Real-time concentration monitoring and intelligent adjusting method and system for oil smoke

Abstract

The invention provides an oil smoke real-time concentration monitoring and intelligent adjusting method and system, and relates to the technical field of intelligent control, and the method comprises the steps: collecting real-time air quality data of a catering area through a plurality of environment sensors, including oil smoke concentration, PM2.5 concentration, carbon monoxide concentration, temperature and humidity; and the control center calculates an oil smoke concentration spatial and temporal distribution field based on real-time data in combination with spatial interpolation, considers smoke exhaust energy consumption, smoke exhaust efficiency and regional oil smoke concentration, and determines an optimal cooperative control strategy by utilizing a multi-target optimization algorithm, and the optimal cooperative control strategy comprises the fan rotating speed of each range hood, the smoke exhaust direction and the air supply parameters of the fresh air system. And the control center issues the control strategy to a range hood controller to adjust operating parameters of the range hood. Dynamic optimization control over the air quality of the catering area can be achieved, the oil smoke concentration is reduced, energy is saved, and the smoke exhaust efficiency is improved.

Description

Technical Field

[0001] The present invention relates to intelligent regulation technology, and in particular to a method and system for real-time oil smoke concentration monitoring and intelligent regulation. Background Art

[0002] Cooking fumes are one of the main air pollutants generated by the catering industry. They are complex in composition and are harmful to human health and the environment. Traditional fume purification methods mainly rely on the filter devices installed on each range hood for purification, and then discharge the purified gas to the outdoors through the exhaust duct. At the same time, in order to ensure the air quality in the catering area, a fresh air system is usually installed for ventilation.

[0003] The prior art has the following defects and deficiencies: 1. Most traditional oil fume purification systems are single-point controlled, with each range hood working independently. There is a lack of overall consideration of the oil fume concentration in the entire catering area, resulting in excessive oil fume concentration in some areas, low purification efficiency, and serious energy waste.

[0004] 2. Existing oil fume purification systems usually use fixed parameter control and cannot dynamically adjust the operating parameters of the range hood according to the actual oil fume concentration, such as fan speed, exhaust direction, etc. It is difficult to adapt to the dynamic changes in oil fume concentration in the catering area, and thus cannot achieve the optimal purification effect.

[0005] 3. The traditional oil fume purification system lacks consideration for the coordinated control of the fresh air system and the range hood. The operating parameters of the fresh air system and the range hood are independent of each other, and it is impossible to achieve overall optimization control of the air quality in the catering area, resulting in energy waste and low purification efficiency. Summary of the invention

[0006] The embodiments of the present invention provide a method and system for real-time oil smoke concentration monitoring and intelligent regulation, which can solve the problems in the prior art.

[0007] According to a first aspect of the embodiments of the present invention, Provides real-time smoke concentration monitoring and intelligent adjustment methods, including: Real-time air quality data in the dining area is collected by using environmental sensors installed in multiple range hoods in the dining area, wherein the environmental sensors include a fume sensor for detecting fume concentration, a PM2.5 sensor for detecting particulate matter concentration, a carbon monoxide sensor for detecting carbon monoxide concentration, a temperature sensor for detecting temperature, and a humidity sensor for detecting humidity; the real-time air quality data is transmitted to a control center via a wireless communication network; The control center determines the initial estimated value of the pollutant concentration based on the real-time air quality data combined with spatial interpolation, calculates the dynamic distribution of the pollutant concentration at any point in the dining area at any time according to the spatial distance attenuation characteristics between environmental sensors, and combines the dynamic distribution of the pollutant concentration with the building structure information and ventilation system layout information of the dining area to calculate the spatiotemporal distribution field of the oil fume concentration in the dining area; According to the spatiotemporal distribution field of oil fume concentration, the exhaust energy consumption and exhaust efficiency of each range hood are taken as optimization targets, and the regional oil fume concentration value in the spatiotemporal distribution field of oil fume concentration is considered. The optimization target is solved in combination with a multi-objective optimization algorithm to determine the optimal collaborative control strategy, which includes the fan speed, exhaust direction and air supply parameters of each range hood and the fresh air system. The control center sends the optimal collaborative control strategy to the controllers of each range hood, and the controllers of each range hood adjust the operating parameters of the corresponding range hood to achieve dynamic optimization control of the air quality in the dining area.

[0008] The control center determines the initial estimated value of the pollutant concentration based on the real-time air quality data combined with spatial interpolation, and calculates the dynamic distribution of the pollutant concentration at any point in the dining area at any time according to the spatial distance attenuation characteristics between environmental sensors, including: Performing outlier elimination and data standardization processing on the real-time air quality data to obtain pre-processed air quality data, based on the pre-processed air quality data, taking the overall pollutant distribution trend of the catering area as a deterministic component and the local pollutant concentration fluctuation as a random component, and calculating the initial estimated value of the pollutant concentration at any location through spatial correlation analysis; Calculate the Euclidean distance between environmental sensors in the catering area, substitute the Euclidean distance into the Gaussian attenuation function to obtain the distance attenuation coefficient, and simultaneously calculate the relative azimuth angles between the environmental sensors to obtain the azimuth weight coefficient, and multiply the distance attenuation coefficient by the azimuth weight coefficient to obtain the spatial correlation weight between the detection points; calculate the Gaussian kernel function value of the time dimension according to the time series data of the environmental sensor, multiply the spatial correlation weight by the Gaussian kernel function value, and obtain the pollutant concentration value of any position point in the catering area at any time through weighted superposition calculation.

[0009] Taking the overall pollutant distribution trend of the catering area as the deterministic component and the local pollutant concentration fluctuation as the random component, the initial estimated value of the pollutant concentration at any location is calculated through spatial correlation analysis, including: Substituting the pollutant concentration observation data into a polynomial fitting function, the polynomial fitting function includes a constant term, a linear term coefficient, and a quadratic term coefficient, and calculating a deterministic trend term that characterizes the overall distribution trend based on the polynomial fitting function; taking the difference between the pollutant concentration observation data and the deterministic trend term as a random fluctuation term, calculating the spatial variability and spatial covariance of the random fluctuation term, and obtaining a random component that characterizes the local fluctuation characteristics; Based on the pollutant concentration observation data, the semivariogram function values ​​between the sample point pairs with different spatial interval distances are calculated, and the semivariogram function values ​​are fitted with the theoretical spherical model to obtain the nugget value, the sill value and the range parameter, wherein the nugget value represents the small-scale variation characteristics, the sill value represents the overall variation degree, and the range parameter represents the scope of spatial correlation influence; and a spatial structure characteristic equation group is constructed according to the nugget value, the sill value and the range parameter; The spatial structure characteristic equation group is combined with the unbiased estimation constraint condition to construct the Kriging weight equation group, and the Kriging weight equation group is solved to obtain the optimal weight coefficient of the observation point; the optimal weight coefficient is weightedly combined with the pollutant concentration observation data to calculate the initial estimate of the pollutant concentration at any location point.

[0010] The dynamic distribution of pollutant concentration is combined with the building structure information and ventilation system layout information of the dining area to calculate the optimal coordinated control strategy of multiple range hoods. The optimal coordinated control strategy includes the fan speed, exhaust direction and air supply parameters of each range hood, including: Acquire the building structure information and ventilation system layout information of the dining area, wherein the building structure information includes the wall position, door and window distribution and indoor partitions, and the ventilation system layout information includes the range hood spatial coordinates, installation angle and exhaust pipe direction; construct a spatial characteristic function based on the building structure information, convert the ventilation system layout information into a ventilation equipment layout function, and fuse the spatial characteristic function with the ventilation equipment layout function to obtain a ventilation environment characteristic model of the dining area; Based on the ventilation environment characteristic model, a flow field motion equation including velocity field, pressure field and external force terms, and a pollutant transport equation including concentration field, diffusion coefficient and source term function are established; the flow field motion equation and the pollutant transport equation are solved jointly to obtain the oil fume diffusion dynamics characteristics, and the spatiotemporal distribution field of the oil fume concentration in the catering area is calculated based on the oil fume diffusion dynamics characteristics.

