Intelligent ventilation method and system based on wind level

Abstract

The invention discloses an intelligent ventilation method and system based on wind levels, and belongs to the field of intelligent ventilation. The method comprises the steps that a window opening strategy is adjusted through wind speed and wind direction information, and the window operation risk caused by building swing under the strong wind condition is avoided; performing strength analysis and life prediction on the window opening mechanism, adjusting a maintenance plan and a replacement period according to a prediction result, and ensuring the reliability of the mechanism under the action of long-term wind power; a central control system is established, and a wind speed and wind direction data processing module, a wind field simulation module, a building swing monitoring module and a window operation decision module are integrated, so that full-automatic management of the intelligent ventilation system is realized; a deep learning algorithm is utilized to train a large amount of collected wind speed and wind direction data, the prediction accuracy of the wind field model is optimized, and the response speed and accuracy of the intelligent ventilation system are improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent ventilation, and in particular relates to an intelligent ventilation method and system based on wind level. Background Art

[0002] The wind speed at the top of a high-rise building is usually much higher than that at a lower level, and the vortex and canyon effect around the building will also cause the wind direction to change dramatically at different heights and locations. This brings the following challenges to the intelligent ventilation system based on wind level: First, how to accurately obtain wind speed, wind direction and temperature data at different floor heights. Traditional rooftop weather stations can only measure single-point data and cannot reflect the actual wind environment of each floor of the building. If the ventilation system switches on all floors are controlled only by the data from the rooftop weather station, it may cause the low-level ventilation system to be insufficiently opened when the wind speed is low, and the high-level ventilation system to be unclosed in time when the wind speed is too high, which may bring safety hazards. Secondly, how to link the opening angle of the ventilation system on different floors with the wind speed and direction. Assuming that at a certain moment, the high-level wind speed reaches level 7, the system will automatically close the high-level ventilation system, but the low-level wind speed may only be level 3 at this time, which is not suitable for ventilation. How to finely control the opening angle of each ventilation system according to the real-time wind conditions of each floor, and maximize the use of natural ventilation while ensuring safety.

[0003] Thirdly, considering the complex shape of tower buildings and the fact that the wind environment is easily affected by surrounding buildings, how to establish an accurate wind field model. This model needs to take into account factors such as the height, shape, spacing, and surrounding terrain of the building in order to predict the wind speed and direction at different floors and locations in real time. Finally, under the action of wind, high-rise buildings will swing to a certain extent, which will affect the opening and closing of the ventilation system. How to ensure that the intelligent ventilation system can still work stably and reliably when the building swings. Under strong winds, the opening and closing mechanism of the ventilation system may be subject to additional stress. How to ensure its structural strength and service life. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides an intelligent ventilation method based on wind level, comprising:

[0005] Deploy wind speed and direction sensors and temperature sensors on multiple floors, and obtain wind speed, wind direction data and temperature data of each floor based on the sensors;

[0006] Based on the machine learning model, the model of wind speed, wind direction data and temperature data of each floor is predicted to obtain the wind speed, wind direction and temperature changes in the short term;

[0007] Constructing a fluid dynamics model to simulate the wind field of the building and the surrounding environment based on the wind speed, wind direction and temperature changes in the short term and the fluid dynamics model;

[0008] A wind field evaluation grade is constructed based on the wind field of the building and the surrounding environment, a ventilation strategy is constructed based on the wind field evaluation grade, and intelligent ventilation is performed on the building based on the ventilation strategy.

[0009] Preferably, the process of acquiring wind speed, wind direction data and temperature data of each floor based on the sensor includes:

[0010] According to the layout of multi-story buildings, wind speed sensors, wind direction sensors and temperature sensors are installed at key locations on each floor to collect real-time wind speed, wind direction and temperature data on each floor through the sensors;

[0011] The wind speed sensor and wind direction sensor are connected to the wireless network, and the collected wind speed, wind direction and temperature data are transmitted to the central processing system in real time based on wireless communication technology;

[0012] The central processing system receives wind speed, wind direction and temperature data from sensors on each floor, parses and processes the data, and obtains real-time wind speed, wind direction and temperature information on each floor.

[0013] Preferably, the process of obtaining short-term wind speed, wind direction and temperature changes includes:

[0014] Obtain historical wind speed, wind direction and temperature data, preprocess the data, remove outliers and noise data, and obtain a wind speed and direction time series data set;

[0015] Constructing an ARIMA model, inputting the wind speed and direction time series data set into the ARIMA model for training, and obtaining a temperature-based wind speed and direction prediction model;

[0016] The wind speed, wind direction data and temperature data of each floor are input into the temperature-based wind speed and wind direction prediction model for calculation to obtain the wind speed, wind direction and temperature changes in the short term.

