Building space intelligent energy-saving control method based on multiple sensors and edge calculation

Abstract

The invention relates to the technical field of building energy saving, in particular to a building space intelligent energy-saving control method based on multiple sensors and edge calculation, which comprises the following steps of: S1, detecting a personnel existence state through fusion of a millimeter wave radar and an infrared sensor; s2, environment temperature and humidity, CO2 and PM2.5 data are collected, and a comfort index is calculated; s3, generating a device control instruction based on a decision engine of edge calculation; s4, the execution layer equipment responds to the instruction and feeds back an execution state; and S5, continuously learning the user behavior data optimization control strategy. Compared with the prior art, the building space intelligent energy-saving control method based on the multiple sensors and the edge calculation realizes automatic and digital upgrading of the space, so that automatic management of energy-consuming equipment in the space is realized, traditional manual management is replaced, the space comfort level is guaranteed, and meanwhile, the energy-saving efficiency of the space is improved. Energy waste in the building operation stage is thoroughly avoided, and the last square meter of energy saving of the building is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of building energy conservation, and particularly to an intelligent energy-saving control method for building spaces based on multi-sensors and edge computing.

Background Art

[0002] The existing building energy-saving methods usually rely on manual management. Methods such as timing control and zonal centralized control cannot adapt to the dynamic changes in personnel distribution, resulting in an irreconcilable contradiction between space comfort and energy conservation.

Summary of the Invention

[0003] To overcome the above problems, the present invention proposes an intelligent energy-saving control method for building spaces based on multi-sensors and edge computing, which can effectively solve the above problems.

[0004] A technical solution provided by the present invention to solve the above technical problems is: to provide an intelligent energy-saving control method for building spaces based on multi-sensors and edge computing, including the following steps:

[0005] Step S1, detecting the presence status of personnel by fusing millimeter-wave radar and infrared sensors;

[0006] In the step S1, it includes step S16 of excluding the influence of interference sources to locate the human body;

[0007] In the step S16, it includes the following steps:

[0008] Step S161, performing mean filtering on the echo signals collected by the radar to remove static noise;

[0009] Step S162, performing fast Fourier transform in the distance and velocity dimensions to obtain high-resolution range images and RD maps;

[0010] Step S163, applying a constant false alarm rate algorithm to the RD map to detect the presence of targets;

[0011] Step S164, searching for peaks in the data, then applying the Capon algorithm to estimate the angle and calculate the coordinate information;

[0012] Step S165, applying a density-based spatial clustering algorithm with noise and binary integration to further remove time-varying clutter and obtain all possible target points;

[0013] Step S166, eliminating false targets by analyzing the multipath position, identifying the presence of the human body, and performing positioning through an FMCW radar;

[0014] Step S2, collecting environmental temperature, humidity, CO2, and PM2.5 data and calculating the comfort index;

[0015] Step S3, the decision-making engine based on edge computing generates device control instructions;

[0016] Step S4, the execution layer device responds to the instructions and feeds back the execution status;

[0017] Step S5, continuously learn the user behavior data to optimize the control strategy.

[0018] Preferably, the step S1 includes the following steps:

[0019] Step S11, the infrared sensor continuously monitors the preset area and outputs digital signals at a frequency of 1 Hz. The millimeter-wave radar is in the default low-power scanning mode;

[0020] Step S12, when the infrared sensor detects the movement of the heat source three times in a row, it sends an interrupt request signal to the edge computing node. After receiving the interrupt, the edge computing node immediately sends a full-power scanning instruction to the millimeter-wave radar through the SPI interface;

[0021] Step S13, millimeter-wave radar data processing and infrared data calibration;

[0022] Step S14, fuse the millimeter-wave radar data and the infrared data to determine the presence status of the person;

[0023] Step S15, exception handling and optimization.

[0024] Preferably, in the step S13, it includes the following steps:

[0025] Step S131, the millimeter-wave radar adopts the FMCW mode, the transmitting frequency is 24 - 24.25 GHz, the bandwidth is 250 MHz, and the receiving end obtains the three-dimensional information of the target's distance, speed, and angle;

[0026] Step S132, the edge computing node extracts effective target points through the CFAR detection algorithm and adopts the DBSCAN clustering algorithm, setting parameters: neighborhood radius ε = 0.5 m, minimum number of points MinPts = 3;

[0027] Step S133, calculate the target speed variance σ 2 <0.01m 2 / s 2 , determine it as a stationary target, and perform micro-Doppler spectrum analysis on the stationary target, that is, breathing detection;

[0028] Step S134, dynamically correct the infrared sensor trigger threshold according to the ambient temperature;

[0029] Step S135, the infrared sensor performs RC low-pass filtering on the short-time pulse signal.

