A method for intelligent air curtain of elevator air conditioner

Abstract

The present application provides an intelligent air curtain method for an elevator air conditioner, comprising: installing a personnel number sensor in a car to obtain personnel data in the car in real time, transmitting the obtained personnel number data to a central control unit, and at the same time, installing temperature and humidity sensors inside and outside the car to obtain temperature and humidity parameters inside and outside the car, and transmitting the obtained temperature and humidity data to the central control unit; installing an external wind pressure sensor to monitor the wind pressure data in real time, analyzing the law of wind pressure change with the height data by combining the height data obtained by the height sensor, and dynamically adjusting the cooling power of the air conditioner in the elevator and the operating power of the air curtain according to this law; setting an opening and closing door sensor in the elevator door machine system to monitor the opening and closing status of the elevator door in real time, transmitting the door status signal to the central control unit, and the central control unit controls the air curtain system to increase the air output at the moment the door is opened according to the door status signal, and reduces the air output after the door is closed.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to an intelligent air curtain method for an elevator air conditioner. Background Art

[0002] During the operation of the elevator, the number of people in the car and the external environmental parameters are constantly changing, which brings challenges to the coordinated control of intelligent air conditioning and air curtains. First of all, the distribution of the number of people in the car is constantly changing, and the distribution of people in the elevator space is not uniform, which also leads to uneven heat generation in each area of ​​the elevator. Taking this as a problem, how to solve the timely air supply and cooling of hot spots in the elevator is also a key issue. At the same time, as the operating height increases, the temperature outside the car gradually decreases, while the temperature inside the car gradually increases due to the breathing and body temperature of the people. The change in the temperature difference between the inside and outside requires the air conditioning system to respond quickly and dynamically adjust. Furthermore, as the height changes, the car wind pressure will also change, and dynamic adjustment according to the wind pressure is also a necessary link. In addition, when the elevator starts and stops frequently, the sudden change of airflow caused by the door switch also affects the stability of the air curtain, and the wind curtain and the door machine control system need to be accurately synchronized. In summary, to achieve intelligent coordinated control, it is necessary to comprehensively consider the dynamic changes of environmental parameters and operating conditions, the scientific nature of airflow organization, and the real-time and robustness of the control system, which is a technical problem that needs to be solved in this field. Summary of the invention

[0003] The present invention provides an elevator air conditioning intelligent air curtain method, which mainly includes:

[0004] By installing a passenger number sensor in the car, the passenger data in the car is obtained in real time, and the obtained passenger number data is transmitted to the central control unit. At the same time, by installing temperature and humidity sensors inside and outside the car, the temperature and humidity parameters inside and outside the car are obtained, and the obtained temperature and humidity data are transmitted to the central control unit;

[0005] According to the real-time data of the number of people obtained in the central control unit, the distribution of people is analyzed, the elevator is divided into areas, and the density of people in each area is obtained. According to the data of the temperature and humidity sensors, the temperature data of each area is analyzed. The temperature data of the target area and the density of people distribution are combined to calculate the required air supply and cooling capacity of the target area, adjust the air conditioning cooling and air curtain air supply, and continuously judge whether the target area reaches the required air supply and temperature;

[0006] A height sensor is installed in the elevator car to obtain the elevator running height data in real time, and the obtained height data is transmitted to the central control unit. The central control unit designs the height-temperature correspondence relationship based on the height data and the internal and external temperature data obtained by the temperature and humidity sensors installed inside and outside, obtains the internal and external temperature parameters corresponding to each floor height, and dynamically adjusts the cooling power of the air-conditioning system according to the corresponding relationship;

[0007] By installing an external wind pressure sensor, the wind pressure data is monitored in real time. By combining the height data obtained by the height sensor, the law of wind pressure changes with height data is analyzed, and the cooling power of the air conditioner in the elevator and the operating power of the air curtain are dynamically adjusted according to this law;

[0008] A door opening and closing sensor is set in the elevator door machine system to monitor the opening and closing status of the elevator door in real time and transmit the door status signal to the central control unit. The central control unit controls the air curtain system to increase the air output at the moment the door is opened and reduce the air output after the door is closed according to the door status signal.

[0009] The central control unit calculates the target control quantity of the air conditioning system and the air curtain system according to the number of people in the car, the temperature difference between the inside and outside, the height of the car, the wind pressure data, and the door status. The target control quantity includes the air conditioning cooling power, the air curtain air outlet speed and the air outlet angle, and sends the control instructions to the air conditioning and air curtain actuators respectively to realize the coordinated control of the air conditioning system and the air curtain system;

[0010] The central control unit studies and trains historical operating data, establishes a mapping relationship between the number of people, temperature and humidity, height, wind pressure, door status parameters and the air conditioning cooling capacity and wind curtain wind speed control quantity, and forms an intelligent control strategy. In actual operation, based on the current parameter input, the trained model is used to predict the target control quantity and adjust the working status of the air conditioner and wind curtain in real time.

[0011] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0012] The present invention discloses an intelligent air curtain method for elevator air conditioning. By installing a variety of sensors inside and outside the car, data such as the number of people, temperature and humidity, height, wind pressure and door status are collected in real time, and these data are transmitted to the central control unit. The central control unit comprehensively analyzes these data and calculates the optimal control parameters of the air conditioning system and the air curtain system, including cooling power, air supply volume, wind speed and wind direction. The present invention also uses machine learning technology to train historical operation data, establish a mapping relationship between each parameter and the control quantity, and form an intelligent control strategy. In actual operation, the optimal control quantity is predicted according to the current input parameters, and the working state of the air conditioner and the air curtain is adjusted in real time. This method can dynamically optimize the air conditioning effect, improve energy utilization efficiency, and improve passenger comfort according to the actual situation of elevator operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The present invention is a flow chart of an elevator air conditioning intelligent air curtain method.

[0014] Figure 2 It is a schematic diagram of an elevator air conditioning intelligent air curtain method of the present invention.

[0015] Figure 3 It is another schematic diagram of an elevator air conditioning intelligent air curtain method of the present invention. DETAILED DESCRIPTION

[0016] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings and embodiments. The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant inventions, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0017] like Figure 1-3 In this embodiment, an elevator air conditioning intelligent air curtain method may specifically include:

[0018] Step S101, by installing a personnel number sensor in the car, real-time personnel data in the car is obtained, and the obtained personnel number data is transmitted to the central control unit; at the same time, by installing temperature and humidity sensors inside and outside the car, the temperature and humidity parameters inside and outside the car are obtained, and the obtained temperature and humidity data are transmitted to the central control unit.