[0017] Preferably, the process of obtaining a temperature-based wind speed and direction prediction model also includes: continuously acquiring real-time wind speed and direction data from sensors on each floor, determining the deviation between the actual wind speed and direction and the predicted value, and if the deviation exceeds a preset threshold, triggering retraining and optimization of the prediction model.

[0018] Preferably, the process of simulating the wind field of the building and the surrounding environment based on the wind speed, wind direction and temperature changes in the short term and the fluid dynamics model includes:

[0019] The computational fluid dynamics model is used to grid the surrounding environment of the building, set boundary conditions and initial conditions, and simulate the wind field distribution around the building by solving the Navier-Stokes equation to obtain wind speed, wind direction and temperature data at different locations.

[0020] Preferably, the process of intelligently ventilating a building based on the ventilation strategy includes:

[0021] Comparing the wind field of the building and the surrounding environment with a preset risk threshold, and constructing a wind field evaluation level based on the comparison result;

[0022] Based on the wind field evaluation level, an opening strategy of the ventilation system under different conditions is constructed, and the building is intelligently ventilated according to the opening strategy of the ventilation system.

[0023] Preferably, it also includes:

[0024] Determine the size, direction and frequency distribution of wind loads based on local meteorological data;

[0025] The fatigue strength theory is used to calculate the stress-strain response of the ventilation system under wind load. The number of stress cycles is counted by the rain flow counting method to obtain the fatigue life distribution curve.

[0026] Based on the fatigue life distribution curve, the strength analysis and life prediction of the ventilation system opening mechanism are performed, and the maintenance plan and replacement cycle of the ventilation system are adjusted according to the prediction results.

[0027] In order to solve the above technical problems, the present invention also provides an intelligent ventilation system based on wind level, comprising:

[0028] A deployment module, used to deploy wind speed and direction sensors and temperature sensors on multiple floors, and obtain wind speed, wind direction data and temperature data of each floor based on the sensors;

[0029] A prediction module, used to predict the wind speed, wind direction data and temperature data of each floor based on a machine learning model to obtain short-term wind speed, wind direction and temperature changes;

[0030] A wind field construction module is used to construct a fluid dynamics model to simulate the wind field of the building and the surrounding environment based on the wind speed, wind direction and temperature changes in the short term and the fluid dynamics model;

[0031] The implementation module is used to construct a wind field evaluation grade based on the wind field of the building and the surrounding environment, construct a ventilation strategy based on the wind field evaluation grade, and perform intelligent ventilation on the building based on the ventilation strategy.

[0032] Compared with the prior art, the present invention has the following advantages and technical effects:

[0033] The present invention deploys wind speed and direction sensors on multiple floors to collect environmental data in real time and transmit it to the central processing system. Combined with the computational fluid dynamics model, the present invention can simulate the wind field around the building and predict the wind conditions at different locations. The machine learning algorithm is used to analyze the wind speed and direction patterns and predict short-term changes. In addition, the present invention also monitors the building swing data, comprehensively considers the wind force and swing factors, and dynamically adjusts the ventilation system opening strategy. The wind field model is optimized through deep learning to improve the prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0035] Figure 1 A schematic diagram of a method flow chart of an embodiment of the present invention;

[0036] Figure 2 Schematic diagram of the system structure of an embodiment of the present invention. DETAILED DESCRIPTION

[0037] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0038] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0039] Embodiment 1

[0040] like Figure 1 As shown, this embodiment provides an intelligent ventilation method based on wind level, including:

[0041] S101. Deploy wind speed and direction sensors on multiple floors and transmit data to the central processing system in real time via wireless networks, so as to obtain wind speed, wind direction and temperature data on each floor and provide accurate environmental information for the intelligent ventilation system.

[0042] According to the layout of multi-story buildings, wind speed sensors, wind direction sensors and temperature sensors are installed at key locations on each floor, and the sensors are used to collect real-time wind speed, wind direction and temperature data on each floor. The wind speed sensors and wind direction sensors are connected to the wireless network, and the collected wind speed, wind direction and temperature data are transmitted to the central processing system in real time using wireless communication technology. The central processing system receives the wind speed, wind direction and temperature data from the sensors on each floor, parses and processes the data, and obtains the real-time wind speed and wind direction information on each floor. According to the preset intelligent ventilation control strategy, combined with the wind speed and wind direction information on each floor, the ventilation demand and ventilation mode of each floor are determined. Machine learning algorithms such as decision trees, support vector machines or neural networks are used to train historical wind speed and wind direction data and ventilation control strategies to establish an intelligent ventilation control model. The real-time wind speed and wind direction information on each floor is input into the intelligent ventilation control model, and the operating parameters of the ventilation equipment on each floor are dynamically adjusted to achieve precise intelligent ventilation control. By continuously collecting wind speed and direction data on each floor and ventilation equipment operation data, we continuously optimize the intelligent ventilation control model and improve the control accuracy and energy efficiency performance of the intelligent ventilation system.