[0011] Based on the ventilation environment characteristic model, a flow field motion equation including velocity field, pressure field and external force terms, and a pollutant transport equation including concentration field, diffusion coefficient and source term function are established; the flow field motion equation and the pollutant transport equation are solved simultaneously to obtain the fume diffusion dynamics characteristics, and the spatiotemporal distribution field of fume concentration in the dining area is calculated according to the fume diffusion dynamics characteristics, including: The flow field motion equation is shown in the following formula: ; Among them, ρ represents the density of the oil smoke mixed gas, represents the velocity field vector, represents the partial derivative of velocity with respect to time, which is the local acceleration, The advection term representing the velocity, the acceleration caused by the fluid motion itself, represents the pressure gradient, is the spatial rate of change of the pressure field, μ represents the dynamic viscosity coefficient, The Laplace operator representing the velocity, represents the external force field vector; The pollutant transport equation is shown in the following formula: ; in, C represents the pollutant concentration field, represents the partial derivative of concentration with respect to time, represents the Laplace operator of concentration, D represents the diffusion coefficient tensor, It represents the second-order derivative of concentration in space, S(x,y,z,t) represents the source function, which describes the rate at which pollutants are generated or disappear. x,y,z represent the spatial coordinates, t represents time, and R(C,T) is the chemical reaction term, which describes the rate at which pollutants participate in chemical reactions and is a function of concentration C and temperature T.

[0012] According to the spatiotemporal distribution field of oil fume concentration, the exhaust energy consumption and exhaust efficiency of each range hood are taken as optimization targets, and the regional oil fume concentration value in the spatiotemporal distribution field of oil fume concentration is considered. The optimization target is solved by combining a multi-objective optimization algorithm, and the optimal collaborative control strategy is determined to include: Construct a total energy consumption objective function for the range hood group, wherein the total energy consumption objective function includes a cubic coefficient and a quadratic coefficient of the fan speed of each range hood, wherein the cubic coefficient represents the aerodynamic loss of the fan, and the quadratic coefficient represents the mechanical friction loss of the fan; obtain the fume concentration and volume flow rate at the inlet and outlet of each range hood, calculate the overall exhaust efficiency of the range hood group, and obtain the exhaust efficiency objective function; determine the maximum fume concentration value of the region according to the spatiotemporal distribution data of the fume concentration; A multi-objective optimization function vector is established based on the total energy consumption objective function and the smoke exhaust efficiency objective function, and each objective function in the multi-objective optimization function vector is divided by a corresponding normalized reference value to obtain a normalized objective function; a fan speed range, a smoke exhaust angle range, and a power consumption range of each range hood are set as equipment operation constraints, and a limited range of the maximum smoke concentration value in the area is set as a concentration control constraint; The normalized objective function is weighted and summed using an adaptive weight coefficient to obtain a comprehensive objective function, wherein the adaptive weight coefficient is dynamically adjusted based on the gradient information of each objective function to ensure the dynamic balance of the weights of each objective function during the optimization process of the comprehensive objective function; the equipment operation constraint condition and the concentration control constraint condition are substituted into the comprehensive objective function, and the comprehensive objective function is iteratively optimized using a gradient descent method; The dominance index of the optimization solution is calculated, and the dominance index characterizes the distance distribution from the optimization solution to each target ideal value through an exponential function. The Pareto optimal solution set is screened based on the dominance index; the Pareto optimal solution set is converted into a control strategy for the range hood group, and the control strategy includes the fan speed and exhaust direction of each range hood and the air supply parameters of the fresh air system.

[0013] The dominance index of the optimization solution is calculated. The dominance index characterizes the distance distribution from the optimization solution to each target ideal value through an exponential function. The Pareto optimal solution set obtained by screening based on the dominance index includes: Each objective function is independently optimized to obtain an ideal value, and the ideal value is constructed as an ideal point vector; the weighted Euclidean distance and Chebyshev distance from the solution of each objective function to the ideal point vector are calculated, and the weighted Euclidean distance and the Chebyshev distance respectively form exponential function terms with corresponding scale parameters, and the sum of the two exponential function terms is used as the basic advantage function; The sum of squares of constraint violations is used as a constraint penalty term, and the constraint penalty term and the penalty factor form an exponential function; the product of the basic advantage function and the exponential function of the constraint penalty term is used as a comprehensive advantage index; Based on the dominance relationship between the comprehensive dominance index and each objective function value, non-dominated solutions are screened out to constitute the Pareto optimal solution set. The dominance relationship requires that the comprehensive dominance index of the dominating solution is greater than the comprehensive dominance index of the dominated solution, and all objective function values ​​of the dominating solution are not greater than the corresponding objective function values ​​of the dominated solution.

[0014] A second aspect of an embodiment of the present invention provides a real-time smoke concentration monitoring and intelligent regulation system, comprising: The first unit is used to collect real-time air quality data in the dining area through environmental sensors installed in multiple range hoods in the dining area, wherein the environmental sensors include a fume sensor for detecting fume concentration, a PM2.5 sensor for detecting particulate matter concentration, a carbon monoxide sensor for detecting carbon monoxide concentration, a temperature sensor for detecting temperature, and a humidity sensor for detecting humidity; and transmit the real-time air quality data to a control center through a wireless communication network; The second unit is used for the control center to determine the initial estimated value of the pollutant concentration based on the real-time air quality data combined with spatial interpolation, calculate the dynamic distribution of the pollutant concentration at any point in the catering area at any time according to the spatial distance attenuation characteristics between environmental sensors, and combine the dynamic distribution of the pollutant concentration with the building structure information and ventilation system layout information of the catering area to calculate the spatiotemporal distribution field of the oil fume concentration in the catering area; The third unit is used to take the exhaust energy consumption and exhaust efficiency of each range hood as the optimization target according to the spatiotemporal distribution field of the oil fume concentration, and consider the regional oil fume concentration value in the spatiotemporal distribution field of the oil fume concentration, and solve the optimization target in combination with the multi-objective optimization algorithm to determine the optimal collaborative control strategy, which includes the fan speed, exhaust direction and air supply parameters of each range hood and the fresh air system; the control center sends the optimal collaborative control strategy to the controllers of each range hood, and the controllers of each range hood adjust the operating parameters of the corresponding range hood to achieve dynamic optimization control of the air quality in the dining area.

[0015] According to a third aspect of the embodiments of the present invention, An electronic device is provided, comprising: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0016] A fourth aspect of the embodiments of the present invention is: A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0017] The beneficial effects of this application are as follows: 1. Real-time monitoring and precise control: This method can monitor the concentrations of pollutants such as oil smoke, PM2.5, and carbon monoxide in the catering area in real time, and combine spatial interpolation and building structure information to accurately calculate the spatiotemporal distribution field of oil smoke concentration, thereby achieving dynamic optimization control of the air quality in the catering area.

[0018] 2. Collaborative control and energy saving and consumption reduction: Through the multi-objective optimization algorithm, comprehensively considering factors such as oil fume concentration, exhaust energy consumption and exhaust efficiency, the optimal collaborative control strategy is determined, which can effectively reduce the energy consumption of the range hood and improve the exhaust efficiency.

