Energy-saving management method and system based on smart building

Abstract

This application provides an energy-saving management method and system based on smart buildings, including: determining the nearest air outlet to the turbulent airflow area based on the location coordinates of the turbulent airflow area; adjusting the airflow velocity and direction of the air outlet to provide directional airflow to the turbulent airflow area; acquiring airflow velocity distribution data of the turbulent airflow area and its surroundings, and modifying the boundary conditions and turbulence parameters of the airflow field based on the results of cleanroom airflow numerical simulation; sending the airflow temperature, airflow humidity, and airflow volume in the airflow strategy to the air conditioning control system to regulate the turbulent airflow area and suppress temperature and humidity fluctuations in the turbulent airflow area; continuously monitoring the particulate matter concentration change in the turbulent airflow area, and determining that the airflow in the turbulent airflow area has stabilized and removing the turbulent airflow area marker when the particulate matter concentration drops below the background value.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to an energy-saving management method and system based on a smart building. Background Art

[0002] In the clean rooms of modern buildings, environmental control is the key to ensuring product quality and production efficiency in fields such as microelectronics and biopharmaceuticals that require extremely high environmental cleanliness. However, the clean room air conditioning automatic control system is facing a complex and severe technical challenge. Since the heat load and humidity load inside the clean room often fluctuate violently, the air conditioning system must have the ability to respond quickly and adjust accurately. However, this requirement has led to a series of chain reactions, namely, the rapid changes in temperature and humidity may cause indoor air flow disorder, destroy the stable laminar state, and then affect the control effect of particulate matter, threatening the core function of the clean room; at the same time, in order to achieve the purpose of rapid response, the system has to increase the operating power of the equipment, resulting in a significant increase in energy consumption. Therefore, the designers of the air conditioning system are caught in a dilemma, that is, how to achieve effective energy saving without compromising the performance of the clean room. Excessive pursuit of rapid response may cause a surge in energy consumption, while excessive emphasis on energy saving may weaken the effect of cleanliness control. This contradiction is particularly prominent in actual operation. For example, when faced with the sudden start-up of the heat source equipment in the clean room, if the air conditioning system does not respond enough, it may cause abnormal room temperature and affect the production process; on the contrary, overreaction will cause energy waste. Therefore, the core difficulty of the clean room air conditioning automatic control system is how to accurately balance the relationship between response speed, energy consumption control and cleanliness maintenance in a constantly changing environment, find the best balance point, and achieve efficient, energy-saving and high-cleanliness environmental control. This is a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present invention provides an energy-saving management method based on a smart building, which mainly includes:

[0004] Obtain the temperature and relative humidity data of different areas in the clean room, calculate the temperature gradient and humidity gradient of each area, and if the temperature gradient or humidity gradient of a certain area exceeds the preset threshold, it is determined that there is heat accumulation or water vapor accumulation in the area, and the area is marked as an abnormal temperature and humidity area;

[0005] Obtain the streamline curvature and flow velocity pulsation data at the boundary layer of the temperature and humidity anomaly area, and determine whether the airflow turbulence in the area exceeds the standard by calculating the streamline curvature change rate and the flow velocity pulsation intensity. If the turbulence exceeds the preset threshold, the area is determined to be an airflow turbulence area, and the position coordinates of the airflow turbulence area are recorded;

[0006] According to the position coordinates of the air flow turbulence area, determine the air supply outlet closest to the air flow turbulence area, and provide directional air supply to the air flow turbulence area by adjusting the air supply speed and air supply direction of the air supply outlet;

[0007] Obtaining the velocity distribution data of the airflow field in the turbulent airflow area and its surroundings, and correcting the boundary conditions and turbulence parameters of the airflow field in combination with the numerical simulation results of the clean room airflow;

[0008] The corrected airflow field is used to simulate the indoor airflow organization forms under different air supply volumes and fresh air ratios. The air supply strategy with the lowest turbulence and the best energy-saving effect is determined by calculating the turbulence and energy consumption indicators under various air supply conditions.

[0009] The air supply temperature, air supply humidity and air supply volume in the air supply strategy are sent to the air conditioning control system to regulate the air flow turbulence area and suppress the temperature and humidity fluctuations in the air flow turbulence area;

[0010] The particle concentration change in the turbulent airflow area is continuously monitored. When the particle concentration drops below the background value, it is determined that the airflow in the turbulent airflow area has returned to stability, and the turbulent airflow area mark is removed.

[0011] The present invention provides an energy-saving management system based on a smart building, which mainly includes:

[0012] A data acquisition module, used to collect temperature and humidity data in the clean room;

[0013] Anomaly detection module, which calculates temperature and humidity gradients and identifies abnormal areas;

[0014] Airflow analysis module, used to analyze the curvature and flow velocity pulsation in turbulent airflow areas;

[0015] The air outlet adjustment module is used to optimize the wind speed and direction of the air outlet in the air turbulence area;

[0016] The simulation and optimization module is used to select the best air supply strategy through airflow field simulation;

[0017] Monitoring and control module, used to monitor particle concentration and adjust airflow stability.

[0018] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0019] The present invention discloses an energy-saving management method based on smart buildings, which significantly improves the performance and efficiency of the clean room air conditioning automatic control system. Faced with the drastic fluctuations of heat and humidity loads in the clean room, as well as the challenges of air flow turbulence and rising energy consumption, the optimal balance between response speed, energy consumption control and cleanliness maintenance is achieved through intelligent monitoring and regulation. Specifically, by integrating temperature and humidity sensors and particle image velocimetry technology, it is possible to monitor and accurately identify abnormal temperature and humidity areas and air flow turbulence areas in the clean room in real time, and timely warn of potential environmental risks. For abnormal temperature and humidity areas, the abnormal areas are automatically marked by calculating the temperature gradient and humidity gradient, and the air supply strategy is optimized based on the results of air flow numerical simulation, and the environmental parameters are effectively regulated to ensure the stability of the production environment and product quality. For the air flow turbulence area, the present invention accurately judges the degree of turbulence by analyzing the rate of change of streamline curvature and the intensity of flow velocity pulsation, and intelligently adjusts the wind speed and direction of the air supply port to directional air supply to suppress turbulence, optimize air flow organization, quickly restore the laminar state, and ensure cleanliness. More importantly, while ensuring the core functions of the clean room, the present invention also focuses on energy conservation and emission reduction. By optimizing the air supply strategy and control parameters, the energy consumption of the air conditioning system is significantly reduced, and efficient and energy-saving environmental control is achieved. In summary, the present invention not only effectively solves the technical problems in the clean room environmental control, but also improves production efficiency and reduces operating costs through intelligent control strategies, providing strong technical support for the sustainable development of clean rooms. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of an energy-saving management method based on smart buildings of the present invention.