[0030] Preferably, in the step S14, it includes the following steps:

[0031] Step S141, use a hardware trigger signal to align the timestamps, ensure the error is < 10 ms, and interpolate and compensate the millimeter-wave radar data (cubic spline interpolation method); establish a radar coordinate system with the position of the millimeter-wave radar as the origin (0, 0, 0), and map the position coordinates (x_ir, y_ir, z_ir) of the infrared sensor to the radar coordinate system through a coordinate transformation matrix.

[0032] Step S142, target association and matching, use the Hungarian Algorithm for optimal matching, and the matching conditions are that the spatial distance is < 0.8 m and the included angle of the velocity directions is < 30°.

[0033] Step S143, calculate the confidence of the millimeter-wave radar and the confidence of the infrared.

[0034] Step S144, use D-S evidence theory for fusion, define the identification framework Θ = {occupied, unoccupied}, then the radar evidence m1(occupied) = C_radar, m1(Θ) = 1 - C_radar, and the infrared evidence m2(occupied) = C_ir, m2(Θ) = 1 - C_ir.

[0035] Step S145, use the Dempster combination rule to calculate the combined confidence:

[0036]

[0037] where K is the conflict factor, K = m1(occupied)m2(unoccupied) + m1(unoccupied)m2(occupied);

[0038] When m(occupied) > 0.85, it is determined that someone is present;

[0039] When 0.6 < m(occupied) ≤ 0.85, maintain the current state;

[0040] When m(occupied) ≤ 0.6, mark it as an unoccupied state.

[0041] Preferably, in the step S143, the confidence of the millimeter-wave radar C_radar = 1 - e^(-N_points / 5), where N_points is the number of clustering points, and the confidence of the infrared C_ir = 1 - |ΔT - 37°C| / 10, where ΔT is the temperature rise amplitude.

[0042] Preferably, in the step S15, when a certain sensor continuously outputs invalid data for 5 times, start the single-sensor mode, adjust the confidence threshold to 1.3 times the original value, and predict the missing data through a Kalman filter.

[0043] Preferably, in step S15, the vital signs are verified once an hour. The millimeter-wave radar detects the chest movement amplitude. If there is no valid breathing signal for 60 consecutive seconds, the infrared sensor is forced to perform a full-power scan.

[0044] Preferably, in step S15, the infrared signal is subjected to spectral analysis to exclude 50Hz power frequency interference, and an adaptive notch filter is added to the radar signal processing chain.

[0045] Preferably, the building space intelligent energy-saving control method based on multi-sensors and edge computing includes a building space intelligent energy-saving control system, and the building space intelligent energy-saving control system includes:

[0046] A perception layer, which includes a millimeter-wave radar system, an infrared sensor array, a temperature and humidity sensor, a gas sensor, and a light sensor;

[0047] An edge computing node, which is used to receive the data of the perception layer and run an AI control algorithm;

[0048] An execution layer, which includes an air-conditioning control module, a lighting control module, and a window curtain driving module;

[0049] A wireless communication module, which uses the Bluetooth Mesh protocol to connect the devices of the perception layer and the execution layer;

[0050] The building space intelligent energy-saving control system determines the presence status of personnel through multi-sensor data fusion and dynamically adjusts the operating parameters of the air conditioner and lighting.

[0051] Preferably, the millimeter-wave radar system includes multiple millimeter-wave radar modules. The millimeter-wave radar modules are installed at the center of the ceiling or on the wall at a height of 2.8m from the ground. Each millimeter-wave radar module covers a fan-shaped area with a radius of 6-8m, and seamless coverage is achieved through multi-node deployment; the millimeter-wave radar module forms an angle of 55° with the wall.

[0052] Compared with the prior art, the building space intelligent energy-saving control method based on multi-sensors and edge computing of the present invention utilizes technologies such as big data, artificial intelligence, the Internet of Things, and edge computing to automate and digitalize the space upgrade, thereby realizing the automatic management of energy-consuming devices in the space, replacing traditional manual management, ensuring the comfort of the space while completely avoiding energy waste during the building operation stage, and opening up the last square meter of building energy conservation.