[0043] Furthermore, to realize intelligent ventilation control in multi-story buildings, it is necessary to analyze the layout of the building and install wind speed sensors, wind direction sensors and temperature sensors at key locations, such as building exterior walls, stairwells, corridors, etc. Assume that a five-story office building has a set of wind speed and direction sensors installed on the four exterior walls of each floor, namely the southeast, northwest and northeast walls, for a total of 20 sets of sensors. These sensors can be ultrasonic or mechanical, and can measure the instantaneous speed and direction of the wind respectively. These sensors transmit the collected data to the central processing system in real time through a wireless network (such as Wi-Fi or ZigBee). For example, data is collected once a second, packaged into JSON format, and published to the central server central processing system through the MQTT protocol. For example, a server will receive data from each sensor, parse the JSON data packet, and obtain the real-time wind speed and wind direction information of each floor in each direction. The wind speed on the west side of the third floor is 3 meters per second, and the wind direction is northwest wind. The central processing system processes the collected data to obtain the real-time wind speed and wind direction information of each floor. Based on the wind speed and direction data in four directions, the average wind speed and dominant wind direction of the third floor can be calculated. Combined with the preset intelligent ventilation control strategy, when the outdoor wind speed is greater than a certain threshold, natural wind is used for ventilation first; when the wind speed is less than the threshold, the mechanical ventilation system is started. If the average wind speed on the third floor is greater than 2 meters per second and the dominant wind direction is northwest wind, the system can determine that natural ventilation is suitable for the third floor.

[0044] S102. Based on the wind speed, wind direction and temperature data obtained from sensors on each floor, the wind speed and wind direction patterns are analyzed through a machine learning algorithm to predict short-term wind speed, wind direction and temperature changes, providing decision support for the dynamic adjustment of the opening angle of the ventilation system.

[0045] The wind speed, wind direction and temperature data collected by sensors on each floor in real time are obtained, and the data are preprocessed to remove outliers and noise data to obtain a normalized wind speed, wind direction and temperature data time series data set. According to the wind speed and direction time series data set, a time series analysis algorithm such as the ARIMA model is used to establish a prediction model for wind speed and direction based on temperature, and the prediction accuracy of the model is improved through model training and optimization. The trained wind speed and wind direction prediction model is used in combination with the real-time wind speed and wind direction data of each floor to predict the wind speed, wind direction and temperature change trends of each floor in the future. According to the predicted wind speed, wind direction and temperature change trends, an optimization algorithm such as a genetic algorithm is used to comprehensively consider factors such as the location and orientation of the ventilation system on each floor to calculate the optimal opening angle of each ventilation system. The calculated optimal opening angle of each ventilation system is used to adjust and control the actual opening angle of the ventilation system on each floor in real time through the building control system to adapt to the dynamic changes of wind speed and direction. Continuously obtain the real-time wind speed and direction data from sensors on each floor, determine the deviation between the actual wind speed and direction and the predicted value, and if the deviation exceeds the preset threshold, trigger the retraining and optimization of the prediction model to improve the accuracy of subsequent predictions. Through the human-computer interaction interface, the real-time status and prediction trend of wind speed and direction on each floor, as well as the optimized opening angle of each ventilation system, are intuitively displayed to provide auxiliary decision support for building managers.

[0046] Furthermore, wind speed and direction sensors collect floor wind information in real time to provide a data basis for intelligent ventilation. For example, on the 10th floor of an office building, the sensor collects wind speed, wind direction and temperature data once a minute, and transmits the data to the central processing system through a wireless network. In the data preprocessing stage, outliers and noise data can be removed. The wind speed at a certain moment suddenly soared to 100 meters per second, which is obviously inconsistent with the actual situation. It can be judged as an outlier and removed. In addition, some noise data generated by sensor failure or environmental interference also need to be filtered. The wind speed value fluctuates frequently in a short period of time. This may be caused by the interference of the sensor, and a filtering algorithm needs to be used for processing. Finally, a normalized wind speed and direction time series data set is obtained. The ARIMA model is used to predict wind speed and direction. Specifically, an ARIMA model is trained using the wind speed and direction data of the past 3 days, and then the model is used to predict the wind speed, wind direction and temperature change trends in the next 24 hours. By adjusting the parameters of the ARIMA model, such as the autoregressive order, the difference order and the moving average order, the prediction accuracy of the model can be optimized. Assume that the wind speed on the 10th floor is predicted to gradually increase in the next hour, and the wind direction will change from easterly to southerly. Based on this prediction result, the optimal opening angle of each ventilation system can be calculated in combination with the location and orientation of the ventilation system on the 10th floor.

