AI-based range hood energy-saving exhaust control method and system

Abstract

The invention discloses an AI-based range hood energy-saving exhaust control method and system, and relates to the technical field of intelligent control. The method comprises the following steps: acquiring temperature, air pressure, airflow disturbance data and lampblack concentration in a kitchen environment; if the temperature data is smaller than a preset temperature threshold value, compensating and correcting the oil smoke concentration according to the temperature data to obtain effective oil smoke concentration; determining initial exhaust air rate according to the air pressure data; the initial exhaust air rate is adjusted according to the airflow disturbance data and the effective oil smoke concentration, and the real-time target exhaust air rate is obtained; converting the real-time target exhaust air volume through a preset air volume-fan rotating speed value mapping table to obtain a fan rotating speed value; and performing exhaust control on the range hood according to the rotating speed value of the fan. According to the method, a distortion compensation and multi-parameter coordinated regulation and control mechanism for oil smoke concentration collection in a low-temperature environment is introduced, air exhaust can be dynamically optimized on the basis of accurately recognizing the oil smoke exhaust state, the energy efficiency is remarkably improved while the smoke exhaust effect is guaranteed, and energy-saving intelligent control over the range hood is achieved.

Description

Technical Field

[0001] This invention belongs to the field of intelligent control technology, specifically relating to an AI-based energy-saving exhaust control method and system for range hoods. Background Technology

[0002] With the escalating global energy crisis and the increasing popularity of low-carbon and environmentally friendly concepts, energy conservation in home appliances has become one of the core trends in the home appliance industry. As an essential kitchen appliance, the range hood's main function is to quickly remove fumes, odors, and harmful gases generated during cooking, ensuring indoor air quality and human health. However, existing range hoods generally suffer from low energy efficiency and insufficient adaptability in ventilation control, making it difficult to balance ventilation performance with energy-saving requirements. Therefore, a highly efficient and intelligent ventilation control technology is urgently needed to address this pain point.

[0003] Patent application CN120521236A discloses an automatic control method and system for a variable frequency range hood, belonging to the field of intelligent control technology for home appliances. The method includes: using a non-contact multimodal sensor to collect real-time data on oil fume concentration, temperature, and airflow disturbance in the kitchen environment; constructing a multivariate fusion model to generate a comprehensive environmental state vector through feature extraction and weight allocation; constructing an LSTM prediction model to abstract the current oil fume distribution state and surrounding environmental characteristics, and combining a multi-objective optimization model of airflow-energy consumption-noise and a genetic algorithm to generate an optimal fan speed control strategy; and adjusting the fan speed in real-time according to the optimal fan speed control strategy.

[0004] However, this method does not consider the condensation of oil fumes in low-temperature operating conditions during range hood exhaust, which weakens the sensitivity of oil fume concentration collection. It does not accurately correct the oil fume concentration for this distorted data, resulting in a significant deviation in the actual air volume calculation. The LSTM prediction model it relies on cannot accurately abstract the oil fume distribution due to the distortion of the input data. The fan speed control strategy generated by the combined air volume-energy consumption-noise multi-objective optimization model and genetic algorithm is also out of touch with actual operating conditions, ultimately resulting in the inability to achieve efficient and energy-saving exhaust control of the range hood. Summary of the Invention

[0005] The purpose of this invention is to solve the problem that the concentration acquisition distortion caused by the condensation of oil fumes under low temperature conditions cannot achieve efficient and energy-saving exhaust control of range hoods, and to propose an AI-based energy-saving exhaust control method and system for range hoods.

[0006] In a first aspect of this invention, an AI-based energy-saving exhaust control method for range hoods is proposed, the method comprising: Real-time collection of temperature, air pressure, airflow disturbance, and oil fume concentration data in the kitchen environment; If the temperature data is less than a preset temperature threshold, the oil fume concentration is compensated and corrected according to the temperature data to obtain an effective oil fume concentration; Determine the initial exhaust volume based on the air pressure data; The initial exhaust volume is adjusted based on the airflow disturbance data and the effective oil fume concentration to obtain the real-time target exhaust volume; The fan speed value is obtained by converting the real-time target exhaust volume using a preset air volume-fan speed value mapping table; The exhaust fan is controlled based on the fan speed value.

[0007] Optionally, the effective oil fume concentration can be obtained by compensating and correcting the oil fume concentration based on the temperature data, including: The temperature difference is obtained by calculating the difference between the temperature data and the preset temperature threshold; via c=1+a×d b The basic deviation correction coefficient is calculated; where c is the basic deviation correction coefficient, a is the temperature influence weighting coefficient, d is the temperature difference, and b is the nonlinear correction index. The coefficient calibration value is obtained by calculating e=ln(x+g); where e is the coefficient calibration value, x is the oil fume concentration, and g is a constant greater than 1. The cumulative deviation coefficient is obtained by multiplying the basic deviation correction coefficient and the coefficient calibration value. The effective oil fume concentration is obtained by correcting the oil fume concentration based on the cumulative deviation coefficient.

[0008] Optionally, determining the initial exhaust volume based on the air pressure data includes: By acquiring continuous sampling values ​​of the air pressure data within a preset time period, air pressure time series data can be obtained; Frequency analysis is performed on the aforementioned air pressure time series data to obtain specific frequency fluctuation components caused by air pressure fluctuations; The amplitude of the specific frequency fluctuation component is compared with a preset amplitude threshold. If the amplitude exceeds a preset amplitude threshold, an exhaust volume modulation signal is generated based on the frequency and phase of the specific frequency fluctuation component to counteract air pressure fluctuations. The exhaust volume modulation signal is converted into an exhaust volume compensation value. The initial exhaust volume is obtained by combining the exhaust volume compensation value and the preset benchmark exhaust volume.

