Ventilation control method and device of natural ventilation air cooling tower

Through multi-point collection and analysis of the internal and external air parameters of natural ventilation air cooling towers, combined with genetic algorithms and CFD analysis, refined control of the air cooling tower environment is achieved, and the problems of unstable cooling efficiency and high energy consumption in traditional control methods are solved. Improves cooling efficiency and energy utilization, and has the ability to learn by itself.

CN118936220AInactive Publication Date: 2024-11-12SHAANXI YULIN ENERGY GRP YANGHUOPAN COAL & ELECTRICITY CO LTD

Patent Information

Application Number
CN202411429759.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-14
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The traditional natural ventilation air-cooling tower control method lacks comprehensive perception and accurate analysis of the air parameters inside and outside the air-cooling tower, and cannot accurately adjust the temperature field, humidity field and air flow field during the cooling process, resulting in unstable cooling efficiency, high energy consumption, and lack of Self-learning and self-optimization ability.

Method used

By conducting multi-point collection and analysis of the internal and external air parameters of natural ventilation air cooling towers, screening and normalizing the temperature and humidity difference and air flow velocity change rate, adjust the opening of the air inlet and outlet, and adjust the angle of the flow guide device. It also dynamically controls the speed of exhaust equipment, and adopts a multi-parameter collaborative control strategy based on genetic algorithm and CFD analysis.

Benefits of technology

It realizes comprehensive perception and refined control of the air-cooled tower environment, improves cooling efficiency and energy utilization, reduces energy consumption, and has self-learning and self-optimization capabilities, significantly improving the system's response speed and adaptability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118936220A_ABST
    Figure CN118936220A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of ventilation adjustment, and discloses a ventilation control method and device for a natural ventilation air cooling tower, which are used for improving the ventilation control efficiency and accuracy of the natural ventilation air cooling tower. The method comprises the following steps: performing multi-point acquisition and analysis on internal and external air parameters of the natural ventilation air cooling tower to obtain spatial distribution characteristic data; screening and normalizing the spatial distribution characteristic data to obtain environmental parameter data; the temperature and humidity difference value and the airflow speed change rate are calculated, and a ventilation parameter adjusting instruction is obtained; adjusting the opening degrees of an air inlet and an air outlet of the natural ventilation air cooling tower to obtain adjusted air flow data; based on the adjusted air flow data, the angle of a flow guiding device in the natural ventilation air cooling tower is adjusted in real time, and optimized air flow distribution data are obtained; and according to the optimized airflow distribution data, the rotating speed of exhaust equipment at the top of the natural ventilation air cooling tower is dynamically controlled, and target exhaust air rate data are obtained.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of ventilation regulation, and in particular to a ventilation control method and device for a natural ventilation air cooling tower. Background Art

[0002] Natural draft air cooling towers are commonly used cooling devices in industrial facilities such as thermal power plants and chemical plants. They use the principle of natural convection to achieve heat exchange and cooling processes. Traditional natural draft air cooling towers usually adopt fixed structural designs and simple control strategies, such as adjusting the fan speed or adjusting the water flow according to the ambient temperature. These control methods mainly rely on a single or limited environmental parameter, such as ambient temperature or humidity, to adjust the operating state of the cooling system. However, this simple control strategy is difficult to cope with complex and changing environmental conditions and load requirements, resulting in unstable cooling efficiency and high energy consumption.

[0003] The deficiencies in the existing technology are mainly reflected in the following aspects: First, the traditional control method lacks comprehensive perception and precise analysis of the air parameters inside and outside the air-cooling tower, and cannot accurately grasp the dynamic changes of the temperature field, humidity field and airflow field during the cooling process. Secondly, the control strategy is relatively simple, and it is difficult to flexibly adjust it according to different working conditions and environmental conditions, resulting in unstable cooling effect. Furthermore, the existing technology lacks precise control of the airflow distribution inside the air-cooling tower, making it difficult to optimize the airflow path, and there is a problem of uneven local cooling. Finally, traditional control makes insufficient use of historical data, lacks self-learning and self-optimization capabilities, and it is difficult to achieve long-term performance improvement. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a ventilation control method and device for a natural ventilation air cooling tower, which are used to improve the efficiency and accuracy of ventilation control of the natural ventilation air cooling tower.

[0005] The present invention provides a ventilation control method for a natural ventilation air-cooling tower, comprising: collecting and analyzing the internal and external air parameters of the natural ventilation air-cooling tower at multiple points to obtain spatial distribution characteristic data; screening and normalizing the spatial distribution characteristic data to obtain environmental parameter data; calculating the temperature and humidity difference and the air flow velocity change rate based on the environmental parameter data to obtain ventilation parameter adjustment instructions; adjusting the openings of the air inlet and the air outlet of the natural ventilation air-cooling tower according to the ventilation parameter adjustment instructions to obtain adjusted air flow data; adjusting the angle of an internal flow guide device of the natural ventilation air-cooling tower in real time based on the adjusted air flow data to obtain optimized air flow distribution data; and dynamically controlling the rotation speed of an exhaust device at the top of the natural ventilation air-cooling tower according to the optimized air flow distribution data to obtain target exhaust volume data.

[0006] The present invention also provides a ventilation control device for a natural ventilation air cooling tower, comprising: a collection module, used for multi-point collection and analysis of internal and external air parameters of the natural ventilation air cooling tower, to obtain spatial distribution characteristic data; a processing module, used for screening and normalizing the spatial distribution characteristic data, to obtain environmental parameter data; a calculation module, used for calculating the temperature and humidity difference and the air flow velocity change rate based on the environmental parameter data, to obtain ventilation parameter adjustment instructions; an adjustment module, used for adjusting the opening of the air inlet and the air outlet of the natural ventilation air cooling tower according to the ventilation parameter adjustment instructions, to obtain adjusted air flow data; an adjustment module, used for real-time adjustment of the angle of the internal guide device of the natural ventilation air cooling tower based on the adjusted air flow data, to obtain optimized air flow distribution data; a control module, used for dynamically controlling the rotation speed of the exhaust equipment at the top of the natural ventilation air cooling tower according to the optimized air flow distribution data, to obtain target exhaust volume data.

[0007] In the technical solution provided by the present invention, by multi-point collection and analysis of air parameters inside and outside the air-cooling tower, spatial distribution characteristic data is obtained, thereby achieving comprehensive and accurate perception of the environment inside and outside the air-cooling tower, and providing a reliable data basis for subsequent control. This multi-point collection method overcomes the limitations of traditional single-point measurement, can capture the unevenness and local characteristics of spatial distribution, and greatly improves the representativeness and accuracy of the data. Secondly, the collected data is screened and normalized to obtain standardized environmental parameter data, effectively removing outliers and noise, improving data quality, and at the same time, normalization is used to make different types of data comparable, creating conditions for subsequent comprehensive analysis. Based on the standardized environmental parameter data, the temperature and humidity difference and the airflow velocity change rate are calculated to obtain ventilation parameter adjustment instructions. This control strategy based on multi-parameter coordination can more comprehensively reflect the operating status of the air-cooling tower, and can better adapt to complex and changeable environmental conditions compared to traditional single parameter control. In addition, by precisely adjusting the opening of the air inlet and outlet, combined with the real-time adjustment of the internal guide device, the fine control of the air flow distribution is achieved, which effectively solves the local dead zone and high-speed zone problems existing in the traditional control method, improves the cooling efficiency and reduces energy consumption. In particular, the angle optimization method of the guide device based on genetic algorithm is introduced, which can find a solution close to the global optimal in a complex nonlinear environment, greatly improving the intelligent level of control. In terms of exhaust volume control, a closed-loop adjustment strategy based on CFD analysis and PID control is adopted, which can not only accurately control the exhaust volume, but also make forward-looking adjustments based on historical data and prediction models, improving the response speed and adaptability of the system. In the entire control process, a large number of data-driven methods are used, such as spatial interpolation, time series analysis, trend prediction, etc., to make the control decision more scientific and reliable. At the same time, by establishing a historical database and continuous data analysis, it has the ability of self-learning and self-optimization, which can continuously improve the control effect. Through multi-parameter collaborative control and refined airflow adjustment, the cooling efficiency and energy utilization rate of the natural ventilation air cooling tower are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0009] Figure 1 This is a flow chart of a ventilation control method for a natural ventilation air cooling tower in an embodiment of the present invention; Figure 2 This is a flow chart of screening and normalizing spatial distribution feature data in an embodiment of the present invention; Figure 3 Schematic diagram of a ventilation control device for a natural ventilation air cooling tower in an embodiment of the present invention.

[0010] Figure numerals: 301, acquisition module; 302, processing module; 303, calculation module; 304, regulation module; 305, adjustment module; 306, control module. DETAILED DESCRIPTION

[0011] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not 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.

[0012] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0013] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0014] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 , Figure 1 Flow chart of a ventilation control method for a natural ventilation air cooling tower according to an embodiment of the present invention. Figure 1As shown, the method includes the following steps: S101, collecting and analyzing the internal and external air parameters of the natural ventilation air cooling tower at multiple points to obtain spatial distribution characteristic data; specifically, installing multiple high-precision temperature sensors, humidity sensors and wind speed sensors at the air inlet, air outlet and key positions around the tower body of the natural ventilation air cooling tower. These sensors should have anti-interference ability and high sampling frequency to ensure the accuracy and real-time performance of the data. At the same time, sensors should also be arranged at different heights and horizontal positions inside the tower body to fully capture the changes in air parameters inside the air cooling tower. These sensors are connected to the central data acquisition system through wireless or wired networks to achieve real-time data transmission. The data acquisition adopts a distributed architecture, which can process the data streams of a large number of sensors at the same time, and perform preliminary data cleaning and preprocessing, such as removing outliers and noise. Subsequently, the preprocessed data is transmitted to a high-performance computing server for in-depth analysis. During the analysis process, the temperature, humidity and wind speed data are first analyzed in time series, and the sliding window technology is used to calculate the change trend and fluctuation characteristics of each parameter. Then, the discrete multi-point data is converted into a continuous three-dimensional spatial distribution field through interpolation algorithms and spatial statistical methods. Using methods such as Kriging interpolation or radial basis function, three-dimensional models of temperature field, humidity field and wind speed field are generated. Next, computational fluid dynamics (CFD) technology is used to combine the generated three-dimensional field model to simulate the dynamic characteristics of airflow inside and outside the air cooling tower. CFD analysis can reveal important information such as the flow path, vortex area and heat transfer efficiency of the airflow. At the same time, through data mining techniques such as principal component analysis (PCA) and cluster analysis, key features are extracted from multidimensional data to identify the main factors affecting the cooling effect and the relationship between parameters.