[0019] 3. Improve the dining environment: Through the coordinated control of the range hood fan speed, exhaust direction and air supply parameters of the fresh air system, the fume concentration in the dining area can be effectively reduced, the dining environment can be improved, and the air quality can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of a method for real-time monitoring and intelligent regulation of oil smoke concentration according to an embodiment of the present invention; Figure 2 It is a structural schematic diagram of the real-time oil fume concentration monitoring and intelligent regulation system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0023] Figure 1 FIG. 1 is a flow chart of a method for real-time monitoring and intelligent regulation of oil smoke concentration according to an embodiment of the present invention. Figure 1 As shown, the method includes: S101. Collecting real-time air quality data in the dining area through environmental sensors installed in multiple range hoods in the dining area, wherein the environmental sensors include a fume sensor for detecting fume concentration, a PM2.5 sensor for detecting particulate matter concentration, a carbon monoxide sensor for detecting carbon monoxide concentration, a temperature sensor for detecting temperature, and a humidity sensor for detecting humidity; transmitting the real-time air quality data to the control center through a wireless communication network; S102. The control center determines the initial estimated value of pollutant concentration based on the real-time air quality data combined with spatial interpolation, calculates the dynamic distribution of pollutant concentration at any point in the catering area at any time according to the spatial distance attenuation characteristics between environmental sensors, and combines the dynamic distribution of pollutant concentration with the building structure information and ventilation system layout information of the catering area to calculate the spatiotemporal distribution field of oil fume concentration in the catering area; S103. According to the spatiotemporal distribution field of oil fume concentration, the exhaust energy consumption and exhaust efficiency of each range hood are taken as optimization targets, and the regional oil fume concentration value in the spatiotemporal distribution field of oil fume concentration is considered, and the optimization target is solved in combination with a multi-objective optimization algorithm to determine the optimal collaborative control strategy, which includes the fan speed, exhaust direction and air supply parameters of each range hood and the fresh air system. The control center sends the optimal collaborative control strategy to the controllers of each range hood, and the controllers of each range hood adjust the operating parameters of the corresponding range hood to achieve dynamic optimization control of the air quality in the dining area.

[0024] In an optional implementation, the control center determines the initial estimated value of the pollutant concentration based on the real-time air quality data combined with spatial interpolation, and calculates the dynamic distribution of the pollutant concentration at any point in the dining area at any time according to the spatial distance attenuation characteristics between environmental sensors, including: Performing outlier elimination and data standardization processing on the real-time air quality data to obtain pre-processed air quality data, based on the pre-processed air quality data, taking the overall pollutant distribution trend of the catering area as a deterministic component and the local pollutant concentration fluctuation as a random component, and calculating the initial estimated value of the pollutant concentration at any location through spatial correlation analysis; Calculate the Euclidean distance between environmental sensors in the catering area, substitute the Euclidean distance into the Gaussian attenuation function to obtain the distance attenuation coefficient, and simultaneously calculate the relative azimuth angles between the environmental sensors to obtain the azimuth weight coefficient, and multiply the distance attenuation coefficient by the azimuth weight coefficient to obtain the spatial correlation weight between the detection points; calculate the Gaussian kernel function value of the time dimension according to the time series data of the environmental sensor, multiply the spatial correlation weight by the Gaussian kernel function value, and obtain the pollutant concentration value of any position point in the catering area at any time through weighted superposition calculation.

[0025] A method for determining the dynamic distribution of air pollutant concentrations in catering areas can monitor and predict pollutant concentrations at any location in the catering area in real time, providing data support for improving the catering environment.

[0026] First, the real-time air quality data from various environmental sensors is preprocessed. This step includes removing outliers and standardizing the data. For example, if the reading of a sensor suddenly soars or drops and deviates significantly from the normal range, it is considered an outlier and needs to be removed. Data standardization is to unify data of different dimensions to the same scale, such as converting the readings of all sensors to values ​​between 0 and 1 for subsequent calculations. Assume that sensors A, B, and C record PM2.5 concentrations of 150, 160, and 500 μg / m³, respectively, and C's reading of 500 is significantly higher, so it is judged as an outlier and removed. The remaining readings of A and B are standardized. Assuming that the normal range of regional PM2.5 concentration is 0-200 μg / m³, the standardized values ​​of A and B are 0.75 and 0.8, respectively.

[0027] Next, determine the initial estimate of the pollutant concentration. The overall pollutant distribution trend in the catering area is taken as the deterministic component, and the local pollutant concentration fluctuation is taken as the random component. The initial estimate of the pollutant concentration at any location is calculated by analyzing the data correlation between environmental sensors, such as whether the trend of sensor readings over time is similar. Assuming that the data of sensors A and B are strongly correlated after analysis, the initial estimate of the pollutant concentration at the target location can be obtained by weighted average based on the readings of A and B, combined with their distance relationship to the target location. Assuming that the target location is closer to A, the reading of A has a greater weight. For example, if the weight of A is 0.7 and the weight of B is 0.3, the initial estimated concentration of the target location is 0.7 * 150 + 0.3 * 160 = 153 μg / m³.

[0028] Then, calculate the spatial association weights between environmental sensors in the dining area. First, calculate the Euclidean distance between sensors, such as the straight-line distance between sensors A and B. Substitute this distance into the Gaussian attenuation function to obtain the distance attenuation coefficient. The greater the distance, the smaller the attenuation coefficient. At the same time, calculate the relative azimuth angles between sensors. For example, based on the north direction, calculate the azimuth angle of sensor A relative to sensor B. Convert the azimuth angle to the azimuth weight coefficient. Multiply the distance attenuation coefficient by the azimuth weight coefficient to obtain the spatial association weight between sensors. Assuming that the distance between A and B is 10 meters, the distance attenuation coefficient is calculated to be 0.8, and the azimuth weight coefficient is 0.9, then the spatial association weight is 0.8 * 0.9 = 0.72.

[0029] Next, calculate the Gaussian kernel function value in the time dimension. Calculate the Gaussian kernel function value in the time dimension based on the time series data of the environmental sensor. The closer the time is to the current moment, the larger the function value. Assuming that the current moment is t, the PM2.5 concentration of sensor A at t-1, t-2, and t-3 are 140, 130, and 120 μg / m³, respectively, the Gaussian kernel function value in the time dimension can be calculated based on these data.

[0030] Finally, calculate the pollutant concentration value at any point in the location at any time. Multiply the spatial association weight by the Gaussian kernel function value in the time dimension, and calculate the pollutant concentration value at any point in the catering area at any time through weighted superposition. Assuming that the spatial association weight between the target location and sensor A is 0.72, and the Gaussian kernel function value of A at the current moment is 0.9, then the contribution of A to the concentration of the target location at the current moment is 0.72 * 0.9 * 150 = 97.2 μg / m³. Add up the contribution values ​​of all sensors to get the final concentration estimate of the target location.

[0031] The beneficial effects of this method are reflected in three aspects: First, the accuracy of pollutant concentration estimation is improved. By combining spatial interpolation and time series analysis, the pollutant concentration at any point in the catering area can be estimated more accurately.

[0032] Secondly, dynamic monitoring of pollutant concentrations has been achieved. The changes in pollutant concentrations in the catering area can be monitored in real time, the pollution source can be discovered in time, and corresponding measures can be taken.

[0033] Finally, it provides data support for the improvement of the catering environment. According to the monitoring results, the ventilation and purification systems in the catering area can be optimized to improve the quality of the catering environment.

[0034] In an optional implementation, the overall pollutant distribution trend of the dining area is taken as a deterministic component, the local pollutant concentration fluctuation is taken as a random component, and the initial estimated value of the pollutant concentration at any location is calculated by spatial correlation analysis, including: Substituting the pollutant concentration observation data into a polynomial fitting function, the polynomial fitting function includes a constant term, a linear term coefficient, and a quadratic term coefficient, and calculating a deterministic trend term that characterizes the overall distribution trend based on the polynomial fitting function; taking the difference between the pollutant concentration observation data and the deterministic trend term as a random fluctuation term, calculating the spatial variability and spatial covariance of the random fluctuation term, and obtaining a random component that characterizes the local fluctuation characteristics; Based on the pollutant concentration observation data, the semivariogram function values ​​between the sample point pairs with different spatial interval distances are calculated, and the semivariogram function values ​​are fitted with the theoretical spherical model to obtain the nugget value, the sill value and the range parameter, wherein the nugget value represents the small-scale variation characteristics, the sill value represents the overall variation degree, and the range parameter represents the scope of spatial correlation influence; and a spatial structure characteristic equation group is constructed according to the nugget value, the sill value and the range parameter; The spatial structure characteristic equation group is combined with the unbiased estimation constraint condition to construct the Kriging weight equation group, and the Kriging weight equation group is solved to obtain the optimal weight coefficient of the observation point; the optimal weight coefficient is weightedly combined with the pollutant concentration observation data to calculate the initial estimate of the pollutant concentration at any location point.