[0009] Optionally, adjusting the initial exhaust volume based on the airflow disturbance data and the effective oil fume concentration to obtain the real-time target exhaust volume includes: The airflow disturbance data is subjected to time-frequency domain feature extraction to obtain a feature vector; The feature vector is input into a pre-trained disturbance pattern classifier to obtain the category to which the current airflow disturbance belongs; the category includes a first category and a second category; the first category is a step pulse type caused by the opening and closing of doors and windows; the second category is a periodic sweep type caused by people walking; The predicted value of oil fume concentration is obtained by predicting the oil fume concentration in the future within a preset time period using historical effective oil fume concentration data and the effective oil fume concentration data. The predicted oil fume concentration is adjusted differentially according to the category to obtain the dynamic compensation air volume; The initial exhaust volume is added to the dynamic compensation air volume to obtain the real-time target exhaust volume.

[0010] Optionally, the dynamically compensated airflow is obtained by differentially adjusting the predicted oil fume concentration value according to the category, including: If the category is the first category and the predicted value of the oil fume concentration is greater than the preset predicted value threshold, then the difference between the predicted value of the oil fume concentration and the effective oil fume concentration data is calculated to obtain the concentration change, the instantaneous peak value is extracted from the airflow disturbance data to obtain the instantaneous peak amplitude, and the instantaneous peak amplitude is multiplied by the concentration change to obtain the dynamic compensation air volume. If the category is the first category, and the predicted value of the oil fume concentration is less than or equal to the preset predicted value threshold, then the instantaneous peak amplitude is multiplied by the preset compensation coefficient to obtain the dynamic compensation air volume; If the category is the second category, the main frequency is extracted from the airflow disturbance data to generate a sine wave compensation signal that is synchronized with the main frequency but opposite in phase. The amplitude of the sine wave compensation signal is positively scaled according to the predicted value of the oil fume concentration to obtain the dynamic compensation air volume.

[0011] In a second aspect of this invention, an AI-based energy-saving exhaust control system for range hoods is proposed, comprising: The data acquisition module is used to collect real-time data on temperature, air pressure, airflow disturbance, and oil fume concentration in the kitchen environment. The condition module is used to compensate and correct the oil fume concentration based on the temperature data to obtain an effective oil fume concentration if the temperature data is less than a preset temperature threshold. The initial determination module is used to determine the initial exhaust volume based on the air pressure data; An adjustment module is used to adjust the initial exhaust volume according to the airflow disturbance data and the effective oil fume concentration to obtain the real-time target exhaust volume; The conversion module is used to convert the real-time target exhaust volume into a fan speed value by using a preset air volume-fan speed value mapping table; The control module is used to control the exhaust of the range hood according to the fan speed value.

[0012] Optionally, the condition module includes: The difference calculation module is used to calculate the temperature difference between the temperature data and the preset temperature threshold. The first coefficient calculation module is used to calculate c = 1 + a × d. b The basic deviation correction coefficient is calculated; where c is the basic deviation correction coefficient, a is the temperature influence weighting coefficient, d is the temperature difference, and b is the nonlinear correction index. The calibration value calculation module is used to calculate the coefficient calibration value using the formula e=ln(x+g); where e is the coefficient calibration value, x is the oil fume concentration, and g is a constant greater than 1. The second coefficient calculation module is used to perform a multiplication operation on the basic deviation correction coefficient and the coefficient calibration value to obtain the cumulative deviation coefficient; The correction module is used to correct the oil fume concentration according to the cumulative deviation coefficient to obtain the effective oil fume concentration.

[0013] Optionally, the initial determination module includes: The time-series data acquisition module is used to acquire continuous sampling values ​​of the air pressure data within a preset time period to obtain air pressure time-series data; The frequency analysis module is used to perform frequency analysis on the air pressure time series data to obtain specific frequency fluctuation components caused by air pressure fluctuations. The comparison module is used to compare the amplitude of the specific frequency fluctuation component with a preset amplitude threshold. The modulation signal generation module is used to generate an exhaust volume modulation signal to offset air pressure fluctuations based on the frequency and phase of the specific frequency fluctuation component if the amplitude exceeds a preset amplitude threshold. The conversion module is used to convert the exhaust volume modulation signal into an exhaust volume compensation value. The synthesis module is used to synthesize the exhaust volume compensation value and the preset benchmark exhaust volume to obtain the initial exhaust volume.

[0014] Optionally, the adjustment module includes: The feature extraction module is used to extract time-frequency domain features from the airflow disturbance data to obtain a feature vector; The category determination module is used to input the feature vector into a pre-trained disturbance pattern classifier to obtain the category to which the current airflow disturbance belongs; the category includes a first category and a second category; the first category is a step pulse type caused by the opening and closing of doors and windows; the second category is a periodic sweeping type caused by people walking. The prediction module is used to predict the oil fume concentration by using historical effective oil fume concentration data and the effective oil fume concentration data in the future within a preset time period to obtain the predicted value of oil fume concentration. The differential adjustment module is used to perform differential adjustment on the predicted value of oil fume concentration according to the category to obtain dynamic compensation air volume; The real-time exhaust volume generation module is used to add the initial exhaust volume to the dynamic compensation air volume to obtain the real-time target exhaust volume.