[0019] The real-time data collected by the sensor for the number of people in the car is obtained, and the real-time data is filtered and denoised by digital signal processing to obtain the processed data for the number of people; the processed data for the number of people is transmitted to the data receiving module of the central control unit through the CAN bus, and the data receiving module of the central control unit parses the received data to obtain the numerical value of the number of people; the data processing module of the central control unit compares the numerical value of the number of people with the preset threshold of the number of people capacity to determine whether the car is overloaded; if the car is overloaded, the alarm module is triggered to issue an alarm prompt, and the programmable logic controller is used to control the elevator to stop running and keep the door open; the car The real-time data collected by the temperature and humidity sensors inside and outside the compartment are smoothed by Kalman filtering to obtain processed temperature and humidity data; the processed temperature and humidity data are transmitted to the data receiving module of the central control unit through the RS485 bus, and the data receiving module of the central control unit parses the received data to obtain temperature and humidity values; the data processing module of the central control unit performs difference calculation on the temperature and humidity values, and compares the difference calculation result with the preset temperature and humidity difference threshold to determine whether to start air conditioning adjustment; if air conditioning adjustment needs to be started, fuzzy control is used to determine the air conditioning operating parameters, and the air conditioning operating parameters include wind speed and temperature setting values.

[0020] Specifically, according to the preset sampling frequency, real-time data is obtained regularly through the number of people sensor in the car, and the original data is filtered and denoised by digital signal processing. Then, the processed data is transmitted to the data receiving module of the central control unit through the CAN bus, and the data is analyzed to obtain the number of people. Using the preset sampling frequency, real-time data is obtained from the temperature and humidity sensors inside and outside the car, and the data is smoothed by Kalman filtering. The processed data is transmitted to the data receiving module of the central control unit through the RS485 bus, and the temperature and humidity values ​​are analyzed. According to the data processing module of the central control unit, the analyzed number of people is compared with the preset personnel capacity threshold to determine whether the car is overloaded. If the car is overloaded, the alarm module is triggered to issue an alarm prompt, and the programmable logic controller is used to control the elevator to stop running and keep the door open until the number of people drops to a safe range. The central control unit data processing module is used to calculate the difference between the analyzed temperature and humidity values ​​inside and outside the car, and the calculation result is compared with the preset temperature and humidity difference threshold to determine whether to start air conditioning adjustment. If adjustment is required, fuzzy control is used to determine the air conditioning operation parameters, including wind speed and temperature setting value, and control instructions are sent to the air conditioner through serial communication. The sensor for the number of people in the car uses an infrared array detector, which collects data every 0.5 seconds. The raw data passes through a bandpass filter with a cutoff frequency of 0.1Hz to 10Hz to filter out high-frequency noise and low-frequency drift. The filtered data is transmitted to the central control unit via the CAN bus at a rate of 500kbps. The data receiving module uses a 16-bit ADC for analog-to-digital conversion with a resolution of 0.01 person / bit. The temperature and humidity sensor uses the DHT22 model with an accuracy of ±0.5℃ and ±2%RH, and is sampled every 2 seconds. The collected temperature and humidity data is Kalman filtered, and the processed data is transmitted to the central control unit via the RS485 bus at a rate of 9600bps. The central control unit uses a 32-bit ARM processor with a main frequency of 200MHz and runs a real-time operating system. The data processing module compares the parsed number of people with the preset threshold of 8 people. If overload is detected, an 85-decibel sound and light alarm is immediately triggered, and the elevator is controlled to stop running and keep the door open through the Siemens S7-200 series PLC. The temperature and humidity difference is calculated by simple subtraction, with the preset temperature difference threshold of 5°C and humidity difference threshold of 10%RH. If the difference exceeds the threshold, the fuzzy control algorithm is started. The input variables of the fuzzy controller are temperature difference and humidity difference, and the output variables are air conditioning wind speed (range 0-5 gears) and temperature setting value (range 18-30°C). The fuzzy rule base contains 25 I F-THEN rules. The control instructions are sent to the air conditioning controller at a rate of 115200bps through the RS232 interface to achieve precise adjustment of the car environment. The response time of the entire system is less than 1 second, and the reliability reaches 99.99%.

[0021] Step S102, according to the personnel quantity data obtained in real time in the central control unit, analyze the personnel distribution, divide the elevator into areas, obtain the personnel density in each area, analyze the temperature data of each area according to the temperature and humidity sensor data, comprehensively analyze the temperature data of the target area and the personnel distribution density, calculate the required air supply and cooling capacity of the target area, adjust the air conditioning cooling and air curtain air supply, and continuously judge whether the target area reaches the required air supply and temperature.

[0022] The K-means clustering algorithm is used to analyze the distribution of people in the elevator, and the elevator space is evenly divided into nine sub-areas to obtain the population density value of each sub-area; according to the population density value, the temperature data of each sub-area is obtained from the temperature and humidity sensor, the temperature weighted value is calculated, and the area with the highest temperature weighted value is selected as the target area; according to the temperature weighted value and the population density of the target area, the cooling capacity required for the target area is calculated using the heat balance equation, wherein the heat balance equation is Q=mcΔT, m is the air mass, c is the specific heat capacity, and ΔT is the temperature difference; the Mamdani fuzzy controller is used to generate the adjustment instructions for air conditioning cooling and air curtain air supply, the input variables are temperature deviation and population density, and the output variables are cooling capacity and air supply; the incremental PID controller is used to adjust the air conditioning cooling capacity and air curtain air supply in real time, and the control quantity u(t) is calculated according to the temperature deviation e(t), and the control quantity u(t)=Kp[e(t)-e(t-1)]+Ki e(t)+Kd[e(t)-2e(t-1)+e(t-2)], where Kp, Ki, Kd are proportional, integral, and differential coefficients; if the temperature deviation between the actual temperature of the target area and the set temperature exceeds a preset threshold, the calculation and adjustment process of the cooling capacity and air supply volume is re-executed.

[0023] Specifically, according to the real-time personnel quantity data obtained by the central control unit, the K-means clustering algorithm is used to analyze the personnel distribution in the elevator. By evenly dividing the elevator space into 9 sub-areas of 3x3, the personnel density value of each sub-area is calculated. The temperature data of each sub-area is obtained from the temperature and humidity sensor, and the temperature weighted value is calculated with the personnel density as the weight, and the area with the highest temperature weighted value is selected as the target area. According to the temperature weighted value and personnel density of the target area, the Q=mcΔT heat balance equation is used to calculate the cooling capacity required for the target area, where m is the air mass, c is the specific heat capacity, and ΔT is the temperature difference. The Mamdan i fuzzy controller is used to generate the adjustment instructions for air conditioning cooling and air curtain air supply. The input variables are temperature deviation and personnel density, and the output variables are cooling capacity and air supply. An incremental PID controller is used to adjust the cooling capacity of the air conditioner and the air supply of the air curtain in real time. The control quantity u(t)=Kp[e(t)-e(t-1)]+Kie(t)+Kd[e(t)-2e(t-1)+e(t-2)] is calculated according to the temperature deviation e(t), where Kp, Ki, and Kd are proportional, integral, and differential coefficients. The actual temperature of the target area is continuously compared with the set temperature through the feedback loop. If the temperature deviation exceeds the preset threshold, the calculation and adjustment process of the cooling capacity and air supply is re-executed. During the operation of the elevator, the central control unit obtains the number of personnel data every 0.5 seconds, and the K-means clustering algorithm with 10 centroids is used to analyze the distribution of personnel. The elevator space is evenly divided into 9 sub-areas of 3x3, and the area of ​​each sub-area is 0.5 square meters. The personnel density value is calculated by dividing the number of people in the sub-area by the area. For example, if 2 people are detected in a sub-area, the density is 4 people / square meter. The temperature and humidity sensor collects data once every 1 second with an accuracy of ±0.1℃. The temperature weighted value is calculated with the population density as the weight. For example, if the temperature in a certain area is 25℃ and the population density is 4 people / square meter, the temperature weighted value is 100℃·person / square meter. The area with the highest temperature weighted value is selected as the target area. Assume that the temperature of the target area is 26℃, the set temperature is 23℃, the air mass is 10kg, and the specific heat capacity is