[0035] A pollutant concentration estimation method for catering areas can comprehensively consider the overall distribution trend and local fluctuation characteristics, thereby more accurately estimating the pollutant concentration at any location.

[0036] First, collect pollutant concentration observation data at multiple locations in the dining area. These data need to include the location information of each sampling point and the corresponding pollutant concentration value. For example, we can choose a 10m x 10m dining area and evenly set 25 sampling points in it, recording the coordinates of each sampling point and the pollutant concentration value at that time. Suppose the collected data is as follows: (1,1) - 12ppm, (1,3) - 15ppm, (1,5) - 18ppm,... (5,5) -25ppm.

[0037] Then, a polynomial containing a constant term, a linear term coefficient, and a quadratic term coefficient is used to fit these observed data. The coefficients of the polynomial can be determined by the least squares method, thereby obtaining a deterministic trend term that can describe the overall distribution trend of the pollutant concentration. For example, the fitted polynomial can be: concentration = 2 + 3 * x + 0.5* y + 0.1 * x * y - 0.2 * x 2 + 0.05 * y 2 , where x and y represent the horizontal and vertical coordinates of the sampling point respectively.

[0038] Next, the difference between the pollutant concentration value and the deterministic trend term at each observation point is calculated to obtain the random fluctuation term. For example, for the sampling point with coordinates (1,1), the observed concentration value is 12ppm, and the deterministic trend term concentration value calculated according to the above polynomial is 5.4ppm, so the random fluctuation term of this point is 12 - 5.4 = 6.6ppm.

[0039] After that, the spatial variability and spatial covariance of the random fluctuation terms are analyzed. Spatial variability refers to the degree of difference between random fluctuation terms at different locations, while spatial covariance describes the correlation between random fluctuation terms at different locations. This spatial correlation can be quantified by calculating the semivariogram value between pairs of sample points at different spatial intervals.

[0040] Fit the calculated semivariogram values ​​to the theoretical spherical model. The theoretical spherical model includes nugget value, sill value and range parameter. The nugget value reflects the variation characteristics within a small scale, the sill value represents the overall degree of variation, and the range parameter indicates the scope of influence of spatial correlation. Assume that the fitted nugget value is 2, the sill value is 10, and the range parameter is 3 meters.

[0041] According to the fitted nugget value, sill value and range parameter, a set of spatial structure characteristic equations is constructed, which describes the spatial correlation of random fluctuation terms between different sampling points.

[0042] The spatial structure characteristic equations are combined with unbiased estimation constraints to construct the Kriging weight equations. The unbiased estimation constraints ensure that the mean of the estimated values ​​is equal to the mean of the true values.

[0043] Solve the Kriging weight equations to obtain the optimal weight coefficients for each observation point. These weight coefficients reflect the contribution of each observation point to the final estimate.

[0044] Finally, the optimal weight coefficient is weighted and combined with the corresponding pollutant concentration observation data to calculate the initial estimated value of the pollutant concentration at any location. For example, to estimate the pollutant concentration at a location with coordinates (2,2), the weight coefficient of each observation point can be calculated based on the distance between the location and each observation point, combined with the obtained nugget value, sill value, and range parameter, and then these weight coefficients are multiplied by the corresponding observed concentration value and summed to obtain the estimated value of the pollutant concentration at the (2,2) location.

[0045] The beneficial effects of this method can be summarized in the following three aspects: Improve estimation accuracy: This method comprehensively considers the overall distribution trend and local fluctuation characteristics, and can more accurately estimate the pollutant concentration at any location. Compared with the method that only considers the overall trend or local fluctuation, the estimation result is closer to the true value.

[0046] Utilization of spatial information: This method makes full use of the spatial correlation information of pollutant concentrations. Through the Kriging interpolation method, the limited observation data can be effectively expanded to the entire catering area to achieve concentration estimation at unsampled locations.

[0047] Strong applicability: This method is applicable to various pollutant concentration distribution situations, whether uniform or non-uniform, and can achieve good estimation results. Especially for situations with large local fluctuations, the advantages of this method are more obvious.

[0048] In an optional implementation, the dynamic distribution of pollutant concentration is combined with the building structure information and ventilation system layout information of the dining area to calculate the optimal coordinated control strategy of multiple range hoods. The optimal coordinated control strategy includes the fan speed, exhaust direction and air supply parameters of each range hood, including: Acquire the building structure information and ventilation system layout information of the dining area, wherein the building structure information includes the wall position, door and window distribution and indoor partitions, and the ventilation system layout information includes the range hood spatial coordinates, installation angle and exhaust pipe direction; construct a spatial characteristic function based on the building structure information, convert the ventilation system layout information into a ventilation equipment layout function, and fuse the spatial characteristic function with the ventilation equipment layout function to obtain a ventilation environment characteristic model of the dining area; Based on the ventilation environment characteristic model, a flow field motion equation including velocity field, pressure field and external force terms, and a pollutant transport equation including concentration field, diffusion coefficient and source term function are established; the flow field motion equation and the pollutant transport equation are solved jointly to obtain the oil fume diffusion dynamics characteristics, and the spatiotemporal distribution field of the oil fume concentration in the catering area is calculated based on the oil fume diffusion dynamics characteristics.

[0049] The coordinated control method of oil fume in the catering area aims to optimize the coordinated work of multiple range hoods, effectively reduce the oil fume concentration in the catering area, and improve the air quality.

[0050] First, obtain the architectural structure information and ventilation system layout information of the dining area. The architectural structure information includes the location of the wall, the distribution of doors and windows, and the indoor partitions, which can be obtained through field measurements, architectural drawings, or 3D modeling software. For example, the location of the wall can be represented by coordinates, the distribution of doors and windows can be represented by their area and position, and the indoor partitions can be represented by their height, length, and position. The ventilation system layout information includes the spatial coordinates of the range hood, the installation angle, and the direction of the exhaust duct, which can be obtained through ventilation system design drawings or field measurements. For example, the spatial coordinates of the range hood can be represented by three-dimensional coordinates, the installation angle can be represented by the angle relative to the horizontal plane, and the direction of the exhaust duct can be represented by the coordinate sequence of the center line of the duct. Assume that the size of a dining area is 10 meters long, 8 meters wide, and 3 meters high. There are two windows on one wall, each with an area of ​​2 square meters, located in the center of the wall. Three range hoods are installed in the area, located at (2, 2,2.5), (5, 2, 2.5), and (8, 2, 2.5), respectively. The installation angles are all vertically downward, and the exhaust ducts all extend vertically upward to the roof.

[0051] Next, a spatial characteristic function is constructed based on the acquired building structure information. The spatial characteristic function is used to describe the geometric features of each location in the dining area. For example, the presence of walls, doors, windows and partitions will affect the flow and diffusion of oil smoke. The dining area can be divided into several small grids, and the spatial characteristic function value of each grid indicates whether the grid is occupied by walls, doors, windows or partitions. For example, if the grid is occupied by a wall, its spatial characteristic function value is 1, otherwise it is 0. At the same time, the ventilation system layout information is converted into a ventilation equipment layout function. The ventilation equipment layout function is used to describe the influence of the position, installation angle and exhaust pipe direction of the range hood on the flow and diffusion of oil smoke. The ventilation equipment layout function value of the grid within the influence range of the range hood can be set to the air volume and direction of the range hood, and the function value of other grids is 0. For example, assuming that the air volume of the range hood is 1 cubic meter per second and the wind direction is vertically upward, the ventilation equipment layout function value of the grid within its influence range is (0, 0, 1). The spatial characteristic function is integrated with the ventilation equipment layout function to obtain the ventilation environment characteristic model of the dining area. The fusion method can be to add or multiply the two function values ​​to obtain a comprehensive function value, which is used to describe the ventilation environment characteristics of various locations in the catering area.