[0015] Optionally, the differentiation adjustment module includes: The first condition module is used to calculate the concentration change by performing a difference calculation between the predicted oil fume concentration and the effective oil fume concentration data if the category is the first category and the predicted oil fume concentration is greater than a preset prediction value threshold; extract the instantaneous peak value of the airflow disturbance data to obtain the instantaneous peak amplitude; and multiply the instantaneous peak amplitude by the concentration change to obtain the dynamic compensation airflow. The second condition module is used to multiply the instantaneous peak amplitude by a preset compensation coefficient to obtain the dynamic compensation air volume if the category is the first category and the predicted value of the oil fume concentration is less than or equal to a preset predicted value threshold. The third condition module is used to extract the main frequency of the airflow disturbance data if the category is the second category, generate a sine wave compensation signal that is synchronized with the main frequency and opposite in phase, and obtain the dynamic compensation air volume by positively scaling the amplitude of the sine wave compensation signal according to the predicted value of the oil fume concentration.

[0016] The beneficial effects of this invention are as follows: This invention proposes an AI-based energy-saving exhaust control method for range hoods. By collecting real-time data on kitchen environment temperature, air pressure, airflow disturbance, and oil fume concentration, the method first compensates and corrects the oil fume concentration when the temperature is below a preset threshold to obtain an effective oil fume concentration. Then, based on air pressure data, it determines the initial exhaust volume and dynamically adjusts the initial exhaust volume by combining airflow disturbance data and the effective oil fume concentration to generate a real-time target exhaust volume. Finally, a preset airflow-fan speed mapping table is used to convert the target exhaust volume into a fan speed value, completing precise exhaust control of the range hood. This method introduces a compensation and correction mechanism for oil fume concentration collection under low-temperature conditions, combined with multi-parameter sensing and dynamic control, achieving accurate identification and intelligent response to oil fume generation. While ensuring effective oil fume extraction, it significantly reduces energy consumption, thereby achieving energy-saving exhaust control of the range hood. Attached Figure Description

[0017] The present invention will now be further described with reference to the accompanying drawings.

[0018] Figure 1 A flowchart illustrating an AI-based energy-saving exhaust control method for a range hood, provided as an embodiment of the present invention; Figure 2 This is a framework diagram of an AI-based energy-saving exhaust control system for range hoods, provided as an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0020] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] This invention provides an AI-based energy-saving exhaust control method for range hoods. See also... Figure 1 , Figure 1 A flowchart illustrating an AI-based energy-saving exhaust control method for a range hood, provided as an embodiment of the present invention. The method includes the following steps: S101 collects real-time data on temperature, air pressure, airflow disturbance, and oil fume concentration in the kitchen environment; S102, If the temperature data is less than the preset temperature threshold, the oil fume concentration is compensated and corrected according to the temperature data to obtain the effective oil fume concentration. S103, determine the initial exhaust volume based on air pressure data; S104, the initial exhaust volume is adjusted based on airflow disturbance data and effective oil fume concentration to obtain the real-time target exhaust volume; S105, the fan speed value is obtained by converting the real-time target exhaust volume through a preset air volume-fan speed value mapping table; S106 controls the exhaust of the range hood based on the fan speed.

[0022] This invention provides an AI-based energy-saving exhaust control method for range hoods. By collecting multi-dimensional data in the kitchen environment, including temperature, air pressure, airflow disturbance, and oil fume concentration in real time, the method automatically compensates and corrects the oil fume concentration when the temperature is detected to be below a set threshold, obtaining a true and effective oil fume concentration value. Subsequently, a base exhaust volume is set based on air pressure information, and dynamic adjustments are made by integrating airflow disturbance and the corrected oil fume concentration to generate a real-time target exhaust volume. Finally, using a preset airflow-speed correspondence, the target exhaust volume is converted into the actual fan speed, achieving precise control of the range hood. This method introduces distortion compensation for oil fume concentration acquisition in low-temperature environments and a multi-parameter collaborative control mechanism. Based on accurate identification of the oil fume emission status, it dynamically optimizes the exhaust strategy, significantly improving energy efficiency while ensuring effective smoke extraction, thus achieving energy-saving intelligent control of the range hood.

[0023] In one implementation, various types of dedicated sensors are deployed on the range hood body and key areas of the kitchen to collect various data in real time: an infrared non-contact temperature sensor is embedded around the air inlet of the range hood to obtain the temperature of the oil fume airflow in the exhaust duct and the overall ambient temperature of the kitchen; a miniature air pressure sensor is installed in the control cavity of the range hood to capture subtle fluctuations in the air pressure in the kitchen in real time; an ultrasonic airflow sensor is deployed at the air inlet grille of the range hood to accurately identify changes in airflow velocity and direction caused by factors such as opening and closing doors and windows and people walking, forming complete time-series data of airflow disturbance; at the same time, an electrochemical oil fume sensor is used, which is installed next to the oil fume circulation duct inside the air inlet grille of the range hood to obtain the oil fume concentration.

[0024] In one implementation, the preset air volume-fan speed value mapping table is a core parameter comparison table pre-calibrated by the range hood control system. Based on the fan's factory performance test and actual working condition calibration, it solidifies the optimal matching relationship between different target exhaust air volumes and corresponding fan speeds, and also reserves environmental factor correction levels. It can quickly convert the real-time target exhaust air volume into executable fan speed commands to achieve precise and efficient exhaust control.