[0024] 1.005 kJ / (kg·℃), then the required cooling capacity Q=10×1.005×(26-23)=30.15 kJ. The input variables "temperature deviation" and "personnel density" of the Mamdan i fuzzy controller are divided into 3 fuzzy sets respectively, and the output variables "cooling capacity" and "air supply volume" are divided into 5 fuzzy sets respectively, with a total of 9 fuzzy rules. The parameters of the incremental PID controller are determined by the Ziegler-Nichols tuning method, assuming Kp=0.6, Ki=0.2, Kd=0.125. The controller calculates the control quantity every 2 seconds. If the temperature deviation exceeds the preset threshold of ±0.5℃, the calculation and adjustment process of cooling capacity and air supply volume is re-executed. The response time of the entire system is less than 3 seconds, and the control accuracy reaches ±0.2℃.

[0025] Step S103, install a height sensor in the elevator car to obtain the elevator running height data in real time, and transmit the obtained height data to the central control unit; the central control unit designs a height-temperature correspondence based on the height data and the internal and external temperature data obtained by the temperature and humidity sensors installed inside and outside, obtains the internal and external temperature parameters corresponding to each floor height, and dynamically adjusts the cooling power of the air-conditioning system according to the corresponding relationship.

[0026] The elevator running height data is obtained, and the elevator running height data is collected by an ultrasonic height sensor with a measurement range of 0-100 meters and an accuracy of ±1mm; the elevator running height data is filtered and filtered by a Butterworth low-pass filter; the filtered elevator running height data is transmitted to a central control unit through a CAN bus; according to the elevator running height data and the temperature data collected by the temperature and humidity sensor, a weighted average method is used for data fusion, and the weight of the weighted average method is determined according to the measurement error of the sensor; a corresponding relationship model between height and internal and external temperature is established by a third-order polynomial regression algorithm; according to the corresponding relationship model between height and internal and external temperature, a cubic spline interpolation algorithm is used to calculate the internal and external temperature parameters at every 0.5-meter height interval; the calculated internal and external temperature parameters are stored in a database of the central control unit; a fuzzy control algorithm is used to calculate the cooling power required by the air-conditioning system according to the elevator running height data and the corresponding internal and external temperature parameters; a PWM signal is generated according to the cooling power, and the duty cycle range of the PWM signal is 0-100%; and the inverter is controlled by the PWM signal to dynamically adjust the cooling power output of the air-conditioning system.

[0027] Specifically, an ultrasonic height sensor with a measurement range of 0-100 meters and an accuracy of ±1mm is used to obtain the elevator running height data in real time. The sampling frequency is 10Hz, and the raw data is filtered through a Butterworth low-pass filter. The processed height data is transmitted to the central control unit through the CAN bus. The temperature data is obtained from the temperature and humidity sensors installed inside and outside the car, and the data is fused in combination with the height data. The weighted average method is used for fusion. The weight is determined according to the measurement error of the sensor, and a third-order polynomial regression algorithm is used to establish a corresponding relationship model between the height and the internal and external temperatures. According to the established height-temperature corresponding relationship model, the cubic spline interpolation algorithm is used to calculate the internal and external temperature parameters at every 0.5-meter height interval. The temperature parameters in the database are updated every 10 seconds, and the calculation results are stored in the database of the central control unit. A 5x5 fuzzy rule base is designed based on the current elevator height and the corresponding internal and external temperature parameters using the fuzzy control algorithm. The input variables are the height and the internal and external temperature difference, and the output variable is the cooling power percentage. The cooling power required by the air-conditioning system is calculated, and the cooling power output of the air-conditioning system is dynamically adjusted by controlling the inverter through PWM signals. The PWM signal frequency is set to 20kHz, and the duty cycle range is 0-100%, corresponding to the output frequency of the inverter 0-60Hz.

[0028] During the operation of the elevator, the ultrasonic height sensor collects elevator height data at a frequency of 10Hz, with a measurement range of 0-100 meters and an accuracy of ±1mm. The raw data is processed by a fourth-order Butterworth low-pass filter, and the cutoff frequency is set to 2Hz to filter out high-frequency noise. The processed height data is transmitted to the central control unit via a 500kbps CAN bus. At the same time, the temperature and humidity sensor collects temperature data inside and outside the car every 0.5 seconds with an accuracy of ±0.1℃. The reciprocal of the measurement error is used as the weight for data fusion. For example, if the internal sensor error is 0.1℃ and the external is 0.2℃, the weight ratio is 2:1. The least squares method is used to fit a third-order polynomial regression model, such as T=a0+a1h+a2h^2+a3h^3, where T is temperature and h is height. The model parameters are updated every 10 seconds to ensure the real-time performance of the model. Based on the updated model, the cubic spline interpolation algorithm is used to calculate the temperature parameters at every 0.5-meter height interval, and the interpolation node is selected as the height of each elevator stop floor. The input variables "height" and "internal and external temperature difference" of the fuzzy controller are divided into 5 fuzzy sets respectively, and the output variable "cooling power percentage" is divided into 5 fuzzy sets, with a total of 25 fuzzy rules. The cooling power percentage obtained after defuzzification controls the inverter through a 20kHz PWM signal, and the duty cycle of 0-100% linearly corresponds to the output frequency of the inverter of 0-60Hz, thereby accurately adjusting the speed and cooling capacity of the air-conditioning compressor. The response time of the entire control process is less than 1 second, and the temperature control accuracy reaches ±0.5℃.

[0029] Step S104, by installing an external wind pressure sensor, real-time monitoring of wind pressure data, by combining the height data obtained by the height sensor, analyzing the law of wind pressure change with height data, and dynamically adjusting the cooling power of the air conditioner in the elevator and the operating power of the air curtain according to this law.