[0015] Finally, all the analysis results are combined to generate spatial distribution characteristic data. These data include information on temperature gradient, humidity distribution, air velocity field, heat transfer efficiency distribution and other dimensions, which fully reflect the spatial distribution characteristics of air parameters inside and outside the air cooling tower.

[0016] For example, in a natural ventilation air-cooling tower of a large thermal power plant, 50 temperature and humidity sensors can be evenly arranged within a 20-meter range around the tower, and 100 wind speed sensors can be installed at different heights inside the tower. These sensors collect data once a second and transmit it to the data center in real time through the 5G network. The data analysis system uses the above method to process these massive data and eventually generates a four-dimensional data model containing temperature field, humidity field, wind speed field and heat transfer efficiency field. This model can intuitively display the cooling effect of the air-cooling tower at different heights and positions, helping operators identify areas with low cooling efficiency, such as local high temperature areas that may appear at the bottom of the tower or airflow dead corners at the top of the tower.

[0017] S102, screen and normalize the spatial distribution feature data to obtain environmental parameter data; specifically, preliminarily screen the collected spatial distribution feature data to remove obvious outliers and noise. This can be done by setting a reasonable threshold range, combined with the moving median method or a statistical anomaly detection algorithm. For example, using the 3σ principle, data points that exceed the range of three standard deviations of the mean value are marked as potential outliers, and the causes of these outliers are further analyzed to determine whether they need to be removed or corrected. Next, the screened data is smoothed to reduce the impact of short-term fluctuations on the overall trend. The moving average method or exponential smoothing method can be used to select an appropriate window size or smoothing parameter according to the characteristics of the data. For example, for temperature data, a 5-minute moving average window is used, while for faster-changing wind speed data, a 1-minute window may be used. Then, for different types of environmental parameters, an appropriate normalization method is selected. For temperature and humidity data, the maximum and minimum normalization method is usually used to map the data to the [0, 1] interval. As for wind speed data, considering its possible large fluctuations, the Z-score standardization method can be used to make the data obey a normal distribution with a mean of 0 and a standard deviation of 1. This processing can make data of different dimensions and scales comparable.

[0018] After normalization, the weight of the influence of different environmental parameters on the cooling effect needs to be considered. The weight coefficient of each parameter can be determined by principal component analysis (PCA) or expert experience, and then the normalized data can be weighted. For example, it may be found that the weight of temperature on the cooling effect is 0.5, humidity is 0.3, and wind speed is 0.2. In order to capture the time dimension characteristics of environmental parameters, the weighted data can be decomposed into trend components, seasonal components, and residual components. This is achieved using the STL (Seasonal and Trend decomposition using Loess) algorithm. Through decomposition, the long-term trend and cyclical changes of environmental parameters can be better understood. In addition, considering the possible complex relationships between environmental parameters, correlation analysis and cluster analysis techniques can be used to explore the potential associations between parameters. For example, the Pearson correlation coefficient between the parameters can be calculated to construct a correlation matrix, or the K-means clustering algorithm can be used to classify environmental states into several types.

[0019] Finally, in order to obtain continuous environmental parameter data, the processed discrete data points can be interpolated. Here, piecewise linear interpolation or spline interpolation methods can be selected to generate smooth environmental parameter curves. In order to further extract signal features, these curves can be wavelet transformed to obtain multi-scale environmental parameter representations.

[0020] For example, suppose that in a large natural ventilation air cooling tower, temperature, humidity and wind speed data are collected from 100 sensor points every minute. First, the 3σ principle is used to screen outliers, and then a 5-minute moving average is applied to the temperature and humidity data, and a 1-minute moving average is applied to the wind speed data. Next, the temperature and humidity data are normalized to the maximum and minimum values, the temperature range of -10℃ to 50℃ is mapped to 0-1, and the wind speed data is Z-score standardized. Then, based on the PCA analysis results, weights of 0.5, 0.3, and 0.2 are assigned to temperature, humidity, and wind speed, respectively. After that, the STL algorithm is used to decompose the 24-hour data into time series and identify the daily variation trend. Through correlation analysis, it is found that temperature and humidity are highly positively correlated in certain periods. Finally, cubic spline interpolation is used to generate continuous environmental parameter curves, and features of different scales are extracted through wavelet transform.

[0021] S103, based on the environmental parameter data, calculate the temperature and humidity difference and the air flow velocity change rate, and obtain the ventilation parameter adjustment instruction; specifically, perform time series analysis on the temperature and humidity data in the environmental parameter data, which can be achieved by using a sliding window technique or an autoregressive integrated moving average (ARIMA) model. Through this analysis, the changing trend of temperature and humidity can be obtained, which is crucial for predicting the temperature and humidity changes in the short term. At the same time, it is necessary to calculate the temperature and humidity difference inside and outside the air cooling tower, which can be achieved by simple numerical subtraction, but it should be noted that the temperature and humidity differences at different heights and positions of the air cooling tower need to be considered, so it may be necessary to calculate the average or weighted average of multiple points. Next, the air flow velocity data needs to be processed. Differentiation operation is performed on the air flow velocity data to obtain the rate of change of the air flow velocity. Numerical differentiation methods such as central difference method or high-order difference method are used here to obtain more accurate results. At the same time, the direction change of the air flow needs to be considered, which can be achieved by calculating the angular change of the wind direction. Combining the air flow velocity change rate with the wind direction change can obtain a more comprehensive description of the dynamic characteristics of the air flow.

[0022] Then, threshold judgment is performed on the calculated temperature and humidity difference and airflow velocity change rate. Using the fuzzy logic control method, continuous values ​​are mapped to discrete language variables, such as "low", "medium", "high", etc. For example, a threshold range for the temperature difference can be set. When the difference is within a certain range, it is judged as "needing slight adjustment", and when it exceeds a certain threshold, it is judged as "needing substantial adjustment". The same method can also be applied to the judgment of humidity difference and airflow velocity change rate.

[0023] Based on these judgment results, a decision tree or rule base is used to generate a preliminary adjustment strategy. For example, if the temperature difference is large and the airflow velocity change rate is small, it may be necessary to increase the fan speed; if the humidity difference is small and the airflow velocity change rate is large, it may be necessary to adjust the guide plate angle, etc. These preliminary strategies need to be further optimized, using genetic algorithms or particle swarm optimization algorithms to find the optimal parameter combination. In the optimization process, constraints such as energy consumption factors and equipment life also need to be considered. A multi-objective optimization model can be constructed, with cooling effect, energy consumption and equipment life as optimization goals, and the balance point can be found through the Pareto optimal solution set. At the same time, the dynamic response characteristics of the system need to be considered, and the model predictive control (MPC) method is used to predict the system behavior in the future based on the current state, and the optimal control sequence is generated accordingly.

[0024] Finally, the optimized control strategy is converted into specific ventilation parameter adjustment instructions. These instructions may include fan speed adjustment value, guide plate angle change, circulating water flow adjustment value, etc. In order to make the instructions smoother and avoid sudden changes causing shocks to the equipment, an interpolation algorithm is used to generate a continuous adjustment curve.

[0025] For example, suppose that at a certain moment, through analysis, it is known that the average temperature difference between the bottom and top of the air-cooling tower is 15°C, the relative humidity difference is 20%, and it is detected that the average airflow velocity in the tower has increased by 0.5m / s in the past 10 minutes. Through fuzzy logic judgment, the temperature difference is judged as "high", the humidity difference is judged as "medium", and the airflow velocity change rate is judged as "medium-high". Based on these judgments, the initial decision is to moderately increase the fan speed and fine-tune the guide plate angle. Then, through genetic algorithm optimization, considering the current high electricity price and increased energy consumption weight, the final optimization solution is to increase the fan speed by 10% and adjust the bottom guide plate angle by 5 degrees. This solution predicts the system response in the next 30 minutes through MPC and confirms that the temperature difference can be reduced to below 10°C within 15 minutes. Finally, a smooth adjustment curve is generated to gradually increase the fan speed to the target value within 5 minutes and complete the adjustment of the guide plate angle within 2 minutes, thereby achieving both effective and smooth ventilation parameter adjustment.

[0026] S104, according to the ventilation parameter adjustment instruction, the opening of the air inlet and outlet of the natural ventilation air cooling tower is adjusted to obtain the adjusted air flow data; specifically, the received ventilation parameter adjustment instruction is parsed to extract the specific opening adjustment amount of the air inlet and outlet. This can be done by using a specific communication protocol and data parsing algorithm to ensure accurate interpretation of the instruction. The parsed data will be converted into a control signal that the actuator can understand, which usually includes the number of rotation steps of the stepper motor or the angle change of the servo motor. Next, the motion trajectory of the actuator is calculated based on the parsed opening adjustment amount. The first step involves kinematic analysis, which needs to consider factors such as the mechanical characteristics, motion limits, acceleration and deceleration of the actuator. Through precise trajectory planning, smooth and efficient opening adjustment can be achieved while avoiding mechanical shock and vibration. The calculated motion trajectory will be converted into a series of discrete control instructions, which will be sent to the motor driver or other actuators.

[0027] During the opening adjustment process, the actual position and motion state of the actuator are monitored in real time. This is usually achieved through encoders or position sensors, which can provide high-precision position feedback. The system compares the real-time position data with the preset target position and calculates the position error. If the error exceeds the preset threshold, the control system will start the error correction mechanism and adjust the movement of the actuator in real time through the PID control algorithm or other advanced control algorithms to ensure that the expected opening position is finally achieved. After the opening adjustment is completed, the actual opening after adjustment is accurately measured and recorded. This can be done with a high-precision position sensor or a visual recognition system. The measured actual opening data will be stored in a historical database for subsequent trend analysis and performance evaluation. At the same time, a series of airflow measurement devices, such as hot wire anemometers, pitot tubes or ultrasonic flow meters, are started to obtain real-time airflow data after adjustment.

[0028] These airflow data need to be processed through statistical analysis, including the calculation of average airflow velocity, flow rate, and uniformity of airflow distribution. During the processing, digital filtering technology may be needed to remove measurement noise and construct a complete airflow field distribution map through interpolation algorithm. In addition, the system may also need to consider the influence of environmental factors, such as changes in wind direction and wind speed, which may require the introduction of meteorological data for compensation calculation. The processed airflow data will be compared with the expected target to calculate the deviation of the adjustment effect. If the deviation exceeds the allowable range, the system will start the adaptive adjustment mechanism and generate new fine-tuning instructions. This process may require multiple iterations until a satisfactory adjustment effect is achieved. The final adjusted airflow data includes not only quantitative flow values, but also multi-dimensional information such as spatial characteristics of airflow distribution and stability indicators.