[0047] The basic idea of ​​the ARIMA model is to treat the data sequence formed by the prediction object over time as a random sequence and use a certain mathematical model to approximate this sequence. Once this model is identified, the future value can be predicted from the past and present values ​​of the time series.

[0048] ARIMA is expressed mathematically as:

[0049]

[0050] Among them, φ represents the coefficient of AR, and θ represents the coefficient of MA;

[0051] The construction process of the ARIMA model includes:

[0052] Obtain the time series data of the observed system;

[0053] Plot the data to see if it is a stationary time series. For non-stationary time series, perform a d-order difference operation to convert it into a stationary time series.

[0054] After the second step, a stationary time series has been obtained. To obtain the autocorrelation coefficient ACF and partial autocorrelation coefficient PACF of the stationary time series, the best layer p and order q can be obtained by analyzing the autocorrelation graph and partial autocorrelation graph.

[0055] From the d, q, and p obtained above, we get the ARIMA model.

[0056] S103. Use computational fluid dynamics (CFD) models to simulate the wind field of buildings and their surroundings, taking into account factors such as building height, shape, spacing and terrain, and predict wind speed, direction and temperature data at different floors and locations.

[0057] A three-dimensional building model is established based on parameters such as building height, building shape, and building spacing, and environmental factors such as the surrounding terrain are incorporated into the model. A computational fluid dynamics (CFD) model is used to grid the building's surrounding environment and set boundary conditions and initial conditions. By solving the Navier-Stokes equation, the wind field distribution around the building is simulated to obtain wind speed, wind direction, and temperature data at different locations. The wind speed and direction data of different floors are extracted and analyzed to determine whether there is excessive wind speed or unreasonable wind direction.

[0058] S104. By real-time monitoring of building swing data and combining wind speed and direction information, the ventilation system opening strategy is adjusted to avoid ventilation system operation risks caused by building swing under strong wind conditions.

[0059] By installing sensors on the building, the swing data of the building, including parameters such as swing amplitude and frequency, is collected in real time, and the collected data is transmitted to the central control system. The current wind speed, wind direction and temperature data are obtained through meteorological sensors, and the wind speed and direction data are compared with the preset risk threshold to determine whether the current situation is in strong wind conditions. If the current wind speed exceeds the preset risk threshold, it is determined to be a strong wind condition. According to the pre-set ventilation system opening strategy, the opening state of each ventilation system under strong wind conditions is determined. The real-time swing data of the building is compared with the preset safety swing threshold to determine whether the swing amplitude of the current building exceeds the safety threshold. If it exceeds the safety threshold, the emergency shutdown mechanism of the ventilation system is triggered. The historical building swing data, wind speed and direction data, and ventilation system opening state data are trained through machine learning algorithms to establish a risk prediction model to predict in real time whether there are safety hazards when the ventilation system is opened under the current wind speed and direction conditions. According to the prediction results of the risk prediction model, the opening strategy of the ventilation system is dynamically adjusted to maximize the use of natural ventilation and save energy while ensuring the safety of the building. Through data visualization technology, the building's real-time swing data, wind speed and direction data, ventilation system opening status and other information are presented to managers in an intuitive manner, facilitating real-time monitoring and risk warning.

[0060] Furthermore, buildings will sway under strong winds. Real-time monitoring of sway data and combining wind speed and direction data to intelligently control the opening state of the ventilation system can effectively improve building safety and energy efficiency. Sensors are the core of intelligent control. By installing sensors such as accelerometers and inclinometers at key locations of buildings (such as roofs, floor centers, etc.), data such as the swing amplitude and frequency of the building can be collected in real time. For example, accelerometers can measure the acceleration changes of buildings in the three directions of X, Y, and Z to calculate the swing amplitude. Inclinometers can measure the inclination angle of the building relative to the vertical direction. These data are transmitted to the central control system for processing by wireless or wired means. The central control system needs to integrate the swing data, wind speed and direction data, and preset risk thresholds. Wind speed and direction data can be obtained through meteorological sensors installed outside the building. For example, ultrasonic wind speed and direction meters can measure wind speed and direction. The risk threshold is a safety indicator set based on factors such as building structural characteristics and geographical location, such as wind speed exceeding 20 meters per second or building swing amplitude exceeding 0.1 degrees. Under strong wind conditions, the ventilation system opening strategy needs to be pre-set.

