Exhibition hall environment control method and system based on sensor group

Abstract

The invention provides an exhibition hall environment control method and system based on a sensor group, and relates to the technical field of indoor environment control, and the method comprises the steps: obtaining exhibition hall environment data through the sensor group, obtaining historical environment data and a historical current response value of an air purifier, and constructing a response prediction model of the air purifier; obtaining a current response value of the air purifier according to the response prediction model of the air purifier; generating a response correction factor for the exhibition hall environment data; correcting the predicted current response value of the air purifier by using a response correction factor to obtain a corrected current response value of the air purifier; the problem of excessive response or low-efficiency operation is avoided, and the balance of energy conservation and efficient purification is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of indoor environment control, and particularly to an exhibition hall environment control method and system based on a sensor group. Background Art

[0002] In recent years, the requirements for air quality in large public places such as exhibition halls have been continuously increasing. Especially in the general environment of epidemic prevention and control and the popularization of fresh air systems, air quality not only affects the comfort of visitors but also plays an important role in health protection. Traditional air purification control systems mostly adjust based on static environmental data and fail to fully consider the complexity of the dynamic environment in the exhibition hall. Some air purification control systems only use fixed sensors to collect environmental data in a single area or at a single moment, lacking real-time evaluation of the overall environmental state and being unable to accurately reflect the air quality changes in various areas of the exhibition hall. At the same time, these systems usually use simple rules or linear models to predict the current response value of the air purifier and fail to effectively model the non-linear relationships such as particulate matter diffusion and crowd activities in the dynamic environment, resulting in poor purification effects or energy waste. Therefore, how to design a system that can fully sense the dynamic environment of the exhibition hall and achieve precise and real-time control based on multi-dimensional data has become a technical problem to be solved urgently.

[0003] In the prior art, the publication number CN116928848A discloses an indoor environment control system based on a sensor group, including a server. The server includes the following modules: a database stores room information and standard information; an acquisition module acquires the indoor environment information of the user's permanent residence and the user's personal identity information; an identification module identifies the user's personal identity information, and an allocation module allocates a room for the user; the acquisition module also acquires the indoor environment information of the room where the user is allocated; a judgment module compares the indoor environment information of the room where the user is allocated with the indoor environment information of the permanent residence. If it meets the standard information, a shutdown information is generated; if not, a startup information is generated; a control module controls the startup of the indoor environment purification equipment in the room where the user is allocated when receiving the startup information, and shuts down the indoor environment purification equipment when receiving the shutdown information. However, the prior art still has defects. The judgment of the indoor environment in the prior art is only obtained by simply comparing the environmental data collected by the sensor with the standard data. However, for indoor environment control, each piece of data is not independent. For example, starting the air purifier will cause changes in indoor humidity, temperature, etc., which are not the objects of the air purifier. Just comparing the data with the standard data will make the analysis result inaccurate, resulting in dissatisfaction among users.

[0004] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The object of the present invention is to provide a method and system for controlling the exhibition hall environment based on a sensor group to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An exhibition hall environment control system based on a sensor group specifically includes:

[0008] Step 1: Obtain the exhibition hall environment data through the sensor group, and the exhibition hall environment data includes moment environment data and period environment data;

[0009] Step 2: Obtain the historical moment environment data and the current response value of the air purifier; use the historical moment environment data as the input, and at the same moment, use the current response value of the air purifier as the label to construct a response prediction model of the air purifier; input the current moment environment data into the response prediction model of the air purifier to obtain the predicted current response value of the air purifier;

[0010] Step 3: Analyze the moment environment data to generate air discomfort, and analyze the period environment data to obtain the particle change trend, personnel circulation, and the change rate of the correlation between the crowd and particles; generate a response correction factor according to the air discomfort, particle change trend, personnel circulation, and the change rate of the correlation between the crowd and particles;

[0011] Step 4: Use the response correction factor to correct the predicted current response value of the air purifier to obtain the corrected current response value of the air purifier.

