization, is the temperature signal of sampling point number i of auxiliary signal b after alignment and reorganization, is the temperature signal of the sampling point number i-1 of the auxiliary signal b after alignment and reorganization, is the maximum value of the temperature signal among all sampling point numbers of the auxiliary signal a after alignment and reorganization, is the most corrected value of the temperature signal among all sampling point numbers of the auxiliary signal a after alignment and reorganization, is the maximum value of the temperature signal among all sampling point numbers of the auxiliary signal b after alignment and reorganization, is the most correct value of the temperature signal among all sampling point numbers of the auxiliary signal b after alignment and reorganization. There are n sampling points on the auxiliary signal a after alignment and reorganization, and there are n sampling points on the auxiliary signal b after alignment and reorganization.

[0043] The first determination module 12 is also used to determine the optimal smoothing result according to the objective function E=min[σ(MSE)+μ(MSE)], wherein σ(MSE) is the mean square error of the mean square errors between all the auxiliary signals after alignment and reorganization, and μ(MSE) is the average value of the mean square errors between all the auxiliary signals after alignment and reorganization.

[0044] For example, firstly, the temperature signal of each temperature sensor is subjected to mean filtering iteration, and the filtering scale L = 3, 5, 7... The upper limit is set to 21, and then the smoothed signal of each iteration is output; then, the first-order derivative of the smoothed signal curve of all mean filtering iterations is calculated, and the segments with derivative values ​​of 0 are temporarily deleted, that is, the signal segments where the temperature has not changed are deleted, and then the remaining signals are reorganized in order. Then, the auxiliary signals of each sensor with the same original length are different in length after reorganization, so the DTW algorithm is used to align the signals, and then the similarity between the auxiliary signals after reorganization is calculated.

[0045]

[0046] Among them, a and b represent any two recombined auxiliary signals for similarity calculation, and i represents the sequence number of any sampling point after the length of the two is aligned by the DTW algorithm. After alignment, there are n one-to-one corresponding sampling points in total. They represent the normalized slope value at the i-th sampling point in the auxiliary signal recombined by a and b respectively, which is the change value of the adjacent signal divided by the range, and can represent the relative trend change of the signal at that point. The mean square error representing the relative trend change between auxiliary signals a and b is denoted as MSE(a, b).

[0047] Formula logic: As mentioned above, even if the positions of temperature sensors are different, when the indoor temperature changes, the change trends of each position are unified, but the change amplitude, duration of change, and change nodes are different. We delete the gently changing segments and only align the trend items and calculate the similarity of the temperature smoothing curves of the changing segments. DTW alignment is to solve the problems of different durations of temperature changes and different change nodes, and the mean square error of the normalized signal slope is calculated when calculating the similarity to solve the problem of different change amplitudes. Then, the trend similarity between the smoothed signal curves can be judged based on the size of the mean square error.

[0048] E = min[σ(MSE) + μ(MSE)]

[0049] Among them, σ 2(MSE) is the mean square error of the mean square errors between all recombined auxiliary signals, and the penalty term is the average of the mean square errors between all recombined auxiliary signals. The former means that the difference between all mean square error values ​​needs to be smaller, and the latter means that all mean square errors need to be smaller. That is, min[σ(MSE)+μ(MSE)], E is the objective function. According to our calculation, if E is the smallest when the low-frequency temperature change trends between all sensors are consistent, then the objective function converges. At this time, it is considered that the smoothing results of the auxiliary signals of each sensor under different mean filtering are relatively optimal.

[0050] Then, the auxiliary signal minus the optimal mean smoothing signal of the signal segment can obtain the residual signal on each sensor auxiliary signal, that is, the noise estimation signal. The mean filter has a higher smoothness, but its denoising accuracy is poorer than that of the Kalman filter, and its real-time performance is also inferior. Therefore, the above operation is only to obtain the noise parameters of the Kalman filter system of different temperature sensors. All the amplitudes of the residual signal on each sensor are combined into a set and converted into a covariance matrix, which is used as the noise covariance matrix of the real-time signal Kalman filter system of the sensor during the temperature adjustment cycle. Then, the baseline term function after the above mean filter smoothing is used as the prediction equation. When the adjustment cycle of each 30-minute interval arrives, the real-time signal is denoised using the Kalman filter.