[0061] S105. Perform strength analysis and life prediction on the ventilation system opening mechanism, adjust the maintenance plan and replacement cycle according to the prediction results, and ensure the reliability of the mechanism under long-term wind action.

[0062] Obtain the structural parameters and material properties of the ventilation system opening mechanism and establish a finite element analysis model; determine the size, direction and frequency distribution of the wind load based on local meteorological data; use fatigue strength theory to calculate the stress-strain response of the mechanism under wind load; use the rain flow counting method to count the number of stress cycles and obtain the fatigue life distribution curve; if the life is lower than the design requirements, optimize the structural dimensions and connection methods of the mechanism; formulate a regular inspection, lubrication and component replacement plan for the mechanism based on the predicted life; establish a full life cycle management system for the mechanism to achieve a balance between reliability and maintenance costs.

[0063] Furthermore, the structural parameters and material properties of the ventilation system opening mechanism are obtained, which is the basis of finite element analysis. It is necessary to know the size, material (such as aluminum alloy, plastic steel, etc.) of the ventilation system frame, and the type and size of the connectors. Material properties include elastic modulus, Poisson's ratio, yield strength, etc. These parameters can be obtained by consulting the product specification, performing physical measurements or material tests. After obtaining these parameters, the geometric model of the ventilation system opening mechanism can be established in the finite element software and the corresponding material properties can be assigned. The finite element analysis model is established to simulate the mechanical behavior of the ventilation system opening mechanism under wind load. The accuracy of the model depends on the meticulousness of the geometric model, the mesh division density and the accuracy of the material properties. For complex connection parts, it is necessary to refine the mesh to ensure the reliability of the calculation results. Determining the size, direction and frequency distribution of the wind load is the key to fatigue analysis. This requires reference to local meteorological data, specifically, querying the multi-year average wind speed, maximum wind speed and wind rose diagram provided by the local meteorological bureau. In addition, it is also necessary to consider the influence of the building's height, shape and surrounding environment on the wind load. Wind loads on high-rise buildings are usually greater than those on low-rise buildings, and obstacles around buildings can change wind speed and direction. Fatigue strength theory is used to calculate the stress-strain response of the mechanism under wind load. Finite element analysis can obtain the stress-strain distribution of the mechanism under different wind loads. According to fatigue strength theory, such as the SN curve (stress-life curve), the fatigue life of the mechanism under cyclic load can be evaluated. If the stress amplitude of a certain part of the mechanism exceeds the fatigue limit of the material, fatigue failure is likely to occur at that part. The number of stress cycles is counted by the rain flow counting method to obtain the fatigue life distribution curve. The rain flow counting method is a commonly used fatigue analysis method that can simplify a complex load spectrum into a series of stress cycles and count the number of each stress cycle. Based on these statistical results, a fatigue life distribution curve can be drawn to predict the fatigue life of the mechanism. If the life is lower than the design requirements, the structural dimensions and connection methods of the mechanism are optimized. Specifically, the thickness of the key parts of the mechanism is increased, or the connection method is changed to improve the strength and stiffness of the mechanism, thereby extending its fatigue life. Higher strength materials can also be considered. According to the predicted life, a regular inspection, lubrication and component replacement plan for the mechanism is formulated. If the life expectancy is 10 years, a plan can be made to inspect every six months, lubricate once a year, and replace key components every five years. Establish a full life cycle management system for the organization to achieve a balance between reliability and maintenance costs.

[0064] S106. By establishing a central control system, integrating wind speed and direction data processing, wind field simulation, building swing monitoring and ventilation system operation decision modules, the fully automated management of the intelligent ventilation system can be achieved.