[0012] Further, the sensor group includes a temperature sensor, a humidity sensor, a carbon dioxide sensor, and a PM2.5 sensor; the moment environment data includes the temperature, humidity, carbon dioxide concentration, and PM2.5 concentration at the current moment; the period environment data includes the time series data of the carbon dioxide concentration and the PM2.5 concentration within the previous 5 minutes before the current moment; the temperature at the current moment is collected by the temperature sensor; the humidity at the current moment is collected by the humidity sensor; the carbon dioxide concentration at the current moment and the time series data of the carbon dioxide concentration within the previous 5 minutes before the current moment are collected by the carbon dioxide sensor; the PM2.5 concentration at the current moment and the time series data of the PM2.5 concentration within the previous 5 minutes before the current moment are collected by the PM2.5 sensor.

[0013] Furthermore, the response prediction model of the air purifier adopts a feedforward neural network, uses the current response value of the air purifier as a label, and trains and optimizes the response prediction model of the air purifier with historical environmental data; specifically including: an input layer, a hidden layer, an output layer and an activation function. The input layer is responsible for receiving historical environmental data; the hidden layer is used to process the historical environmental data; it consists of multiple layers, each layer contains 4 nodes, and the nodes of each hidden layer are connected to the previous layer through weights, which are used to perform feature abstraction and non-linear transformation on the input historical environmental data; by using the ReLu activation function, a non-linear relationship is introduced so that the model can fit complex feature relationships; an independent neuron is set in the output layer, which is responsible for converting the local and high-level feature representations extracted by the hidden layer for outputting the current response value of the air purifier; the root mean square error loss function is adopted; the input data is calculated once through the network to obtain the output result, the loss function is calculated according to the predicted value and the true value, the gradient of the loss function with respect to each weight and bias is calculated by the chain rule, and the weights and biases of the network are updated using the gradient descent algorithm to minimize the loss function.

[0014] Furthermore, the specific logic for generating air discomfort is: comparing the temperature and humidity at the current moment with the preset ideal temperature and humidity, and analyzing the comparison result to generate air discomfort; the specific formula for analyzing and generating air discomfort is:

[0015] AC = |T - T 0 ||W - W 0 |

[0016] where AC is air discomfort, T is the temperature at the current moment, W is the humidity at the current moment, T 0 is the ideal temperature, and W 0 is the ideal humidity;

[0017] The PM2.5 concentration time series data and the carbon dioxide concentration time series data are divided into n segments of equal length, which contain n + 1 time nodes in total, and each time node is sorted in chronological order;

[0018] The specific logic for generating the particle change trend is: calculating the dynamic change trend of each segment of PM2.5 concentration time series data through the sign function, and taking the average value as the particle change trend; the specific formula for generating the particle change trend is:

[0019]

[0020] where SK is the particle change trend, n is the number of time series data segments, Cpm2.5 i+1 is the PM2.5 concentration at the i + 1th time node, and Cpm2.5 iis the PM2.5 concentration at the i-th time node, where i is the index of the time node of the time series data;

[0021] The personnel fluidity is obtained by analyzing the time series data of carbon dioxide concentration; the specific logic is: calculate the change rate of carbon dioxide concentration at each time node, and the average value of the absolute value of the change rate of carbon dioxide concentration at each time node can be obtained; the specific formula for generating the personnel fluidity is:

[0022]

[0023] where RC is the personnel fluidity, CO 2 (i + 1) is the carbon dioxide concentration at the (i + 1)-th time node, CO 2 (i) is the carbon dioxide concentration at the i-th time node, and t is the time interval between the (i + 1)-th time node and the i-th time node;

[0024] The specific logic for obtaining the associated change rate of population particles is: obtain the average value of the carbon dioxide concentration when the museum is closed as the carbon dioxide reference concentration, calculate the crowding degree at each time node according to the carbon dioxide concentration and the carbon dioxide reference concentration at each time node, calculate the associated particle change rate according to the crowding degree and the PM2.5 concentration at each time node, and the specific formula for calculating the associated change rate of population particles is:

[0025]

[0026] where RS is the associated change rate of population particles, COD i is the crowding degree at the i-th time node, COD i+1 is the crowding degree at the i-th time node, CO 2B is the carbon dioxide reference concentration, Cpm2.5 i-1 is the PM2.5 concentration at the (i - 1)-th time node.