Description of the Drawings

[0053] Figure 1 It is a flowchart of the building space intelligent energy-saving control method based on multi-sensors and edge computing of the present invention.

Detailed Embodiments

[0054] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0055] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are limited to the relative positions on the specified views, rather than absolute positions.

[0056] In addition, in the present invention, descriptions such as "first" and "second" are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present invention, the meaning of "a plurality" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0057] Please refer to Figure 1 , the intelligent energy-saving control method for building spaces based on multi-sensors and edge computing of the present invention includes the following steps:

[0058] Step S1, detecting the presence status of personnel through the fusion of millimeter-wave radar and infrared sensors;

[0059] Step S2, collecting environmental temperature and humidity, CO2, and PM2.5 data and calculating the comfort index;

[0060] Step S3, generating device control instructions based on the decision-making engine of edge computing;

[0061] Step S4, the execution layer device responds to the instruction and feeds back the execution status;

[0062] Step S5, continuously learning user behavior data to optimize the control strategy.

[0063] The said step S1 includes the following steps:

[0064] Step S11, the infrared sensor continuously monitors a preset area (such as the door area, corridor entrance), outputs a digital signal at a frequency of 1 Hz (1 = there is a heat source moving, 0 = none), and the millimeter-wave radar is default in a low-power scanning mode (duty cycle 10%, only monitors moving targets).

[0065] Step S12, when the infrared sensor detects the movement of the heat source continuously three times (that is, outputs three "1" signals, lasting for 3 seconds), send an interrupt request signal to the edge computing node. After receiving the interrupt, the edge computing node immediately sends a full-power scanning instruction to the millimeter-wave radar through the SPI interface.

[0066] Step S13, millimeter-wave radar data processing and infrared data calibration.

[0067] In the said step S13, the following steps are included:

[0068] Step S131, the millimeter-wave radar adopts the FMCW (Frequency Modulated Continuous Wave) mode, with a transmission frequency of 24 - 24.25 GHz, a bandwidth of 250 MHz, and the receiving end obtains the three-dimensional information of the target's distance (Δf calculation), speed (Doppler frequency shift), and angle (phase difference).

[0069] Step S132, the edge computing node extracts effective target points through the CFAR (Constant False Alarm Rate) detection algorithm (false alarm rate < 10^-6), and adopts the DBSCAN clustering algorithm, setting the parameters: neighborhood radius ε = 0.5 m, minimum number of points MinPts = 3.

[0070] Step S133, calculate the target speed variance σ 2 <0.01m 2 / s 2 , determine it as a stationary target, and perform micro-Doppler spectrum analysis (frequency range 0.1 - 0.5 Hz) on the stationary target, that is, respiration detection.

[0071] Step S134, dynamically correct the infrared sensor trigger threshold according to the ambient temperature (obtained from the ambient sensor).

[0072] Step S135, the infrared sensor performs RC low-pass filtering (cutoff frequency 0.2 Hz) on the short-time pulse signal (duration < 0.5 s).

[0073] Step S14, fuse the millimeter-wave radar data and the infrared data to determine the presence status of personnel.

[0074] In the said step S14, the following steps are included:

[0075] Step S141, use the hardware trigger signal to align the timestamps to ensure the error < 10 ms, and perform interpolation compensation on the millimeter-wave radar data (cubic spline interpolation method); establish a radar coordinate system with the position of the millimeter-wave radar as the origin (0, 0, 0), and map the position coordinates (x_ir, y_ir, z_ir) of the infrared sensor to the radar coordinate system through the coordinate transformation matrix.

[0076] Step S142, target association and matching, adopt the Hungarian Algorithm for optimal matching, and the matching condition is that the spatial distance < 0.8 m and the included angle of the speed direction < 30°.

[0077] Step S143, calculate the millimeter-wave radar confidence and the infrared confidence.

[0078] In step S143, the confidence level of the millimeter-wave radar \(C_{radar}=1 - e^{(-N_{points} / 5)}\) (\(N_{points}\) is the number of clustering points), and the confidence level of the infrared \(C_{ir}=1-\vert\Delta T - 37^{\circ}C\vert / 10\) (\(\Delta T\) is the temperature rise amplitude).