[0065] The wind speed, wind direction and temperature data around the building are obtained, and the wind speed and wind direction data are preprocessed by filtering, normalizing and other preprocessing through the data processing module to obtain standardized wind speed and wind direction data. The preprocessed wind speed and wind direction data are input into the wind field simulation module, and the wind field around the building is numerically simulated by the computational fluid dynamics algorithm to obtain the wind pressure distribution data at various positions on the surface of the building. According to the wind pressure distribution data obtained by the wind field simulation, the deformation and vibration of the building under the action of wind load are calculated by the finite element analysis method to determine whether the building is in a safe state. If the deformation or vibration of the building exceeds the preset threshold, an alarm is triggered. The wind pressure data at various positions on the surface of the building are input into the ventilation system operation decision module, and through machine learning algorithms such as support vector machines or neural networks, it is determined whether each ventilation system should be opened or closed under the current wind pressure conditions to achieve the purpose of ventilation and heat dissipation. According to the output results of the ventilation system operation decision module, the switch state of each ventilation system is controlled, and the automatic switch of the ventilation system is realized through actuators such as motors or pneumatic devices, and the state of the ventilation system is fed back to the central control system. The central control system coordinates and optimizes the entire intelligent ventilation system based on the output results of modules such as wind speed and direction data processing, wind field simulation, building swing monitoring, and ventilation system operation decision-making, ensuring that the building is in a safe and comfortable state. The human-computer interaction interface displays information such as wind environment parameters around the building, building vibration conditions, and ventilation system switch status in real time, and can adjust the control strategy and threshold parameters of the intelligent ventilation system according to user settings, realizing the integration of artificial intelligence and human wisdom.

[0066] Furthermore, in the intelligent ventilation system, it is first necessary to obtain the wind speed, wind direction data and temperature data around the building. Specifically, an anemometer and wind vane installed on the top of the building are used to monitor the wind status in real time. These devices can provide accurate wind speed and wind direction information, which is the basic data for the operation of the entire system. Through the data processing module, these raw data are filtered and normalized to eliminate the influence of environmental noise and equipment errors to ensure the accuracy and consistency of the data. The Kalman filter algorithm can effectively smooth the wind speed data, remove random fluctuations, and output more stable and reliable wind speed information. Next, these processed data are input into the wind field simulation module. The wind field conditions around the building are simulated by the computational fluid dynamics (CFD) algorithm. This step can help understand the flow characteristics of wind on the surface of the building. The simulation results may show that the wind speed is particularly high in a corner of the building. This information is crucial for the calculation of wind pressure distribution. Through this simulation, the wind pressure distribution data at various locations on the surface of the building can be obtained, which is the basis for evaluating the deformation and vibration of the building affected by wind. Using the obtained wind pressure distribution data, the deformation and vibration of the building under wind load are calculated by the finite element analysis method. If the simulation shows that the wind pressure on the west side of the building is much greater than that on the east side, the structure on the west side may need to be reinforced to prevent excessive deformation or vibration caused by wind. If the deformation or vibration of the building exceeds the preset safety threshold, the system will automatically trigger an alarm to notify maintenance personnel to conduct inspections and necessary structural reinforcement. In addition, the wind pressure data on the surface of the building is also used in the ventilation system operation decision module. Through machine learning algorithms, such as support vector machines or neural networks, the system can automatically decide whether each ventilation system should be opened or closed according to the current wind pressure conditions to optimize ventilation and temperature control in the building. In the process of typhoons affecting buildings, different wind field characteristics (such as wind direction, wind speed distribution, wind pressure changes, etc.) may be caused. The wind field evaluation level takes into account the specific location, structural form and surrounding environment of the building, which are closely related to the actual impact of the typhoon. Therefore, when the typhoon level is high, the wind field around the building may become more intense, resulting in increased wind pressure and unstable airflow. In this case, the wind field evaluation level can help determine the ventilation needs in the building, especially when the wind speed is high, the ventilation system may need to be closed or adjusted to avoid unnecessary risks. In conditions of low wind speed or stable wind field, the ventilation system can be turned on to maintain air circulation.

[0067] When the wind pressure is low, the system may decide to turn on the ventilation system to enhance indoor ventilation; when the wind pressure is too high, the ventilation system is turned off to avoid damage to indoor items. Finally, the central control system coordinates and optimizes the entire intelligent ventilation system based on the output results of modules such as wind speed and direction data processing, wind field simulation, building swing monitoring, and ventilation system operation decision-making.

[0068] S107. Use deep learning algorithms to train the large amount of wind speed and direction data collected, optimize the prediction accuracy of the wind field model, and improve the response speed and accuracy of the intelligent ventilation system.

[0069] Obtain a large amount of historical and real-time data on wind speed and direction, and preprocess the data, including data cleaning and normalization, to obtain a data set suitable for training. According to the characteristics of the wind field model, select a suitable deep learning algorithm, such as long short-term memory network (LSTM) or convolutional neural network (CNN), to build a prediction model. Divide the preprocessed data set into a training set and a validation set, use the training set to train the deep learning model, and continuously adjust the model parameters through the back propagation algorithm to optimize the model performance. During the training process, use cross-validation and other methods to evaluate the model, monitor the training effect of the model, and avoid the occurrence of overfitting or underfitting problems. After the training is completed, use the validation set data to test the model, evaluate the prediction accuracy and generalization ability of the model, and further optimize and adjust the model according to the evaluation results. Integrate the optimized deep learning model into the intelligent ventilation system, obtain wind speed and direction data in real time, input it into the model for prediction, and obtain the wind speed, direction and temperature change trend in the future. According to the prediction results, the intelligent ventilation system timely adjusts the operating parameters of the ventilation equipment, such as fan speed, damper opening, etc., to achieve rapid response and precise control of the ventilation system, and improve the system's operating efficiency and energy-saving effects.