[0027] Furthermore, the specific logic for generating the response correction factor is: based on air discomfort, particle change trend, personnel fluidity, and the associated change rate of population particles, and combine the above three data to generate a response correction factor for correcting the current response value of the air purifier; generate the response correction factor:

[0028]

[0029] where XR is the response correction factor, AC is air discomfort, SK is the particle change trend, RC is the personnel fluidity, and RS is the associated change rate of population particles.

[0030] Further, the specific logic for calculating the corrected current response value is as follows: The predicted current response value of the air purifier is corrected using a response correction factor to obtain the corrected current response value of the air purifier; the specific formula for calculating the corrected current response value is:

[0031] px′ = px(1 + XR%)

[0032] where px′ is the corrected current response value, px is the predicted current response value of the air purifier, and XR is the response correction factor.

[0033] The present invention further provides an exhibition hall environment control system based on a sensor group. The method is used to implement the exhibition hall environment control method based on the sensor group. The specific steps include:

[0034] A data acquisition module, configured to obtain exhibition hall environment data through the sensor group. The exhibition hall environment data includes moment environment data and period environment data;

[0035] A response prediction module, configured to obtain historical moment environment data and the current response value of the historical air purifier; construct a response prediction model of the air purifier with the historical moment environment data as the input and the current response value of the air purifier at the same moment as the label; input the current moment environment data into the response prediction model of the air purifier to obtain the predicted current response value of the air purifier;

[0036] A correction calculation module, configured to analyze the moment environment data to generate air discomfort, analyze the period environment data to obtain the particle change trend, personnel circulation, and the associated change rate of the crowd particles; generate a response correction factor based on the air discomfort, particle change trend, personnel circulation, and the associated change rate of the crowd particles;

[0037] A response correction module, configured to use the response correction factor to correct the predicted current response value of the air purifier to obtain the corrected current response value of the air purifier.

[0038] Compared with the prior art, the beneficial effects of the present invention are:

[0039] The present invention comprehensively analyzes the exhibition hall environment data to generate a response correction factor, and dynamically adjusts the current response value of the air purifier through the response correction factor, avoiding problems of over-response or inefficient operation, and achieving a balance between energy conservation and efficient purification. At the same time, the correlation modeling between the particle change trend and crowd activities enables the purifier response to actively adapt to complex environmental changes, thereby effectively improving the energy utilization efficiency. Description of the Drawings

[0040] Figure 1 Schematic diagram of the overall system structure of the present invention.

[0041] Figure 2 Schematic diagram of the overall method flow of the present invention. Specific implementation manners

[0042] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below in conjunction with specific embodiments.

[0043] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this term cover the elements or objects listed after this term and their equivalents, without excluding other elements or objects. The terms such as "connected" or "linked" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left" and "right" are only used to represent relative position relationships, and when the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0044] Embodiment:

[0045] Step 1: Obtain the exhibition hall environment data through a sensor group, where the exhibition hall environment data includes moment environment data and period environment data;

[0046] The sensor group includes a temperature sensor, a humidity sensor, a carbon dioxide sensor and a PM2.5 sensor; the moment environment data includes the temperature, humidity, carbon dioxide concentration and PM2.5 concentration at the current moment; the period environment data includes the time series data of the carbon dioxide concentration and the PM2.5 concentration within the previous 5 minutes before the current moment; the temperature at the current moment is collected by the temperature sensor; the humidity at the current moment is collected by the humidity sensor; the carbon dioxide concentration at the current moment and the time series data of the carbon dioxide concentration within the previous 5 minutes before the current moment are collected by the carbon dioxide sensor; the PM2.5 concentration at the current moment and the time series data of the PM2.5 concentration within the previous 5 minutes before the current moment are collected by the PM2.5 sensor.