[0029] For example, suppose a large natural ventilation air cooling tower receives a ventilation parameter adjustment instruction, requiring the air inlet opening to increase from 60% to 75% and the air outlet opening to decrease from 50% to 45%. First, this instruction is converted into a specific control instruction that the air inlet actuator needs to rotate 1500 steps clockwise and the air outlet actuator needs to rotate 500 steps counterclockwise. During the execution process, the encoder monitors the position in real time and finds that the air inlet is slightly stuck after rotating 1200 steps. The PID control algorithm is immediately started, and the problem is successfully overcome by increasing the motor torque and fine-tuning the speed, and finally accurately reaches the target position.

[0030] After the adjustment, airflow data was collected through 50 hot-wire anemometers distributed at different locations of the air-cooling tower. After 5 minutes of continuous sampling and data processing, the average air intake increased by 15% and the air outlet decreased by 8%. At the same time, through the constructed three-dimensional airflow field distribution map, it was found that although the overall airflow volume reached the expected level, local vortices appeared in the northwest corner of the tower, affecting the cooling efficiency. Based on this, the system generated a fine-tuning instruction and slightly adjusted the angle of the guide plate near the area. After re-measurement and verification, the adjusted airflow volume data with uniform distribution and total amount in line with expectations was finally obtained.

[0031] S105. Based on the adjusted air flow data, the angle of the internal flow guide device of the natural ventilation air cooling tower is adjusted in real time to obtain optimized air flow distribution data; specifically, the adjusted air flow data is deeply analyzed and processed. This usually involves using computational fluid dynamics (CFD) technology, combined with spatial interpolation algorithms such as Kriging or radial basis function method, to construct a high-precision three-dimensional air flow field model. This model can intuitively display the air flow distribution inside the air cooling tower, including multiple dimensions such as velocity field, pressure field and temperature field.

[0032] Based on this three-dimensional model, image processing and pattern recognition techniques are used to identify abnormal areas in the airflow distribution, such as airflow dead zones, high-speed zones, or eddy zones. Edge detection algorithms, cluster analysis, or deep learning target detection methods may be used. After identifying these abnormal areas, the specific areas and adjustment ranges that need to be adjusted are calculated, which usually requires a comprehensive judgment based on the expert knowledge base and historical data. Next, based on the identified abnormal areas, an optimization algorithm is used to calculate the best adjustment plan for the guide device. This involves the use of genetic algorithms, particle swarm optimization algorithms, or gradient descent methods, etc. The optimization objectives are usually multidimensional, and not only the uniformity of airflow distribution must be considered, but also factors such as cooling efficiency and energy consumption. The optimization algorithm generates a series of possible adjustment plans, predicts the effect of each plan through numerical simulation, and finally selects the best plan.

[0033] After the adjustment plan is determined, it is converted into specific execution instructions, such as the angle change of the guide plate. These instructions are sent to the actuator, usually a stepper motor or a servo motor. During the execution process, the actual position of the guide device is monitored in real time, and PID control or more advanced adaptive control algorithms are used to ensure accurate execution. At the same time, in order to avoid the airflow turbulence that may be caused by sudden and large adjustments, the system usually adopts a gradual adjustment strategy to decompose large adjustments into a series of small continuous adjustments. After the adjustment is completed, a new round of airflow data collection will be started immediately. This is usually achieved through a large number of sensors distributed inside the air cooling tower, such as hot wire anemometers, pressure sensors, etc. The collected raw data will be pre-processed by filtering, denoising, etc., and then the three-dimensional airflow field model will be constructed again. This new model is compared with the model before adjustment to calculate the degree of improvement, such as the improvement of airflow uniformity and the reduction of dead zone area.

[0034] If the improvement effect is not ideal, an iterative optimization mechanism will be initiated. This involves the use of reinforcement learning algorithms, such as Q-learning or deep Q networks, to gradually optimize and adjust strategies through continuous attempts and learning. At the same time, dynamic changes in the external environment, such as sudden changes in wind direction and wind speed, are considered, and responses are made in advance through the prediction model to achieve feedforward control. Finally, when the airflow distribution reaches the preset optimization target, a detailed optimized airflow distribution data report will be generated. This report includes not only the final airflow distribution data, but also the changes in various indicators throughout the optimization process, such as the cooling efficiency improvement curve, energy consumption change trend, etc. These data will be stored in the historical database for subsequent trend analysis and system performance evaluation.

[0035] For example, suppose that in a large natural ventilation air-cooling tower, after the initial airflow adjustment, CFD analysis found that there was an obvious low-speed area in the northwest corner of the bottom of the tower, with an area of ​​about 50 square meters and an average airflow speed 40% lower than the surrounding area. First, it was calculated that the three sets of guide plates near the area needed to be adjusted, and the angles needed to be changed by +5 degrees, -3 degrees, and +7 degrees respectively. When performing these adjustments, a 5-minute progressive adjustment strategy was adopted, with adjustments made every 30 seconds, and each adjustment was 10% of the target value. During the adjustment process, a guide plate encountered slight resistance when performing a +4-degree adjustment, and the adaptive control algorithm was immediately started, and the adjustment was successfully completed by increasing the motor torque and fine-tuning the speed. After the adjustment was completed, the system once again conducted a comprehensive airflow data collection, and after processing, it was found that the area of ​​the low-speed zone was reduced to 20 square meters, and the average airflow speed increased by 25%. However, this improvement has not yet reached the preset target.

[0036] The second round of optimization was immediately launched. This time, the influence of adjacent areas was taken into account, and the guide plates in a larger range around were fine-tuned. After three rounds of iterative optimization, the low-speed area was completely eliminated, the overall airflow distribution uniformity was improved by 35%, the cooling efficiency was improved by 8%, and the energy consumption was reduced by 3%. The entire optimization process took 30 minutes and generated a detailed report, including the specific parameters adjusted at each step, the dynamic change diagram of the airflow distribution, and the comparative analysis before and after optimization. This optimized airflow distribution data is not only directly used for current operation control, but also provides a valuable reference for rapid response in similar situations in the future.

[0037] S106. According to the optimized air flow distribution data, dynamically control the rotation speed of the exhaust equipment at the top of the natural ventilation air cooling tower to obtain target exhaust volume data.

[0038] Specifically, the optimized air flow distribution data is deeply analyzed, which usually involves the use of computational fluid dynamics (CFD) technology combined with numerical simulation methods to build an accurate dynamic model of the air flow in the tower. This model not only includes the air flow velocity field and pressure field, but also needs to consider the temperature distribution and humidity changes, so as to fully reflect the thermodynamic state inside the air cooling tower. Based on this dynamic model, predictive control algorithms such as model predictive control (MPC) or adaptive predictive control are used to estimate the required exhaust volume. The process needs to consider multiple factors, including the current cooling load, environmental conditions, energy consumption targets, etc. The prediction algorithm generates a series of possible exhaust volume schemes and evaluates the effect and energy consumption of each scheme through numerical simulation.

[0039] Next, the estimated exhaust volume demand is compared with the current operating status of the exhaust equipment to calculate the speed change that needs to be adjusted. In this process, the dynamic characteristics of the exhaust equipment, such as start-up time, acceleration limit, etc., are considered to ensure the executability of the speed control command. At the same time, in order to avoid equipment wear and energy consumption fluctuations caused by frequent and large speed adjustments, the system usually uses smoothing processing to generate a progressive speed control curve. The generated speed control command will be converted into a specific electrical control signal and transmitted to the drive motor of the exhaust fan through a frequency converter or other power electronic equipment. During the speed control process, the operating parameters of the motor, including speed, current, temperature, etc., are monitored in real time, and closed-loop control algorithms, such as PID control or fuzzy PID control, are used to ensure that the actual speed accurately matches the target value. At the same time, parameters such as bearing vibration and noise are monitored to prevent mechanical failures caused by speed regulation.

[0040] After the speed regulation is executed, a series of measurement and verification procedures will be started immediately. This usually involves using a variety of flow measurement techniques, such as hot wire anemometers, ultrasonic flow meters or pressure differential methods, to perform high-precision flow measurements at the exhaust outlet. The measurement data will undergo real-time digital filtering and statistical processing to eliminate the influence of instantaneous fluctuations and obtain stable and reliable exhaust volume data. The actual exhaust volume measured is compared with the target value and the deviation is calculated. If the deviation exceeds the preset threshold, the adaptive control mechanism will be activated and a correction instruction will be generated. This process may require multiple iterations until the error between the actual exhaust volume and the target value falls within an acceptable range. In this process, its internal model is continuously updated to improve the accuracy of future control.

[0041] At the same time, it continuously monitors changes in environmental parameters, such as external wind speed and air pressure, and responds quickly to these changes through a feedforward control mechanism. For example, if strong winds are detected, the exhaust strategy will be adjusted in advance to maintain a stable cooling effect. Finally, a complete exhaust control report is generated, including target exhaust volume data, actual exhaust volume data, speed adjustment process, energy consumption analysis and other information. These data are not only used for current operation monitoring, but also stored in the historical database for long-term performance analysis and optimization.

[0042] For example, suppose that in a natural ventilation air-cooling tower of a large thermal power plant, the system calculates the need to increase the exhaust volume from 200,000 m³ / h to 230,000 m³ / h based on the optimized air flow distribution data to cope with the sudden increase in heat load. First, a 10-minute progressive speed regulation scheme is generated through the MPC algorithm, gradually increasing the exhaust fan speed from the original 600 rpm to 690 rpm. During the execution process, it was detected that the vibration of the exhaust fan bearing increased slightly at 650rpm. The system immediately started the vibration suppression algorithm and successfully controlled the vibration within a safe range by fine-tuning the speed and adjusting the blade angle. At the same time, the system also noticed that the external wind speed had a slight upward trend, and the subsequent speed regulation curve was corrected in real time based on this.

[0043] Finally, after three rounds of fine-tuning, the system stabilized the exhaust volume at 228,500m³ / h, with an error of only 0.65%, while controlling the energy consumption increase within 5%. Throughout the process, the system continuously monitored the temperature distribution changes inside the air-cooling tower and confirmed that the cooling effect achieved the expected goal. The generated report shows that compared with traditional fixed speed control, this dynamic control method not only improves cooling efficiency, but also saves about 3% of total energy consumption during full-day operation. These data provide valuable decision-making support for power plant managers and also point out the direction for future system optimization.