[0052] Then, the flow field motion equation and pollutant transport equation are established based on the ventilation environment characteristic model. The flow field motion equation describes the flow velocity and pressure changes of the air in the dining area, which includes the velocity field, pressure field and external force terms. The pollutant transport equation describes the change of the oil fume concentration in the dining area, which includes the concentration field, diffusion coefficient and source term function. The source term function represents the emission and position of the range hood. For example, assuming that the emission of each range hood is 0.1 kg per second, the source term function value of the grid located at the position of the range hood is 0.1, and the function value of other grids is 0. The flow field motion equation and the pollutant transport equation are solved simultaneously to obtain the oil fume diffusion dynamics characteristics, and the spatiotemporal distribution field of the oil fume concentration in the dining area is calculated based on the oil fume diffusion dynamics characteristics. The solution method can use numerical simulation methods, such as finite difference method or finite element method, to discretize the dining area into several grids, and then iteratively calculate the velocity, pressure and oil fume concentration of each grid.

[0053] Finally, the optimal collaborative control strategy of multiple range hoods is calculated based on the calculated spatiotemporal distribution field of oil fume concentration. The optimal collaborative control strategy includes the fan speed, exhaust direction and air supply parameters of each range hood. The optimization goal can be to minimize the average oil fume concentration in the dining area or to maximize the oil fume emission efficiency. The optimization algorithm can use intelligent optimization algorithms such as genetic algorithm and particle swarm algorithm. For example, suppose that after optimization calculation, the optimal fan speeds of the three range hoods are 800, 900 and 1000 revolutions per minute, respectively, the exhaust direction is kept vertically upward, and the air supply volume of the fresh air system is 5 cubic meters per second.

[0054] The beneficial effects of this method are reflected in three aspects: Improve oil fume purification efficiency: By precisely controlling the fan speed, exhaust direction and fresh air parameters of the range hood, oil fume can be captured and discharged more effectively, reducing the residence time of oil fume in the catering area, thereby improving the oil fume purification efficiency.

[0055] Reduce energy consumption: By optimizing the coordinated work of range hoods, it is possible to avoid overworking of range hoods, reduce energy consumption, and save operating costs.

[0056] Improve indoor air quality: By effectively controlling the concentration of oil smoke, the indoor air quality of the catering area can be significantly improved, and the comfort and health level of customers and employees can be enhanced.

[0057] In an optional implementation, based on the ventilation environment characteristic model, a flow field motion equation including a velocity field, a pressure field and an external force term, and a pollutant transport equation including a concentration field, a diffusion coefficient and a source term function are established; the flow field motion equation and the pollutant transport equation are solved simultaneously to obtain the fume diffusion dynamics characteristics, and the spatiotemporal distribution field of the fume concentration in the dining area is calculated according to the fume diffusion dynamics characteristics, including: The flow field motion equation is shown in the following formula: ; Among them, ρ represents the density of the oil smoke mixed gas, represents the velocity field vector, represents the partial derivative of velocity with respect to time, which is the local acceleration, The advection term representing the velocity, the acceleration caused by the fluid motion itself, represents the pressure gradient, is the spatial rate of change of the pressure field, μ represents the dynamic viscosity coefficient, The Laplace operator representing the velocity, represents the external force field vector; The pollutant transport equation is shown in the following formula: ; in, C represents the pollutant concentration field, represents the partial derivative of concentration with respect to time, represents the Laplace operator of concentration, D represents the diffusion coefficient tensor, It represents the second-order derivative of concentration in space, S(x,y,z,t) represents the source function, which describes the rate at which pollutants are generated or disappear. x,y,z represent the spatial coordinates, t represents time, and R(C,T) is the chemical reaction term, which describes the rate at which pollutants participate in chemical reactions and is a function of concentration C and temperature T.

[0058] The simulation method of oil fume diffusion in the catering area can accurately predict the concentration distribution of oil fume in the catering area, providing a scientific basis for optimizing the design of ventilation systems and improving indoor air quality. This method establishes a dynamic model of oil fume diffusion based on the characteristics of the ventilation environment, and performs simulation calculations based on actual data.

[0059] First, a mathematical model describing the diffusion of oil smoke is constructed. The model contains two core equations: the flow field motion equation and the pollutant transport equation. The flow field motion equation describes the movement law of the oil smoke mixed gas in a ventilation environment, taking into account the influence of factors such as oil smoke density, velocity change, pressure gradient, viscous resistance and external force. The pollutant transport equation describes the change of oil smoke concentration over time and space, taking into account factors such as concentration change, diffusion effect, emission from oil smoke sources and possible chemical reactions.

[0060] Secondly, determine the input parameters required by the model. These parameters include the geometric dimensions of the ventilation environment, the location and air volume of the vents, the emission intensity and location of the fume source, the density and viscosity of the fume mixed gas, and the ambient temperature. For example, a dining area with a length of 10 meters, a width of 5 meters, and a height of 3 meters can be set, in which there is a fume source located in the corner with an emission intensity of 0.1 kg per second.

[0061] Next, the dining area is discretized into a series of small computational units, and the flow field motion equations and pollutant transport equations are solved on each unit. Numerical calculation methods, such as the finite volume method or the finite element method, can be used to transform the equations into a set of algebraic equations, and then the velocity, pressure, and fume concentration values ​​of each unit are obtained by iterative solution. For example, the dining area can be divided into 1,000 small cubes, each with a side length of 0.5 meters.

[0062] Then, based on the calculation results, the spatiotemporal distribution field of the concentration of oil smoke in the catering area is obtained. The calculation results can be visualized, such as drawing oil smoke concentration contour maps or three-dimensional concentration distribution maps, so as to intuitively understand the diffusion of oil smoke. For example, the distribution map of oil smoke concentration on the horizontal and vertical planes at different time points can be drawn.

[0063] Finally, the effectiveness of the ventilation system can be evaluated based on the simulation results and optimized. For example, the location and air volume of the vents can be adjusted based on the simulation results to reduce the fume concentration and improve the indoor air quality.

[0064] The beneficial effects of this method are reflected in the following three aspects: First, the accuracy of fume diffusion prediction is improved. This method takes into account the influence of multiple factors on fume diffusion, such as the geometry of the ventilation environment, ventilation parameters, emission characteristics of the fume source, and the physical and chemical properties of the fume, so that the concentration distribution of fume in the catering area can be predicted more accurately.

[0065] Second, it provides a scientific basis for the optimal design of ventilation systems. By simulating the diffusion of oil smoke under different ventilation schemes, the optimal ventilation parameters, such as the location of the vents, air volume and wind speed, can be found to minimize the concentration of oil smoke and improve ventilation efficiency.

[0066] Third, it helps to improve the indoor air quality in the dining area. By optimizing the ventilation system design, the diffusion of oil smoke can be effectively controlled, the harm of oil smoke to human health can be reduced, and a more comfortable and healthy dining environment can be created.

[0067] In an optional implementation, according to the spatiotemporal distribution field of oil fume concentration, the exhaust energy consumption and exhaust efficiency of each range hood are taken as optimization targets, and the regional oil fume concentration value in the spatiotemporal distribution field of oil fume concentration is considered, and the optimization target is solved in combination with a multi-objective optimization algorithm, and the optimal collaborative control strategy is determined, including: Construct a total energy consumption objective function for the range hood group, wherein the total energy consumption objective function includes a cubic coefficient and a quadratic coefficient of the fan speed of each range hood, wherein the cubic coefficient represents the aerodynamic loss of the fan, and the quadratic coefficient represents the mechanical friction loss of the fan; obtain the fume concentration and volume flow rate at the inlet and outlet of each range hood, calculate the overall exhaust efficiency of the range hood group, and obtain the exhaust efficiency objective function; determine the maximum fume concentration value of the region according to the spatiotemporal distribution data of the fume concentration; A multi-objective optimization function vector is established based on the total energy consumption objective function and the smoke exhaust efficiency objective function, and each objective function in the multi-objective optimization function vector is divided by a corresponding normalized reference value to obtain a normalized objective function; a fan speed range, a smoke exhaust angle range, and a power consumption range of each range hood are set as equipment operation constraints, and a limited range of the maximum smoke concentration value in the area is set as a concentration control constraint; The normalized objective function is weighted and summed using an adaptive weight coefficient to obtain a comprehensive objective function, wherein the adaptive weight coefficient is dynamically adjusted based on the gradient information of each objective function to ensure the dynamic balance of the weights of each objective function during the optimization process of the comprehensive objective function; the equipment operation constraint condition and the concentration control constraint condition are substituted into the comprehensive objective function, and the comprehensive objective function is iteratively optimized using a gradient descent method; The dominance index of the optimization solution is calculated, and the dominance index characterizes the distance distribution from the optimization solution to each target ideal value through an exponential function. The Pareto optimal solution set is screened based on the dominance index; the Pareto optimal solution set is converted into a control strategy for the range hood group, and the control strategy includes the fan speed and exhaust direction of each range hood and the air supply parameters of the fresh air system.