[0021] Figure 2 A schematic diagram of an energy-saving management method and system based on a smart building of the present invention.

[0022] Figure 3 This is another schematic diagram of an energy-saving management method and system based on a smart building of the present invention. DETAILED DESCRIPTION

[0023] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] like Figure 1-3 In this embodiment, a smart building-based energy-saving management method and system may specifically include:

[0025] S101. Obtain the temperature and relative humidity data of different areas of the clean room, calculate the temperature gradient and humidity gradient of each area, and if the temperature gradient or humidity gradient of a certain area exceeds a preset threshold, it is determined that heat accumulation or water vapor accumulation exists in the area, and the area is marked as an abnormal temperature and humidity area.

[0026] Acquire real-time temperature and relative humidity data collected by temperature and humidity sensors pre-arranged in different areas of the clean room; use the finite difference method to calculate the temperature gradient and humidity gradient between adjacent areas based on the real-time temperature and relative humidity data; determine whether the temperature gradient and humidity gradient exceed the preset temperature gradient threshold and humidity gradient threshold for the temperature gradient and humidity gradient; if the temperature gradient or humidity gradient exceeds the corresponding preset threshold, determine that the area is an abnormal temperature and humidity area; use a two-dimensional coordinate system to record the position information of the abnormal temperature and humidity area, and determine the boundary range of the abnormal temperature and humidity area by an integration method; calculate the duration of the abnormal condition of the abnormal temperature and humidity area based on the continuously collected temperature and humidity data; generate an abnormal report based on the position information, abnormal gradient value, boundary range and duration of the abnormal temperature and humidity area; store the abnormal report in a preset database.

[0027] Specifically, real-time temperature and relative humidity data are obtained from temperature and humidity sensors pre-arranged in different areas of the clean room, and the finite difference method is used to calculate the temperature and humidity change rates between adjacent areas to obtain the temperature gradient and humidity gradient of each area. The calculated gradient value is compared with the preset temperature gradient threshold and humidity gradient threshold to determine the temperature and humidity conditions of each area. If the temperature gradient or humidity gradient of a certain area exceeds the corresponding preset threshold, it is determined that there is heat accumulation or water vapor accumulation in the area, and the area is marked as an abnormal temperature and humidity area. The location information of the abnormal area is recorded using a two-dimensional coordinate system, and the maximum temperature gradient and humidity gradient values ​​of the area are recorded. The area of ​​the marked abnormal temperature and humidity area is calculated, and the boundary range of the abnormal area is determined by the integral method. According to the continuously collected data, the duration of the abnormal condition is calculated to determine the severity of the abnormal condition. Based on the location information, abnormal value, area and duration of the abnormal area, an abnormal report is generated, including the specific coordinates of the abnormal area, the abnormal gradient value, the size of the affected range and the duration. The generated abnormal report data is stored in the preset database for subsequent statistical analysis and early warning. 20 temperature and humidity sensors are pre-arranged in the clean room to form a 4x5 grid distribution. Each sensor collects temperature and relative humidity data every 5 seconds. The DHT22 digital temperature and humidity sensor is used, with a temperature measurement range of -40℃ to 80℃ and an accuracy of ±0.5℃; the relative humidity measurement range is 0-100% and an accuracy of ±2%. After acquiring the data, the temperature gradient and humidity gradient are calculated using the central difference method. For example, for the temperature gradient at position (i, j), the calculation formula is gradient x=(T(i+1, j)-T(i-1, j)) / (2dx), gradient y=(T(i, j+1)-T(i, j-1)) / (2dy), where T(i, j) represents the temperature value at position (i, j) on a two-dimensional grid. The grid can be used to simulate the temperature distribution in a region, where i and j are the row and column coordinates of the grid point; T(i+1, j) and T(i-1, j) represent the temperature values ​​of the points on the grid that are in the same row as position (i, j) but adjacent to the right and left, respectively. The coordinates of these points are The labels are (i+1, j) and (i-1, j); T(i, j+1) and T(i, j-1) represent the temperature values ​​of the points in the same column as position (i, j) on the grid but adjacent to it downward and upward, and the coordinates of these points are (i, j+1) and (i, j-1) respectively; dx represents the sensor spacing or grid step length along the x-axis direction, i.e. the horizontal direction, in the grid, which is the distance between position (i, j) and its adjacent x-direction points; dy represents the sensor spacing or grid step length along the y-axis direction, i.e. the vertical direction, in the grid, which is the distance between position (i, j) and its adjacent y-direction points. Set the temperature gradient threshold to 0.5℃ / m and the humidity gradient threshold to 2% / m. If the temperature gradient or humidity gradient of a certain area exceeds the threshold, it is marked as an abnormal area.The abnormal area is recorded using a two-dimensional coordinate system. The origin (0,0) is set at the lower left corner of the clean room. The x and y coordinates represent the horizontal and vertical distances from the origin, respectively, in meters. The maximum temperature gradient and humidity gradient values ​​of the abnormal area are recorded. The discretization method is used to calculate the area of ​​the abnormal area. The area is divided into small grids of 0.1m*0.1m. The number of abnormal grids is counted and multiplied by the area of ​​the small grid. The duration of the abnormality is determined by continuous data collection. If 10 consecutive samples (50 seconds) show abnormalities, it is determined to be a continuous abnormality. Generate an abnormality report, including the coordinate range of the abnormal area (2.5,3.0) to (3.5,4.0), the maximum temperature gradient of 0.7℃ / m, the maximum humidity gradient of 2.5% / m, the affected area of ​​1.2 square meters, and the duration of 120 seconds. The abnormality report is stored in the "anomaly_records" table of the MySQL database, including timestamp, location, gradient value, area, and duration fields.

[0028] S102. Obtain the streamline curvature and flow velocity pulsation data at the boundary layer of the temperature and humidity anomaly area, and determine whether the airflow turbulence in the area exceeds the standard by calculating the streamline curvature change rate and the flow velocity pulsation intensity. If the turbulence exceeds the preset threshold, the area is determined to be an airflow turbulence area, and the position coordinates of the airflow turbulence area are recorded.

[0029] The thickness of the boundary layer is determined according to the temperature gradient method, and the airflow velocity and direction data are obtained at the boundary layer; the airflow streamline is fitted by the least square method to obtain the streamline curvature, and the streamline curvature is used to calculate the streamline curvature value; the airflow velocity collected by the hot wire anemometer array is obtained, the flow velocity root mean square value is calculated, and the flow velocity pulsation intensity is obtained; the Reynolds number is calculated according to the streamline curvature value and the flow velocity pulsation intensity, and it is determined whether the Reynolds number exceeds the preset turbulence threshold; if the Reynolds number exceeds the preset turbulence threshold, the area is determined to be an airflow turbulence area; the Kriging interpolation algorithm is used to determine the precise boundary of the airflow turbulence area; data is obtained from the hot wire anemometer array, and the three-dimensional spatial coordinates of the airflow turbulence area are calculated by the triangulation principle; the position information of the airflow turbulence area is recorded, and the position information includes the center coordinates and the boundary coordinates.