[0044] By executing the above steps, the air parameters inside and outside the air cooling tower are collected and analyzed at multiple points to obtain spatial distribution characteristic data, which realizes the comprehensive and accurate perception of the environment inside and outside the air cooling tower, and provides a reliable data basis for subsequent control. This multi-point collection method overcomes the limitations of traditional single-point measurement, can capture the unevenness and local characteristics of spatial distribution, and greatly improves the representativeness and accuracy of the data. Secondly, the collected data is screened and normalized to obtain standardized environmental parameter data, effectively removing outliers and noise, improving data quality, and making different types of data comparable through normalization, creating conditions for subsequent comprehensive analysis. Based on the standardized environmental parameter data, the temperature and humidity difference and the air flow velocity change rate are calculated to obtain the ventilation parameter adjustment instructions. This control strategy based on multi-parameter coordination can more comprehensively reflect the operating status of the air cooling tower. Compared with the traditional single parameter control, it can better adapt to complex and changing environmental conditions. In addition, by precisely adjusting the opening of the air inlet and outlet, combined with the real-time adjustment of the internal guide device, the fine control of the air flow distribution is achieved, which effectively solves the local dead zone and high-speed zone problems existing in the traditional control method, improves the cooling efficiency and reduces energy consumption. In particular, the angle optimization method of the guide device based on genetic algorithm is introduced, which can find a solution close to the global optimal in a complex nonlinear environment, greatly improving the intelligent level of control. In terms of exhaust volume control, a closed-loop adjustment strategy based on CFD analysis and PID control is adopted, which can not only accurately control the exhaust volume, but also make forward-looking adjustments based on historical data and prediction models, improving the response speed and adaptability of the system. In the entire control process, a large number of data-driven methods are used, such as spatial interpolation, time series analysis, trend prediction, etc., to make the control decision more scientific and reliable. At the same time, by establishing a historical database and continuous data analysis, it has the ability of self-learning and self-optimization, which can continuously improve the control effect. Through multi-parameter collaborative control and refined airflow adjustment, the cooling efficiency and energy utilization rate of the natural ventilation air cooling tower are significantly improved.

[0045] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) multiple temperature sensors are set at the air inlet, air outlet and around the tower body of the natural ventilation air cooling tower to obtain multi-point temperature data, and multiple humidity sensors are set at the air inlet, air outlet and around the tower body of the natural ventilation air cooling tower to obtain multi-point humidity data; (2) multiple wind speed sensors are set at the air inlet, air outlet and around the tower body of the natural ventilation air cooling tower to obtain multi-point air flow velocity data, and time series sampling is performed on the multi-point temperature data to obtain temperature change trend data; (3) time series sampling is performed on the multi-point humidity data to obtain humidity change trend data, and time series sampling is performed on the multi-point air flow velocity data to obtain air flow velocity change trend data; (4) temperature and humidity gradients and air flow velocity gradients are calculated based on the temperature change trend data, humidity change trend data and air flow velocity change trend data to obtain spatial distribution characteristic data.

[0046] Specifically, to achieve multi-parameter monitoring and data analysis of natural ventilation air-cooling towers, it is first necessary to deploy a high-precision sensor network at key locations. Temperature sensors, humidity sensors, and wind speed sensors are strategically set at the air inlet, outlet, and around the tower body. The temperature sensor uses a thermocouple or platinum resistance type to ensure that the measurement accuracy in the range of -50°C to 150°C reaches ±0.1°C. The humidity sensor uses a capacitive or polymer film type, with a relative humidity measurement range of 0-100% and an accuracy of ±2%RH. The wind speed sensor uses a hot wire or ultrasonic type, with a measurement range of 0-50m / s and an accuracy of ±0.1m / s.

[0047] These sensors are connected to the data acquisition system through industrial-grade wireless networks or wired connections to achieve high-frequency and synchronous data acquisition. Data acquisition adopts a distributed architecture, and each acquisition node is equipped with a high-performance processor and large-capacity storage, which can perform preliminary data cleaning and preprocessing locally, such as removing outliers and high-frequency noise. The data acquisition frequency is dynamically adjusted according to the characteristics of parameter changes. The typical sampling interval for temperature and humidity data is 1 minute, and the wind speed data can reach 1 time per second to capture instantaneous changes.

[0048] Time series sampling of the collected multi-point temperature, humidity and air velocity data is a key step in obtaining the trend of change. Using sliding window technology, the window size is flexibly set according to the characteristics of the data. Temperature and humidity data usually use a window of 30 minutes to 2 hours, and wind speed data may use a window of 5-15 minutes. In each sliding window, statistical features such as mean, variance, slope and kurtosis are calculated to construct a multidimensional feature vector. At the same time, the autoregressive integrated moving average (ARIMA) model is used to model the time series to capture the periodicity and trend characteristics of the data.

[0049] The parameter selection of the ARIMA model is automatically optimized through grid search and AIC criteria. For example, for temperature data, an ARIMA (2,1,1) model may be obtained, indicating the use of 2nd-order autoregressive terms, 1st-order differences, and 1st-order moving average terms. After the model training is completed, 1 hour of historical data is used to predict the change trend in the next 30 minutes to obtain the temperature change trend data. The humidity and wind speed data also go through this process to obtain their respective change trend data. Based on the obtained temperature, humidity, and airflow velocity change trend data, calculating the spatial gradient is a key step in obtaining the distribution characteristics. A three-dimensional interpolation algorithm, such as the Kriging method or the radial basis function method, is used to interpolate discrete measurement point data into a continuous three-dimensional space field. For the temperature field T (x, y, z), calculate its spatial gradient: , represents the temperature field, , , They represent the three coordinate axes of the three-dimensional space, ∂T / ∂x represents the rate of change of temperature in the x direction, ∂T / ∂y represents the rate of change of temperature in the y direction, and ∂T / ∂z represents the rate of change of temperature in the z direction. The same is true for the humidity field H(x,y,z) and the wind speed field V(x,y,z). These gradient vectors provide information about the direction and rate of change of parameters in space.

[0050] In order to comprehensively characterize the spatial distribution characteristics, a multidimensional feature vector F is constructed: , is the gradient of the temperature field, is the gradient of the humidity field, the structure and Similar, but Indicates humidity, is the gradient of the wind speed field, represents the wind speed vector field, is the curl of the wind velocity field, which indicates the rotational characteristics of the airflow. is the divergence of the wind velocity field, indicating the convergence or divergence trend of the airflow; curl (V) represents the curl of the wind velocity field, reflecting the rotational characteristics of the airflow; It represents the divergence of the wind speed field and indicates the convergence or divergence trend of the airflow. This eigenvector F is the final spatial distribution characteristic data, which fully describes the distribution state and change trend of the environmental parameters inside and outside the air cooling tower.

[0051] For example, the natural ventilation air-cooling tower of a large thermal power plant is 150 meters high and has a bottom diameter of 120 meters. A total of 50 temperature sensors, 50 humidity sensors and 30 wind speed sensors are arranged around the tower. In a typical data collection process, the temperature data range is 15°C to 45°C, the relative humidity range is 30% to 80%, and the wind speed range is 0.5m / s to 12m / s. After ARIMA model analysis, it is predicted that the bottom temperature of the tower will rise by 2°C and the top temperature will drop by 1°C in 30 minutes, forming a significant temperature gradient. The humidity trend shows that the humidity in the tower will increase by 5%RH as a whole. Wind speed analysis found that a vortex structure with an intensity of 2m / s was formed in the tower. The maximum temperature gradient calculated after spatial interpolation is 0.2°C / m, which appears in the middle of the tower. The maximum humidity gradient is 1%RH / m, which is located near the air inlet. The divergence analysis of the wind speed field shows that there is a significant airflow convergence phenomenon at the top of the tower, and the maximum divergence value reaches .

[0052] In a specific embodiment, if Figure 2 As shown, the process of executing step S102 can specifically include the following steps: S201, performing outlier detection on multi-point temperature data, multi-point humidity data and multi-point airflow velocity data, eliminating data points that exceed a preset range, obtaining preliminary screened environmental data, and performing moving average processing on the preliminary screened environmental data to obtain smoothed environmental data; S202, performing maximum and minimum value normalization processing on the temperature data in the smoothed environmental data to obtain normalized temperature data, and performing maximum and minimum value normalization processing on the humidity data in the smoothed environmental data to obtain normalized humidity data; S203, performing Z-score normalization processing on the airflow velocity data in the smoothed environmental data to obtain standardized airflow velocity data, and performing moving average processing on the environmental data according to the normalized temperature data and the normalized humidity data. The weight coefficient of each parameter is calculated based on the data and the standardized air flow velocity data to obtain weighted environmental parameter data; S204, principal component analysis is performed on the weighted environmental parameter data to extract the main features to obtain environmental characteristic data after dimensionality reduction, and time series decomposition is performed on the environmental characteristic data after dimensionality reduction to separate trend, seasonality and residual components to obtain multi-scale environmental characteristic data; S205, the correlation matrix of each parameter is calculated according to the multi-scale environmental characteristic data to obtain the correlation data between parameters, and cluster analysis is performed on the environmental characteristics based on the correlation data between parameters to obtain the environmental state classification result; S206, piecewise linear interpolation is performed on each environmental parameter according to the environmental state classification result to obtain a continuous environmental parameter curve, and the continuous environmental parameter curve is subjected to wavelet transformation to obtain environmental parameter data.

[0053] Specifically, outlier detection is first performed, and statistical methods such as the interquartile range (IQR) are used to identify and remove abnormal data points. For temperature data, the normal range is set to -10°C to 60°C, and data points outside this range are marked as abnormal. The normal range of humidity data is set to 0% to 100%, and the normal range of airflow velocity data is 0m / s to 30m / s. In this way, obvious erroneous data caused by factors such as sensor failure and electromagnetic interference are eliminated, and the environmental data after preliminary screening is obtained. Then, the environmental data after preliminary screening is processed by moving average to smooth short-term fluctuations and highlight long-term trends. The simple moving average method with a 5-minute window is used to take the average value of the data within 2.5 minutes before and after each data point. This processing effectively reduces the random noise in the data and obtains smoothed environmental data.

[0054] Normalization of the smoothed environmental data is to eliminate the dimension differences between different parameters and make the data comparable. The temperature and humidity data are normalized using the maximum and minimum values ​​to map the data to the [0,1] interval. For example, the temperature data from -10°C to 60°C is mapped to the range of 0 to 1.

[0055] For air velocity data, considering its possible large volatility, the Z-score standardization method is used. The calculation method is: Z=(X-μ) / σ, where X is the original data, μ is the mean value, and σ is the standard deviation. This method makes the mean of air velocity data become 0 and the standard deviation become 1, effectively dealing with the influence of outliers. Next, the weight coefficient of each parameter is calculated based on the normalized temperature, humidity and standardized air velocity data. The entropy weight method is used to determine the weight. This method assigns weights based on the degree of variation of the data. The greater the variation, the higher the weight of the indicator. The information entropy of each indicator is calculated, and then the weight is obtained by subtracting the information entropy from 1. The obtained weight is multiplied by the normalized / standardized data to obtain the weighted environmental parameter data. The weighted environmental parameter data is subjected to principal component analysis (PCA) to reduce the dimension and extract the main features. PCA projects the original data onto a set of orthogonal bases through linear transformation, and selects several principal components with the largest variance contribution as new features. Usually, the principal components with a cumulative variance contribution of more than 85% are selected. This step effectively reduces the dimension of the data while retaining the main information of the data.