[0025] In one embodiment, the effective oil fume concentration is obtained by compensating and correcting the oil fume concentration based on temperature data, including: The temperature difference is calculated by comparing the temperature data with the preset temperature threshold. via c=1+a×d b The basic deviation correction coefficient is calculated; where c is the basic deviation correction coefficient, a is the temperature influence weighting coefficient, d is the temperature difference, and b is the nonlinear correction index. The coefficient calibration value is obtained by calculating e=ln(x+g); where e is the coefficient calibration value, x is the oil fume concentration, and g is a constant greater than 1. The cumulative deviation coefficient is obtained by multiplying the basic deviation correction factor and the coefficient calibration value. The effective oil fume concentration is obtained by correcting the oil fume concentration based on the cumulative deviation coefficient.

[0026] In one implementation, the preset temperature threshold is set by technicians, and its value can be 20 degrees; a and b are obtained by fitting multiple sets of oil fume concentration detection experiments under low temperature conditions.

[0027] In one implementation, the formula c = 1 + a × d is used. bThe calculation of the basic deviation correction coefficient utilizes a nonlinear model to fit the influence of temperature on the concentration of cooking fumes. Because the temperature-induced interference with cooking fume concentration is not a linear relationship, the impact of condensation at low temperatures accelerates as the temperature decreases. A single linear correction cannot cover the complex effects of temperature deviation; therefore, a weighting coefficient 'a' and a nonlinear exponent 'b' are used to adapt to the error characteristics of different temperature ranges, thereby more accurately matching the correlation between temperature deviation and concentration error and improving the accuracy of the basic correction. The calibration value calculated using e=ln(x+g) is then adjusted based on the actual concentration of cooking fumes. The actual impact of temperature interference varies with different cooking fume concentrations; for example, the condensation error differs between high-concentration and low-concentration cooking fumes. A basic correction based solely on temperature does not consider the differences in cooking fume concentration, potentially leading to insufficient correction at high concentrations and overcorrection at low concentrations. This correction process simultaneously adapts to both temperature deviation and cooking fume concentration levels, making the calibration more closely aligned with actual operating conditions.

[0028] In one implementation, the effective oil fume concentration is calculated by first effective oil fume concentration = real-time collected oil fume concentration ÷ cumulative deviation coefficient.

[0029] In one embodiment, determining the initial exhaust volume based on air pressure data includes: By acquiring continuous sampling values ​​of air pressure data within a preset time period, air pressure time series data can be obtained. Frequency analysis of air pressure time series data yields specific frequency fluctuation components caused by air pressure fluctuations. The amplitude of a specific frequency fluctuation component is compared with a preset amplitude threshold. If the amplitude exceeds the preset amplitude threshold, an exhaust volume modulation signal is generated based on the frequency and phase of the specific frequency fluctuation component to counteract the air pressure fluctuation. The exhaust volume modulation signal is converted into an air volume to obtain the exhaust volume compensation value; The initial exhaust volume is obtained by combining the exhaust volume compensation value and the preset baseline exhaust volume.

[0030] In one implementation, the air pressure fluctuation is caused by gas resonance in the building flue or by external gusts of wind. The preset time period and preset amplitude threshold are set by technicians. The preset time period is determined based on the response delay of the range hood exhaust volume control, and its value can be 1 second. The preset amplitude threshold can be 0.01 kPa.

[0031] In one implementation, during the operation of the range hood, the periodic rotation of the fan impeller generates an airflow pulsation in the form of an approximately sinusoidal wave. The frequency of this pulsation is proportional to the fan speed, constituting an inherent periodic signal in the exhaust airflow. Against this backdrop, frequency analysis algorithms such as Fourier transform are used to perform spectral decomposition on the collected air pressure time-series data, converting the time-domain air pressure signal into a superposition of different frequency components in the frequency domain. Furthermore, specific frequency ranges corresponding to the resonance of building flue gas and the impact of external gusts are selected from the spectrum. For example, the typical frequency range of flue resonance is 2-5Hz, and the typical frequency range of gust impact is 0.5-2Hz. The fluctuation components within these ranges are extracted to obtain the specific frequency fluctuation components generated by air pressure fluctuations, and the frequency, phase, and amplitude information of these components are determined.

[0032] In one implementation, based on the principle of fluctuation cancellation, a modulation signal with the same frequency and opposite phase as the specific frequency fluctuation component is generated. Specifically, the frequency of the modulation signal is the same as the frequency of the air pressure fluctuation component, and the phase difference is π. At the same time, the amplitude of the modulation signal needs to match the interference level of the air pressure fluctuation. For example, when the air pressure fluctuation amplitude is 0.3 kPa, the amplitude of the modulation signal corresponds to an adjustment range of 15% of the baseline exhaust volume, ultimately forming an exhaust volume modulation signal that can cancel the air pressure fluctuation.

[0033] In one implementation, the actual pot projection area and fire intensity level corresponding to the current stove area are first analyzed and identified based on the collected airflow disturbance data. Then, by querying the pre-configured mapping relationship table of "pot projection area - fire intensity level - exhaust volume", the minimum required exhaust volume corresponding to the current identification result is matched from the table. This exhaust volume is the preset benchmark exhaust volume.