[0030] Specifically, an array of hot-wire anemometers is installed around the temperature and humidity anomaly area, and the temperature gradient method is used to determine the thickness of the boundary layer. The airflow velocity and direction data are collected at the boundary layer. The airflow streamlines are fitted by the least squares method, and the streamline curvature is calculated to obtain the streamline curvature value. The specific process is to take the coordinates (xi, yi) of n adjacent measuring points, fit the quadratic curve y=ax^2+bx+c, and calculate the curvature k=|2a| / (1+(2ax+b)^2)^(3 / 2). The airflow velocity data collected by the hot-wire anemometer is used to calculate the root mean square value of the velocity to obtain the velocity pulsation intensity. The calculation method is to find the standard deviation of the velocity data over a period of time. The Reynolds number is calculated based on the streamline curvature and the velocity pulsation intensity as an indicator of airflow turbulence. The calculated Reynolds number is compared with the preset turbulence threshold, and the threshold is set based on historical data statistics. If the Reynolds number exceeds the preset threshold, the area is determined to be an airflow turbulence area. Using the installed hot-wire anemometer array data, the three-dimensional spatial coordinates of the airflow turbulence zone are calculated by triangulation principle. The Kriging interpolation algorithm is used to determine the precise boundary of the turbulent area. The location information of the airflow turbulence zone, including the center coordinates and boundary coordinates, is recorded. The location coordinates, Reynolds value, boundary layer thickness data and corresponding timestamp information of the airflow turbulence zone are stored in a relational database for subsequent analysis of the time trend of the turbulent condition. 16 hot-wire anemometers are arranged around the temperature and humidity anomaly area to form a 4x4 matrix array, with each anemometer spaced 0.5 meters apart. The boundary layer thickness is determined by the temperature gradient method. When the vertical temperature gradient is less than 0.1℃ / cm, it is determined to be the top of the boundary layer. At the boundary layer, each hot-wire anemometer collects airflow velocity and direction data at a frequency of 100Hz for 60 seconds. The least squares method is applied to the collected data to fit the airflow streamlines. The coordinates of the five adjacent measuring points are taken to fit the quadratic curve y=ax^2+bx+c. The curvature k=|2a| / (1+(2ax+b)^2)^(3 / 2) is calculated to obtain the streamline curvature value. x and y represent the horizontal and vertical coordinates in the two-dimensional coordinate system, where x represents the horizontal position and y represents the vertical position. The coordinates correspond to points on the airflow path. a, b, and c are the coefficients of the quadratic curve y=ax2+bx+c. These coefficients will be obtained from the given measuring point data through the least squares method. Specifically, a is the coefficient of the quadratic term, which determines the opening direction and width of the curve. b is the coefficient of the linear term, which affects the inclination of the curve. c is the constant term, which determines the vertical position of the curve. 2a is the coefficient of the second derivative, which reflects the acceleration change of the curve in the x direction, that is, the degree of curvature. 1+(2ax+b)^2 represents the square of the slope of the tangent line of the curve plus 1, which is used to standardize the calculation of the curvature. At the same time, the standard deviation of the velocity data within 60 seconds is calculated to obtain the flow velocity pulsation intensity.Substitute the streamline curvature and velocity pulsation intensity into the Reynolds number calculation formula Re=(ρvL) / μ, where ρ is the air density (1.225kg / m^3), v is the average velocity, L is the characteristic length (taken as 0.5 meters), and μ is the air dynamic viscosity (1.81×10^-5Pa·s). The calculated Reynolds number is used as an index of airflow turbulence. By analyzing historical data, the Reynolds number threshold is set to 2300. If the calculated Reynolds number exceeds 2300, the area is judged to be an airflow turbulence area. Using the triangulation principle, the three-dimensional spatial coordinates of the airflow turbulence area are calculated based on the known position of the hot wire anemometer and the measured airflow velocity vector. The Kriging interpolation algorithm is used to spatially interpolate the Reynolds number with a grid interval of 0.1 meters to determine the precise boundary of the turbulent area. Record the center coordinates and boundary coordinates of the turbulent area, such as the center coordinates (2.5, 3.0, 1.5) meters, the boundary range is 2.0-3.0 meters for x, 2.5-3.5 meters for y, and 1.0-2.0 meters for z. Store the location coordinates of the turbulent area, the Reynolds value of 2500, the boundary layer thickness of 0.15 meters, and the corresponding timestamp information in the "turbulence_records" table of the MySQL database for subsequent analysis of the time trend of the turbulence condition.

[0031] S103, determining the air supply outlet closest to the turbulent airflow area according to the position coordinates of the turbulent airflow area, and performing directional air supply to the turbulent airflow area by adjusting the air supply speed and air supply direction of the air supply outlet.

[0032] Obtain the position coordinates of the air turbulence zone and the position coordinates of all air outlets in the clean room, and use the Euclidean distance formula to calculate the distance between the air turbulence zone and the air outlet; select the air outlet corresponding to the minimum distance value as the target air outlet according to the distance. Calculate the direction vector of the line connecting the center point of the air turbulence zone and the target air outlet to obtain the ideal air supply direction; calculate the horizontal angle and the vertical angle according to the direction vector. Adjust the horizontal and vertical louver angles of the target air outlet; if the error between the actual air supply direction and the ideal air supply direction is greater than the preset threshold, continue to adjust the horizontal and vertical louver angles. Obtain the Reynolds value of the air turbulence zone; determine the air supply speed adjustment amount according to the difference between the Reynolds value and the preset threshold. Use the PID control algorithm, and use the difference between the Reynolds value and the preset threshold as input; determine the PID parameters by the Ziegler-Nichols tuning method; calculate the air supply parameter adjustment amount according to the PID control algorithm.