[0056] The time series decomposition of the environmental characteristic data after dimensionality reduction was performed using the STL (Seasonal and Trend decomposition using Loess) method. The STL method decomposes the time series into trend components, seasonal components, and residual components. The trend component reflects the long-term trend of the data, the seasonal component captures the periodic change pattern, and the residual component contains random fluctuations. This decomposition allows a deeper understanding of the different time scale characteristics of environmental parameters. Based on the multi-scale environmental characteristic data, the correlation matrix between the parameters is calculated. The Pearson correlation coefficient is used to measure the degree of linear correlation between parameters. The correlation coefficient matrix provides a quantitative description of the strength of association between parameters, which helps to understand the interaction between different environmental factors.

[0057] Cluster analysis is performed using the correlation data between parameters, and the K-means clustering algorithm is used. The K-means algorithm distributes data points into K clusters through iterative optimization, so that the similarity of data points within a cluster is maximized and the similarity between clusters is minimized. The clustering results reflect the classification of different environmental states and help identify typical environmental patterns. Finally, according to the environmental state classification results, each environmental parameter is piecewise linearly interpolated to obtain a continuous environmental parameter curve. The interpolation process ensures the continuity of the data, which is convenient for subsequent analysis and control. The continuous environmental parameter curve is subjected to wavelet transform, and the Daubechies wavelet is selected to effectively capture the multi-scale characteristics of the signal and obtain the final environmental parameter data.

[0058] For example, in a typical data processing process of a natural ventilation air cooling tower in a large thermal power plant, the original data contains 24-hour data from 100 temperature sensors, 100 humidity sensors, and 60 wind speed sensors, with a sampling interval of 1 minute. Outlier detection removes 3% of temperature data points, 2% of humidity data points, and 5% of wind speed data points. Moving average processing smoothes the data and reduces short-term fluctuations by about 40%. After normalization, the temperature data range is mapped from the original 10°C to 50°C to 0 to 1, and the humidity data is mapped from 20% to 90% to 0 to 1. After the air flow velocity data is standardized by Z-score, 95% of the data falls within the range of -2 to 2. The entropy weight method calculates the weights of temperature, humidity, and wind speed to be 0.4, 0.3, and 0.3, respectively. After PCA analysis, the first three principal components are selected, and the cumulative variance contribution rate reaches 92%. Time series decomposition finds that the temperature data has an obvious daily variation cycle with an amplitude of about 5°C. Correlation analysis shows that the correlation coefficient between temperature and humidity is -0.7, showing a strong negative correlation. K-means clustering divides the environmental state into four categories, corresponding to typical conditions such as high temperature and low humidity during the day, low temperature and high humidity at night, mild transition period, and extreme weather. The final environmental parameter data contains the main features of the original data, while greatly reducing the amount of data, providing high-quality input for subsequent control decisions.

[0059] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) performing a time series analysis on the temperature data in the environmental parameter data, calculating the temperature change trend, and obtaining the temperature trend data, and performing a time series analysis on the humidity data in the environmental parameter data, calculating the humidity change trend, and obtaining the humidity trend data; (2) calculating the temperature and humidity difference inside and outside the tower based on the temperature trend data and the humidity trend data, obtaining the temperature and humidity gradient data, and performing a differential operation on the airflow velocity data in the environmental parameter data, obtaining the airflow velocity change rate data; (3) performing a threshold judgment on the temperature and humidity gradient data, determining the temperature and humidity adjustment requirements, obtaining the temperature and humidity adjustment parameters, and performing a threshold judgment on the airflow velocity change rate data, determining (4) According to the temperature and humidity adjustment parameters and the airflow adjustment parameters, the air inlet opening adjustment amount is calculated through fuzzy logic reasoning to obtain the preliminary air inlet adjustment instruction, and according to the temperature and humidity adjustment parameters and the airflow adjustment parameters, the air outlet opening adjustment amount is calculated through fuzzy logic reasoning to obtain the preliminary air outlet adjustment instruction; (5) Comprehensively analyze the preliminary air inlet adjustment instruction and the preliminary air outlet adjustment instruction, calculate the weight coefficient of each adjustment instruction, obtain the weighted adjustment instruction, and generate a continuous adjustment curve based on the weighted adjustment instruction through the interpolation algorithm to obtain the smoothed adjustment instruction; (6) Based on the smoothed adjustment instruction, the ventilation parameter adjustment instruction is generated through the operating status and historical data of the natural ventilation air cooling tower.

[0060] Specifically, to achieve efficient control of natural ventilation air cooling towers, it is first necessary to conduct in-depth time series analysis of environmental parameter data. When performing time series analysis on temperature data, the autoregressive integrated moving average (ARIMA) model is used. The ARIMA model predicts future temperature change trends by identifying the autocorrelation, trend, and seasonal characteristics of the data. The specific steps include data stationarity test, differential processing, model parameter determination, and model fitting. The order of the model is determined by analyzing the autocorrelation function (ACF) and partial autocorrelation function (PACF). After the model is fitted, the root mean square error (RMSE) is used to evaluate the model accuracy. The ARIMA model is also applied to humidity data, but considering the changing characteristics of humidity, logarithmic transformation may be required to improve the prediction accuracy.

[0061] Based on the obtained temperature trend data and humidity trend data, calculate the temperature and humidity difference inside and outside the tower. The temperature difference is directly obtained by subtracting the temperature inside and outside the tower, while the humidity difference needs to consider the calculation of absolute humidity. Through the temperature and humidity difference, calculate the temperature and humidity gradient, which represents the rate of change of temperature or humidity per unit distance. Perform numerical differentiation on the airflow velocity data and use the central difference method to calculate the rate of change of the airflow velocity. The central difference method takes into account the data of two time points before and after, providing a more accurate estimate of the rate of change.

[0062] Next, thresholds are determined for the temperature and humidity gradient data and the airflow velocity change rate data. The temperature gradient threshold is set to 0.5°C / m, the humidity gradient threshold is set to 2%RH / m, and the airflow velocity change rate threshold is set to 0.1m / s². Data points that exceed these thresholds are marked as requiring adjustment. Based on the marking results, the temperature and humidity adjustment requirements and the airflow adjustment requirements are determined, and the corresponding adjustment parameters are generated.

[0063] Based on the obtained adjustment parameters, fuzzy logic reasoning is applied to calculate the opening adjustment amount of the air inlet and outlet. Fuzzy logic control includes three steps: fuzzification, reasoning and defuzzification. First, the temperature and humidity adjustment parameters and airflow adjustment parameters are converted into fuzzy sets, such as "low", "medium" and "high". Then, reasoning is performed according to the preset fuzzy rules, such as "if the temperature is high and the humidity is low, increase the opening of the air inlet". Finally, the fuzzy output is converted into a specific opening adjustment amount by the centroid method. The obtained preliminary adjustment instructions for the air inlet and outlet are comprehensively analyzed, and the weight coefficients of each adjustment instruction are calculated. The weight calculation takes into account factors such as current environmental conditions, energy consumption targets and equipment limitations. The weight coefficients are applied to the preliminary adjustment instructions to obtain weighted adjustment instructions. In order to achieve a smooth transition, the cubic spline interpolation algorithm is used to generate a continuous adjustment curve to avoid sudden changes causing shocks to the equipment.

[0064] Finally, the smoothing adjustment instructions are fine-tuned based on the current operating status and historical operating data of the natural ventilation air cooling tower. The operating status includes equipment load, energy consumption level, etc., and the historical data provides a reference for the best control strategy under similar environmental conditions. Through this step, the final ventilation parameter adjustment instructions are generated to ensure that the control strategy meets current needs and is supported by historical experience.

[0065] For example, in a typical summer afternoon, the environmental parameter data of a natural ventilation air-cooling tower of a large thermal power plant showed that the temperature inside the tower was 40°C and the relative humidity was 30%, while the temperature outside the tower was 35°C and the relative humidity was 40%. ARIMA model analysis predicts that in the next 2 hours, the temperature inside the tower will rise to 42°C and the humidity will drop to 25%, while the temperature and humidity outside the tower will remain basically unchanged. The calculated temperature gradient is 0.67°C / m (assuming the tower is 75m high), which exceeds the preset threshold. The air flow velocity data shows that the average velocity from the bottom of the tower to the top of the tower increases from 2m / s to 5m / s, with a change rate of 0.04m / s². The fuzzy logic control system judges the temperature gradient as "high", the humidity gradient as "medium", and the air flow velocity change rate as "low". Based on the fuzzy rules, the system deduces that the air inlet opening needs to be increased and the air outlet opening needs to be slightly reduced. The specific calculation results are that the air inlet opening increases by 15% and the air outlet opening decreases by 5%. Taking into account the current high temperature weather conditions and equipment load conditions, the system assigns a weight of 0.7 to the air inlet adjustment command and a weight of 0.3 to the air outlet adjustment command. Through cubic spline interpolation, the system generates a smooth curve that is gradually adjusted within 10 minutes. Combined with historical data analysis, it is found that under similar weather conditions, a slight increase in the adjustment range is more effective. Therefore, the final ventilation parameter adjustment command increases the air inlet opening to 18% and decreases the air outlet opening to 6%, and the adjustment process is completed within 15 minutes. This adjustment strategy takes into account both current environmental needs and historical experience. It is expected to effectively improve the cooling effect of the air cooling tower while avoiding drastic fluctuations in equipment.

[0066] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) parsing the ventilation parameter adjustment instruction, extracting the inlet and outlet opening adjustment amount, obtaining the opening adjustment parameter, and calculating the inlet and outlet actuator movement trajectory according to the opening adjustment parameter to obtain the actuator control instruction; (2) converting the actuator control instruction into a motor drive signal to control the movement of the inlet and outlet actuator to obtain actual opening data, and comparing the actual opening data with the target opening, calculating the opening error, and obtaining the opening correction value; (3) fine-tuning the actuator according to the opening correction value so that the inlet and outlet openings reach the target value, obtaining the calibrated opening data, and recording and storing the calibrated opening data to establish an opening adjustment history database to obtain. to the opening adjustment trend data; (4) installing air flow sensors at the air inlet and outlet to measure the adjusted air flow speed and direction to obtain real-time air flow data, and performing statistical analysis on the real-time air flow data to calculate the average air flow speed and flow rate to obtain comprehensive air flow parameters; (5) performing numerical calculations on the comprehensive air flow parameters to obtain air flow distribution prediction data, and performing comparative analysis on the air flow distribution prediction data and historical data to obtain the air flow change trend, and evaluating the air flow change trend to obtain the air flow adjustment effect evaluation results; (6) analyzing the air flow adjustment effect evaluation results to obtain the adjustment optimization parameters, and modifying the ventilation parameter adjustment instructions according to the adjustment optimization parameters to obtain the updated ventilation parameter adjustment instructions, and executing the updated ventilation parameter adjustment instructions to obtain the adjusted air flow data.