[0070] Furthermore, a large amount of historical and real-time data on wind speed and direction is obtained, which is the basis for forecasting wind speed and direction. Specifically, wind speed and direction data of the past few years, as well as current real-time data, are obtained from weather stations, sensors on buildings, and public meteorological data platforms. These data may contain time, wind speed, wind direction, and temperature information. Data cleaning is to remove noise and errors in the data, such as missing values, outliers, etc. Interpolation and mean methods can be used to fill missing values, and statistical methods can be used to identify and remove outliers. Normalization is to convert data of different dimensions into a unified range, such as between 0 and 1, to facilitate model training and learning. Commonly used normalization methods include minimum and maximum normalization, standardization, etc. Assuming that the original wind speed data range is 0 to 50 meters per second, it can be converted to between 0 and 1 through minimum and maximum normalization. The choice of deep learning model depends on the characteristics of the data and the requirements of the prediction task. Long short-term memory network (LSTM) is suitable for processing time series data and can capture the time dependence and long-term trends in wind speed and direction data. Convolutional neural networks (CNNs) are better at capturing spatial features and can be used to extract spatial patterns in wind field data. If you want to predict the trend of wind speed, wind direction and temperature changes in the future, LSTM is a more appropriate choice. Specifically, use a network containing multiple LSTM layers, the input is the wind speed and wind direction data sequence in the past period of time, and the output is the predicted value of wind speed and wind direction in the future. Divide the preprocessed data set into a training set and a validation set. Specifically, 80% of the data is used for training and 20% of the data is used for validation. Use the training set to train the deep learning model, and continuously adjust the model parameters through the back propagation algorithm so that the model can learn the rules in the wind speed and wind direction data. During the training process, it is necessary to select a suitable loss function and optimizer. Specifically, use the mean square error as the loss function and use the Adam optimizer to adjust the model parameters. Cross-validation can help evaluate the generalization ability of the model. Specifically, use K-fold cross-validation to divide the training set into K parts, use K-1 parts for training each time, and 1 part for validation. Finally, take the average of the K validation results as the performance indicator of the model. This can effectively avoid the occurrence of overfitting or underfitting problems. If the model performs well on the training set but performs poorly on the validation set, it means that the model may be overfitting and needs to adjust the model structure or increase the training data. After training, the model is tested using the validation set data to evaluate the model's prediction accuracy and generalization ability. Some common indicators can be used to evaluate the performance of the model, such as root mean square error (RMSE), mean absolute error (MAE), etc. Based on the evaluation results, the model can be further optimized and adjusted, such as adjusting the model's hyperparameters, increasing training data, etc. Assume that the RMSE of the model on the validation set is 0.5 meters per second, which means that the model's prediction error is 0.5 meters per second on average.The optimized deep learning model is integrated into the intelligent ventilation system to obtain wind speed and direction data in real time, input it into the model for prediction, and obtain the wind speed, direction and temperature change trend in the future. Specifically, the wind speed and direction data are collected every once in a while, input into the model for prediction, and the predicted value of wind speed and direction in the next hour is obtained. According to the prediction results, the intelligent ventilation system can adjust the operating parameters of the ventilation equipment in time, such as fan speed, damper opening, etc., to achieve rapid response and precise control of the ventilation system. Specifically, if it is predicted that the wind speed will increase in the next hour, the fan speed can be increased in advance to ensure the indoor ventilation effect. This can improve the operating efficiency and energy-saving effect of the system, and adjust the fan speed in advance according to the prediction results to avoid unnecessary energy consumption.

[0071] Embodiment 2

[0072] like Figure 2 As shown, in this embodiment, a wind level-based intelligent ventilation system is provided, including:

[0073] A deployment module, used to deploy wind speed and direction sensors on multiple floors, and obtain wind speed, wind direction data and temperature data of each floor based on the sensors;

[0074] A prediction module, used to predict the wind speed, wind direction data and temperature data of each floor based on a machine learning model to obtain short-term wind speed, wind direction and temperature changes;

[0075] A wind field construction module is used to construct a fluid dynamics model to simulate the wind field of the building and the surrounding environment based on the wind speed, wind direction and temperature changes in the short term and the fluid dynamics model;

[0076] The implementation module is used to construct a wind field evaluation grade based on the wind field of the building and the surrounding environment, construct a ventilation strategy based on the wind field evaluation grade, and perform intelligent ventilation on the building based on the ventilation strategy.