[0079] Step S144, using D-S evidence theory for fusion, defining the identification framework \(\Theta=\{occupied, unoccupied\}\), then the radar evidence \(m1(occupied)=C_{radar}\), \(m1(\Theta)=1 - C_{radar}\), and the infrared evidence \(m2(occupied)=C_{ir}\), \(m2(\Theta)=1 - C_{ir}\).

[0080] Step S145, using the Dempster combination rule to calculate the combined confidence level:

[0081]

[0082] Among them, \(K\) is the conflict factor, \(K = m1(occupied)m2(unoccupied)+m1(unoccupied)m2(occupied)\);

[0083] When \(m(occupied)>0.85\), it is determined that there is someone present;

[0084] When \(0.6 < m(occupied)\leq0.85\), the current state is maintained;

[0085] When \(m(occupied)\leq0.6\), it is marked as the unoccupied state.

[0086] Step S15, anomaly handling and optimization.

[0087] In step S15, when a certain sensor continuously outputs invalid data for 5 times, the single-sensor mode is started, the confidence level threshold is adjusted to 1.3 times the original value, and the missing data is predicted through a Kalman filter.

[0088] In step S15, a vital sign verification is performed once an hour. The millimeter-wave radar detects the chest movement amplitude (accuracy \(\pm1mm\)). If there is no valid breathing signal for 60 consecutive seconds, the infrared sensor is forced to perform a full-power scan.

[0089] In step S15, the infrared signal is spectroscopically analyzed to exclude 50Hz power frequency interference, and an adaptive notch filter is added to the radar signal processing chain.

[0090] In step S1, it also includes step S16, excluding the influence of interference sources to locate the human body.

[0091] Due to the interference of the indoor environment and complex objects such as green plants and curtain swings, the radar signal will undergo penetration, reflection, refraction, and scattering, and the echo signal will contain noise, clutter, and multipath interference with different characteristics.

[0092] In step S16, the following steps are included:

[0093] Step S161: Perform mean filtering on the echo signal collected by the radar to remove static noise.

[0094] Step S162: Perform fast Fourier transform in the range and velocity dimensions to obtain a high-resolution range profile (HRRP) and an RD map.

[0095] Step S163: Apply the constant false alarm rate (CFAR) algorithm to the RD map to detect the presence of targets.

[0096] Step S164: Search for peaks in the data, then apply the Capon algorithm to estimate the angle and calculate the coordinate information.

[0097] Step S165: Apply the density-based spatial clustering of applications with noise (DBSCAN) algorithm and binary integration to further remove time-varying clutter and obtain all possible target points.

[0098] Step S166: Eliminate false targets by analyzing the multipath position, identify the presence of the human body, and perform positioning through the FMCW radar.

[0099] Through step S16, in a variety of complex environments, such as in environments with green plants, curtains, or furniture, the real target human body can be located with a small error. At the same time, when the human body is behind the potted plant and beside the potted plant, the detection probability can be improved.

[0100] In step S2, through data preprocessing such as temperature and humidity compensation, CO2 concentration correction, and PM2.5 mass concentration conversion, combined with a multi-parameter fusion model, the comfort index (ECI) calculation formula is obtained:

[0101] ECI = w T ×(T norm ) + w H ×(H norm ) + w C ×(C norm ) + w P ×(P norm )

[0102] Among them, w T 、w H 、w C 、w P are the weight coefficients of temperature, humidity, CO2 concentration, and PM2.5 concentration respectively.

[0103] The building space intelligent energy-saving control method based on multi-sensors and edge computing of the present invention includes a building space intelligent energy-saving control system. The building space intelligent energy-saving control system includes:

[0104] The sensing layer, which includes a millimeter-wave radar system, an infrared sensor array, a temperature and humidity sensor, a gas sensor, and a light sensor;

[0105] Edge computing nodes, which are used to receive sensing layer data and run AI control algorithms;

[0106] The execution layer, which includes an air-conditioning control module, a lighting control module, and a window curtain driving module;

[0107] A wireless communication module, which uses the Bluetooth Mesh protocol to connect the devices of the sensing layer and the execution layer;

[0108] The intelligent energy-saving control system for building spaces determines the presence status of personnel through multi-sensor data fusion and dynamically adjusts the operating parameters of air conditioners and lights.