[0068] The coordinated control method of the range hood group can effectively reduce the overall fume concentration in the kitchen, while minimizing energy consumption and improving exhaust efficiency. This method mainly analyzes the spatiotemporal distribution of fume concentration, combines a multi-objective optimization algorithm, and dynamically adjusts the operating parameters of each range hood to ultimately achieve the optimal coordinated control effect.

[0069] First, obtain the spatiotemporal distribution data of the fume concentration in the kitchen area. For example, a sensor network can be used to monitor the fume concentration in different locations in the kitchen in real time and record the changes in concentration data over time. Assuming a 5x5 meter kitchen area, a sensor is set every 1 meter, a total of 25 sensors, and data is collected once a second, a fume concentration dataset containing temporal and spatial information can be obtained.

[0070] Next, we construct an energy consumption model for the range hood fleet. The energy consumption of a range hood is mainly determined by the fan speed. The higher the fan speed, the greater the energy consumption. The energy consumption model can be simplified to the weighted sum of the cubic and quadratic terms of the fan speed. The cubic term represents the aerodynamic loss, and the quadratic term represents the mechanical friction loss. For example, the energy consumption of a range hood can be expressed as 0.001* speed³ + 0.01 * speed², where the coefficients 0.001 and 0.01 can be adjusted according to actual conditions.

[0071] Then, the exhaust efficiency model of the range hood group is constructed. The exhaust efficiency can be expressed by the ratio of the difference in oil fume concentration between the inlet and outlet of the range hood to the oil fume concentration at the air inlet. For example, if the oil fume concentration at the air inlet of a range hood is 100ppm and the oil fume concentration at the air outlet is 10ppm, then its exhaust efficiency is (100-10) / 100 = 90%.

[0072] According to the spatiotemporal distribution data of oil fume concentration, determine the area that needs to be focused on, for example, the area with the highest oil fume concentration. Assuming that in the above 5x5 meter kitchen area, the oil fume concentration in the upper left corner is the highest, the maximum oil fume concentration value in this area is used as the control target.

[0073] Based on the above energy consumption model, smoke exhaust efficiency model and regional maximum oil smoke concentration value, a multi-objective optimization function vector is constructed. The vector contains three objective functions: minimizing total energy consumption, maximizing overall smoke exhaust efficiency, and minimizing regional maximum oil smoke concentration value.

[0074] In order to facilitate comparison and calculation, each objective function is normalized by dividing it by its corresponding reference value. For example, assuming that the reference value of total energy consumption is 1000W, the total energy consumption objective function is divided by 1000 to obtain the normalized energy consumption objective function.

[0075] Set the operating constraints of the range hood, including the fan speed range (e.g., 0-1000 rpm), smoke exhaust angle range (e.g., 0-90 degrees), and power consumption range (e.g., 0-500W). At the same time, set the concentration control constraints, for example, the maximum smoke concentration value in the area cannot exceed 50ppm.

[0076] The normalized objective functions are weighted and summed using adaptive weight coefficients to obtain a comprehensive objective function. The weight coefficients are dynamically adjusted according to the gradient information of each objective function to ensure the dynamic balance of the weights of each objective function during the optimization process.

[0077] The operation constraints and concentration control constraints are substituted into the comprehensive objective function, and the gradient descent method is used to iteratively optimize the comprehensive objective function to find the optimal solution.

[0078] Calculate the dominance index of the optimization solution, which is used to measure the distance distribution from the optimization solution to the ideal value of each target. For example, an exponential function can be used to calculate the dominance index. The Pareto optimal solution set is obtained based on the dominance index. Each solution in the Pareto optimal solution set cannot improve any objective function value without reducing the values ​​of other objective functions.

[0079] Finally, the Pareto optimal solution set is converted into a control strategy for the range hood group, which includes the fan speed, exhaust direction and air supply parameters of each range hood. For example, an optimal solution can be: the speed of range hood 1 is 500 rpm, the exhaust direction is 45 degrees; the speed of range hood 2 is 600 rpm, the exhaust direction is 60 degrees; the air supply volume of the fresh air system is 1000m³ / h.

[0080] Beneficial effects: 1. Energy saving and consumption reduction: By optimizing the operating parameters of the range hood group, the overall energy consumption can be effectively reduced and energy can be saved.

[0081] 2. Improve exhaust efficiency: Through coordinated control, the overall exhaust efficiency of the range hood group can be improved, and kitchen fumes can be removed faster.

[0082] 3. Improve the kitchen environment: By reducing the maximum oil fume concentration in the area, the kitchen air quality can be effectively improved to create a more comfortable cooking environment.

[0083] In an optional implementation, a dominance index of the optimization solution is calculated, wherein the dominance index characterizes the distance distribution from the optimization solution to each target ideal value through an exponential function, and the Pareto optimal solution set obtained by screening based on the dominance index includes: Each objective function is independently optimized to obtain an ideal value, and the ideal value is constructed as an ideal point vector; the weighted Euclidean distance and Chebyshev distance from the solution of each objective function to the ideal point vector are calculated, and the weighted Euclidean distance and the Chebyshev distance respectively form exponential function terms with corresponding scale parameters, and the sum of the two exponential function terms is used as the basic advantage function; The sum of squares of constraint violations is used as a constraint penalty term, and the constraint penalty term and the penalty factor form an exponential function; the product of the basic advantage function and the exponential function of the constraint penalty term is used as a comprehensive advantage index; Based on the dominance relationship between the comprehensive dominance index and each objective function value, non-dominated solutions are screened out to constitute the Pareto optimal solution set. The dominance relationship requires that the comprehensive dominance index of the dominating solution is greater than the comprehensive dominance index of the dominated solution, and all objective function values ​​of the dominating solution are not greater than the corresponding objective function values ​​of the dominated solution.

[0084] A Pareto optimal solution set screening method based on dominance index is used to screen the optimal solution set from the solution sets of multiple objective functions. The core idea of ​​this method is to measure the quality of each solution through a comprehensive dominance index, which comprehensively considers the distance between the objective function value and the ideal value and the constraint violation.

[0085] First, each objective function is optimized independently to obtain the optimal value of each objective function. These optimal values ​​are combined into an ideal point vector. This ideal point vector represents the ideal situation where all objective functions are optimal at the same time.

[0086] Next, calculate the distance from each solution to the ideal point vector. Two distance metrics are used here: weighted Euclidean distance and Chebyshev distance. Weighted Euclidean distance takes into account the importance of each objective function and adjusts the impact of different objective functions on the distance through weight coefficients. Chebyshev distance focuses on the maximum deviation between the solution and the ideal point vector on each objective function. Substitute these two distances and the corresponding scale parameters into the exponential function to obtain two exponential function terms. The sum of these two exponential function terms constitutes the basic dominance function.

[0087] Then, consider the impact of the constraints. The sum of the squares of the constraint violations is taken as the constraint penalty term. The constraint violation refers to the degree to which the solution violates the constraint. Substitute the constraint penalty term and the penalty factor into the exponential function to obtain the exponential function of the constraint penalty term.

[0088] Multiply the basic advantage function by the exponential function of the constraint penalty term to get the comprehensive advantage index. This index comprehensively reflects the distance to the ideal point vector and the constraint violation. The larger the value, the better the solution.

[0089] Finally, based on the dominance relationship between the comprehensive dominance index and each objective function value, non-dominated solutions are selected to form the Pareto optimal solution set. The judgment criteria for the dominance relationship are: the comprehensive dominance index of the dominating solution is greater than the comprehensive dominance index of the dominated solution, and all objective function values ​​of the dominating solution are not greater than the corresponding objective function values ​​of the dominated solution. If these two conditions are met, the dominating solution is said to dominate the dominated solution. The selected non-dominated solutions are the Pareto optimal solution set.