[0047] All the exhibition hall environment data described above are processed by maximum-minimum normalization.

[0048] Step 2: Obtain the environmental data at historical moments and the current response values of the historical air purifiers; use the environmental data at historical moments as the input, and at the same time, use the current response values of the air purifiers as the labels to construct a response prediction model for the air purifiers; input the environmental data at the current moment into the response prediction model of the air purifiers to obtain the predicted current response values of the air purifiers;

[0049] The response prediction model of the air purifier adopts a feedforward neural network. Using the current response values of the air purifiers as the labels, train and optimize the response prediction model of the air purifier with the environmental data at historical moments; specifically including: an input layer, a hidden layer, an output layer, and an activation function. The input layer is responsible for receiving the environmental data at historical moments; the hidden layer is used to process the environmental data at historical moments; it consists of multiple layers, each layer contains 4 time nodes, and the time nodes of each hidden layer are connected to the previous layer through weights, which are used to abstract features and perform non-linear transformations on the input environmental data at historical moments; by using the ReLu activation function, introduce non-linear relationships so that the model can fit complex feature relationships; there is an independent neuron in the output layer, which is responsible for converting the local and high-level feature representations extracted by the hidden layer for outputting the current response values of the air purifiers; adopt the root mean square error loss function; calculate the input data through the network once to obtain the output result, calculate the loss function based on the predicted value and the true value, calculate the gradient of the loss function with respect to each weight and bias through the chain rule, and use the gradient descent algorithm to update the weights and biases of the network to minimize the loss function.

[0050] Step 3: Analyze the environmental data at a moment to generate air discomfort, analyze the environmental data in a time period to obtain the particle change trend, personnel circulation, and the change rate of the particle association of the crowd; generate a response correction factor based on the air discomfort, particle change trend, personnel circulation, and the change rate of the particle association of the crowd;

[0051] Furthermore, the specific logic for generating air discomfort is: compare the temperature and humidity at the current moment with the preset ideal temperature and humidity, and analyze the comparison results to generate air discomfort; the specific formula for analyzing and generating air discomfort is:

[0052] AC = |T - T 0 ||W - W 0 |

[0053] where AC is the air discomfort, T is the temperature at the current moment, W is the humidity at the current moment, T 0 is the ideal temperature, and W 0 is the ideal humidity; the air discomfort represents the air discomfort of the current environment. The larger the value, the greater the deviation of the temperature and humidity of the environment from the ideal state, and people may feel uncomfortable in such an environment. |T - T 0|Represents the absolute difference between the current temperature and the ideal temperature. The larger the difference, the less comfortable the current temperature is. W - W 0 |Represents the absolute difference between the current humidity and the ideal humidity. The larger the difference, the more the current humidity deviates from the ideal state and the comfort level decreases. The formula quantifies the air comfort level as a numerical value, providing an important basis for subsequent environmental control analysis and adjustment.

[0054] Divide the time - series data of PM2.5 concentration and the time - series data of carbon dioxide concentration into n segments of equal length, which contains n + 1 time nodes, and sort each time node in chronological order;

[0055] The specific logic for generating the particulate change trend is: calculate the dynamic change trend of the PM2.5 concentration time - series data for each segment through the sign function and sum it as the particulate change trend; the specific formula for generating the particulate change trend is:

[0056]