[0109] The millimeter-wave radar system includes multiple millimeter-wave radar modules. The millimeter-wave radar modules are installed at the center of the ceiling or on the wall at a height of 2.8 m from the ground. Each millimeter-wave radar module covers a fan-shaped area with a radius of 6 - 8 m, and seamless coverage is achieved through multi-node deployment. The millimeter-wave radar module forms an angle of 55° with the wall, which is conducive to expanding the coverage range of millimeter-wave signals and enabling them to cover a larger human body area before reflection.

[0110] The millimeter-wave radar module uses a 24 GHz or 60 GHz frequency band, has a beam width of 80°, and supports micro-motion detection, such as respiration and finger motion detection.

[0111] The infrared sensor array is installed above the door frame for monitoring the entry and exit of personnel, or installed on top of the workstation partition for monitoring local presence.

[0112] The infrared sensor array uses a 120° wide-angle infrared probe, has a detection distance of 5 m, and a sampling frequency of 1 Hz.

[0113] The temperature and humidity sensor, gas sensor, and light sensor are installed on the wall at a height of 1.5 m from the ground (simulating the human perception height), and one set is deployed for every 50 ㎡.

[0114] One to two edge computing nodes are set on each floor of the building, adjusted according to the area, and installed in the weak current well or the ceiling.

[0115] Compared with the prior art, the intelligent energy-saving control method for building spaces based on multi-sensors and edge computing of the present invention utilizes technologies such as big data, artificial intelligence, Internet of Things, and edge computing to automate and digitalize the upgrade of spaces, thereby realizing the automatic management of energy-consuming devices in the spaces, replacing traditional manual management, and while ensuring the comfort of the spaces, completely avoiding energy waste during the building operation stage and achieving energy conservation for the last square meter of the building.

[0116] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any modifications, equivalent substitutions, and improvements made within the concept of the present invention shall be included in the patent protection scope of the present invention.

Claims

1. An intelligent energy-saving control method for building spaces based on multi-sensors and edge computing, characterized in that, It includes the following steps: Step S1, detecting the presence status of a person through the fusion of millimeter-wave radar and infrared sensor; In the step S1, it includes step S16, excluding the influence of interference sources to locate the human body; In the step S16, it includes the following steps: Step S161, performing mean filtering on the echo signals collected by the radar to remove static noise; Step S162, performing fast Fourier transform in the distance and velocity dimensions to obtain high-resolution range images and RD maps; Step S163, applying the constant false alarm rate algorithm to the RD map to detect the presence of targets; Step S164, searching for peaks in the data, then applying the Capon algorithm to estimate the angles and calculating the coordinate information; Step S165, applying the density-based spatial clustering of applications with noise algorithm and binary integration to further remove time-varying clutter and obtain all possible target points; Step S166, eliminating false targets by analyzing the multipath positions, identifying the presence of the human body, and performing positioning through the FMCW radar; Step S2, collecting environmental temperature and humidity, CO2, and PM2.5 data and calculating the comfort index; Step S3, generating device control instructions based on the edge computing decision engine; Step S4, the execution layer device responds to the instructions and feeds back the execution status; Step S5, continuously learning user behavior data to optimize the control strategy.

2. The intelligent energy-saving control method for building space based on multi-sensors and edge computing according to claim 1, wherein The step S1 includes the following steps: Step S11, the infrared sensor continuously monitors the preset area and outputs digital signals at a frequency of 1 Hz, and the millimeter-wave radar defaults to the low-power scanning mode; Step S12, when the infrared sensor detects the movement of a heat source continuously three times, it sends an interrupt request signal to the edge computing node. After receiving the interrupt, the edge computing node immediately sends a full-power scanning instruction to the millimeter-wave radar through the SPI interface; Step S13, millimeter-wave radar data processing and infrared data calibration; Step S14, fusing the millimeter-wave radar data and the infrared data to determine the presence status of the person; Step S15, anomaly handling and optimization.

3. The intelligent energy-saving control method for building space based on multi-sensors and edge computing according to claim 2, characterized in that, In the step S13, it includes the following steps: Step S131, the millimeter-wave radar adopts the FMCW mode, with a transmission frequency of 24 - 24.25 GHz, a bandwidth of 250 MHz, and the receiving end obtains the three-dimensional information of the target's distance, velocity, and angle; Step S132, the edge computing node extracts effective target points through the CFAR detection algorithm and adopts the DBSCAN clustering algorithm, setting parameters: neighborhood radius ε = 0.5 m, minimum number of points MinPts = 3; Step S133, calculate the target speed variance σ 2 <0.01m 2 / s 2 , determine it as a stationary target, and perform micro-Doppler spectrum analysis on the stationary target, that is, respiration detection; Step S134, dynamically correcting the infrared sensor trigger threshold according to the environmental temperature; Step S135, the infrared sensor performs RC low-pass filtering on the short-time pulse signal.