[0030] A piezoelectric wind pressure sensor is used to monitor the external wind pressure data in real time, and the wind pressure data is processed by a second-order Butterworth low-pass filter; a support vector regression algorithm is used to establish a nonlinear relationship model between wind pressure and height based on the processed wind pressure data and the height data obtained by the height sensor; an adaptive piecewise linear interpolation algorithm is used to calculate the wind pressure change rate in different height intervals using the nonlinear relationship model between wind pressure and height; a 5-layer fuzzy neural network algorithm is used to input height, wind pressure and wind pressure change rate, and output cooling power and wind curtain power; according to the output result of the fuzzy neural network algorithm, the inverter is controlled by PWM signal to dynamically adjust the power output of the air conditioner and wind curtain. The cutoff frequency of the second-order Butterworth low-pass filter is set to 10Hz, and the processed wind pressure data is transmitted to the central control unit via the RS485 bus. The support vector regression algorithm uses a radial basis function kernel function, and optimizes the model parameters through cross-validation to obtain a regular function of wind pressure changing with height. The adaptive piecewise linear interpolation algorithm dynamically adjusts the segmentation points according to the size of the wind pressure change rate, and stores the calculation results in the database of the central control unit. The fuzzy neural network algorithm uses the back propagation algorithm for online learning, and the learning rate is set to 0.01. The PWM signal controls the inverter to adjust the cooling power required by the air conditioning system and the operating power of the air curtain according to the current elevator height, wind pressure data and wind pressure change rate.

[0031] Specifically, a piezoelectric wind pressure sensor with a measurement range of 0-1000Pa and an accuracy of ±1Pa is used to monitor the external wind pressure data in real time. The sampling frequency is 50Hz, and the signal is processed by a second-order Butterworth low-pass filter with a cutoff frequency of 10Hz. The processed wind pressure data is transmitted to the central control unit through the RS485 bus. Combined with the height data obtained by the height sensor, the support vector regression algorithm is used to establish a nonlinear relationship model between wind pressure and height. The radial basis function kernel function is used to optimize the model parameters through cross-validation to obtain the regular function of wind pressure changing with height. According to the established wind pressure-height relationship model, the adaptive piecewise linear interpolation algorithm is used to calculate the wind pressure change rate in different height intervals. The segmentation point is dynamically adjusted according to the size of the wind pressure change rate, and the calculation results are stored in the database of the central control unit. The model parameters are updated every 5 minutes. Using a 5-layer fuzzy neural network algorithm, the input layer includes height, wind pressure and wind pressure change rate, and the output layer is cooling power and wind curtain power. Back propagation algorithm is used for online learning, and the learning rate is set to 0.01. According to the current elevator height, wind pressure data and wind pressure change rate, the cooling power required by the air conditioning system and the operating power of the wind curtain are calculated, and the inverter is controlled by PWM signal to dynamically adjust the power output of the air conditioner and wind curtain. During the operation of the elevator, the piezoelectric wind pressure sensor collects external wind pressure data at a frequency of 50Hz, with a measurement range of 0-1000Pa and an accuracy of ±1Pa. The original data is processed by a second-order Butterworth low-pass filter with a cutoff frequency of 10Hz to filter out high-frequency noise. The processed wind pressure data is transmitted to the central control unit via a 9600bps RS485 bus. At the same time, the height sensor collects elevator height data every 0.1 second with an accuracy of ±1cm. The support vector regression algorithm with C=10 and γ=0.1 is used to establish a nonlinear relationship model between wind pressure and height, and 5-fold cross validation is used to optimize the model parameters. The adaptive piecewise linear interpolation algorithm dynamically adjusts the segmentation points according to the size of the wind pressure change rate. When the wind pressure change rate exceeds 5Pa / m, a segmentation point is added. The model parameters are updated every 5 minutes to ensure the real-time performance of the model. The fuzzy neural network adopts a 5-layer structure, with 3 neurons in the input layer (height, wind pressure, wind pressure change rate), 10 and 8 neurons in the hidden layer, and 2 neurons in the output layer (cooling power, wind curtain power). The back propagation algorithm with a learning rate of 0.01 is used for online learning, and the network weights are updated every 100ms. According to the output of the fuzzy neural network, the inverter is controlled by a 20kHz PWM signal, and the duty cycle of 0-100% linearly corresponds to the output frequency of the inverter of 0-50Hz, thereby accurately adjusting the speed of the air-conditioning compressor and the wind curtain fan. The response time of the entire control process is less than 0.5 seconds, and the wind pressure adaptability control accuracy reaches ±5Pa.

[0032] Step S105, a door opening and closing sensor is set in the elevator door machine system to monitor the opening and closing status of the elevator door in real time, and transmit the door status signal to the central control unit; the central control unit controls the air curtain system to increase the air output at the moment the door is opened according to the door status signal, and reduces the air output after the door is closed.

[0033] The method comprises the following steps: obtaining a change signal of the magnetic field strength of an elevator door measured by a Hall effect sensor, wherein the Hall effect sensor is installed in the elevator door machine system; processing the change signal by a second-order Butterworth low-pass filter to obtain a processed digital signal; transmitting the processed digital signal to a central control unit through a CAN bus, wherein the digital signal includes a door status signal; processing the door status signal by a state machine algorithm according to the door status signal received by the central control unit to determine the state of the elevator door; if the rate of change of the magnetic field strength exceeds a preset threshold, the door is determined to be in an open state; if the magnetic field strength is stable at a minimum value, the door is determined to be in a fully open state; calculating the required air volume of the air curtain system by a fuzzy control algorithm according to the result of the elevator door state recognition; calculating the control amount of the air curtain system by an incremental PID controller according to the target air volume value; controlling the frequency converter by a PWM signal to adjust the speed of the air curtain motor to achieve dynamic adjustment of the air volume of the air curtain; and adaptively adjusting the PID parameters by using the MIT rule to dynamically adjust the parameters of the PID controller according to the rate of change of the control error.