[0057] Where SK is the particulate change trend, n is the number of segments of time - series data, Cpm2.5 i+1 is the PM2.5 concentration at the i + 1 - th time node, Cpm2.5 i is the PM2.5 concentration at the i - th time node, and i is the index of the time node of the time - series data; the particulate change trend reflects the overall change trend of the PM2.5 concentration. The larger its value, the greater the upward trend of the PM2.5 concentration. The PM2.5 concentration is the main object of the air purifier. The generation of the particulate change trend can provide an important basis for the control and adjustment of the air purifier. sgn(Cpm2.5 i+1 -Cpm2.5 i ) reflects the direction trend of the PM2.5 concentration data for each segment through the sign function; the larger its value, the greater the upward trend of the PM2.5 concentration for each segment; the summation operation is to reflect the overall trend change of the PM2.5 concentration reflected by the accumulation of each segment trend; if the sum is positive, it indicates that the upward trend is dominant and the particulate matter concentration continues to increase. If the sum is negative, it indicates that the downward trend is dominant and the particulate matter concentration continues to decrease.

[0058] The personnel mobility is obtained by analyzing the time - series data of carbon dioxide concentration; the specific logic is: calculate the change rate of carbon dioxide concentration at each time node, and the mean value of the absolute values of the change rates of carbon dioxide concentration at each time node can be obtained; the specific formula for generating the personnel mobility is:

[0059]

[0060] Where RC is the personnel mobility, CO 2(i + 1) is the carbon dioxide concentration at the (i + 1)-th time node, CO 2 (i) is the carbon dioxide concentration at the i-th time node, and t is the time interval between the (i + 1)-th time node and the i-th time node;

[0061] The carbon dioxide concentration in the exhibition hall is mainly affected by the population in the hall. The more people there are in the hall, the higher the carbon dioxide concentration. The personnel circulation reflects the population flow in the exhibition hall by analyzing the carbon dioxide concentration; the personnel circulation at each time node is evaluated by differentiating the carbon dioxide concentration at the position of the time node. Since the carbon dioxide concentration is not the main target of the air purifier, the calculated value of the personnel circulation explores the movement of people, whether the population in the exhibition hall is increasing or decreasing; therefore, the differential result of the carbon dioxide concentration is expressed in absolute value.

[0062] The specific logic for obtaining the associated change rate of population particles is as follows: Obtain the average value of the carbon dioxide concentration when the museum closes as the carbon dioxide reference concentration. Calculate the crowding degree of the population at each time node based on the carbon dioxide concentration and the carbon dioxide reference concentration at each time node. Calculate the associated particle change rate based on the crowding degree of the population and the PM2.5 concentration at each time node. The specific formula for calculating the associated change rate of population particles is:

[0063]

[0064] where RS is the associated change rate of population particles, COD i is the crowding degree of the population at the i-th time node, COD i+1 is the crowding degree of the population at the i-th time node, CO 2B is the carbon dioxide reference concentration, Cpm2.5 i-1 is the PM2.5 concentration at the (i - 1)-th time node.

[0065] The crowding degree of the population is the change in the carbon dioxide concentration at the current time node relative to the reference concentration. Since the carbon dioxide concentration is mainly affected by the population in the exhibition hall, the crowding degree of the population can reflect the degree of crowding of the population in the exhibition hall. The larger the value, the greater the degree of crowding and the higher the population density in the exhibition hall; the associated change rate of population particles reflects the association between the population and the change in PM2.5 concentration. The smaller the value, the stronger the association between the associated change rate of population particles and the change in PM2.5 concentration; in the formula, if the association between the population and the change in PM2.5 concentration is strong, then the ratio of each change in PM2.5 concentration to the crowding degree of the population is closer to a fixed value. If the change in PM2.5 concentration is only associated with the population, then the value should be 0, but this situation is impossible in real life. Taking the absolute value is because no matter Whether the value increases or decreases indicates that the association between PM2.5 concentration changes and the population is weakening. For When performing the summation operation to evaluate the association between PM2.5 concentration changes and the population for this PM2.5 concentration data, dividing by the number of segments n of the time series data is to calculate the association between PM2.5 concentration changes and the population for each segment.