4. The intelligent energy-saving control method for building space based on multi-sensors and edge computing according to claim 2, characterized in that, In the step S14, it includes the following steps: Step S141, using the hardware trigger signal to align the timestamps to ensure the error < 10 ms, interpolating and compensating the millimeter-wave radar data (cubic spline interpolation method); establishing a radar coordinate system with the position of the millimeter-wave radar as the origin (0, 0, 0), and mapping the position coordinates (x_ir, y_ir, z_ir) of the infrared sensor to the radar coordinate system through the coordinate transformation matrix; Step S142, target correlation matching, using the Hungarian Algorithm for optimal matching, with the matching conditions being a spatial distance < 0.8m and an angle between the velocity directions < 30°; Step S143, calculate the confidence of the millimeter-wave radar and the confidence of the infrared sensor; Step S144, use D-S evidence theory for fusion. Define the identification framework Θ = {occupied, unoccupied}. Then the radar evidence m1(occupied) = C_radar, m1(Θ) = 1 - C_radar, and the infrared evidence m2(occupied) = C_ir, m2(Θ) = 1 - C_ir; Step S145, use Dempster's combination rule to calculate the combined confidence: where K is the conflict factor, K = m1(occupied)m2(unoccupied) + m1(unoccupied)m2(occupied); When m(occupied) > 0.85, it is determined that someone is present; When 0.6 < m(occupied) ≤ 0.85, the current state is maintained; When m(occupied) ≤ 0.6, it is marked as the unoccupied state.

5. The intelligent energy-saving control method for building space based on multi-sensors and edge computing according to claim 4, characterized in that In the said Step S143, the confidence of the millimeter-wave radar C_radar = 1 - e^(-N_points / 5), where N_points is the number of clustering points, and the confidence of the infrared sensor C_ir = 1 - |ΔT - 37°C| / 10, where ΔT is the temperature rise amplitude.

6. The intelligent energy-saving control method for building space based on multi-sensors and edge computing according to claim 2, wherein In the said Step S15, when a certain sensor outputs invalid data continuously for 5 times, the single-sensor mode is started, the confidence threshold is adjusted to 1.3 times the original value, and the missing data is predicted through a Kalman filter.

7. The intelligent energy-saving control method for building space based on multi-sensors and edge computing according to claim 2, wherein, In the said Step S15, a vital sign verification is performed once an hour. The millimeter-wave radar detects the chest movement amplitude. If there is no valid breathing signal for 60 consecutive seconds, the infrared sensor is forced to perform a full-power scan.

8. The intelligent energy-saving control method for building space based on multi-sensors and edge computing according to claim 2, characterized in that, In the said Step S15, the infrared signal is spectrally analyzed to exclude 50Hz power frequency interference, and an adaptive notch filter is added to the radar signal processing chain.

9. The intelligent energy-saving control method for building space based on multi-sensors and edge computing according to claim 1, characterized in that, The building space intelligent energy-saving control method based on multi-sensor and edge computing includes a building space intelligent energy-saving control system, and the building space intelligent energy-saving control system includes: The sensing layer, which includes a millimeter-wave radar system, an infrared sensor array, a temperature and humidity sensor, a gas sensor, and a light sensor; The edge computing node, which is used to receive the data of the sensing layer and run the AI control algorithm; The execution layer, which includes an air-conditioning control module, a lighting control module, and a window curtain drive module; The wireless communication module, which uses the Bluetooth Mesh protocol to connect the devices of the sensing layer and the execution layer; The building space intelligent energy-saving control system determines the presence state of people through multi-sensor data fusion and dynamically adjusts the operating parameters of the air conditioner and lighting.

10. The intelligent energy-saving control method for building space based on multi-sensors and edge computing according to claim 9, wherein The millimeter-wave radar system includes multiple millimeter-wave radar modules. The millimeter-wave radar modules are installed at the center of the ceiling or on the wall at a height of 2.8m from the ground. Each millimeter-wave radar module covers a fan-shaped area with a radius of 6 - 8m, and seamless coverage is achieved through multi-node deployment; the millimeter-wave radar module forms an angle of 55° with the wall.

Images