[0051] The smart home control system based on the Internet of Things in this embodiment includes: an acquisition module for acquiring the temperature signal monitored by each indoor temperature sensor, wherein a plurality of the temperature sensors are arranged indoors; a first determination module for determining the optimal smoothing result of the temperature signal curve of each temperature sensor based on the trend consistency according to each temperature signal, wherein the temperature signal curve includes the temperature signal composition curve; a second determination module for subtracting the optimal smoothing result from the temperature signal to determine the residual signal of each temperature sensor; a processing module for forming a set of all amplitudes of the residual signal, using the covariance matrix converted from the set as the noise covariance matrix of the Kalman filter system, and denoising the real-time temperature signal. By constructing an objective function according to the consistency of the indoor temperature change trend, the optimal smoothing result of each temperature sensor on the basis of ensuring the trend consistency is obtained, and then the residual signal of each temperature sensor data is obtained, which is used as the system noise covariance matrix of the Kalman filter. Then, the real-time temperature signal of the adjustment cycle is denoised and smoothed to obtain accurate temperature monitoring data. Then, by calculating the correlation coefficient between the temperature change and the distance between the air-conditioning outlet position and each temperature sensor, the monitoring data error that may be caused by external factors such as the living room window orientation, the sunlight exposure area, and the air-conditioning outlet direction is eliminated. Finally, the convolution method is used to interpolate the temperature at the location of the personnel and automatically adjust to the expected temperature value. Compared with the traditional method of denoising multiple sensors separately using default parameters, the present invention takes into account the actual indoor electrical environment and the different noise interference of sensors at different locations. Therefore, it has better adaptive ability, optimizes the system noise setting, and can also greatly improve the convergence speed of the Kalman filter, thereby improving the real-time performance of the denoising algorithm.

[0052] Figure 3 is a schematic diagram of the structure of a smart home control system based on the Internet of Things according to another embodiment of the present invention, Figure 4 FIG. 1 is a schematic diagram of the structure of an air conditioner and a temperature sensor of a smart home control system based on the Internet of Things according to an embodiment of the present invention. Figure 3 and Figure 4 As shown, on the basis of the above embodiment, the smart home control system based on the Internet of Things also includes the interpolation module 15, which is used to perform temperature interpolation processing on the target position according to the correlation coefficient between the temperature change and the distance, and adjust the air conditioner according to the processing result so that the target position obtains the expected temperature value, the temperature change includes the difference between the air outlet temperature of the air conditioner and the temperature monitored by each of the temperature sensors, and the distance includes the distance between the air outlet position of the air conditioner and each of the temperature sensors.

[0053] For example, the real-time signal after denoising can get an accurate temperature value, and then the temperature change between the air conditioner position and each sensor position, as well as the correlation coefficient between the temperature change and the distance, can be obtained based on the temperature value. This is because the basic principle of interpolation is to assign different weight coefficients to adjacent temperature values ​​based on the distance and then weight them, which is actually a convolution operation. However, the relationship between temperature change and distance is not completely linearly related, and there must be other factors that affect it, such as the direction of the living room window, the area of ​​sunlight, the direction of the air conditioner vents, etc. Therefore, it is necessary to obtain different correlation coefficients for different rooms and calibrate the impact of these factors.

[0054] Specifically, the interpolation module 15 is specifically used to determine the temperature difference ΔI between the air conditioner display temperature and the temperature monitored by the temperature sensor P through the air conditioner display temperature and the temperature monitored by the temperature sensor P. p ;

[0055] ΔI p =I p -I o , where o represents the air conditioner display temperature, p represents any temperature sensor, I p -I o Represents the temperature difference ΔI between the air conditioner and the pth temperature sensor during the adjustment cycle p .

[0056] According to the formula Determine the correlation coefficient ρ, where o represents the air conditioner display temperature, p represents any temperature sensor, and I p -I o Represents the temperature difference ΔI between the air conditioner and the pth temperature sensor during the adjustment cycle p , Represents the average value of the temperature difference between all temperature sensors and the air conditioner display. represents the distance between the air conditioner and the pth temperature sensor, Represents the average distance between the air conditioner and the temperature sensor. Represents the covariance of the temperature difference between the air conditioner and all sensors and the distance between the control and all sensors. Divided by the mean square error product of the two, that is The correlation coefficient ρ between temperature change and distance is obtained.

[0057] In order to realize the human-centered intelligent air conditioning adjustment function, it is necessary to obtain temperature interpolation based on the current location of the person.

[0058] The home video surveillance system in the Internet of Things system is used to determine the location of the personnel within the adjustment period, and then the temperature is adaptively interpolated according to the distance between the personnel location and each sensor.

[0059] Furthermore, the interpolation module 15 is used to calculate the value according to the formula Determine the temperature interpolation at the target location R, where R represents the location of the personnel, the number of temperature sensors is N, and ρ represents the correlation coefficient between the temperature change at the air-conditioning temperature and the distance at different temperature sensor locations. represents the distance between the target position and the pth temperature sensor, To normalize the distance result, the temperature interpolation weight is assigned according to the distance between the personnel position and the sensor. Then the correlation coefficient is multiplied by the distance weight and then multiplied by the temperature value of each temperature sensor to obtain the temperature interpolation value of the personnel position.

[0060] According to the temperature interpolation, the difference between the position and the expected adjustment temperature can be obtained, and the expected temperature value can be reached by adjusting the position where the personnel are located. Taking summer as an example, generally speaking, the optimal indoor temperature is 26 degrees. R When the temperature is higher than the expected temperature, the air conditioner temperature is automatically adjusted to lower the temperature so that the interpolation result at the personnel's location approaches or reaches the expected temperature of 26 degrees.