[0067] Specifically, in the control process of the natural ventilation air cooling tower, the analysis and execution of the ventilation parameter adjustment instructions are key links. First, the ventilation parameter adjustment instructions are analyzed to extract the opening adjustment amount of the air inlet and outlet. This process involves instruction decoding and parameter extraction, which can be expressed as: ,in, represents the i-th opening adjustment amount, is the decoding function, is the adjustment instruction. According to the extracted opening adjustment amount, the motion trajectory of the actuator is calculated. The motion trajectory calculation takes into account the kinematic characteristics of the actuator and can be described by the following equation: ,here, is the angle of the actuator at time t, is the initial angle, is the angular velocity function.

[0068] The calculated motion trajectory is converted into a motor drive signal to control the movement of the actuator. The actual opening data is obtained through a high-precision encoder, compared with the target opening, and the opening error is calculated: ,in, is the opening error, is the actual opening, is the target opening. Based on the opening error, the actuator is fine-tuned according to the opening correction value so that the inlet and outlet openings reach the target values, and the calibrated opening data is obtained. The calibrated opening data is recorded and stored in the historical database for analyzing the opening adjustment trend. At the same time, airflow sensors such as hot wire anemometers are installed at the inlet and outlet to measure the adjusted airflow parameters. The real-time airflow data is statistically analyzed, and the average airflow velocity and flow rate are calculated to obtain the comprehensive airflow parameters.

[0069] The comprehensive airflow parameters are numerically calculated and the airflow distribution is predicted using computational fluid dynamics (CFD) methods. CFD calculations are based on the Navier-Stokes equation: ,in, is the fluid density, is the velocity vector, For pressure, is the dynamic viscosity, is the volume force.

[0070] Compare the prediction results with historical data, analyze the airflow change trend, and evaluate the adjustment effect. Based on the evaluation results, calculate the adjustment optimization parameters and correct the ventilation parameter adjustment instructions. Finally, execute the updated adjustment instructions to obtain the adjusted air flow data. For example, the natural ventilation air cooling tower of a large thermal power plant receives an adjustment instruction, requiring the air inlet opening to increase from 60% to 75% and the air outlet opening to decrease from 50% to 45%. After parsing the instruction, it is calculated that the air inlet actuator needs to rotate 30° and the air outlet actuator needs to rotate -10°. According to the S-shaped acceleration and deceleration curve, the motion trajectory of the actuator is generated and converted into a pulse sequence of the stepper motor. After the adjustment is executed, the encoder feedback shows that the actual air inlet opening is 74.2% and the air outlet is 45.5%. After three fine-tunings, the opening error is finally controlled within ±0.1%. After the adjustment is completed, the air flow sensor measures the average wind speed of the air inlet to be 5.2m / s and the air outlet to be 7.8m / s. CFD calculation predicts that a vortex zone with a diameter of about 10m is formed in the tower. Comparing this result with historical data, it was found that the vortex area was 15% smaller than usual, indicating a good regulation effect. Based on this evaluation, a slight correction was made to the ventilation parameter adjustment command, and the air inlet opening was slightly increased by 0.5% to further optimize the air flow distribution. After executing the corrected command, the air inlet flow rate was finally measured to increase by 12% and the air outlet flow rate was reduced by 7%, achieving the expected regulation target.

[0071] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) performing spatial interpolation on the adjusted air flow data to construct a three-dimensional airflow field to obtain an airflow distribution diagram, and identifying the airflow dead zone and high-speed zone in the air-cooling tower based on the airflow distribution diagram to obtain airflow uneven distribution area data; (2) performing numerical analysis on the airflow uneven distribution area data to obtain the angle adjustment amount of the guide device, and parameterizing the angle adjustment amount of the guide device to obtain an initial angle adjustment parameter, and performing actuator input on the initial angle adjustment parameter to obtain a guide plate rotation control instruction, and adjusting the guide plate according to the guide plate rotation control instruction to obtain actual angle adjustment data; (3) comparing the actual angle adjustment data with the initial angle adjustment parameter, calculating the angle adjustment error, obtaining the angle correction value, and executing the guide device according to the angle correction value. The guiding mechanism is fine-tuned to make the guide plate angle reach the target value and obtain the calibrated angle data; (4) Multiple airflow sensors are arranged around the guide device to measure the local airflow velocity and direction after adjustment to obtain local airflow data, and the local airflow data is statistically analyzed to calculate the average airflow velocity and direction to obtain comprehensive airflow parameters; (5) The comprehensive airflow parameters are visualized to generate an airflow streamline diagram to obtain a visualization result of the airflow distribution, and the airflow distribution visualization result is compared with the historical data to analyze the change trend of the airflow distribution and obtain the airflow regulation effect evaluation data; (6) The airflow regulation effect evaluation data is processed by genetic algorithm to obtain the guide device angle optimization parameters, and the guide device angle optimization parameters are converted to obtain the optimization regulation strategy, and the guide device is adjusted according to the optimization regulation strategy to obtain the optimized airflow distribution data.

[0072] Specifically, to achieve optimal control of air flow distribution in a natural ventilation air cooling tower, it is first necessary to spatially interpolate the adjusted air flow data to construct a three-dimensional air flow field. This process uses the Kriging interpolation method, and its basic formula is: ,in, is the value of the point to be estimated, is the value of the known point, is the weight coefficient. Through this method, a continuous airflow distribution map is obtained, and the threshold method is used to identify the airflow dead zone (speed less than 0.5m / s) and the high-speed zone (speed greater than 8m / s), thereby obtaining the data of the uneven airflow distribution area. The data of the uneven airflow distribution area is numerically analyzed, and the gradient descent method is used to calculate the angle adjustment of the guide device. The iterative formula is: ,here, It is The angle of the iteration, is the learning rate, is the gradient of the objective function with respect to the angle. The obtained angle adjustment is parameterized and converted into the control instruction of the actuator.

[0073] After the adjustment is performed, the actual angle adjustment data is obtained through the high-precision encoder, compared with the initial angle adjustment parameters, and the angle adjustment error is calculated. The PID control algorithm is used for fine-tuning, and its control law is: ,in, It is the control output, representing the control signal to the actuator of the guide device, which is used to adjust the angle of the guide plate. is the error, which is the difference between the target angle and the actual angle. In this scenario, it represents the difference between the current angle of the deflector and the desired angle. is the proportional coefficient, which is used to adjust the control system’s response to the current error. is the integral coefficient, which is used to eliminate the steady-state error of the system. is the differential coefficient, which is used to predict the changing trend of the error. Airflow sensors such as hot wire anemometers are arranged around the guide device to measure the local airflow speed and direction. The collected local airflow data are statistically analyzed, the average airflow speed and direction are calculated, and the comprehensive airflow parameters are obtained. The streamline method is used to visualize the comprehensive airflow parameters and generate an airflow streamline diagram. The equation of the streamline can be expressed as: , where (u, v, w) are the three components of the velocity vector. The airflow distribution visualization results are compared with historical data, the airflow distribution change trend is analyzed, and the airflow regulation effect evaluation data is obtained. The genetic algorithm is applied to these evaluation data to optimize the angle of the guide device. The fitness function of the genetic algorithm is defined as the airflow uniformity index. Through iterative optimization such as crossover and mutation operations, the optimization parameters of the guide device angle are obtained.

[0074] Finally, the optimization parameters are converted into specific adjustment strategies, and the adjustments are performed to obtain the optimized airflow distribution data.

[0075] For example, a large natural ventilation air-cooling tower is 150 meters high, with a bottom diameter of 100 meters and is equipped with 50 sets of adjustable guide plates. The initial airflow data showed that there was a low-speed area of ​​about 20 square meters (average speed 0.3m / s) and a high-speed area of ​​15 square meters (average speed 9.5m / s) in the middle of the tower. The three-dimensional airflow field was constructed by Kriging interpolation, and these two unevenly distributed areas were identified. The gradient descent method was used to analyze these areas, and it was calculated that the guide plates near the low-speed area needed to be adjusted inward by 5°, and the guide plates near the high-speed area needed to be adjusted outward by 3°. After the adjustment was performed, the encoder feedback showed that the actual adjustment angles were 4.8° and 3.2°, respectively. Through PID control, the setting =0.8, =0.1, =0.05, after three iterations, the angle error was controlled within ±0.1°. After adjustment, the 20 hot wire anemometers arranged around the guide plate measured an average airflow velocity of 5.8m / s, and the directional deviation was within the range of ±5°. The generated streamline diagram shows that the area of ​​the low-speed zone is reduced to 5 square meters and the high-speed zone disappears. Compared with historical data, the airflow uniformity is improved by 25%. During the genetic algorithm optimization process, the population size was set to 100, 500 generations were iterated, the crossover probability was 0.8, and the mutation probability was 0.1. The final optimization scheme recommends fine-tuning some guide plates by 1-2°. After executing this optimization scheme, the standard deviation of the airflow velocity in the entire tower body was finally measured to be reduced to 0.5m / s, achieving the optimization goal.

[0076] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) analyzing the optimized air flow distribution data, calculating the air flow velocity field and pressure field inside the air cooling tower, obtaining the dynamic characteristic data of the air flow inside the tower, and estimating the required exhaust volume based on the dynamic characteristic data of the air flow inside the tower to obtain a preliminary exhaust volume estimation value; (2) comparing the preliminary exhaust volume estimation value with the current exhaust equipment operating parameters, calculating the exhaust volume difference, obtaining the exhaust equipment speed regulation requirement, and generating a speed regulation instruction based on the exhaust equipment speed regulation requirement to obtain the exhaust equipment initial speed data; (3) smoothing the exhaust equipment initial speed data to generate a continuous speed change curve to obtain a smooth speed regulation instruction, and sending the smooth speed regulation instruction to the exhaust equipment frequency conversion controller to adjust the motor speed to obtain the actual exhaust equipment operating data; (4 ) Install an air flow sensor at the outlet of the exhaust equipment to measure the actual exhaust speed and flow rate to obtain real-time exhaust data, and perform statistical analysis on the real-time exhaust data to calculate the average exhaust speed and flow rate to obtain comprehensive exhaust parameters; (5) Compare the comprehensive exhaust parameters with the target exhaust volume to obtain exhaust volume deviation data, and analyze and process the exhaust volume deviation data to obtain an exhaust volume correction value, and adjust the exhaust equipment speed according to the exhaust volume correction value to obtain adjusted speed data, and perform execution control on the adjusted speed data to obtain calibrated exhaust volume data; (6) Record and store the calibrated exhaust volume data to obtain exhaust volume adjustment history data, and perform trend analysis on the exhaust volume adjustment history data to obtain the exhaust volume change trend, and perform predictive calculation based on the exhaust volume change trend to obtain the target exhaust volume data.