[0034] Specifically, a Hall effect sensor with a measurement range of 0-5000 Gauss and an accuracy of ±1 Gauss is installed in the elevator door machine system. The sampling frequency is 100Hz. The switch state of the elevator door is monitored in real time by measuring the change in the magnetic field strength of the door body. A second-order Butterworth low-pass filter is used for signal processing, and the cut-off frequency is set to 20Hz. The processed digital signal is transmitted to the central control unit through the CAN bus. The central control unit uses a state machine algorithm to process the received door state signal and defines three states: closed, opening, and fully open. When the magnetic field strength change rate exceeds 100 Gauss / second, it is determined to be in the opening state; when the magnetic field strength is stable at the minimum value, it is determined to be in the fully open state. The edge detection algorithm is used to identify the moment of door opening and closing, calculate the door opening and opening speed, and store the recognition results in the temporary cache of the control unit. According to the door state recognition results, the fuzzy control algorithm is used to calculate the required air volume of the air curtain system. A 7x7 fuzzy rule base is designed. The input variables "door opening" and "switching speed" are divided into 7 fuzzy sets respectively, and the output variable "air volume of the air curtain" is divided into 7 fuzzy sets. Fast fuzzy reasoning is realized by table lookup method to obtain the target air volume value of the air curtain system. The incremental PID controller is used to calculate the control quantity of the air curtain system according to the target air volume value. The inverter is controlled by PWM signal to adjust the speed of the air curtain motor to realize the dynamic adjustment of the air volume of the air curtain. At the same time, the MIT rule is used to perform adaptive adjustment of PID parameters. The Kp, Ki, and Kd parameters are dynamically adjusted according to the change rate of the control error. The adjustment step size is set to 0.01, 0.005, and 0.002 to improve the control accuracy and response speed. During the operation of the elevator, the Hall effect sensor collects the magnetic field strength data of the door body at a frequency of 100Hz. The measurement range is 0-5000 Gauss, and the accuracy reaches ±1 Gauss. The raw data is processed by a second-order Butterworth low-pass filter with a cutoff frequency of 20 Hz to filter out high-frequency noise. The processed data is transmitted to the central control unit via a 500 kbps CAN bus. The state machine algorithm determines the door state based on the rate of change of the magnetic field strength. When the rate of change exceeds 100 gauss / second, it is judged as being in the middle of opening; when the magnetic field strength drops below 500 gauss and the rate of change is less than 10 gauss / second, it is judged as being in the fully open state. The edge detection algorithm uses the first-order difference method to calculate the difference in magnetic field strength between two adjacent samples. When the absolute value of the difference exceeds 50 gauss, it is identified as an opening or closing moment. In the fuzzy control algorithm, the door opening is divided into 7 levels (0%, 16.7%, 33.3%, 50%, 66.7%, 83.3%, 100%), the opening and closing speed is divided into 7 levels (0-0.1m / s, 0.1-0.2m / s, 0.2-0.3m / s, 0.3-0.4m / s, 0.4-0.5m / s, 0.5-0.6m / s, >0.6m / s), and the air curtain air volume is divided into 7 levels (0-50m 3 / h、50-100m3 / h、100-150m 3 / h、150-200m 3 / h、200-250m 3 / h、250-300m 3 / h,>300m 3 / h). The initial parameters of the PID controller are set to Kp = 0.5, Ki = 0.2, Kd = 0.1, and a 20kHz PWM signal is used to control the inverter. The duty cycle of 0-100% linearly corresponds to the speed of the air curtain motor of 0-3000rpm. The MIT rule calculates the parameter adjustment amount every 100ms, and the control error change rate threshold is set to ±5% / s. The response time of the entire system is less than 50ms, and the air volume control accuracy reaches ±5m 3 / h.

[0035] Step S106, the central control unit calculates the target control quantity of the air-conditioning system and the air curtain system according to the number of people in the car, the temperature difference between the inside and outside, the car height, the wind pressure data, and the door status. The target control quantity includes the air-conditioning cooling power, the air outlet speed and the air outlet angle of the air curtain, and sends the control instructions to the air-conditioning and air curtain actuators respectively to realize the coordinated control of the air-conditioning system and the air curtain system.

[0036] Acquire multi-source sensor data, and use DS evidence theory to perform data fusion according to the set credibility weights of each sensor; if the data of the personnel quantity sensor, temperature sensor, height sensor, wind pressure sensor and door status sensor are obtained, set the credibility weights of each sensor to 0.3, 0.25, 0.2, 0.15 and 0.1 respectively; perform noise suppression and state estimation on the fused sensor data through the Kalman filter to obtain the optimized environmental parameters; according to the optimized environmental parameters, use a 5-layer fully connected neural network to establish a multi-input and multi-output nonlinear mapping model, the input layer of the 5-layer fully connected neural network includes the optimized environmental parameters, and the hidden The number of neurons in each layer is 64, 32, and 16 respectively, the activation function adopts ReLU, and the output layer is the air conditioning refrigeration power, the air curtain air outlet speed and the air outlet angle; based on the nonlinear mapping model, the model predictive control algorithm is adopted to predict the system state change in the future period of time, and the optimal control sequence is calculated by the quadratic programming solver, from which the optimal control quantity at the current moment is extracted as the target control quantity of the air conditioning system and the air curtain system; the target control quantity is sent to the air conditioning and air curtain actuators respectively through the CAN bus, and the incremental PID controller is adopted to realize the adjustment of the air conditioning compressor speed and the air curtain motor speed. The initial parameters of the PID controller are set to Kp, Ki, and Kd, and the Ziegler-Nichols tuning method is adopted for parameter tuning.

[0037] Specifically, DS evidence theory is used for multi-source data fusion, and the credibility weight of each sensor is set, 0.3 for the number of personnel sensor, 0.25 for the temperature sensor, 0.2 for the height sensor, 0.15 for the wind pressure sensor, and 0.1 for the door state sensor. The Kalman filter is used to suppress noise and estimate the state of each sensor data to obtain the optimized environmental parameters. According to the fused environmental parameters, a multi-input and multi-output nonlinear mapping model is established using a 5-layer fully connected neural network. The input layer includes the optimized environmental parameters, the number of neurons in the hidden layer is 64, 32, and 16 respectively, the activation function uses ReLU, and the output layer is the air conditioning cooling power, wind curtain air outlet speed and wind outlet angle. Online training is performed every 10 seconds, the learning rate is set to 0.001, and the network parameters are updated by the back propagation algorithm. The model predictive control algorithm is used. Based on the established nonlinear mapping model, the prediction time domain is set to 30 seconds and the control time domain is 10 seconds. The system state changes in the future are predicted, and the optimal control sequence is calculated by the quadratic programming solver. The optimal control quantity at the current moment is extracted as the target control quantity of the air conditioning system and the wind curtain system. The calculated target control quantity is sent to the air conditioner and air curtain actuators respectively through the CAN bus. The incremental PID controller is used to achieve precise adjustment of the air conditioner compressor speed and the air curtain motor speed. The initial parameters of the PID controller are set to Kp=0.5, Ki=0.2, and Kd=0.1. The Ziegler-Nichols tuning method is used for parameter tuning. The control period is set to 100ms. At the same time, a stepper motor is used to control the angle of the air curtain guide plate to achieve coordinated control of the air conditioning system and the air curtain system.