[0077] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent ventilation method based on wind level, characterized in that: include: Deploy wind speed and direction sensors and temperature sensors on multiple floors, and obtain wind speed, wind direction data and temperature data of each floor based on the sensors; Based on the machine learning model, the model of wind speed, wind direction data and temperature data of each floor is predicted to obtain the wind speed, wind direction and temperature changes in the short term; Constructing a fluid dynamics model to simulate the wind field of the building and the surrounding environment based on the wind speed, wind direction and temperature changes in the short term and the fluid dynamics model; A wind field evaluation grade is constructed based on the wind field of the building and the surrounding environment, a ventilation strategy is constructed based on the wind field evaluation grade, and intelligent ventilation is performed on the building based on the ventilation strategy.

2. The method according to claim 1, characterized in that The process of obtaining wind speed, wind direction data and temperature data of each floor based on the sensor includes: According to the layout of multi-story buildings, wind speed sensors, wind direction sensors and temperature sensors are installed at key locations on each floor to collect real-time wind speed, wind direction and temperature data on each floor through the sensors; The wind speed sensor and wind direction sensor are connected to the wireless network, and the collected wind speed, wind direction and temperature data are transmitted to the central processing system in real time based on wireless communication technology; The central processing system receives wind speed, wind direction and temperature data from sensors on each floor, parses and processes the data, and obtains real-time wind speed, wind direction and temperature information on each floor.

3. The method according to claim 1, characterized in that: The process of obtaining short-term wind speed, wind direction and temperature changes includes: Obtain historical wind speed, wind direction and temperature data, preprocess the data, remove outliers and noise data, and obtain a wind speed and direction time series data set; Constructing an ARIMA model, inputting the wind speed and direction time series data set into the ARIMA model for training, and obtaining a temperature-based wind speed and direction prediction model; The wind speed, wind direction data and temperature data of each floor are input into the temperature-based wind speed and wind direction prediction model for calculation to obtain the wind speed, wind direction and temperature changes in the short term.

4. The method according to claim 3, characterized in that: The process of obtaining a temperature-based wind speed and direction prediction model also includes: continuously acquiring real-time wind speed and direction data from sensors on each floor, determining the deviation between the actual wind speed and direction and the predicted value, and triggering retraining and optimization of the prediction model if the deviation exceeds a preset threshold.

5. The method according to claim 1, characterized in that The process of simulating the wind field of the building and the surrounding environment based on the wind speed, wind direction and temperature changes in the short term and the fluid dynamics model includes: The computational fluid dynamics model is used to grid the surrounding environment of the building, set boundary conditions and initial conditions, and simulate the wind field distribution around the building by solving the Navier-Stokes equation to obtain wind speed, wind direction and temperature data at different locations.

6. The method according to claim 1, characterized in that The process of intelligently ventilating a building based on the ventilation strategy includes: Comparing the wind field of the building and the surrounding environment with a preset risk threshold, and constructing a wind field evaluation level based on the comparison result; Based on the wind field evaluation level, an opening strategy of the ventilation system under different conditions is constructed, and the building is intelligently ventilated according to the opening strategy of the ventilation system.

7. The method according to claim 1, characterized in that Also includes: Determine the size, direction and frequency distribution of wind loads based on local meteorological data; The fatigue strength theory is used to calculate the stress-strain response of the ventilation system under wind load. The number of stress cycles is counted by the rain flow counting method to obtain the fatigue life distribution curve. Based on the fatigue life distribution curve, the strength analysis and life prediction of the ventilation system opening mechanism are performed, and the maintenance plan and replacement cycle of the ventilation system are adjusted according to the prediction results.

8. An intelligent ventilation system based on wind level, characterized in that: include: A deployment module, used to deploy wind speed and direction sensors and temperature sensors on multiple floors, and obtain wind speed, wind direction data and temperature data of each floor based on the sensors; A prediction module, used to predict the wind speed, wind direction data and temperature data of each floor based on a machine learning model to obtain short-term wind speed, wind direction and temperature changes; A wind field construction module is used to construct a fluid dynamics model to simulate the wind field of the building and the surrounding environment based on the wind speed, wind direction and temperature changes in the short term and the fluid dynamics model; The implementation module is used to construct a wind field evaluation grade based on the wind field of the building and the surrounding environment, construct a ventilation strategy based on the wind field evaluation grade, and perform intelligent ventilation on the building based on the ventilation strategy.

Images