[0034] In one embodiment, adjusting the initial exhaust volume based on airflow disturbance data and effective oil fume concentration to obtain the real-time target exhaust volume includes: Feature vectors are obtained by performing time-frequency domain feature extraction on airflow disturbance data; The feature vector is input into a pre-trained disturbance pattern classifier to obtain the category to which the current airflow disturbance belongs; the category includes a first category and a second category; the first category is a step pulse type caused by the opening and closing of doors and windows; the second category is a periodic sweep type caused by people walking. The predicted value of oil fume concentration is obtained by predicting the oil fume concentration in the future within a preset time period using historical effective oil fume concentration data and effective oil fume concentration data. The predicted fume concentration is adjusted differently according to the category to obtain the dynamic compensation air volume; The initial exhaust volume is added to the dynamically compensated air volume to obtain the real-time target exhaust volume.

[0035] In one implementation, the airflow disturbance data collected by the range hood is subjected to time-frequency domain feature extraction using methods such as short-time Fourier transform. The dominant frequency is determined by the frequency corresponding to the peak value of the spectrum. For example, the dominant frequency of step pulse-type disturbances caused by opening and closing doors and windows is relatively high, while the dominant frequency of periodic scanning-type disturbances caused by people walking is relatively low. The energy spectral density distribution is obtained by calculating the energy proportion in different frequency intervals, and the number of abrupt changes is obtained by counting the number of sudden changes in flow velocity or flow direction in the data. The extracted dominant frequency, energy spectral density distribution, and number of abrupt changes are arranged according to a preset dimension to form a feature vector containing time-frequency domain information, providing a data foundation for subsequent classification.

[0036] In one implementation, the pre-trained disturbance pattern classifier is trained based on support vector machines or neural networks. The training data includes labeled samples of two types of disturbances: door and window opening / closing - step pulse type and people walking - periodic scanning type. The classifier compares the feature distribution of the input feature vector with that of the training samples to determine the feature matching degree of the current airflow disturbance. If the feature vector has few abrupt change points, concentrated main frequency, and pulse-like energy spectrum, it is classified as the first category. If the feature vector contains periodic frequency components and the energy spectrum is continuously scanning, it is classified as the second category. Finally, the class result of the current airflow disturbance is output.

[0037] In one implementation, historical effective oil fume concentration data, including time-series records of concentration under different operating conditions, is first compiled. Combined with the currently obtained real-time effective oil fume concentration data, a time series prediction algorithm (LSTM) model is used for prediction. The historical effective oil fume concentration data is used as the model input to learn the pattern of concentration change over time, such as the rapid increase in concentration during stir-frying and the stable concentration during stewing. Based on the current trend of effective concentration, the changes in oil fume concentration within a preset time period are predicted, and finally, the predicted value of oil fume concentration is output, providing a forward-looking basis for subsequent air volume compensation.

[0038] In one embodiment, the dynamic compensation airflow is obtained by differentially adjusting the predicted oil fume concentration based on the category, including: If the category is the first category and the predicted value of the oil fume concentration is greater than the preset predicted value threshold, the difference between the predicted value of the oil fume concentration and the effective oil fume concentration data is calculated to obtain the concentration change, the instantaneous peak value is extracted from the airflow disturbance data to obtain the instantaneous peak amplitude, and the instantaneous peak amplitude is multiplied by the concentration change to obtain the dynamic compensation air volume. If the category is Category 1 and the predicted value of the oil fume concentration is less than or equal to the preset predicted value threshold, then the instantaneous peak amplitude is multiplied by the preset compensation coefficient to obtain the dynamic compensation air volume. If the category is the second category, the main frequency is extracted from the airflow disturbance data to generate a sine wave compensation signal that is synchronized with the main frequency but opposite in phase. The amplitude of the sine wave compensation signal is positively scaled according to the predicted value of oil fume concentration to obtain the dynamic compensation air volume.

[0039] In one implementation, the preset prediction threshold is an empirical value determined by a comprehensive analysis of the accuracy range of the oil fume concentration sensor, the physical characteristics of oil fume diffusion, and the exhaust efficiency of the range hood; the preset compensation coefficient is an empirical coefficient determined based on measured data, used to convert the instantaneous peak amplitude of airflow disturbance into a conservative compensation air volume that matches the current level of interference in the kitchen environment when the predicted oil fume concentration is low.

[0040] In one implementation, the relative magnitude of the impending increase in oil fume is first quantified by the difference between the predicted concentration and the current effective concentration, resulting in a dimensionless concentration change coefficient. Then, the instantaneous peak amplitude of the airflow disturbance is extracted and mapped to an airflow impact compensation base reflecting the impact of door and window opening and closing using a preset intensity-airflow conversion coefficient. This base has airflow dimensions. The concentration change coefficient is multiplied by the airflow impact compensation base to obtain the comprehensive compensation airflow. This result reflects both the demand ratio of oil fume changes and matches the absolute intensity of airflow disturbance. Through this process, when the predicted oil fume concentration exceeds a threshold, the system simultaneously responds to the impending large increase in oil fume and the accompanying airflow impact, achieving precise airflow compensation that balances the oil fume diffusion trend and instantaneous airflow disturbance. This avoids insufficient compensation or energy waste that may result from adjusting solely based on concentration or airflow, making the dynamic compensation airflow more closely match the actual risks of oil fume diffusion and interference.

[0041] In one implementation, when the predicted value of the oil fume concentration does not exceed the preset predicted value threshold, it indicates that the increase in oil fume generation is within the controllable range of the background. At this time, based on the instantaneous peak amplitude of the step pulse event in the airflow disturbance data, it is converted into an air volume compensation coefficient through a predefined intensity-compensation mapping function, and then multiplied by a preset benchmark compensation air volume unit to directly output the dynamic compensation air volume, thereby simplifying the calculation and maintaining the stability of the airflow in the capture area in the low oil fume increment scenario.