[0038] During the operation of the elevator, the DS evidence theory fusion algorithm is executed every 100ms to comprehensively process the data from the five sensors. The accuracy of the personnel number sensor is ±1 person, and the accuracy of the temperature sensor is

[0039] ±0.1℃, the height sensor accuracy is ±1cm, the wind pressure sensor accuracy is ±1Pa, and the door state sensor is a Boolean value. The process noise covariance matrix Q and the measurement noise covariance matrix R of the Kalman filter are set by empirical values, and the initial state estimation error covariance P0 is set to the unit matrix. The input layer of the 5-layer neural network has 5 neurons, corresponding to 5 types of sensor data, and the output layer has 3 neurons, corresponding to the air conditioning cooling power (0-5000W), wind curtain air speed (0-20m / s) and air outlet angle (0-90°). The Adam optimizer is used for training, and the batch size is set to 32. The prediction time domain of the model predictive control algorithm is divided into 10 sampling points within 30 seconds on average, and the control time domain is divided into 5 sampling points within 10 seconds on average. The CVXOPT solver is used to solve the quadratic programming problem. The target control quantity is transmitted to the actuator through the 500kbps CAN bus. The control cycle of the PID controller is 100ms. The improved Ziegler-Nichols tuning method is used to determine the critical proportional gain and critical cycle through the 25% overshoot law. The speed range of the air conditioning compressor is 600-6000rpm, the speed range of the wind curtain motor is 0-3000rpm, the step angle of the stepper motor of the wind curtain guide plate is 1.8°, and the maximum rotation angle is 90°. The response time of the entire system is less than 500ms, and the control accuracy reaches: temperature ±0.5℃, wind speed ±0.2m / s, angle ±2°.

[0040] Step S107, the central control unit studies and trains the historical operation data, establishes a mapping relationship between the number of people, temperature and humidity, height, wind pressure, door status parameters and the air conditioning cooling capacity and wind curtain wind speed control quantity, and forms an intelligent control strategy; in actual operation, according to the current parameter input, the trained model is used to predict the target control quantity, and the working status of the air conditioner and wind curtain is adjusted in real time.

[0041] Obtain historical operation data including the number of people, temperature and humidity, height, wind pressure, door status parameters, and the corresponding air conditioning cooling capacity and wind curtain wind speed control quantity, and the historical operation data is stored in the Inf l uxDB time series database; according to the historical operation data, a 3-layer LSTM network is used for learning and training, each layer of the 3-layer LSTM network contains 128 neurons, and the dropout rate is 0.2, to build a nonlinear mapping relationship between input parameters and control quantities; receiving the current parameter input collected in real time, and performing forward reasoning through the trained LSTM model to obtain the predicted air conditioning cooling capacity and wind curtain wind speed control quantity; using the sliding time window method to achieve continuous prediction, the window size of the sliding time window method is 1 hour, and the sliding interval is 5 minutes; for the predicted control quantity, the fuzzy controller is used for optimization and adjustment, the fuzzy controller contains a 7x7 rule base, the input variables "prediction deviation" and "change rate" are each divided into 7 fuzzy sets, and the output variable "control correction value" is divided into 7 fuzzy sets; if the output result of the fuzzy controller meets the preset conditions, it is sent to the air conditioner and wind curtain actuators through the field bus to achieve real-time adjustment of the working status of the air conditioner and wind curtain.

[0042] Specifically, the InfluxDB time series database is used to store historical operation data, including the number of people, temperature and humidity, height, wind pressure, door status parameters, and the corresponding air conditioning cooling capacity and wind curtain wind speed control quantity. The data cleaning standards are as follows: the deviation of the number of people is ≤3 people, the temperature deviation is ≤2℃, and the humidity deviation is ≤1℃.

[0043] ≤5%, height deviation ≤0.5m, wind pressure deviation ≤10Pa, outliers and redundant data are eliminated, and the minimum-maximum normalization method is used for data standardization to form a standardized training data set. A three-layer LSTM network is used to learn and train the processed historical data, with 128 neurons in each layer and a dropout rate of 0.2. A nonlinear mapping relationship between input parameters and control quantities is constructed, and network parameters are optimized by the back propagation algorithm. The Adam optimizer is used with a learning rate of 0.001. The early stopping method is used to prevent overfitting, and the patience is set to 10 epochs to obtain a trained intelligent control model. According to the current parameter input collected in real time, the trained LSTM model is used for forward reasoning to predict the optimal air conditioning cooling capacity and wind curtain wind speed control quantity. Continuous prediction is achieved through the sliding time window method. The window size is set to 1 hour and slides every 5 minutes. The reliability of the prediction result is evaluated by calculating the confidence interval of the predicted value to improve the real-time and adaptability of the model. The predicted target control quantity is optimized and adjusted through the fuzzy controller. A 7x7 rule base is designed. The input variables "prediction deviation" and "change rate" are divided into 7 fuzzy sets respectively, and the output variable "control correction value" is divided into 7 fuzzy sets. The centroid method is used for defuzzification to obtain the final control instruction, which is sent to the air conditioner and air curtain actuators through the field bus to achieve real-time adjustment.

[0044] InfluxDB stores historical operation data at a sampling rate of 1 second. Each record contains 7 input parameters and 2 output parameters. During the data cleaning process, if 5 consecutive records exceed the set threshold, they are marked as abnormal and removed. Minimum-maximum normalization maps all parameters to the interval [0,1]. The LSTM network input layer has 7 neurons, corresponding to 7 input parameters, and the output layer has 2 neurons, corresponding to the air conditioning cooling capacity and wind curtain wind speed, respectively. During the training process, the mini-batch gradient descent method with a batch size of 64 is used, the initial learning rate is 0.001, and it decays by 10% every 50 epochs. The early stopping method terminates the training when the validation set loss has not improved for 10 consecutive epochs. The sliding time window uses 3600 data points within 1 hour and slides once every 5 minutes (300 data points). The confidence interval of the prediction result is calculated by the bootstrap method, and the sampling is repeated 100 times. The 2.5% and 97.5% quantiles are taken as the 95% confidence interval. The input variables "prediction deviation" and "rate of change" of the fuzzy controller are divided into 7 triangular membership functions, corresponding to "negative large", "negative medium", "negative small", "zero", "positive small", "positive medium", and "positive large". The output variable "control correction value" also uses 7 triangular membership functions. The fuzzy rules use the Mamdan i reasoning method, such as "IF the prediction deviation is positive large AND the rate of change is positive medium THEN the control correction value is positive large". The defuzzification uses the centroid method, and the calculated control correction value is sent to the actuator through the CAN bus with a period of 100ms. The response time of the entire system is less than 1 second, and the control accuracy reaches: temperature ±0.3℃, wind speed

[0045] ±0.1m / s.