[0090] For example, suppose there are two objective functions, objective function 1 and objective function 2, and a constraint. Through independent optimization, the ideal value of objective function 1 is 10, and the ideal value of objective function 2 is 20. The ideal point vector is (10, 20). Now there is a solution A, whose objective function 1 has a value of 12, whose objective function 2 has a value of 25, and satisfies the constraint. Another solution B, whose objective function 1 has a value of 15, whose objective function 2 has a value of 22, and violates the constraint by 2.

[0091] First, the weighted Euclidean distance and Chebyshev distance from solution A and solution B to the ideal point vector are calculated, assuming that the weight coefficients are both 1 and the scale parameters are both 1. Then the calculated distance values ​​are substituted into the exponential function to obtain two exponential function terms, and they are added together to obtain the basic advantage function.

[0092] For solution B, the constraint penalty term needs to be calculated. The square sum of the violation amount 2 and 4 is taken as the constraint penalty term. Assuming the penalty factor is 1, substitute 4 into the exponential function to obtain the exponential function of the constraint penalty term.

[0093] The basic advantage function of solution A and the basic advantage function of solution B are multiplied by the exponential function of the constraint penalty term to obtain the comprehensive advantage index of solution A and solution B respectively.

[0094] Finally, compare the comprehensive dominance index and objective function values ​​of solution A and solution B to determine the dominance relationship between them. If the comprehensive dominance index of solution A is greater than that of solution B, and the values ​​of objective function 1 and objective function 2 of solution A are both less than or equal to the corresponding objective function values ​​of solution B, then solution A dominates solution B.

[0095] Beneficial effects: 1. Improved screening efficiency: This method quickly evaluates the pros and cons of solutions through a comprehensive dominance index, avoiding tedious pair-by-pair comparisons and thus improving screening efficiency.

[0096] 2. Improved screening quality: This method comprehensively considers the distance between the objective function value and the ideal value and the constraint violation, and can more comprehensively evaluate the pros and cons of the solution, thereby screening out a better solution set.

[0097] 3. Enhanced adaptability of the method: By introducing weight coefficients and scale parameters, this method can flexibly adjust the importance of different objective functions and distance metrics according to the characteristics of actual problems, thereby enhancing the adaptability of the method.

[0098] Figure 2 FIG. 1 is a schematic diagram of the structure of the oil fume real-time concentration monitoring and intelligent regulation system according to an embodiment of the present invention. Figure 2 As shown, the system comprises: The first unit is used to collect real-time air quality data in the dining area through environmental sensors installed in multiple range hoods in the dining area, wherein the environmental sensors include a fume sensor for detecting fume concentration, a PM2.5 sensor for detecting particulate matter concentration, a carbon monoxide sensor for detecting carbon monoxide concentration, a temperature sensor for detecting temperature, and a humidity sensor for detecting humidity; and transmit the real-time air quality data to a control center through a wireless communication network; The second unit is used for the control center to determine the initial estimated value of the pollutant concentration based on the real-time air quality data combined with spatial interpolation, calculate the dynamic distribution of the pollutant concentration at any point in the catering area at any time according to the spatial distance attenuation characteristics between environmental sensors, and combine the dynamic distribution of the pollutant concentration with the building structure information and ventilation system layout information of the catering area to calculate the spatiotemporal distribution field of the oil fume concentration in the catering area; The third unit is used to take the exhaust energy consumption and exhaust efficiency of each range hood as the optimization target according to the spatiotemporal distribution field of the oil fume concentration, and consider the regional oil fume concentration value in the spatiotemporal distribution field of the oil fume concentration, and solve the optimization target in combination with the multi-objective optimization algorithm to determine the optimal collaborative control strategy, which includes the fan speed, exhaust direction and air supply parameters of each range hood and the fresh air system; the control center sends the optimal collaborative control strategy to the controllers of each range hood, and the controllers of each range hood adjust the operating parameters of the corresponding range hood to achieve dynamic optimization control of the air quality in the dining area.

[0099] According to a third aspect of the embodiments of the present invention, An electronic device is provided, comprising: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0100] A fourth aspect of the embodiments of the present invention is: A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0101] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for real-time monitoring and intelligent adjustment of oil smoke concentration, characterized in that: include: Real-time air quality data in the dining area is collected by using environmental sensors installed in multiple range hoods in the dining area, wherein the environmental sensors include a fume sensor for detecting fume concentration, a PM2.5 sensor for detecting particulate matter concentration, a carbon monoxide sensor for detecting carbon monoxide concentration, a temperature sensor for detecting temperature, and a humidity sensor for detecting humidity; the real-time air quality data is transmitted to a control center via a wireless communication network; The control center determines the initial estimated value of the pollutant concentration based on the real-time air quality data combined with spatial interpolation, calculates the dynamic distribution of the pollutant concentration at any point in the dining area at any time according to the spatial distance attenuation characteristics between environmental sensors, and combines the dynamic distribution of the pollutant concentration with the building structure information and ventilation system layout information of the dining area to calculate the spatiotemporal distribution field of the oil fume concentration in the dining area; According to the spatiotemporal distribution field of oil fume concentration, the exhaust energy consumption and exhaust efficiency of each range hood are taken as optimization targets, and the regional oil fume concentration value in the spatiotemporal distribution field of oil fume concentration is considered. The optimization target is solved in combination with a multi-objective optimization algorithm to determine the optimal collaborative control strategy, which includes the fan speed, exhaust direction and air supply parameters of each range hood and the fresh air system. The control center sends the optimal collaborative control strategy to the controllers of each range hood, and the controllers of each range hood adjust the operating parameters of the corresponding range hood to achieve dynamic optimization control of the air quality in the dining area.

2. The method according to claim 1, characterized in that The control center determines the initial estimated value of the pollutant concentration based on the real-time air quality data combined with spatial interpolation, and calculates the dynamic distribution of the pollutant concentration at any point in the dining area at any time according to the spatial distance attenuation characteristics between environmental sensors, including: Performing outlier elimination and data standardization processing on the real-time air quality data to obtain pre-processed air quality data, based on the pre-processed air quality data, taking the overall pollutant distribution trend of the catering area as a deterministic component and the local pollutant concentration fluctuation as a random component, and calculating the initial estimated value of the pollutant concentration at any location through spatial correlation analysis; Calculate the Euclidean distance between environmental sensors in the catering area, substitute the Euclidean distance into the Gaussian attenuation function to obtain the distance attenuation coefficient, and simultaneously calculate the relative azimuth angles between the environmental sensors to obtain the azimuth weight coefficient, and multiply the distance attenuation coefficient by the azimuth weight coefficient to obtain the spatial correlation weight between the detection points; calculate the Gaussian kernel function value of the time dimension according to the time series data of the environmental sensor, multiply the spatial correlation weight by the Gaussian kernel function value, and obtain the pollutant concentration value of any position point in the catering area at any time through weighted superposition calculation.

3. The method according to claim 2, characterized in that Taking the overall pollutant distribution trend of the catering area as the deterministic component and the local pollutant concentration fluctuation as the random component, the initial estimated value of the pollutant concentration at any location is calculated through spatial correlation analysis, including: Substituting the pollutant concentration observation data into a polynomial fitting function, the polynomial fitting function includes a constant term, a linear term coefficient, and a quadratic term coefficient, and calculating a deterministic trend term that characterizes the overall distribution trend based on the polynomial fitting function; taking the difference between the pollutant concentration observation data and the deterministic trend term as a random fluctuation term, calculating the spatial variability and spatial covariance of the random fluctuation term, and obtaining a random component that characterizes the local fluctuation characteristics; Based on the pollutant concentration observation data, the semivariogram function values ​​between the sample point pairs with different spatial interval distances are calculated, and the semivariogram function values ​​are fitted with the theoretical spherical model to obtain the nugget value, the sill value and the range parameter, wherein the nugget value represents the small-scale variation characteristics, the sill value represents the overall variation degree, and the range parameter represents the scope of spatial correlation influence; and a spatial structure characteristic equation group is constructed according to the nugget value, the sill value and the range parameter; The spatial structure characteristic equation group is combined with the unbiased estimation constraint condition to construct the Kriging weight equation group, and the Kriging weight equation group is solved to obtain the optimal weight coefficient of the observation point; the optimal weight coefficient is weightedly combined with the pollutant concentration observation data to calculate the initial estimate of the pollutant concentration at any location point.