[0077] Specifically, to achieve precise control of the exhaust volume of the natural ventilation air cooling tower, it is necessary to first conduct an in-depth analysis of the optimized air flow distribution data. The computational fluid dynamics (CFD) method is used to calculate the air flow velocity field and pressure field inside the air cooling tower. This process is based on the Navier-Stokes equation: ,in, is the velocity vector, is the fluid density, For pressure, is the kinematic viscosity, is the acceleration of gravity. By solving this equation, the dynamic characteristic data of the airflow in the tower is obtained. Based on the obtained dynamic characteristic data, the required exhaust volume is estimated using the principle of mass conservation. The mass conservation equation can be expressed as: , integrate this equation over the entire air cooling tower volume to obtain a preliminary exhaust volume estimate. Compare the preliminary exhaust volume estimate with the current exhaust equipment operating parameters, calculate the exhaust volume difference, and obtain the exhaust equipment speed regulation requirement. According to the fan characteristic curve, convert the exhaust volume difference into a speed regulation amount, generate a speed regulation instruction, and obtain the exhaust equipment initial speed data.

[0078] The initial speed data is smoothed and the continuous speed curve is generated by cubic spline interpolation. The cubic spline function S(x) satisfies: ,in, , , , is the coefficient to be determined, is a node. By solving the coefficient, the smooth speed adjustment command is obtained. The smooth speed adjustment command is sent to the frequency converter to adjust the motor speed. An air flow sensor such as a hot wire anemometer is installed at the outlet of the exhaust equipment to measure the actual exhaust speed and flow rate to obtain real-time exhaust data. The data is statistically analyzed to calculate the average exhaust speed and flow rate to obtain comprehensive exhaust parameters.

[0079] The comprehensive exhaust parameters are compared with the target exhaust volume, and the exhaust volume deviation is calculated. The proportional-integral-differential (PID) control algorithm is used to process the deviation data to obtain the exhaust volume correction value. The exhaust equipment speed is adjusted according to the obtained correction value, and the calibrated exhaust volume data is obtained after the control is executed. These data are recorded and stored to form a historical database for exhaust volume adjustment. Trend analysis is performed on historical data, and time series analysis methods such as the autoregressive integrated moving average (ARIMA) model are used to predict future exhaust volume change trends, and finally the target exhaust volume data is obtained.

[0080] For example, the natural ventilation air cooling tower of a large thermal power plant is 150 meters high and 80 meters in diameter at the top. CFD analysis shows that a large-scale vortex structure is formed in the tower, with a maximum velocity field of 8m / s and a minimum of 0.5m / s. Through mass conservation calculation, the required exhaust volume is estimated to be 500,000 m³ / h. The current exhaust equipment operating parameters show that the actual exhaust volume is 450,000 m³ / h, and the calculated exhaust volume difference is 50,000 m³ / h.

[0081] According to the fan characteristic curve, this difference is converted into a speed adjustment amount, and the fan speed needs to be increased from the current 600 rpm to 660 rpm. The cubic spline interpolation method is used to generate a smooth speed change curve, and the speed adjustment process from 600 rpm to 660 rpm is decomposed into 100 discrete points, each corresponding to a speed value. After executing the speed adjustment command, the hot wire anemometer installed at the exhaust port measured an average wind speed of 15.5 m / s. Considering that the exhaust port area is 10 m², the actual exhaust volume is calculated to be 558,000 m³ / h. Compared with the target value of 500,000 m³ / h, the deviation is 58,000 m³ / h. The wind speed reduction parameter analysis is performed, and it is calculated that the fan speed needs to be reduced by 20 rpm. After performing this adjustment, the exhaust volume is finally measured to be stable at 505,000 m³ / h, and the error is controlled within 1%. The data of this process is recorded in the historical database and analyzed by time series using the ARIMA (2,1,1) model. The model predicts that in the next 4 hours, the exhaust air demand will increase by about 2% due to the gradual increase in ambient temperature. Based on this prediction, the target exhaust air volume for the next adjustment cycle is set at 515,000 m³ / h, providing forward-looking guidance for the next round of control.

[0082] The embodiment of the present invention also provides a ventilation control device for a natural ventilation air cooling tower, such as Figure 3 As shown, the ventilation control device of a natural ventilation air cooling tower specifically includes: an acquisition module 301, which is used to perform multi-point acquisition and analysis of the internal and external air parameters of the natural ventilation air cooling tower to obtain spatial distribution characteristic data; a processing module 302, which is used to screen and normalize the spatial distribution characteristic data to obtain environmental parameter data; a calculation module 303, which is used to calculate the temperature and humidity difference and the air flow velocity change rate based on the environmental parameter data to obtain a ventilation parameter adjustment instruction; an adjustment module 304, which is used to adjust the opening of the air inlet and the air outlet of the natural ventilation air cooling tower according to the ventilation parameter adjustment instruction to obtain adjusted air flow data; an adjustment module 305, which is used to adjust the angle of the internal guide device of the natural ventilation air cooling tower in real time based on the adjusted air flow data to obtain optimized air flow distribution data; a control module 306, which is used to dynamically control the rotation speed of the exhaust equipment on the top of the natural ventilation air cooling tower according to the optimized air flow distribution data to obtain target exhaust volume data.

[0083] Through the collaborative work of the above modules, the air parameters inside and outside the air cooling tower are collected and analyzed at multiple points to obtain spatial distribution characteristic data, which realizes the comprehensive and accurate perception of the environment inside and outside the air cooling tower, and provides a reliable data basis for subsequent control. This multi-point collection method overcomes the limitations of traditional single-point measurement, can capture the unevenness and local characteristics of spatial distribution, and greatly improves the representativeness and accuracy of the data. Secondly, the collected data is screened and normalized to obtain standardized environmental parameter data, effectively removing outliers and noise, improving data quality, and making different types of data comparable through normalization, creating conditions for subsequent comprehensive analysis. Based on the standardized environmental parameter data, the temperature and humidity difference and the air flow velocity change rate are calculated to obtain the ventilation parameter adjustment instructions. This control strategy based on multi-parameter collaboration can more comprehensively reflect the operating status of the air cooling tower. Compared with traditional single parameter control, it can better adapt to complex and changing environmental conditions. In addition, by precisely adjusting the opening of the air inlet and outlet, combined with the real-time adjustment of the internal guide device, the fine control of the air flow distribution is achieved, which effectively solves the local dead zone and high-speed zone problems existing in the traditional control method, improves the cooling efficiency and reduces energy consumption. In particular, the angle optimization method of the guide device based on genetic algorithm is introduced, which can find a solution close to the global optimal in a complex nonlinear environment, greatly improving the intelligent level of control. In terms of exhaust volume control, a closed-loop adjustment strategy based on CFD analysis and PID control is adopted, which can not only accurately control the exhaust volume, but also make forward-looking adjustments based on historical data and prediction models, improving the response speed and adaptability of the system. In the entire control process, a large number of data-driven methods are used, such as spatial interpolation, time series analysis, trend prediction, etc., to make the control decision more scientific and reliable. At the same time, by establishing a historical database and continuous data analysis, it has the ability of self-learning and self-optimization, which can continuously improve the control effect. Through multi-parameter collaborative control and refined airflow adjustment, the cooling efficiency and energy utilization rate of the natural ventilation air cooling tower are significantly improved.

[0084] The above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the embodiments, a person skilled in the art should understand that the specific implementation modes of the present invention can still be modified or replaced by equivalents, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A ventilation control method for a natural ventilation air cooling tower, characterized in that: include: Collect and analyze the internal and external air parameters of the natural ventilation air cooling tower at multiple points to obtain spatial distribution characteristic data; The spatial distribution characteristic data are screened and normalized to obtain environmental parameter data; based on the environmental parameter data, the temperature and humidity difference and the air flow velocity change rate are calculated to obtain ventilation parameter adjustment instructions; according to the ventilation parameter adjustment instructions, the openings of the air inlet and the air outlet of the natural ventilation air cooling tower are adjusted to obtain adjusted air flow data; based on the adjusted air flow data, the angle of the internal flow guide device of the natural ventilation air cooling tower is adjusted in real time to obtain optimized air flow distribution data; according to the optimized air flow distribution data, the rotation speed of the exhaust equipment at the top of the natural ventilation air cooling tower is dynamically controlled to obtain target exhaust volume data.

2. The ventilation control method for a natural ventilation air cooling tower according to claim 1, characterized in that: The steps of collecting and analyzing the internal and external air parameters of the natural ventilation air cooling tower at multiple points to obtain spatial distribution characteristic data include: setting multiple temperature sensors at the air inlet, air outlet and around the tower body of the natural ventilation air cooling tower to obtain multi-point temperature data, and setting multiple humidity sensors at the air inlet, air outlet and around the tower body of the natural ventilation air cooling tower to obtain multi-point humidity data; setting multiple wind speed sensors at the air inlet, air outlet and around the tower body of the natural ventilation air cooling tower to obtain multi-point air flow velocity data, and performing time series sampling on the multi-point temperature data to obtain temperature change trend data; performing time series sampling on the multi-point humidity data to obtain humidity change trend data, and performing time series sampling on the multi-point air flow velocity data to obtain air flow velocity change trend data; calculating the temperature and humidity gradient and the air flow velocity gradient according to the temperature change trend data, the humidity change trend data and the air flow velocity change trend data to obtain spatial distribution characteristic data.