[0066] The specific logic for generating the response correction factor is as follows: Based on air discomfort, particle change trend, personnel mobility, and the change rate of the association between the population and particles, and combining the above three data to generate a response correction factor for correcting the current response value of the air purifier; generating the response correction factor:

[0067]

[0068] Among them, XR is the response correction factor, AC is air discomfort, SK is the particle change trend, RC is personnel mobility, and RS is the change rate of the association between the population and particles.

[0069] The current response value of the air purifier predicted using the above response prediction model of the air purifier can only reflect the control current response value required for the air purifier to adjust the exhibition hall air to satisfaction in a static state. However, in real life, this static state is almost impossible, so it is necessary to consider the changes in the personnel in the exhibition hall. The response correction factor reflects the correction of the current response value of the air purifier considering the above personnel changes. The larger its value, the greater the correction intensity, indicating that the influence of personnel activities or environmental changes on the current response value of the air purifier is greater, and the purifier requires a higher adjustment intensity. This formula combines air quality and personnel activities to dynamically correct the current response value of the purifier; it considers the disturbance of personnel mobility on air quality in real life, thereby more accurately regulating the air purifier.

[0070] Air discomfort reflects the degree of air discomfort in the exhibition hall. The larger its value, the greater the degree of air discomfort, and usually the more intense the population activities in the exhibition hall, which also intensifies the population flow, resulting in the need for a greater response from the air purifier to adjust; the particle change trend reflects the overall change trend of PM2.5 concentration. The larger its value, the greater the upward trend of PM2.5 concentration. As the main target of the air purifier, PM2.5 naturally requires a greater response. Personnel mobility and the change rate of the association between the population and particles jointly reflect the impact of population flow on PM2.5 concentration in the exhibition hall. In the formula, SK + RC summarizes the comprehensive impact of personnel mobility and particulate matter change trends on air quality, ensuring that the correction factor is sensitive to multi-dimensional characteristics. It has the effect of amplifying the changing trend of particulate matter and the impact of population flow on air quality. The larger its value, the greater the impact of the changing trend of particulate matter and population flow on air quality. Moreover, the exponential function is always greater than zero, avoiding the situation of negative-negative getting positive with SK+RS+RC. It reflects the impact of population flow on air quality. The larger its value, the greater the impact of population flow on air quality. Among them, under normal circumstances, an increase in population flow will cause an increase in particulate matter in the air. The increase in particulate matter in the air is not only affected by population flow. Among them, the population-particle correlation change rate reflects the correlation between the population and the change in PM2.5 concentration. The smaller its value, the stronger the correlation between the population-particle correlation change rate and the change in PM2.5 concentration; The personnel circulation and population particles are passed through The mathematical form reflects the change in particulate matter in the air caused by the change in the flow of people, thus accurately reflecting the impact of population flow on air quality.

[0071] Step 4: Use the response correction factor to correct the predicted current response value of the air purifier to obtain the corrected current response value of the air purifier.

[0072] The specific logic for calculating the corrected current response value is: Use the response correction factor to correct the predicted current response value of the air purifier to obtain the corrected current response value of the air purifier; The specific formula for calculating the corrected current response value is:

[0073] px′=px(1+XR%)

[0074] Where, px′ is the corrected current response value, px is the predicted current response value of the air purifier, and XR is the response correction factor.

[0075] The present invention further provides an exhibition hall environment control system based on a sensor group. The method is used to implement the exhibition hall environment control method based on the sensor group. The specific steps include:

[0076] Step 1: Obtain the exhibition hall environment data through the sensor group. The exhibition hall environment data includes moment environment data and period environment data;

[0077] Step 2: Obtain the historical moment environment data and the current response value of the historical air purifier; Use the historical moment environment data as the input, and at the same moment, use the current response value of the air purifier as the label to construct the response prediction model of the air purifier; Input the current moment environment data into the response prediction model of the air purifier to obtain the predicted current response value of the air purifier;

[0078] Step 3: Analyze the moment environmental data to generate air discomfort, analyze the period environmental data to obtain the particulate change trend, personnel fluidity, and the change rate of the correlation between the crowd and particulate matter; generate a response correction factor based on the air discomfort, particulate change trend, personnel fluidity, and the change rate of the correlation between the crowd and particulate matter.