4. The method according to claim 1, characterized in that: The dynamic distribution of pollutant concentration is combined with the building structure information and ventilation system layout information of the dining area to calculate the optimal coordinated control strategy of multiple range hoods. The optimal coordinated control strategy includes the fan speed, exhaust direction and air supply parameters of each range hood, including: Acquire the building structure information and ventilation system layout information of the dining area, wherein the building structure information includes the wall position, door and window distribution and indoor partitions, and the ventilation system layout information includes the range hood spatial coordinates, installation angle and exhaust pipe direction; construct a spatial characteristic function based on the building structure information, convert the ventilation system layout information into a ventilation equipment layout function, and fuse the spatial characteristic function with the ventilation equipment layout function to obtain a ventilation environment characteristic model of the dining area; Based on the ventilation environment characteristic model, a flow field motion equation including velocity field, pressure field and external force terms, and a pollutant transport equation including concentration field, diffusion coefficient and source term function are established; the flow field motion equation and the pollutant transport equation are solved jointly to obtain the oil fume diffusion dynamics characteristics, and the spatiotemporal distribution field of the oil fume concentration in the catering area is calculated based on the oil fume diffusion dynamics characteristics.

5. The method according to claim 4, characterized in that Based on the ventilation environment characteristic model, a flow field motion equation including velocity field, pressure field and external force terms, and a pollutant transport equation including concentration field, diffusion coefficient and source term function are established; the flow field motion equation and the pollutant transport equation are solved simultaneously to obtain the fume diffusion dynamics characteristics, and the spatiotemporal distribution field of fume concentration in the dining area is calculated according to the fume diffusion dynamics characteristics, including: The flow field motion equation is shown in the following formula: ； Among them, ρ represents the density of the oil smoke mixed gas, represents the velocity field vector, represents the partial derivative of velocity with respect to time, which is the local acceleration, The advection term representing the velocity, the acceleration caused by the fluid motion itself, represents the pressure gradient, is the spatial rate of change of the pressure field, μ represents the dynamic viscosity coefficient, The Laplace operator representing the velocity, represents the external force field vector; The pollutant transport equation is shown in the following formula: ； in, C represents the pollutant concentration field, represents the partial derivative of concentration with respect to time, represents the Laplace operator of concentration, D represents the diffusion coefficient tensor, It represents the second-order derivative of concentration in space, S(x,y,z,t) represents the source function, which describes the rate at which pollutants are generated or disappear. x,y,z represent the spatial coordinates, t represents time, and R(C,T) is the chemical reaction term, which describes the rate at which pollutants participate in chemical reactions and is a function of concentration C and temperature T.

6. The method according to claim 1, characterized in that According to the spatiotemporal distribution field of oil fume concentration, the exhaust energy consumption and exhaust efficiency of each range hood are taken as optimization targets, and the regional oil fume concentration value in the spatiotemporal distribution field of oil fume concentration is considered. The optimization target is solved by combining a multi-objective optimization algorithm, and the optimal collaborative control strategy is determined to include: Construct a total energy consumption objective function for the range hood group, wherein the total energy consumption objective function includes a cubic coefficient and a quadratic coefficient of the fan speed of each range hood, wherein the cubic coefficient represents the aerodynamic loss of the fan, and the quadratic coefficient represents the mechanical friction loss of the fan; obtain the fume concentration and volume flow rate at the inlet and outlet of each range hood, calculate the overall exhaust efficiency of the range hood group, and obtain the exhaust efficiency objective function; determine the maximum fume concentration value of the region according to the spatiotemporal distribution data of the fume concentration; A multi-objective optimization function vector is established based on the total energy consumption objective function and the smoke exhaust efficiency objective function, and each objective function in the multi-objective optimization function vector is divided by a corresponding normalized reference value to obtain a normalized objective function; a fan speed range, a smoke exhaust angle range, and a power consumption range of each range hood are set as equipment operation constraints, and a limited range of the maximum smoke concentration value in the area is set as a concentration control constraint; The normalized objective function is weighted and summed using an adaptive weight coefficient to obtain a comprehensive objective function, wherein the adaptive weight coefficient is dynamically adjusted based on the gradient information of each objective function to ensure the dynamic balance of the weights of each objective function during the optimization process of the comprehensive objective function; the equipment operation constraint condition and the concentration control constraint condition are substituted into the comprehensive objective function, and the comprehensive objective function is iteratively optimized using a gradient descent method; The dominance index of the optimization solution is calculated, and the dominance index characterizes the distance distribution from the optimization solution to each target ideal value through an exponential function. The Pareto optimal solution set is screened based on the dominance index; the Pareto optimal solution set is converted into a control strategy for the range hood group, and the control strategy includes the fan speed and exhaust direction of each range hood and the air supply parameters of the fresh air system.

7. The method according to claim 6, characterized in that The dominance index of the optimization solution is calculated. The dominance index characterizes the distance distribution from the optimization solution to each target ideal value through an exponential function. The Pareto optimal solution set obtained by screening based on the dominance index includes: Each objective function is independently optimized to obtain an ideal value, and the ideal value is constructed as an ideal point vector; the weighted Euclidean distance and Chebyshev distance from the solution of each objective function to the ideal point vector are calculated, and the weighted Euclidean distance and the Chebyshev distance respectively form exponential function terms with corresponding scale parameters, and the sum of the two exponential function terms is used as the basic advantage function; The sum of squares of constraint violations is used as a constraint penalty term, and the constraint penalty term and the penalty factor form an exponential function; the product of the basic advantage function and the exponential function of the constraint penalty term is used as a comprehensive advantage index; Based on the dominance relationship between the comprehensive dominance index and each objective function value, non-dominated solutions are screened out to constitute the Pareto optimal solution set. The dominance relationship requires that the comprehensive dominance index of the dominating solution is greater than the comprehensive dominance index of the dominated solution, and all objective function values ​​of the dominating solution are not greater than the corresponding objective function values ​​of the dominated solution.

8. A real-time oil smoke concentration monitoring and intelligent regulation system, used to implement the method according to any one of claims 1 to 7, characterized in that: include: The first unit is used to collect real-time air quality data in the dining area through environmental sensors installed in multiple range hoods in the dining area, wherein the environmental sensors include a fume sensor for detecting fume concentration, a PM2.5 sensor for detecting particulate matter concentration, a carbon monoxide sensor for detecting carbon monoxide concentration, a temperature sensor for detecting temperature, and a humidity sensor for detecting humidity; and transmit the real-time air quality data to a control center through a wireless communication network; The second unit is used for the control center to determine the initial estimated value of the pollutant concentration based on the real-time air quality data combined with spatial interpolation, calculate the dynamic distribution of the pollutant concentration at any point in the catering area at any time according to the spatial distance attenuation characteristics between environmental sensors, and combine the dynamic distribution of the pollutant concentration with the building structure information and ventilation system layout information of the catering area to calculate the spatiotemporal distribution field of the oil fume concentration in the catering area; The third unit is used to take the exhaust energy consumption and exhaust efficiency of each range hood as the optimization target according to the spatiotemporal distribution field of the oil fume concentration, and consider the regional oil fume concentration value in the spatiotemporal distribution field of the oil fume concentration, and solve the optimization target in combination with the multi-objective optimization algorithm to determine the optimal collaborative control strategy, which includes the fan speed, exhaust direction and air supply parameters of each range hood and the fresh air system; the control center sends the optimal collaborative control strategy to the controllers of each range hood, and the controllers of each range hood adjust the operating parameters of the corresponding range hood to achieve dynamic optimization control of the air quality in the dining area.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images