[0046] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An elevator air conditioning intelligent air curtain method, characterized in that: The method comprises: By installing a passenger number sensor in the car, the passenger data in the car is obtained in real time, and the obtained passenger number data is transmitted to the central control unit. At the same time, by installing temperature and humidity sensors inside and outside the car, the temperature and humidity parameters inside and outside the car are obtained, and the obtained temperature and humidity data are transmitted to the central control unit; According to the real-time data of the number of people obtained in the central control unit, the distribution of people is analyzed, the elevator is divided into areas, and the density of people in each area is obtained. According to the data of the temperature and humidity sensors, the temperature data of each area is analyzed. The temperature data of the target area and the density of people distribution are combined to calculate the required air supply and cooling capacity of the target area, adjust the air conditioning cooling and air curtain air supply, and continuously judge whether the target area reaches the required air supply and temperature; A height sensor is installed in the elevator car to obtain the elevator running height data in real time, and the obtained height data is transmitted to the central control unit. The central control unit designs the height-temperature correspondence relationship based on the height data and the internal and external temperature data obtained by the temperature and humidity sensors installed inside and outside, obtains the internal and external temperature parameters corresponding to each floor height, and dynamically adjusts the cooling power of the air-conditioning system according to the corresponding relationship; By installing an external wind pressure sensor, the wind pressure data is monitored in real time. By combining the height data obtained by the height sensor, the law of wind pressure changes with height data is analyzed, and the cooling power of the air conditioner in the elevator and the operating power of the air curtain are dynamically adjusted according to this law; A door opening and closing sensor is set in the elevator door machine system to monitor the opening and closing status of the elevator door in real time and transmit the door status signal to the central control unit. The central control unit controls the air curtain system to increase the air output at the moment the door is opened and reduce the air output after the door is closed according to the door status signal. The central control unit calculates the target control quantity of the air conditioning system and the air curtain system according to the number of people in the car, the temperature difference between the inside and outside, the height of the car, the wind pressure data, and the door status. The target control quantity includes the air conditioning cooling power, the air curtain air outlet speed and the air outlet angle, and sends the control instructions to the air conditioning and air curtain actuators respectively to realize the coordinated control of the air conditioning system and the air curtain system; The central control unit studies and trains historical operating data, establishes a mapping relationship between the number of people, temperature and humidity, height, wind pressure, door status parameters and the air conditioning cooling capacity and wind curtain wind speed control quantity, and forms an intelligent control strategy. In actual operation, based on the current parameter input, the trained model is used to predict the target control quantity and adjust the working status of the air conditioner and wind curtain in real time.

2. The method according to claim 1, characterized in that The method comprises: installing a personnel number sensor in the car to obtain personnel data in the car in real time, transmitting the obtained personnel number data to the central control unit, and installing temperature and humidity sensors inside and outside the car to obtain temperature and humidity parameters inside and outside the car, and transmitting the obtained temperature and humidity data to the central control unit, including: Acquire real-time data collected by a sensor for the number of people in the car, and use digital signal processing to filter and reduce noise on the real-time data to obtain processed data on the number of people; The processed personnel quantity data is transmitted to the central control unit data receiving module through the CAN bus, and the central control unit data receiving module analyzes the received data to obtain the personnel quantity value; The data processing module of the central control unit compares the number of people with a preset threshold of the number of people to determine whether the car is overloaded; If the car is overloaded, the alarm module is triggered to issue an alarm prompt, and the programmable logic controller is used to control the elevator to stop running and keep the door open; Acquire real-time data collected by temperature and humidity sensors inside and outside the car, and use Kalman filtering to smooth the real-time data to obtain processed temperature and humidity data; The processed temperature and humidity data are transmitted to the data receiving module of the central control unit via the RS485 bus, and the data receiving module of the central control unit analyzes the received data to obtain temperature and humidity values; The data processing module of the central control unit performs difference calculation on the temperature and humidity values, and compares the difference calculation result with the preset temperature and humidity difference threshold value to determine whether to start air conditioning adjustment; If the air conditioning needs to be started, the fuzzy control is used to determine the air conditioning operation parameters, which include the wind speed and the temperature setting value.

3. The method according to claim 1, characterized in that The personnel distribution is analyzed based on the personnel quantity data obtained in real time in the central control unit, the elevator is divided into regions, the personnel density of each region is obtained, the temperature data of each region is analyzed based on the temperature and humidity sensor data, the temperature data of the target region and the personnel distribution density are integrated, the required air supply and cooling capacity of the target region are calculated, the air conditioning cooling and air curtain air supply are adjusted, and the target region is continuously judged whether the required air supply and temperature are achieved, including: The K-means clustering algorithm is used to analyze the distribution of people in the elevator, and the elevator space is evenly divided into nine sub-areas to obtain the density of people in each sub-area. According to the population density value, the temperature data of each sub-area is obtained from the temperature and humidity sensor, a temperature weighted value is calculated, and the area with the highest temperature weighted value is selected as the target area; According to the temperature weighted value and the personnel density of the target area, the required cooling capacity of the target area is calculated by using a heat balance equation, wherein the heat balance equation is Q=mcΔT, where m is the air mass, c is the specific heat capacity, and ΔT is the temperature difference; The Mamdani fuzzy controller is used to generate the adjustment instructions for air conditioning cooling and air curtain air supply. The input variables are temperature deviation and personnel density, and the output variables are cooling capacity and air supply volume. The incremental PID controller is used to adjust the air conditioning cooling capacity and the air curtain air supply in real time. The control quantity u(t) is calculated according to the temperature deviation e(t). The control quantity u(t)=Kp[e(t)-e(t- 1)]+Kie(t)+Kd[e(t)-2e(t-1)+e(t-2)], where Kp, Ki, Kd are proportional, integral and differential coefficients; If the temperature deviation between the actual temperature of the target area and the set temperature exceeds a preset threshold, the calculation and adjustment process of the cooling capacity and the air supply volume is re-executed.

4. The method according to claim 1, characterized in that: The height sensor is installed in the elevator car to obtain the elevator running height data in real time, and the obtained height data is transmitted to the central control unit. The central control unit designs the height-temperature correspondence relationship according to the height data and the internal and external temperature data obtained by the temperature and humidity sensors installed inside and outside, obtains the internal and external temperature parameters corresponding to each floor height, and dynamically adjusts the cooling power of the air-conditioning system according to the corresponding relationship, including: Acquire elevator running height data, where the elevator running height data is collected by an ultrasonic height sensor with a measuring range of 0-100 meters and an accuracy of ±1mm; Performing filtering processing on the elevator running height data, using a Butterworth low-pass filter to filter the elevator running height data; The filtered elevator running height data is transmitted to the central control unit via the CAN bus; According to the elevator running height data and the temperature data collected by the temperature and humidity sensor, a weighted average method is used to perform data fusion, and the weight of the weighted average method is determined according to the measurement error of the sensor; The corresponding relationship model between height and internal and external temperatures was established using the third-order polynomial regression algorithm; According to the correspondence model between the height and the internal and external temperatures, a cubic spline interpolation algorithm is used to calculate the internal and external temperature parameters at every 0.5 meter height interval; storing the calculated internal and external temperature parameters in a database of the central control unit; Calculate the cooling power required by the air-conditioning system according to the elevator running height data and the corresponding internal and external temperature parameters by using a fuzzy control algorithm; Generate a PWM signal according to the cooling power, wherein the duty cycle range of the PWM signal is 0-100%; The PWM signal is used to control the frequency converter to dynamically adjust the cooling power output of the air-conditioning system.