[0033] Specifically, according to the position coordinates (x1, y1, z1) of the air turbulence zone, the Euclidean distance formula d=√((x2-x1)^2+(y2-y1)^2+(z2-z1)^2) is used to calculate the distance between the area and all air outlets in the clean room. The position coordinates of a certain air outlet (x2, y2, z2), d is the Euclidean distance between the air outlet and the air turbulence zone, and the air outlet corresponding to the minimum distance is selected as the target air outlet. By calculating the direction vector v=(x2-x1, y2-y1, z2-z1) of the line connecting the center point of the air turbulence zone and the target air outlet, the ideal air supply direction is obtained, and the horizontal angle θ=arctan((y2-y1) / (x2-x1)) and the vertical angle φ=arcsin((z2-z1) / d) are calculated. The servo motor is used to adjust the horizontal and vertical louver angles of the air outlet, adjusting 0.1 degrees each time until the error between the actual air supply direction and the ideal air supply direction is less than 0.5 degrees. Based on the Reynolds value of the air flow turbulence zone, a fuzzy control algorithm is designed, with the input being the difference between the Reynolds number and the preset threshold, and the output being the air supply speed adjustment. The fuzzy rule base contains rules such as "if the Reynolds number is too large, increase the wind speed". The speed of the blower is adjusted by the frequency converter in 0.1Hz steps to achieve precise control of the air supply speed of the target air outlet. The PID control algorithm is used, with the difference between the Reynolds number and the preset threshold in the air flow turbulence zone as input to calculate the adjustment amount of the air supply parameters. The PID parameters are determined by the Ziegler-Nichols tuning method. The optimized air supply parameters are applied to the target air outlet to achieve directional air supply to the air flow turbulence zone. A hot wire anemometer is used to monitor the Reynolds number of the air flow turbulence zone in real time. When the difference between the monitoring value and the preset threshold is greater than 5%, the air supply parameter adjustment process is retriggered. 20 air outlets are arranged in the clean room, and the position coordinates of each air outlet are pre-stored in the database. Assuming that the center coordinates of the detected air flow turbulence zone are (5.2, 3.8, 2.1) meters, the Euclidean distance between this point and all air outlets is automatically calculated. The calculation results show that the coordinates of the nearest air outlet are (4.5, 4.0, 2.5) meters, and the distance is 0.82 meters. This air outlet is then determined as the target air outlet. The ideal horizontal angle of the air supply direction is calculated to be -53.13 degrees, and the vertical angle is -28.07 degrees. The servo motor equipped with the air outlet begins to adjust the shutter angle, adjusting 0.1 degrees each time. After 47 fine-tuning, the error between the actual air supply direction and the ideal direction is reduced to within 0.4 degrees. The Reynolds number of the turbulent airflow area is 2500, and the preset threshold is 2300. The fuzzy control algorithm takes 200 as input, and after querying the fuzzy rule base, the output wind speed adjustment is +0.5m / s. The original speed of the blower is 30Hz, and the inverter gradually adjusts it to 35Hz, and the air supply speed increases from the original 5m / s to 5.5m / s.The PID control algorithm uses 200 as input, and the parameters Kp=0.6, Ki=0.2, Kd=0.1 obtained by the Ziegler-Nichols tuning method are calculated to fine-tune the air supply angle by 0.2 degrees and increase the wind speed by another 0.1m / s. These parameters are applied to the target air outlet to achieve directional air supply. The hot wire anemometer continuously monitors the turbulent air flow area. After 5 minutes, it is detected that the Reynolds number drops to 2320, which is less than 5% from the preset threshold of 2300. The current air supply parameters are temporarily kept unchanged and monitoring continues. If subsequent monitoring finds that the Reynolds number exceeds 2300 again, the adjustment process will be retriggered.

[0034] S104, obtaining the velocity distribution data of the airflow field in the turbulent airflow area and its surrounding area, and modifying the boundary conditions and turbulence parameters of the airflow field in combination with the numerical simulation results of the clean room airflow.

[0035] Oil droplets are uniformly injected into the turbulent airflow area and its surrounding area as tracer particles through an atomizer; a camera is used to capture a motion image sequence of the tracer particles at a preset frame rate, and the motion image sequence is subjected to median filtering to remove background noise; based on the motion image sequence after the background noise is removed, the tracer particles are segmented using the Otsu adaptive threshold method to obtain particle images; a fast Fourier transform cross-correlation analysis is performed on the particle image, the displacement vector of the particle in two adjacent frames of the image is calculated, and the instantaneous velocity distribution of the airflow field is solved based on the known time interval and spatial calibration results; the instantaneous velocity distribution is compared with the pre-set airflow numerical simulation results, and the boundary conditions and turbulence parameters are iteratively optimized until the relative error between the simulation results and the measured data is less than a preset threshold; an uncertainty analysis is performed on the measured data, and if the data point exceeds the preset standard deviation range, the abnormal value is eliminated to obtain the final airflow field velocity distribution data.

[0036] Specifically, oil droplets with a particle size of 1-5 microns were selected as tracer particles and uniformly injected into the turbulent airflow area and its surrounding area through an atomizer. A high-speed camera was used to capture a sequence of particle motion images at a rate of 1000 frames per second, and background noise was removed by median filtering. The tracer particles were segmented using the Otsu adaptive threshold method to obtain a clear particle image. A calibration plate was used for spatial calibration to establish the correspondence between pixel coordinates and actual spatial coordinates. Fast Fourier transform cross-correlation analysis was performed on the processed particle image sequence to calculate the displacement vector of the particles in two adjacent frames of the image. The instantaneous velocity distribution of the airflow field was solved based on the known time interval and spatial calibration results. The discrete velocity vector was converted into a continuous velocity field distribution by the cubic spline interpolation algorithm, the velocity gradient was calculated by the second-order central difference method, and the turbulent kinetic energy and dissipation rate were solved in combination with the Reynolds stress tensor. The velocity field data obtained by particle image velocity measurement was compared with the pre-calculated airflow numerical simulation results, and the error function was fitted using the least squares method. The boundary condition parameters and k-ε turbulence coefficient were iteratively optimized by the L-BFGS-B algorithm until the relative error between the simulation results and the measured data was less than 5%. The accuracy of the modified model was evaluated by five-fold cross validation, and the root mean square error and correlation coefficient were calculated. The uncertainty analysis of the measured data was performed, and the outliers beyond the range of 3 times the standard deviation were eliminated to ensure the data quality. In the clean room, silicone oil droplets with an average particle size of 3 microns were selected as tracer particles, and were uniformly injected into the observation area of ​​10x10x5 meters through a six-nozzle atomizer, and the particle concentration reached 10^5 pieces / m^3. A Phantomv2640 high-speed camera was used with a resolution of 1920x1080 pixels. The images were continuously collected at a rate of 1000 frames / second for 5 seconds to obtain a 5000-frame particle image sequence. The median filter with a 3x3 pixel window was used to remove the background noise, and then the Otsu adaptive threshold method was applied to binarize the image and extract the particle contour. A 10x10cm calibration plate was placed, and the calibration points were identified through image processing. The mapping relationship between pixel coordinates and actual space coordinates was established, and the spatial resolution reached 0.1mm / pixel. For the processed image sequence, a 32x32 pixel interrogation window was selected, and the displacement vector of the particle group in two adjacent frames was calculated using the cross-correlation algorithm accelerated by fast Fourier transform. According to the frame interval of 1ms, the instantaneous velocity field was obtained with a spatial resolution of 16x16 pixels. The discrete velocity field of the 16x16 grid was interpolated into a continuous distribution of 1x1 pixels using the cubic spline interpolation algorithm. The velocity gradient was calculated using the second-order central difference method, and the grid spacing was 5 pixels. Combined with the known air density (1.225kg / m^3) and dynamic viscosity (1.81x10^-5Pa·s), the Reynolds stress tensor was calculated, and then the turbulent kinetic energy k and dissipation rate ε were solved. The measured velocity field was compared with the numerical simulation results of the airflow established in advance using ANSYSFluent software, and the error function was constructed using the least squares method.The velocity distribution of the inlet boundary conditions and the eddy viscosity proportional coefficient Cμ in k-ε turbulence, the generation coefficient C1ε and the destruction rate coefficient C2ε that control the turbulent dissipation rate ε were iteratively optimized by the L-BFGS-B algorithm until the relative error between the simulation results and the measured data was less than 5%. The accuracy of the revised model was evaluated by five-fold cross validation, and the root mean square error was calculated to be 0.05m / s and the correlation coefficient was 0.95. The uncertainty analysis of the measured data was performed with a standard deviation of 0.03m / s. Outliers exceeding 3 times the standard deviation were eliminated, and 95% of the valid data points were retained.