3. The ventilation control method for a natural ventilation air cooling tower according to claim 2, characterized in that: The step of screening and normalizing the spatial distribution characteristic data to obtain environmental parameter data includes: performing outlier detection on the multi-point temperature data, multi-point humidity data and multi-point airflow velocity data, eliminating data points that exceed a preset range, obtaining preliminary screened environmental data, and performing moving average processing on the preliminary screened environmental data to obtain smoothed environmental data; performing maximum and minimum value normalization processing on the temperature data in the smoothed environmental data to obtain normalized temperature data, and performing maximum and minimum value normalization processing on the humidity data in the smoothed environmental data to obtain normalized humidity data; performing Z-score normalization processing on the airflow velocity data in the smoothed environmental data to obtain standardized airflow velocity data, and performing moving average processing on the preliminary screened environmental data to obtain smoothed environmental data. Normalized humidity data and standardized air flow velocity data are used to calculate the weight coefficient of each parameter to obtain weighted environmental parameter data; principal component analysis is performed on the weighted environmental parameter data to extract main features to obtain environmental characteristic data after dimensionality reduction, and time series decomposition is performed on the environmental characteristic data after dimensionality reduction to separate trend, seasonality and residual components to obtain multi-scale environmental characteristic data; a correlation matrix of each parameter is calculated based on the multi-scale environmental characteristic data to obtain parameter correlation data, and cluster analysis is performed on environmental characteristics based on the parameter correlation data to obtain an environmental state classification result; piecewise linear interpolation is performed on each environmental parameter based on the environmental state classification result to obtain a continuous environmental parameter curve, and the continuous environmental parameter curve is subjected to wavelet transformation to obtain the environmental parameter data.

4. The ventilation control method for a natural ventilation air cooling tower according to claim 1, characterized in that: The step of calculating the temperature and humidity difference and the air flow velocity change rate based on the environmental parameter data to obtain the ventilation parameter adjustment instruction comprises: performing a time series analysis on the temperature data in the environmental parameter data, calculating the temperature change trend, and obtaining the temperature trend data, and performing a time series analysis on the humidity data in the environmental parameter data, calculating the humidity change trend, and obtaining the humidity trend data; calculating the temperature and humidity difference inside and outside the tower according to the temperature trend data and the humidity trend data, and obtaining the temperature and humidity gradient data, and performing a differential operation on the air flow velocity data in the environmental parameter data, and obtaining the air flow velocity change rate data; performing a threshold judgment on the temperature and humidity gradient data, determining the temperature and humidity adjustment requirements, and obtaining the temperature and humidity adjustment parameters, and performing a differential operation on the air flow velocity data in the environmental parameter data, and obtaining the air flow velocity change rate data. Threshold judgment is performed to determine the airflow adjustment demand and obtain the airflow adjustment parameters; according to the temperature and humidity adjustment parameters and the airflow adjustment parameters, the air inlet opening adjustment amount is calculated through fuzzy logic reasoning to obtain the preliminary air inlet adjustment instruction, and according to the temperature and humidity adjustment parameters and the airflow adjustment parameters, the air outlet opening adjustment amount is calculated through fuzzy logic reasoning to obtain the preliminary air outlet adjustment instruction; the preliminary air inlet adjustment instruction and the preliminary air outlet adjustment instruction are comprehensively analyzed to calculate the weight coefficient of each adjustment instruction to obtain the weighted adjustment instruction, and according to the weighted adjustment instruction, a continuous adjustment curve is generated through an interpolation algorithm to obtain a smoothing adjustment instruction; according to the smoothing adjustment instruction, the ventilation parameter adjustment instruction is generated through the operating status and historical data of the natural ventilation air cooling tower.

5. The ventilation control method for a natural ventilation air cooling tower according to claim 1, characterized in that: The steps of adjusting the opening of the air inlet and the air outlet of the natural ventilation air cooling tower according to the ventilation parameter adjustment instruction to obtain the adjusted air flow data include: parsing the ventilation parameter adjustment instruction, extracting the air inlet and air outlet opening adjustment amount, obtaining the opening adjustment parameter, and calculating the motion trajectory of the air inlet and air outlet actuator according to the opening adjustment parameter to obtain the actuator control instruction; converting the actuator control instruction into a motor drive signal, controlling the movement of the air inlet and air outlet actuator to obtain the actual opening data, and comparing the actual opening data with the target opening, calculating the opening error, and obtaining the opening correction value; fine-tuning the actuator according to the opening correction value to make the air inlet and air outlet opening reach the target value, obtaining the calibrated opening data, and recording and storing the calibrated opening data to establish the opening adjustment Historical database, to obtain opening adjustment trend data; install air flow sensors at the air inlet and outlet, measure the adjusted air flow speed and direction, obtain real-time air flow data, and perform statistical analysis on the real-time air flow data, calculate the average air flow speed and flow, and obtain comprehensive air flow parameters; perform numerical calculation on the comprehensive air flow parameters, obtain air flow distribution prediction data, and compare and analyze the air flow distribution prediction data with historical data to obtain the air flow change trend, and evaluate the air flow change trend to obtain the air flow adjustment effect evaluation result; analyze the air flow adjustment effect evaluation result to obtain the adjustment optimization parameters, and modify the ventilation parameter adjustment instruction according to the adjustment optimization parameters to obtain the updated ventilation parameter adjustment instruction, and execute the updated ventilation parameter adjustment instruction to obtain the adjusted air flow data.

6. The ventilation control method for a natural ventilation air cooling tower according to claim 1, characterized in that: The method comprises the steps of adjusting the angle of the internal guide device of the natural ventilation air cooling tower in real time based on the adjusted air flow data to obtain the optimized air flow distribution data, including: performing spatial interpolation on the adjusted air flow data to construct a three-dimensional air flow field to obtain an air flow distribution diagram, and identifying the air flow dead zone and high-speed zone in the air cooling tower according to the air flow distribution diagram to obtain the data of the uneven air flow distribution area; performing numerical analysis on the data of the uneven air flow distribution area to obtain the angle adjustment amount of the guide device, and parameterizing the angle adjustment amount of the guide device to obtain the initial angle adjustment parameter, and performing actuator input on the initial angle adjustment parameter to obtain the guide plate rotation control instruction, and adjusting the guide plate according to the guide plate rotation control instruction to obtain the actual angle adjustment data; comparing the actual angle adjustment data with the initial angle adjustment parameter, calculating the angle adjustment error, obtaining the angle correction value, and calculating the angle adjustment error based on the actual angle adjustment data. The actuator of the guide device is fine-tuned according to the angle correction value so that the guide plate angle reaches the target value and the calibrated angle data is obtained; multiple airflow sensors are arranged around the guide device to measure the adjusted local airflow velocity and direction to obtain local airflow data, and the local airflow data is statistically analyzed to calculate the average airflow velocity and flow direction to obtain comprehensive airflow parameters; the comprehensive airflow parameters are visualized to generate an airflow streamline diagram to obtain a visualization result of the airflow distribution, and the airflow distribution visualization result is compared with historical data to analyze the change trend of the airflow distribution to obtain airflow regulation effect evaluation data; the airflow regulation effect evaluation data is processed by genetic algorithm to obtain the guide device angle optimization parameters, and the guide device angle optimization parameters are converted into a strategy to obtain an optimization regulation strategy, and the guide device is adjusted according to the optimization regulation strategy to obtain optimized airflow distribution data.

7. The ventilation control method for a natural ventilation air cooling tower according to claim 6, characterized in that: The steps of dynamically controlling the rotation speed of the exhaust device at the top of the natural ventilation air-cooling tower according to the optimized airflow distribution data to obtain the target exhaust volume data include: analyzing the optimized airflow distribution data, calculating the airflow velocity field and pressure field inside the air-cooling tower, obtaining the dynamic characteristic data of the airflow in the tower, and estimating the required exhaust volume according to the dynamic characteristic data of the airflow in the tower to obtain a preliminary exhaust volume estimation value; comparing the preliminary exhaust volume estimation value with the current exhaust equipment operating parameters, calculating the exhaust volume difference, obtaining the exhaust equipment speed regulation demand, and generating a speed regulation instruction according to the exhaust equipment speed regulation demand to obtain the exhaust equipment initial speed data; smoothing the exhaust equipment initial speed data to generate a continuous speed change curve to obtain a smooth speed regulation instruction, and sending the smooth speed regulation instruction to the exhaust equipment frequency conversion controller to adjust the motor speed to obtain the actual exhaust equipment operating speed. data; an airflow sensor is installed at the outlet of the exhaust equipment to measure the actual exhaust speed and flow rate to obtain real-time exhaust data, and the real-time exhaust data is statistically analyzed to calculate the average exhaust speed and flow rate to obtain comprehensive exhaust parameters; the comprehensive exhaust parameters are compared and calculated with the target exhaust volume to obtain exhaust volume deviation data, and the exhaust volume deviation data is analyzed and processed to obtain an exhaust volume correction value, and the exhaust equipment speed is adjusted according to the exhaust volume correction value to obtain adjusted speed data, and the adjusted speed data is executed and controlled to obtain calibrated exhaust volume data; the calibrated exhaust volume data is recorded and stored to obtain exhaust volume adjustment history data, and trend analysis is performed on the exhaust volume adjustment history data to obtain the exhaust volume change trend, and a prediction calculation is performed according to the exhaust volume change trend to obtain the target exhaust volume data.

8. A ventilation control device for a natural draft air cooling tower, used to execute the ventilation control method for a natural draft air cooling tower according to any one of claims 1 to 7, characterized in that: include: The acquisition module is used to collect and analyze the internal and external air parameters of the natural ventilation air cooling tower at multiple points to obtain spatial distribution characteristic data; A processing module, used for screening and normalizing the spatial distribution characteristic data to obtain environmental parameter data; A calculation module is used to calculate the temperature and humidity difference and the air flow velocity change rate based on the environmental parameter data to obtain ventilation parameter adjustment instructions; an adjustment module is used to adjust the opening of the air inlet and outlet of the natural ventilation air cooling tower according to the ventilation parameter adjustment instructions to obtain adjusted air flow data; an adjustment module is used to adjust the angle of the internal guide device of the natural ventilation air cooling tower in real time based on the adjusted air flow data to obtain optimized air flow distribution data; a control module is used to dynamically control the rotation speed of the exhaust equipment at the top of the natural ventilation air cooling tower according to the optimized air flow distribution data to obtain target exhaust volume data.

Citation Information

Patent Citations

  • Intelligent anti-freezing control system of wet cooling tower

    CN115493424A

  • Assembly type angle-adjustable cooling tower air guide device

    CN116294767A

  • Intelligent adjusting system for environmental adaptability of water-cooling machine

    CN118517777A

  • Energy-saving management method and system based on smart building

    CN118705740A

  • Temperature self-adaptive adjusting method and system for warmer

    CN118729379A

Cited By

  • Fuel cell air compressor flow self-adaptive control method and system

    CN119712519A

  • Operating parameter optimization control method and system of air condenser

    CN119860679A

  • An operating parameter optimization control method and system for an air condenser

    CN119860679B

  • Control system and method for gas purification

    CN120085556A

  • Anti-explosion pressure relief ball valve

    CN120292311A