[0079] Step 4: Use the response correction factor to correct the predicted current response value of the air purifier to obtain the corrected current response value of the air purifier.

[0080] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0081] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0082] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0083] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all of them should be covered by the protection scope of the present application.

Claims

1. A method for controlling exhibition hall environment based on a sensor group, characterized in that: The specific steps include: Step 1: Acquire exhibition hall environment data through a sensor group, wherein the exhibition hall environment data includes moment environment data and time period environment data; Step 2: Obtain historical environmental data and historical current response values ​​of air purifiers; historical environmental data is used as input, and the current response value of the air purifier at the same time is used as a label to build a response prediction model of the air purifier; the current environmental data is input into the response prediction model of the air purifier to obtain the predicted current response value of the air purifier; Step 3: Analyze the environmental data at each moment to generate air discomfort, and analyze the environmental data for each period to obtain the particle change trend, personnel flow and the change rate of the correlation between people and particles; generate a response correction factor based on the air discomfort, particle change trend, personnel flow and the change rate of the correlation between people and particles; Step 4: Use the response correction factor to correct the predicted current response value of the air purifier to obtain the corrected current response value of the air purifier.

2. The method for controlling exhibition hall environment based on a sensor group according to claim 1, characterized in that: The sensor group includes a temperature sensor, a humidity sensor, a carbon dioxide sensor and a PM2.5 sensor; the momentary environmental data includes the temperature, humidity, carbon dioxide concentration and PM2.5 concentration at the current moment; the period environmental data includes the carbon dioxide concentration and PM2.5 concentration time series data within 5 minutes before the current moment; the temperature at the current moment is collected by a temperature sensor; the humidity at the current moment is collected by a humidity sensor; the carbon dioxide concentration at the current moment and the carbon dioxide concentration time series data within 5 minutes before the current moment are collected by a carbon dioxide sensor; the PM2.5 concentration at the current moment and the PM2.5 concentration time series data within 5 minutes before the current moment are collected by a PM2.5 sensor.

3. The method for controlling exhibition hall environment based on a sensor group according to claim 1, characterized in that: The response prediction model of the air purifier adopts a feedforward neural network, takes the current response value of the air purifier as a label, and uses the environmental data at historical moments to train and optimize the response prediction model of the air purifier; specifically, it includes: an input layer, a hidden layer, an output layer, and an activation function. The input layer is responsible for receiving the environmental data at historical moments; the hidden layer is used to process the environmental data at historical moments; it is composed of multiple layers, each layer contains 4 nodes, and the nodes of each hidden layer are connected to the previous layer through weights, which are used to perform feature abstraction and nonlinear transformation on the input environmental data at historical moments; by using the ReLu activation function, nonlinear relationships are introduced so that the model can fit complex feature relationships; an independent neuron is set in the output layer, which is responsible for converting the local and high-level feature representations extracted by the hidden layer to output the current response value of the air purifier; the root mean square error loss function is adopted; the input data is calculated once through the network to obtain the output result, the loss function is calculated according to the predicted value and the true value, the gradient of the loss function for each weight and bias is calculated by the chain rule, and the weights and biases of the network are updated using the gradient descent algorithm to minimize the loss function.