5. The method according to claim 1, characterized in that The method includes installing an external wind pressure sensor to monitor wind pressure data in real time, analyzing the law of wind pressure changing with height data by combining the height data obtained by the height sensor, and dynamically adjusting the cooling power of the air conditioner in the elevator and the operating power of the air curtain according to the law, including: A piezoelectric wind pressure sensor is used to monitor external wind pressure data in real time, and the wind pressure data is processed by a second-order Butterworth low-pass filter; Based on the processed wind pressure data and the height data obtained by the height sensor, a nonlinear relationship model between wind pressure and height is established using the support vector regression algorithm; The nonlinear relationship model between wind pressure and height is used to calculate the wind pressure change rate in different height intervals using an adaptive piecewise linear interpolation algorithm; Adopting a 5-layer fuzzy neural network algorithm, inputting height, wind pressure and wind pressure change rate, outputting cooling power and air curtain power; According to the output result of the fuzzy neural network algorithm, the frequency converter is controlled by PWM signal to dynamically adjust the power output of the air conditioner and the air curtain; The cutoff frequency of the second-order Butterworth low-pass filter is set to 10 Hz, and the processed wind pressure data is transmitted to the central control unit via the RS485 bus; The support vector regression algorithm adopts radial basis function kernel function, optimizes model parameters through cross-validation, and obtains the regular function of wind pressure changing with height; The adaptive piecewise linear interpolation algorithm dynamically adjusts the segmentation points according to the wind pressure change rate, and stores the calculation results in the database of the central control unit; The fuzzy neural network algorithm adopts the back propagation algorithm for online learning, and the learning rate is set to 0.01; The PWM signal controls the frequency converter to adjust the cooling power required by the air conditioning system and the operating power of the air curtain according to the current elevator height, wind pressure data and wind pressure change rate.

6. The method according to claim 1, characterized in that The door opening and closing sensor is arranged in the elevator door machine system to monitor the opening and closing state of the elevator door in real time, and transmit the door state signal to the central control unit. The central control unit controls the air curtain system to increase the air volume at the moment the door is opened according to the door state signal, and reduces the air volume after the door is closed, including: Obtain a signal of a change in the magnetic field strength of an elevator door body measured by a Hall effect sensor, wherein the Hall effect sensor is installed in the elevator door machine system; Using a second-order Butterworth low-pass filter to process the change signal to obtain a processed digital signal; transmitting the processed digital signal to a central control unit via a CAN bus, wherein the digital signal includes a door status signal; According to the door status signal received by the central control unit, the door status signal is processed by a state machine algorithm to determine the state of the elevator door; If the rate of change of the magnetic field strength exceeds a preset threshold, it is determined to be in the on state; If the magnetic field strength is stable at the minimum value, it is determined to be in a fully open state; According to the elevator door status recognition results, the fuzzy control algorithm is used to calculate the air volume required by the air curtain system; An incremental PID controller is used to calculate the control quantity of the air curtain system according to the target air volume value; The inverter is controlled by PWM signal to adjust the speed of the air curtain motor to achieve dynamic adjustment of the air curtain air volume; The MIT rule is used to perform adaptive adjustment of PID parameters, and the parameters of the PID controller are dynamically adjusted according to the change rate of the control error.

7. The method according to claim 1, characterized in that The central control unit calculates the target control quantity of the air conditioning system and the air curtain system according to the number of people in the car, the temperature difference between the inside and outside, the height of the car, the wind pressure data, and the door status. The target control quantity includes the air conditioning cooling power, the air curtain air outlet speed and the air outlet angle, and sends the control instructions to the air conditioning and air curtain actuators respectively to realize the coordinated control of the air conditioning system and the air curtain system, including: Acquire multi-source sensor data and use DS evidence theory to perform data fusion according to the set credibility weights of each sensor; If the data of the personnel quantity sensor, temperature sensor, height sensor, wind pressure sensor and door status sensor are obtained, the credibility weight of each sensor is set to 0.3, 0.25, 0.2, 0.15 and 0.1 respectively; The fused sensor data is subjected to noise suppression and state estimation through the Kalman filter to obtain the optimized environmental parameters; According to the optimized environmental parameters, a 5-layer fully connected neural network is used to establish a multi-input and multi-output nonlinear mapping model, wherein the input layer of the 5-layer fully connected neural network includes the optimized environmental parameters, the number of neurons in the hidden layer is 64, 32 and 16 respectively, the activation function adopts ReLU, and the output layer is the air conditioning cooling power, the air curtain outlet speed and the air outlet angle; Based on the nonlinear mapping model, a model predictive control algorithm is used to predict the system state changes in the future, and the optimal control sequence is calculated by a quadratic programming solver, from which the optimal control quantity at the current moment is extracted as the target control quantity of the air conditioning system and the air curtain system; The target control quantity is sent to the air conditioner and air curtain actuators respectively through the CAN bus, and an incremental PID controller is used to adjust the air conditioner compressor speed and the air curtain motor speed. The initial parameters of the PID controller are set to Kp, Ki and Kd, and the Ziegler-Nichols tuning method is used for parameter tuning.

8. The method according to claim 1, characterized in that The central control unit learns and trains historical operation data, establishes mapping relationships between the number of personnel, temperature and humidity, height, wind pressure, door status parameters and the air conditioning cooling capacity and wind curtain wind speed control quantity, and forms an intelligent control strategy. In actual operation, according to the current parameter input, the trained model is used to predict the target control quantity and adjust the working status of the air conditioner and wind curtain in real time, including: Obtain historical operation data including the number of people, temperature and humidity, height, wind pressure, door status parameters, and corresponding air conditioning cooling capacity and wind curtain wind speed control quantity, and the historical operation data is stored in the InfluxDB time series database; According to the historical operation data, a 3-layer LSTM network is used for learning and training, wherein each layer of the 3-layer LSTM network contains 128 neurons and a dropout rate of 0.2, to construct a nonlinear mapping relationship between input parameters and control quantities; Receive the current parameter input collected in real time, perform forward reasoning through the trained LSTM model, and obtain the predicted air conditioning cooling capacity and wind curtain wind speed control value; A sliding time window method is used to achieve continuous prediction. The window size of the sliding time window method is 1 hour and the sliding interval is 5 minutes; The predicted control quantity is optimized and adjusted by a fuzzy controller, which contains a 7x7 rule base, and the input variables "prediction deviation" and "change rate" are divided into 7 fuzzy sets respectively, and the output variable "control correction value" is divided into 7 fuzzy sets; If the output result of the fuzzy controller meets the preset conditions, it is sent to the air conditioner and air curtain actuators through the field bus to achieve real-time adjustment of the working status of the air conditioner and air curtain.

Images