[0037] S105. Using the corrected airflow field, simulate the indoor airflow organization form under different air supply volumes and fresh air ratios, and determine the air supply strategy with the lowest turbulence and the best energy-saving effect by calculating the turbulence and energy consumption indicators under each air supply condition.

[0038] Generate a combination of air supply conditions according to the range of air supply volume and fresh air ratio; use the finite volume method to perform numerical simulation on the combination of air supply conditions to obtain the indoor air flow field distribution. Calculate the energy consumption index of the combination of air supply conditions through the energy balance equation; wherein the energy balance equation includes the power of the air supply fan, heat transfer of the heat exchanger and fresh air processing. Construct a multi-objective optimization function with the average turbulence and comprehensive energy consumption index as optimization targets. Use the NSGA-II algorithm to solve the Pareto optimal solution set of the multi-objective optimization function; from the Pareto optimal solution set, select the solution with the lowest turbulence and the lowest energy consumption as the optimal air supply strategy.

[0039] Specifically, according to the revised air flow field model, based on the GB50073-2013 clean room design specification and historical operation data, the air supply volume range is set to 500-2000m³ / h, the step size is 100m³ / h, the fresh air ratio range is 20%-100%, and the step size is 10%, to generate different air supply condition combinations. The finite volume method is used to perform numerical simulation on each combination and calculate the indoor air flow field distribution. The Reynolds stress model is used to calculate the turbulent kinetic energy k and dissipation rate ε under each air supply condition. The turbulence value τ of each area is calculated by the turbulence formula τ=k / (0.5*U²), where U is the average flow rate. The average turbulence of the entire space is calculated as an evaluation index. An energy balance equation including the power of the air supply fan, heat transfer of the cold and heat exchanger, and fresh air treatment is established. Considering the fan efficiency curve, heat exchanger characteristics, and outdoor meteorological parameters, the energy consumption index under different air supply conditions is calculated. The comprehensive energy consumption index is obtained by weighted summation, and the weight coefficient is determined based on the initial investment and operating cost of the equipment. A multi-objective optimization function was constructed, with average turbulence and comprehensive energy consumption as optimization targets. The NSGA-II algorithm was used to solve the Pareto optimal solution set, with the population size set to 100, the number of iterations to 500, the crossover probability to 0.9, and the mutation probability to 0.1. From the obtained Pareto frontier, the solution with the lowest turbulence and the lowest energy consumption was selected as the optimal air supply strategy. The CFD verification simulation of the optimal air supply strategy was performed using ANSYS Fluent software to compare the deviations between the actual turbulence and energy consumption and the optimization results. The OpenMP parallel computing framework was used to distribute the solution process to multiple CPU cores to improve the computing efficiency. The adaptive grid method was used to encrypt the grid in the area where the air flow changed drastically to optimize the allocation of computing resources. In a clean workshop with an area of ​​1000 square meters, according to the GB50073-2013 specification and the operating data of the past year, the air supply volume range was determined to be 500-2000m³ / h and the fresh air ratio range was 20%-100%. The 3D model was established using ANSYS Fluent software, with a mesh size of 2 million tetrahedral units. The finite volume method and SIMPLE algorithm were used to solve the Navier-Stokes equations and calculate the airflow field distribution of 16 different air supply combinations. The Reynolds stress model was used to calculate the turbulence parameters, where the average value of the turbulent kinetic energy k was 0.05m² / s², and the average value of the dissipation rate ε was 0.01m² / s³. The calculated average turbulence degree τ ranged from 0.02 to 0.08. An energy balance equation was established, including the power curve of the 15kW air supply fan, the COP curve of the 50kW chiller, and the efficiency curve of the fresh air heat recovery device. The calculation weights of the comprehensive energy consumption index were 30% for fan power consumption, 50% for cooling and heating load, and 20% for fresh air treatment. The NSGA-II algorithm was used for multi-objective optimization, and parallel calculation was performed on the Inteli7-10700KCPU, which took about 4 hours.The obtained Pareto front contains 20 non-dominated solutions, from which the solution with an average turbulence of 0.03 and a comprehensive energy consumption index of 85kWh / day is selected as the optimal air supply strategy. The corresponding air supply volume of this strategy is 1200m³ / h and the fresh air ratio is 60%. ANSYS Fluent was used for verification simulation, and the results showed that the actual average turbulence was 0.031, which deviated from the optimization result by 2.5%; the actual comprehensive energy consumption was 86.2kWh / day, with a deviation of 1.4%. The adaptive grid method was used to reduce the grid size to 2cm near the supply and return air outlets and 10cm in other areas. The total number of grids was reduced to 1.5 million, and the calculation time was shortened by 25%.

[0040] S106: Send the air supply temperature, air supply humidity, and air supply volume in the air supply strategy to the air conditioning control system to regulate the air flow turbulence area to suppress temperature and humidity fluctuations in the air flow turbulence area.