4. The method for controlling exhibition hall environment based on a sensor group according to claim 1, characterized in that: The specific logic for generating air unsuitability is: compare the current temperature and humidity with the preset ideal temperature and humidity, and analyze the comparison results to generate air unsuitability; the specific formula for analyzing and generating air unsuitability is: AC=|T-T0||W-W0| Where AC is the air unsuitability, T is the temperature at the current moment, W is the humidity at the current moment, T0 is the ideal temperature, and W0 is the ideal humidity; The PM2.5 concentration time series data and the carbon dioxide concentration time series data are divided into n segments of equal length, including a total of n+1 time nodes, and each time node is sorted in chronological order; The specific logic for generating the particle change trend is: the dynamic change trend of each section of PM2.5 concentration time series data is calculated through the symbolic function, and the average value is obtained as the particle change trend; the specific formula for generating the particle change trend is: Among them, SK is the particle change trend, n is the number of time series data segments, Cpm2.5 i+1 is the PM2.5 concentration at the i+1th time node, Cpm2.5 i is the PM2.5 concentration at the i-th time node, i is the index of the time node of the time series data; The personnel mobility is obtained by analyzing the time series data of carbon dioxide concentration; the specific logic is: calculate the change rate of carbon dioxide concentration at each time node, and obtain the average of the absolute value of the change rate of carbon dioxide concentration at each time node; the specific formula for generating personnel mobility is: Among them, RC is the personnel mobility, CO2(i+1) is the carbon dioxide concentration at the i+1th time node, CO2(i) is the carbon dioxide concentration at the i-th time node, and t is the time interval between the i+1th time node and the i-th time node; The specific logic for obtaining the crowd particle correlation change rate is as follows: the average value of the carbon dioxide concentration when the museum is closed is obtained as the carbon dioxide baseline concentration, the crowd congestion at each time node is calculated based on the carbon dioxide concentration at each time node and the carbon dioxide baseline concentration, and the correlation particle change rate is calculated based on the crowd congestion and PM2.5 concentration at each time node. The specific formula for calculating the crowd particle correlation change rate is as follows: Among them, RS is the population particle correlation change rate, COD i is the crowd congestion degree at the i-th time node, COD i+1 is the crowd congestion degree at the i-th time node, CO 2B is the baseline concentration of carbon dioxide, Cpm2.5 i-1 is the PM2.5 concentration at the i-1th time node.

5. The method for controlling exhibition hall environment based on a sensor group according to claim 1, characterized in that: The specific logic for generating the response correction factor is as follows: based on air unsuitability, particle change trend, personnel mobility and population-particle correlation change rate, the response correction factor for correcting the current response value of the air purifier is generated by combining the above three data; Generate the response correction factor: Among them, XR is the response correction factor, AC is the air discomfort, SK is the particle change trend, RC is the personnel mobility, and RS is the population-particle association change rate.

6. The method for controlling exhibition hall environment based on a sensor group according to claim 1, characterized in that: The specific logic for calculating the corrected current response value is: using the response correction factor to correct the predicted current response value of the air purifier to obtain the corrected current response value of the air purifier; the specific formula for calculating the corrected current response value is: px′=px(1+XR%) Wherein, px′ is the corrected current response value, px is the predicted current response value of the air purifier, and XR is the response correction factor.

7. An exhibition hall environment control system based on a sensor group, characterized in that: The method is used to implement the exhibition hall environment control method based on a sensor group according to any one of claims 1 to 6, and the specific steps include: A data acquisition module, used to obtain exhibition hall environment data through a sensor group, wherein the exhibition hall environment data includes moment environment data and time period environment data; The response prediction module is used to obtain the environmental data at historical moments and the current response values ​​of the air purifier at historical moments; the environmental data at historical moments are used as input, and the current response values ​​of the air purifier are recorded as labels to construct the response prediction model of the air purifier; the environmental data at the current moment is input into the response prediction model of the air purifier to obtain the predicted current response value of the air purifier; The correction calculation module is used to analyze the environmental data at each moment to generate air discomfort, and analyze the environmental data during each period to obtain the particle change trend, personnel flow and the change rate of the correlation between the crowd and particles; and generate a response correction factor based on the air discomfort, particle change trend, personnel flow and the change rate of the correlation between the crowd and particles; The response correction module is used to correct the predicted current response value of the air purifier using the response correction factor to obtain a corrected current response value of the air purifier.

Images