[0061] In this embodiment, due to the electromagnetic interference between indoor electrical appliances and the different interference levels, when the IoT system with multiple temperature sensors performs temperature interpolation, there is a problem of large interpolation error. We solve the problem of different duration, amplitude and nodes of temperature changes at different locations by performing mean smoothing iteration on the non-regulated periodic signals of multiple sensors, and make the subsequent temperature interpolation results more accurate for the intelligent control feedback results of indoor temperature, and the control effect is more outstanding.

Claims

1. A smart home control system based on the Internet of Things, characterized in that: include: An acquisition module, used for acquiring a temperature signal monitored by each indoor temperature sensor, wherein a plurality of indoor temperature sensors are arranged; A first determination module, configured to determine, according to each of the temperature signals, an optimal smoothing result of a temperature signal curve of each temperature sensor based on trend consistency, wherein the temperature signal curve includes the temperature signal component curve; A second determination module, configured to subtract the optimal smoothing result from the temperature signal to determine a residual signal of each temperature sensor; A processing module, used to group all amplitudes of the residual signal into a set, use the covariance matrix converted from the set as the noise covariance matrix of the Kalman filter system, and perform denoising on the real-time temperature signal; an interpolation module, configured to perform temperature interpolation processing on a target position according to a correlation coefficient between a temperature variation and a distance, and to adjust the air conditioner according to the processing result so that the target position obtains an expected temperature value, wherein the temperature variation comprises a difference between an air outlet temperature of the air conditioner and a temperature monitored by each of the temperature sensors, and the distance comprises a distance between an air outlet position of the air conditioner and each of the temperature sensors; The first determination module is specifically used to perform mean filtering iteration on the temperature signal of each sensor within a period of time, and output the smoothed signal of each iteration; perform first-order derivative on the curve composed of the smoothed signals, delete the sections with the first-order derivative of 0 from the curve of the smoothed signals, and reorganize the remaining signals in order to obtain the reorganized auxiliary signal; Performing alignment processing on the reorganized auxiliary signals using a dynamic time normalization (DTW) algorithm, and performing similarity processing on the processing results to determine the similarity between the aligned and reorganized auxiliary signals; and determining the optimal smoothing result of the temperature signal curve of each temperature sensor according to the similarity; The first determination module is further used to align trend items of the curve composed of the recombined auxiliary signals using a DTW algorithm to determine each aligned recombined auxiliary signal; The similarity between the aligned and reorganized auxiliary signals is determined according to the mean square error of the relative trend change between the aligned and reorganized auxiliary signals.

2. The smart home control system based on the Internet of Things according to claim 1 is characterized in that: The first determination module is specifically based on the formula Determine the similarity MSE(a,b) between the auxiliary signal a after alignment and reorganization and the auxiliary signal b after alignment and reorganization, is the temperature signal of sampling point number i of auxiliary signal a after alignment and reorganization, is the temperature signal of sampling point number i-1 of auxiliary signal a after alignment and reorganization, is the temperature signal of sampling point number i of auxiliary signal b after alignment and reorganization, is the temperature signal of sampling point number i-1 of auxiliary signal b after alignment and reorganization, is the maximum value of the temperature signal among all sampling point numbers of the auxiliary signal a after alignment and reorganization, is the most corrected value of the temperature signal among all sampling point numbers of the auxiliary signal a after alignment and reorganization, is the maximum value of the temperature signal among all sampling point numbers of the auxiliary signal b after alignment and reorganization, is the minimum value of the temperature signal among all sampling point numbers of the auxiliary signal b after alignment and reorganization. There are n sampling points on the auxiliary signal a after alignment and reorganization, and there are n sampling points on the auxiliary signal b after alignment and reorganization.

3. The smart home control system based on the Internet of Things according to claim 2 is characterized in that: The first determination module is also used to determine the optimal smoothing result according to the objective function E=min[σ(MSE)+μ(MSE)], wherein σ(MSE) is the mean square error of the mean square errors between all the auxiliary signals after alignment and reorganization, and μ(MSE) is the average value of the mean square errors between all the auxiliary signals after alignment and reorganization.

4. The smart home control system based on the Internet of Things according to claim 3 is characterized in that: The interpolation module is specifically used to determine the temperature difference ΔI between the air conditioner display temperature and the temperature monitored by the temperature sensor P through the air conditioner display temperature and the temperature monitored by the temperature sensor P. p According to the formula Determine the correlation coefficient ρ, where o represents the air conditioner display temperature, p represents any temperature sensor, ΔI p For I p -I o Represents the temperature difference ΔI between the air conditioner and the pth temperature sensor during the adjustment cycle p , Represents the average value of the temperature difference between all temperature sensors and the air conditioner display. represents the distance between the air conditioner and the pth temperature sensor, Represents the average distance between the air conditioner and the temperature sensor.

5. The smart home control system based on the Internet of Things according to claim 4 is characterized in that: The interpolation module is used to calculate the value according to the formula Determine the temperature interpolation of the target position R, where the number of temperature sensors is N, and ρ represents the correlation coefficient between the temperature change and the distance from the air-conditioning temperature to different temperature sensor positions. represents the distance between the target position and the pth temperature sensor, is the distance normalization result.

Images