[0041] The optimized air supply strategy parameters are obtained and transmitted to the programmable logic controller of the air conditioning control system through industrial Ethernet; according to the air supply strategy parameters, the temperature and humidity data of the air flow turbulence area are processed by the Kalman filter to obtain the main trend of temperature and humidity fluctuations after eliminating high-frequency noise and outliers. The fuzzy PID control algorithm is used to perform closed-loop control on the air supply parameters. The fuzzy PID control algorithm dynamically adjusts the PID parameters based on the main trend of temperature and humidity fluctuations. The speed of the air supply fan is continuously adjusted by the variable frequency speed regulation method, and the opening control value of the electronic expansion valve and the proportional adjustment value of the humidifier are determined according to the dynamically adjusted PID parameters. Combined with weather forecast data, the ARIMA model is used for load forecasting to determine the feedforward compensation value of temperature and humidity control. If the standard deviation of temperature and humidity fluctuations exceeds the preset threshold, the adaptive adjustment of the control temperature and humidity parameters is triggered.

[0042] Specifically, the optimized air supply strategy parameters are transmitted to the programmable logic controller of the air conditioning control system through industrial Ethernet, the AES encryption algorithm is used to encrypt the data, the CRC check is used to ensure data integrity, and the supply air temperature, supply air humidity and supply air volume setting values ​​are written into the corresponding data registers. The temperature and humidity data of the air turbulence area are processed by the Kalman filter to eliminate high-frequency noise and outliers, and the main trend of temperature and humidity fluctuations is extracted as the input signal of the control system. The fuzzy PID control algorithm is used to perform closed-loop control of the air supply parameters, and the PID parameters are dynamically adjusted according to the real-time temperature and humidity feedback signals of the air turbulence area. The fuzzy rule base contains 25 rules, which are designed based on temperature deviation, humidity deviation and their change rate. The Mamdani reasoning method is used to optimize the membership function of the fuzzy set through the genetic algorithm. The speed of the air supply fan is continuously adjusted by the variable frequency speed regulation technology, combined with the opening control of the electronic expansion valve and the proportional adjustment of the humidifier to achieve smooth changes in the air supply parameters. The multivariable decoupling control method is used to establish the coupling matrix between the air supply temperature, humidity and air volume, and the independent adjustment of each parameter is achieved through feedback linearization. Combined with weather forecast data, a load forecasting model based on the ARIMA model is constructed to realize feedforward compensation of temperature and humidity control. The standard deviation of temperature and humidity fluctuations is calculated in real time. When the standard deviation exceeds the set threshold, the adaptive adjustment of the control parameters is triggered. In a 1000 square meter clean room, the air conditioning control system receives the optimized air supply strategy parameters through the Siemens S7-1500PLC. Data transmission adopts industrial Ethernet, uses the AES-256 encryption algorithm, and the key length is 256 bits to ensure data security. The CRC-32 check algorithm is used for data integrity verification, and the error rate is less than 10^-6. The air supply temperature is set to 22.5℃, the relative humidity is 45%, and the air volume is 10000m³ / h. These parameters are written into the DB100 data block of the PLC. The Kalman filter is used to process the data from 20 temperature and humidity sensors, with a sampling frequency of 1Hz, a process noise covariance of 0.01, and a measurement noise covariance of 0.1, which effectively reduces 95% of the high-frequency noise. The fuzzy rule base of the fuzzy PID controller contains 25 rules. The input variables are temperature deviation -2℃ to +2℃, humidity deviation -5% to +5% and their change rate. The output is the PID parameter adjustment amount. Through genetic algorithm optimization, the population size is 100, 500 iterations, crossover probability is 0.8, and mutation probability is 0.01, and the optimal membership function is obtained. The inverter controls the speed of the blower in the range of 20-50Hz, and the response time is less than 5 seconds. The opening range of the electronic expansion valve is 10-100%, and the step accuracy is 0.1%. The adjustment range of the ultrasonic humidifier is 0-10kg / h, and the accuracy is 0.1kg / h. The multivariable decoupling control adopts a 3x3 coupling matrix, and the independent adjustment of parameters is achieved through SVD decomposition. The control cycle is 1 second.The ARIMA (2,1,1) model is used for 24-hour load forecasting, with a forecast error within ±5%. The standard deviation thresholds for temperature and humidity fluctuations are set at 0.5°C and 2%RH. When the thresholds are exceeded, the control gain automatically increases by 20%, so that the clean room temperature fluctuation is controlled within ±0.3°C and the humidity fluctuation is controlled within ±2%RH, meeting ISOClass5 cleanliness requirements.

[0043] S107, continuously monitoring the change of particle concentration in the turbulent airflow area, and when the particle concentration drops below the background value, determining that the airflow in the turbulent airflow area has returned to stability, and removing the turbulent airflow area mark.

[0044] According to the set sampling volume and measurement frequency, multiple laser particle counters are arranged in the airflow turbulence zone to collect particle concentration data of four particle size ranges; the particle concentration data are preprocessed to obtain the average concentration value and standard deviation of each particle size range; the change trend of the particle concentration is predicted by the quadratic exponential smoothing method, and if the predicted value is lower than the preset threshold for a preset number of consecutive cycles, the recalculation of the background value is triggered; the airflow stability is judged based on the Mamdani fuzzy logic controller, the input variable of the controller is the relative deviation of the particle concentration from the background value and its change rate, and the output is the stability score; if the stability score is higher than the preset value and exceeds the preset duration, the airflow turbulence zone mark is released and the regional status information is updated.