[0042] In one implementation, when the category is the second category, the dominant frequency of the airflow disturbance data is extracted to generate a standard sine wave compensation signal that is synchronized with the dominant frequency but out of phase. Then, based on the predicted value of the oil fume concentration, the corresponding amplitude scaling gain coefficient is determined according to a preset concentration-gain mapping function. This coefficient is a dimensionless proportional value. The amplitude scaling gain coefficient is multiplied by a reference compensation amplitude with a unit of air volume to obtain the scaled compensation amplitude. The instantaneous value of the standard sine wave compensation signal is multiplied by the scaled compensation amplitude to obtain a dynamic compensation air volume that varies with time and whose amplitude adapts to the oil fume concentration. The higher the predicted value of the oil fume concentration, the larger the gain coefficient, the larger the scaled compensation amplitude, and the stronger the dynamic compensation air volume. The lower the predicted value of the oil fume concentration, the smaller the gain coefficient, the smaller the scaled compensation amplitude, and the weaker the dynamic compensation air volume. This approach is adopted because airflow interference caused by personnel movement is periodic, and its fluctuations need to be offset by signals of the same frequency but opposite phase. Furthermore, the absolute magnitude of the compensation airflow required varies depending on the concentration of cooking fumes. High-concentration scenarios require a larger compensation airflow to maintain effective capture. Therefore, this process, by separating the periodic waveform and amplitude modulation, can not only accurately offset the impact of periodic airflow fluctuations, but also adapt the amplitude of the compensation airflow to the current level of cooking fumes, thus balancing anti-interference and exhaust efficiency.

[0043] Based on the same inventive concept, this invention also provides an AI-based energy-saving exhaust control system for range hoods. See also... Figure 2 , Figure 2 A framework diagram of an AI-based energy-saving exhaust control system for range hoods, provided for embodiments of the present invention, includes: The data acquisition module is used to collect real-time data on temperature, air pressure, airflow disturbance, and oil fume concentration in the kitchen environment. The condition module is used to compensate and correct the oil fume concentration based on the temperature data if the temperature data is less than the preset temperature threshold, so as to obtain the effective oil fume concentration. The initial determination module is used to determine the initial exhaust volume based on the air pressure data; The adjustment module is used to adjust the initial exhaust volume based on airflow disturbance data and effective oil fume concentration to obtain the real-time target exhaust volume; The conversion module is used to convert the real-time target exhaust volume into the fan speed value by using a preset air volume-fan speed value mapping table; The control module is used to control the exhaust of the range hood based on the fan speed.

[0044] This invention provides an AI-based energy-saving exhaust control system for range hoods. By collecting multi-dimensional data in the kitchen environment, including temperature, air pressure, airflow disturbance, and oil fume concentration, the system automatically compensates for and corrects the oil fume concentration when the temperature is detected to be below a set threshold, obtaining a true and effective oil fume concentration value. Subsequently, a base exhaust volume is set based on air pressure information, and dynamic adjustments are made by considering airflow disturbance and the corrected oil fume concentration to generate a real-time target exhaust volume. Finally, using a preset airflow-speed correspondence, the target exhaust volume is converted into the actual fan speed, achieving precise control of the range hood. This method introduces distortion compensation for oil fume concentration acquisition under low-temperature conditions and a multi-parameter collaborative control mechanism. Based on accurate identification of the oil fume emission status, it dynamically optimizes the exhaust strategy, significantly improving energy efficiency while ensuring effective smoke extraction, thus achieving energy-saving intelligent control of the range hood.

[0045] In one embodiment, the condition module includes: The difference calculation module is used to calculate the temperature difference between the temperature data and the preset temperature threshold. The first coefficient calculation module is used to calculate c = 1 + a × d. b The basic deviation correction coefficient is calculated; where c is the basic deviation correction coefficient, a is the temperature influence weighting coefficient, d is the temperature difference, and b is the nonlinear correction index. The calibration value calculation module is used to calculate the coefficient calibration value using the formula e=ln(x+g); where e is the coefficient calibration value, x is the oil fume concentration, and g is a constant greater than 1. The second coefficient calculation module is used to perform multiplication operations on the basic deviation correction coefficient and the coefficient calibration value to obtain the cumulative deviation coefficient; The correction module is used to correct the oil fume concentration based on the cumulative deviation coefficient to obtain the effective oil fume concentration.

[0046] In one embodiment, the initial determination module includes: The time-series data acquisition module is used to obtain the continuous sampling values ​​of air pressure data within a preset time period to obtain air pressure time-series data; The frequency analysis module is used to perform frequency analysis on air pressure time series data to obtain specific frequency fluctuation components caused by air pressure fluctuations. The comparison module is used to compare the amplitude of a specific frequency fluctuation component with a preset amplitude threshold. The modulation signal generation module is used to generate an exhaust volume modulation signal to offset air pressure fluctuations based on the frequency and phase of a specific frequency fluctuation component if the amplitude exceeds a preset amplitude threshold. The conversion module is used to convert the exhaust volume modulation signal into an exhaust volume compensation value. The synthesis module is used to synthesize the exhaust volume compensation value and the preset baseline exhaust volume to obtain the initial exhaust volume.

[0047] In one embodiment, the adjustment module includes: The feature extraction module is used to extract time-frequency domain features from airflow disturbance data to obtain feature vectors; The category determination module is used to input the feature vector into a pre-trained disturbance pattern classifier to obtain the category to which the current airflow disturbance belongs; the category includes a first category and a second category; the first category is a step pulse type caused by the opening and closing of doors and windows; the second category is a periodic sweeping type caused by people walking. The prediction module is used to predict the oil fume concentration by using historical effective oil fume concentration data and effective oil fume concentration data for a preset time period in the future. The differential adjustment module is used to dynamically compensate the air volume by differentially adjusting the predicted value of oil fume concentration according to the category. The real-time exhaust volume generation module is used to add the initial exhaust volume to the dynamically compensated air volume to obtain the real-time target exhaust volume.

[0048] In one embodiment, the differentiation adjustment module includes: The first condition module is used to calculate the difference between the predicted oil fume concentration and the effective oil fume concentration data to obtain the concentration change if the category is the first category and the predicted oil fume concentration is greater than the preset prediction value threshold. It also extracts the instantaneous peak value of the airflow disturbance data to obtain the instantaneous peak value amplitude and multiplies the instantaneous peak value amplitude with the concentration change to obtain the dynamic compensation air volume. The second condition module is used to obtain the dynamic compensation air volume by multiplying the instantaneous peak amplitude by the preset compensation coefficient if the category is the first category and the predicted value of the oil fume concentration is less than or equal to the preset predicted value threshold. The third condition module is used to extract the main frequency of the airflow disturbance data if the category is the second category, generate a sine wave compensation signal that is synchronized with the main frequency and opposite in phase, and obtain the dynamic compensation air volume by positively scaling the amplitude of the sine wave compensation signal according to the predicted value of oil fume concentration.

[0049] The foregoing has described one embodiment of the present invention in detail, but this content is merely a preferred embodiment and should not be considered as limiting the scope of the present invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the scope of the claims of this invention.

Claims

1. An AI-based energy-saving exhaust control method for range hoods, characterized in that, The method includes: Real-time collection of temperature, air pressure, airflow disturbance, and oil fume concentration data in the kitchen environment; If the temperature data is less than a preset temperature threshold, the oil fume concentration is compensated and corrected according to the temperature data to obtain an effective oil fume concentration; Determine the initial exhaust volume based on the air pressure data; The initial exhaust volume is adjusted based on the airflow disturbance data and the effective oil fume concentration to obtain the real-time target exhaust volume; The fan speed value is obtained by converting the real-time target exhaust volume using a preset air volume-fan speed value mapping table; The exhaust fan is controlled based on the fan speed value.

2. The AI-based energy-saving exhaust control method for range hoods according to claim 1, characterized in that, The effective oil fume concentration is obtained by compensating and correcting the oil fume concentration based on the temperature data, including: The temperature difference is obtained by calculating the difference between the temperature data and the preset temperature threshold; via c=1+a×d b The basic deviation correction coefficient is calculated; where c is the basic deviation correction coefficient, a is the temperature influence weighting coefficient, d is the temperature difference, and b is the nonlinear correction index. The coefficient calibration value is obtained by calculating e=ln(x+g); where e is the coefficient calibration value, x is the oil fume concentration, and g is a constant greater than 1. The cumulative deviation coefficient is obtained by multiplying the basic deviation correction coefficient and the coefficient calibration value. The effective oil fume concentration is obtained by correcting the oil fume concentration based on the cumulative deviation coefficient.

3. The AI-based energy-saving exhaust control method for range hoods according to claim 1, characterized in that, Determining the initial exhaust volume based on the aforementioned air pressure data includes: By acquiring continuous sampling values ​​of the air pressure data within a preset time period, air pressure time-series data can be obtained; Frequency analysis is performed on the aforementioned air pressure time series data to obtain specific frequency fluctuation components caused by air pressure fluctuations; The amplitude of the specific frequency fluctuation component is compared with a preset amplitude threshold. If the amplitude exceeds a preset amplitude threshold, an exhaust volume modulation signal is generated based on the frequency and phase of the specific frequency fluctuation component to counteract air pressure fluctuations. The exhaust volume modulation signal is converted into an exhaust volume compensation value. The initial exhaust volume is obtained by combining the exhaust volume compensation value and the preset benchmark exhaust volume.

4. The AI-based energy-saving exhaust control method for range hoods according to claim 1, characterized in that, The real-time target exhaust volume is obtained by adjusting the initial exhaust volume based on the airflow disturbance data and the effective oil fume concentration, including: The airflow disturbance data is subjected to time-frequency domain feature extraction to obtain a feature vector; The feature vector is input into a pre-trained disturbance pattern classifier to obtain the category to which the current airflow disturbance belongs; the category includes a first category and a second category; the first category is a step pulse type caused by the opening and closing of doors and windows; the second category is a periodic sweeping type caused by people walking; The predicted value of oil fume concentration is obtained by predicting the oil fume concentration in the future within a preset time period using historical effective oil fume concentration data and the effective oil fume concentration data. The predicted oil fume concentration is adjusted differentially according to the category to obtain the dynamic compensation air volume; The initial exhaust volume is added to the dynamic compensation air volume to obtain the real-time target exhaust volume.