[0045] Specifically, multiple laser particle counters were arranged in the air turbulence area, and the sampling volume and measurement frequency were set according to the ISO14644 standard. For ISOClass5 clean rooms, the sampling volume was set to 1 cubic foot and the measurement frequency was 1 time / minute. The particle concentration data of four particle size ranges of 0.3μm, 0.5μm, 1μm and 5μm were continuously collected, and the real-time data was transmitted to the central controller through the RS-485 interface. The collected particle concentration data were preprocessed using a 5-minute sliding window algorithm to eliminate the influence of instantaneous fluctuations and calculate the average concentration value and standard deviation of each particle size range. The quadratic exponential smoothing method was used to predict the trend of particle concentration changes, with the smoothing parameter α set to 0.3 and the trend smoothing parameter β set to 0.2. When the predicted value was lower than the preset threshold for three consecutive cycles, the background value was triggered to be recalculated. The kernel density estimation method was used to determine the new background concentration level, and the bandwidth selection adopted the Silverman rule. Based on the Mamdani type fuzzy logic controller, the airflow stability is judged. The input variable is the relative deviation of the particle concentration from the background value, ranging from -100% to +100% and its rate of change, ranging from -10% / min to +10% / min, and the output is the stability score, ranging from 0 to 100. 25 fuzzy rules are designed and the centroid method is used for defuzzification. Weights are assigned to particles in different particle size ranges, 0.3μm is 0.4, 0.5μm is 0.3, 1μm is 0.2, and 5μm is 0.1, and the comprehensive stability score is calculated. When the score is higher than 80 and lasts for 15 minutes, the airflow turbulence zone mark is automatically removed and the regional status information is updated. At the same time, the 3σ criterion is used for anomaly detection to identify possible sensor failures or external interference. In a 100-square-meter ISOClass5 clean room, four TSI9110 laser particle counters are arranged, the sampling volume is set to 1 cubic foot, and the measurement frequency is 1 time / minute. Each counter transmits the particle concentration data of four particle size ranges of 0.3μm, 0.5μm, 1μm and 5μm to the central controller through the RS-485 interface at a baud rate of 9600. The controller uses a 5-minute sliding window algorithm to calculate the average and standard deviation every 5 data points. For example, the 5-minute average concentration of 0.3μm particles is 3520 / m³, and the standard deviation is 150 / m³. In the quadratic exponential smoothing method, the initial value S0 is set to 3500, the initial trend value b0 is set to -10, the smoothing parameter α is 0.3, and the trend smoothing parameter β is 0.2. The predicted concentration of 0.3μm particles at the next moment is 3480 / m³. When the predicted value is lower than the threshold of 3600 / m³ for three consecutive times, the background value is triggered to be recalculated. The kernel density estimation uses a Gaussian kernel function with a bandwidth of h = 1.06σn^(-1 / 5)≈62, where σ is the standard deviation of the data and n is the sample size, and the new background value is 3450 / m³.The fuzzy logic controller input variable "relative deviation of concentration" is set to 5 fuzzy sets, "rate of change" is set to 3 fuzzy sets, and the output "stability score" is set to 5 fuzzy sets. 25 fuzzy rules are defined by IF-THEN statements, such as "IF relative deviation is zero AND rate of change is negative THEN stability is high". After defuzzification, the stability score is 85. The weights [0.4, 0.3, 0.2, 0.1] are applied to the four particle size ranges to obtain a comprehensive score of 83. The score remains above 80 for 15 consecutive minutes, and the airflow turbulence zone mark is automatically removed, and the regional status in the database is updated. At the same time, the 3σ anomaly detection found that the concentration of 5μm particles suddenly increased to 950 / m³, exceeding the mean + 3σ (200+3*30=290), triggering an alarm to indicate that there may be a pollution source.

[0046] The above only lists some preferred embodiments of the present invention, but the present invention is not limited thereto, and many improvements and changes can be made. As long as the improvements and changes are made on the basis of the basic principles of the present invention, they should be regarded as falling within the protection scope of the present invention.

Claims

1. An energy-saving management method based on smart buildings, characterized in that: The method comprises: Obtain the temperature and relative humidity data of different areas in the clean room, calculate the temperature gradient and humidity gradient of each area, and if the temperature gradient or humidity gradient of a certain area exceeds the preset threshold, it is determined that there is heat accumulation or water vapor accumulation in the area, and the area is marked as an abnormal temperature and humidity area; Obtain the streamline curvature and flow velocity pulsation data at the boundary layer of the temperature and humidity anomaly area, and determine whether the airflow turbulence in the area exceeds the standard by calculating the streamline curvature change rate and the flow velocity pulsation intensity. If the turbulence exceeds the preset threshold, the area is determined to be an airflow turbulence area, and the position coordinates of the airflow turbulence area are recorded; According to the position coordinates of the air flow turbulence area, determine the air supply outlet closest to the air flow turbulence area, and provide directional air supply to the air flow turbulence area by adjusting the air supply speed and air supply direction of the air supply outlet; Obtaining the velocity distribution data of the airflow field in the turbulent airflow area and its surroundings, and correcting the boundary conditions and turbulence parameters of the airflow field in combination with the numerical simulation results of the clean room airflow; The corrected airflow field is used to simulate the indoor airflow organization forms under different air supply volumes and fresh air ratios. The air supply strategy with the lowest turbulence and the best energy-saving effect is determined by calculating the turbulence and energy consumption indicators under various air supply conditions. The air supply temperature, air supply humidity and air supply volume in the air supply strategy are sent to the air conditioning control system to regulate the air flow turbulence area and suppress the temperature and humidity fluctuations in the air flow turbulence area; Continuously monitoring the change of particle concentration in the turbulent airflow area, and when the particle concentration drops below the background value, determining that the airflow in the turbulent airflow area has returned to stability, and removing the turbulent airflow area mark; The method of obtaining the velocity distribution data of the airflow field in the airflow turbulence area and its surrounding area, and correcting the boundary conditions and turbulence parameters of the airflow field in combination with the numerical simulation results of the clean room airflow, includes: Evenly injecting oil droplets as tracer particles into the turbulent airflow area and its surrounding area through an atomizer; Using a camera to capture a motion image sequence of the tracer particles at a preset frame rate, and performing a median filter on the motion image sequence to remove background noise; According to the motion image sequence after background noise is removed, the tracer particles are segmented using the Otsu adaptive threshold method to obtain a particle image; Perform fast Fourier transform cross-correlation analysis on particle images, calculate the displacement vector of particles in two adjacent frames of images, and solve the instantaneous velocity distribution of the airflow field based on the known time interval and spatial calibration results; The instantaneous velocity distribution is compared with the pre-set airflow numerical simulation results, and the boundary conditions and turbulence parameters are iteratively optimized until the relative error between the simulation results and the measured data is less than a preset threshold; Perform uncertainty analysis on the measured data, and if the data point exceeds the preset standard deviation range, remove the abnormal value that exceeds the preset standard deviation range to obtain the final airflow field velocity distribution data; The modified airflow field is used to simulate the indoor airflow organization forms under different air supply air volumes and fresh air ratios, and the air supply strategy with the lowest turbulence and the best energy-saving effect is determined by calculating the turbulence and energy consumption indicators under each air supply condition, including: Generate a combination of air supply conditions according to the air supply volume range and fresh air ratio range; Finite volume method is used to numerically simulate the air supply condition combination and obtain the indoor air flow field distribution; Calculate the energy consumption index of the air supply condition combination through the energy balance equation; The energy balance equation includes the power of the blower, the heat transfer of the heat exchanger and the fresh air treatment; Taking the average turbulence degree and comprehensive energy consumption index as optimization targets, a multi-objective optimization function is constructed; Using NSGA-II algorithm to solve the Pareto optimal solution set of the multi-objective optimization function; From the Pareto optimal solution set, the solution with the lowest turbulence and the lowest energy consumption is selected as the optimal air supply strategy.