5. The AI-based energy-saving exhaust control method for range hoods according to claim 4, characterized in that, The dynamically compensated airflow is obtained by differentially adjusting the predicted oil fume concentration value according to the category, including: If the category is the first category and the predicted value of the oil fume concentration is greater than the preset predicted value threshold, then the difference between the predicted value of the oil fume concentration and the effective oil fume concentration data is calculated to obtain the concentration change, the instantaneous peak value is extracted from the airflow disturbance data to obtain the instantaneous peak amplitude, and the instantaneous peak amplitude is multiplied by the concentration change to obtain the dynamic compensation air volume. If the category is the first category, and the predicted value of the oil fume concentration is less than or equal to the preset predicted value threshold, then the instantaneous peak amplitude is multiplied by the preset compensation coefficient to obtain the dynamic compensation air volume; If the category is the second category, the main frequency is extracted from the airflow disturbance data to generate a sine wave compensation signal that is synchronized with the main frequency but opposite in phase. The amplitude of the sine wave compensation signal is positively scaled according to the predicted value of the oil fume concentration to obtain the dynamic compensation air volume.

6. An AI-based energy-saving exhaust control system for range hoods, characterized in that, The system includes: The data acquisition module is used to collect real-time data on temperature, air pressure, airflow disturbance, and oil fume concentration in the kitchen environment. The condition module is used to compensate and correct the oil fume concentration based on the temperature data to obtain an effective oil fume concentration if the temperature data is less than a preset temperature threshold. The initial determination module is used to determine the initial exhaust volume based on the air pressure data; An adjustment module is used to adjust the initial exhaust volume according to the airflow disturbance data and the effective oil fume concentration to obtain the real-time target exhaust volume; The conversion module is used to convert the real-time target exhaust volume into a fan speed value by using a preset air volume-fan speed value mapping table; The control module is used to control the exhaust of the range hood according to the fan speed value.

7. The AI-based energy-saving exhaust control system for a range hood according to claim 6, characterized in that, The condition module includes: The difference calculation module is used to calculate the temperature difference between the temperature data and the preset temperature threshold. The first coefficient calculation module is used to calculate c = 1 + a × d. b The basic deviation correction coefficient is calculated; where c is the basic deviation correction coefficient, a is the temperature influence weighting coefficient, d is the temperature difference, and b is the nonlinear correction index. The calibration value calculation module is used to calculate the coefficient calibration value using the formula e=ln(x+g); where e is the coefficient calibration value, x is the oil fume concentration, and g is a constant greater than 1. The second coefficient calculation module is used to perform a multiplication operation on the basic deviation correction coefficient and the coefficient calibration value to obtain the cumulative deviation coefficient; The correction module is used to correct the oil fume concentration according to the cumulative deviation coefficient to obtain the effective oil fume concentration.

8. The AI-based energy-saving exhaust control system for a range hood according to claim 6, characterized in that, The initial determination module includes: The time-series data acquisition module is used to acquire continuous sampling values ​​of the air pressure data within a preset time period to obtain air pressure time-series data; The frequency analysis module is used to perform frequency analysis on the air pressure time series data to obtain specific frequency fluctuation components caused by air pressure fluctuations. The comparison module is used to compare the amplitude of the specific frequency fluctuation component with a preset amplitude threshold. The modulation signal generation module is used to generate an exhaust volume modulation signal to offset air pressure fluctuations based on the frequency and phase of the specific frequency fluctuation component if the amplitude exceeds a preset amplitude threshold. The conversion module is used to convert the exhaust volume modulation signal into an exhaust volume compensation value. The synthesis module is used to synthesize the exhaust volume compensation value and the preset benchmark exhaust volume to obtain the initial exhaust volume.

9. The AI-based energy-saving exhaust control system for a range hood according to claim 6, characterized in that, The adjustment module includes: The feature extraction module is used to extract time-frequency domain features from the airflow disturbance data to obtain a feature vector; The category determination module is used to input the feature vector into a pre-trained disturbance pattern classifier to obtain the category to which the current airflow disturbance belongs; the category includes a first category and a second category; the first category is a step pulse type caused by the opening and closing of doors and windows; the second category is a periodic sweeping type caused by people walking. The prediction module is used to predict the oil fume concentration by using historical effective oil fume concentration data and the effective oil fume concentration data in the future within a preset time period to obtain the predicted value of oil fume concentration. The differential adjustment module is used to perform differential adjustment on the predicted value of oil fume concentration according to the category to obtain dynamic compensation air volume; The real-time exhaust volume generation module is used to add the initial exhaust volume to the dynamic compensation air volume to obtain the real-time target exhaust volume.

10. The AI-based energy-saving exhaust control system for a range hood according to claim 9, characterized in that, The differential adjustment module includes: The first condition module is used to calculate the concentration change by performing a difference calculation between the predicted oil fume concentration and the effective oil fume concentration data if the category is the first category and the predicted oil fume concentration is greater than a preset prediction value threshold; extract the instantaneous peak value of the airflow disturbance data to obtain the instantaneous peak amplitude; and multiply the instantaneous peak amplitude by the concentration change to obtain the dynamic compensation airflow. The second condition module is used to multiply the instantaneous peak amplitude by a preset compensation coefficient to obtain the dynamic compensation air volume if the category is the first category and the predicted value of the oil fume concentration is less than or equal to a preset predicted value threshold. The third condition module is used to extract the main frequency of the airflow disturbance data if the category is the second category, generate a sine wave compensation signal that is synchronized with the main frequency and opposite in phase, and obtain the dynamic compensation air volume by positively scaling the amplitude of the sine wave compensation signal according to the predicted value of the oil fume concentration.

Images