2. The method according to claim 1, wherein: The method of obtaining temperature and relative humidity data of different areas of the clean room, calculating the temperature gradient and humidity gradient of each area, and if the temperature gradient or humidity gradient of a certain area exceeds a preset threshold, determining that heat accumulation or water vapor accumulation exists in the area, and marking the area as an abnormal temperature and humidity area, includes: Obtain real-time temperature and relative humidity data collected by temperature and humidity sensors pre-placed in different areas of the clean room; Calculating the temperature gradient and humidity gradient between adjacent areas using a finite difference method based on the real-time temperature and relative humidity data; For the temperature gradient and humidity gradient, determine whether they exceed a preset temperature gradient threshold and humidity gradient threshold; If the temperature gradient or humidity gradient exceeds the corresponding preset threshold, the area is determined to be an abnormal temperature and humidity area; Using a two-dimensional coordinate system to record the location information of the abnormal temperature and humidity area, and determining the boundary range of the abnormal temperature and humidity area by an integration method; Calculate the duration of the abnormal condition in the abnormal temperature and humidity area according to the continuously collected temperature and humidity data; Generate an abnormal report based on the location information, abnormal gradient value, boundary range and duration of the abnormal temperature and humidity area; The abnormal report is stored in a preset database.

3. The method according to claim 1, wherein: The method of obtaining the streamline curvature and flow velocity pulsation data at the boundary layer of the temperature and humidity anomaly area, and determining whether the airflow turbulence in the area exceeds the standard by calculating the streamline curvature change rate and the flow velocity pulsation intensity, and if the turbulence exceeds the preset threshold, determining that the area is an airflow turbulence area, and recording the position coordinates of the airflow turbulence area, includes: Determine the boundary layer thickness based on the temperature gradient method and obtain the airflow velocity and direction data at the boundary layer; Fitting the airflow streamline by the least square method to obtain the curvature of the streamline, and the curvature of the streamline is used to calculate the curvature value of the streamline; Obtaining the air flow velocity collected by the hot wire anemometer array, calculating the root mean square value of the flow velocity, and obtaining the flow velocity pulsation intensity; Calculating the Reynolds number according to the streamline curvature value and the flow velocity pulsation intensity, and determining whether the Reynolds number exceeds a preset turbulence threshold; If the Reynolds number exceeds the preset turbulence threshold, the area is determined to be a turbulent airflow area; Using Kriging interpolation algorithm to determine the precise boundary of the airflow turbulence zone; Acquire data from a hot wire anemometer array and calculate the three-dimensional spatial coordinates of the airflow turbulence area by triangulation principle; The position information of the airflow turbulence area is recorded, and the position information includes the center coordinates and the boundary coordinates.

4. The method according to claim 1, wherein: The method of determining the air outlet closest to the turbulent airflow area according to the position coordinates of the turbulent airflow area, and providing directional air supply to the turbulent airflow area by adjusting the air supply speed and air supply direction of the air supply outlet comprises: Obtain the position coordinates of the air flow turbulence area and the position coordinates of all air outlets in the clean room, and use the Euclidean distance formula to calculate the distance between the air flow turbulence area and the air outlet; According to the distance, the air outlet corresponding to the minimum distance is selected as the target air outlet; Calculate the direction vector of the line connecting the center point of the airflow turbulence area and the target air supply outlet to obtain the ideal air supply direction; Calculate the horizontal angle and the vertical angle according to the direction vector; Adjusting the horizontal and vertical louver angles of the target air outlet; If the error between the actual air supply direction and the ideal air supply direction is greater than a preset threshold, continue to adjust the horizontal and vertical louver angles; Obtaining the Reynolds value of the airflow turbulence area; Determining the air supply speed adjustment amount according to the difference between the Reynolds value and a preset threshold; A PID control algorithm is used, with the difference between the Reynolds value and a preset threshold as input; Determine PID parameters by Ziegler-Nichols tuning method; The air supply parameter adjustment amount is calculated according to the PID control algorithm.

5. The method according to claim 1, wherein: The air supply temperature, air supply humidity, and air supply volume in the air supply strategy are sent to the air conditioning control system to regulate the air flow turbulence area to suppress the temperature and humidity fluctuations in the air flow turbulence area, including: Obtain optimized air supply strategy parameters and transmit them to the programmable logic controller of the air conditioning control system via industrial Ethernet; According to the air supply strategy parameters, the Kalman filter is used to process the temperature and humidity data in the airflow turbulence area to obtain the main trend of temperature and humidity fluctuations after eliminating high-frequency noise and outliers; A fuzzy PID control algorithm is used to perform closed-loop control on the air supply parameters. The fuzzy PID control algorithm dynamically adjusts the PID parameters based on the main trends of the temperature and humidity fluctuations. The speed of the blower is continuously adjusted by the variable frequency speed regulation method, and the opening control value of the electronic expansion valve and the proportional adjustment value of the humidifier are determined according to the dynamically adjusted PID parameters; Combined with weather forecast data, the ARIMA model is used to forecast load to determine the feedforward compensation value for temperature and humidity control; If the standard deviation of the temperature and humidity fluctuations exceeds the preset threshold, the adaptive adjustment of the control temperature and humidity parameters is triggered.

6. The method according to claim 1, wherein: The method of continuously monitoring the change of the particle concentration in the turbulent airflow area and determining that the airflow in the turbulent airflow area has recovered to be stable when the particle concentration drops below the background value, and removing the turbulent airflow area mark, comprises: According to the set sampling volume and measurement frequency, multiple laser particle counters are arranged through the airflow turbulence area to collect particle concentration data in four particle size ranges; Preprocessing the particle concentration data to obtain an average concentration value and a standard deviation for each particle size range; The changing trend of the particle concentration is predicted by using a quadratic exponential smoothing method. If the predicted value is lower than a preset threshold for a preset number of consecutive cycles, recalculation of the background value is triggered. The airflow stability is judged based on a Mamdani-type fuzzy logic controller, wherein the input variable of the controller is the relative deviation of the particle concentration from the background value and its rate of change, and the output is the stability score; If the stability score is higher than a preset value and lasts longer than a preset time, the airflow turbulence area mark is removed and the area status information is updated.

7. An energy-saving management system based on smart buildings, characterized in that: The system implements the energy-saving management method according to any one of claims 1 to 6, and the system comprises: A data acquisition module, used to collect temperature and humidity data in the clean room; Anomaly detection module, which calculates temperature and humidity gradients and identifies abnormal areas; Airflow analysis module, used to analyze the curvature of streamlines and flow velocity pulsation in turbulent airflow areas; The air outlet adjustment module is used to optimize the wind speed and direction of the air outlet in the air turbulence area; The simulation and optimization module is used to select the best air supply strategy through airflow field simulation; Monitoring and control module, used to monitor particle concentration and adjust airflow stability.

Images