Method and system for controlling indoor air quality through five-constant system

Abstract

The invention discloses a method and system for controlling indoor air quality through a five-constant system. The method comprises the steps that a sample data processing module obtains historical air quality information and corresponding overall sample data of purification equipment output frequency data in the historical day in different time periods; associating the output frequency data of the purification equipment in different time periods every day with the historical air quality information to obtain associated sample data; the model calculation and processing module inputs the current divided time period and the current historical air quality information into a target machine learning model for model calculation and processing, calculates the set similarity between the current divided time period and the current historical air quality information and the associated sample data, and then determines a matching result with the highest similarity; and the air quality feedback regulation module directly controls the purification equipment to carry out intervention control treatment according to a matching result output by the model calculation processing module. The method has the technical advantages of high adaptability, energy conservation, high efficiency, high control intelligence and the like.

Description

Technical Field

[0001] The invention relates to the field of air quality feedback control, and in particular to a method for controlling indoor air quality using a five-constant system. Background Art

[0002] In our daily life at home, we often encounter some embarrassing situations. For example, we feel cold and shivering when we get up in the morning, feel stuffy indoors during lunch break, and feel too cold when we go to bed at night. This uneven temperature makes us feel uncomfortable and even causes colds and other diseases.

[0003] During the rainy season, high humidity is also a headache. You always feel sticky and your furniture and clothes are prone to dampness and mold.

[0004] Most of our time is spent indoors, so indoor temperature, humidity and air quality have a great impact on our health and comfort. In the air purification industry, in order to solve this problem, people have developed the "Five Constant Systems", which are constant temperature, constant humidity, constant oxygen, constant cleanliness and constant quietness processing systems that keep indoor air fresh, hence the name Five Constant Systems.

[0005] Among them, Hengjie specifically means: through the purification function of the fresh air system (hereinafter referred to as purification equipment), formaldehyde, C02 and other harmful substances in the indoor air can be discharged; at the same time, the triple filter of the fresh air system can filter out outdoor dust, haze, PM2.5, TOVC and other harmful substances, keeping the air clean all day long and the indoor air fresh and natural.

[0006] During further research, it was found that the Hengjie feedback control to ensure indoor air quality is to receive remote control commands sent by users, but this control method cannot meet the modern technical requirements of electrical appliance intelligence and autonomous control. Therefore, the control and adjustment of air purification equipment in the existing technology still faces the problem of insufficient intelligent control. Summary of the invention

[0007] The object of the present invention is to provide a method for controlling indoor air quality by a five-constant system, comprising the following steps:

[0008] The data acquisition module is pre-installed at multiple detection locations in the room: the data acquisition module collects indoor air quality information in real time;

[0009] The preprocessing module and the data acquisition module are combined in a one-to-one correspondence and installed at multiple detection positions: the preprocessing module is connected to the data acquisition module in real time for communication and collects the indoor air quality information sent by the data acquisition module; the preprocessing module also preprocesses the indoor air quality information to obtain preprocessed target data, and then sends the preprocessed target data to the sample data processing module;

[0010] The sample data processing module obtains the above historical air quality information and the overall sample data of the corresponding daily time-divided purification equipment output frequency data; associates the daily time-divided purification equipment output frequency data with the historical air quality information to obtain the associated sample data; trains the associated sample data as a training set for the machine learning model to finally obtain the target machine learning model; the overall sample data divides the 24-hour range of each day into 48 time periods;

[0011] The time segment information collection module collects the current time and determines the time segment to which the current time belongs;

[0012] The model calculation processing module inputs the current time segment and the current historical air quality information into the target machine learning model for model calculation processing, calculates the set similarity of the current time segment and the current historical air quality information with the associated sample data, and then determines the matching result with the highest similarity; the matching result includes the purification equipment output frequency data corresponding to the current time segment as the target control data;

[0013] The air quality feedback adjustment module directly controls the purification equipment to perform intervention control processing according to the matching results output by the model calculation processing module;

[0014] The secondary monitoring and processing module detects the indoor air quality information within a preset time after the current moment, and determines whether the indoor air quality information within the preset time after the intervention meets the standard control quality. If it is judged to be compliant, the control ends; if it is judged to be compliant, the human-computer interaction module is started;

[0015] The human-computer interaction module directly sends an alarm message, and then receives a manual intervention control instruction input by the user; analyzes the manual intervention control instruction, and starts a control operation to adjust the output frequency of the purification equipment.

[0016] Preferably, as an implementable embodiment; the pretreatment comprises:

[0017] The collected indoor air quality information is used for outlier detection, and if an outlier is detected, the current outlier is removed;

[0018] Use a preset smoothing method to smooth the indoor air quality information data after removing abnormal values;

[0019] The smoothed indoor air quality data is subjected to feature extraction of the time period average value to obtain the preprocessed target data.

[0020] Preferably, as an implementable solution, the preset smoothing method is a Savitzky-Golay filtering method.

[0021] Preferably, as an implementable solution, the data acquisition module includes a PM2.5 sensor, a CO2 sensor, and a formaldehyde detection sensor; the indoor air quality information includes PM2.5 concentration information, CO2 concentration information, and formaldehyde concentration information.

[0022] Preferably, as an implementable solution, the human-computer interaction module includes an embedded touch screen; the embedded touch screen is used for manual intervention control instructions input by the user.

[0023] Preferably, as an implementable method, the current time segment and the current historical air quality information are input into the target machine learning model for model calculation processing, and the set similarity between the current time segment and the current historical air quality information and the associated sample data is calculated, specifically including:

[0024] Input the current time period information and the current historical air quality information as query samples into the target machine learning model;

[0025] Obtain previously trained associated sample data from the sample data processing module, including the air quality information of each historical time period and the corresponding purification equipment output frequency;

[0026] Use the cosine similarity algorithm to calculate the similarity between the current query sample and the historical associated samples;

[0027] From the historical associated samples, select the associated sample with the highest similarity to the current query sample;

[0028] The output frequency of the purification equipment and the corresponding time period corresponding to the best matching associated sample are extracted as target control data.

[0029] Preferably, as an implementable solution, the human-computer interaction module also receives the purification equipment control satisfaction information sent by the current user, and determines the time period corresponding to the purification equipment control satisfaction information currently sent; the purification equipment control satisfaction information includes "good", "satisfied" and "unsatisfied";

[0030] The indoor air quality information corresponding to the "good" result reflected in the purification equipment control satisfaction information and the corresponding purification equipment output frequency data in the current time period are taken as typical samples and stored in the overall sample data.

[0031] Preferably, as an implementable embodiment, the preset time is 15 minutes to minutes.

[0032] Preferably, as an implementable solution, the data acquisition module is communicatively connected with the preprocessing module via a Zigbee module.

[0033] Accordingly, the present invention provides a control system for indoor air quality by a five-constant system, which implements indoor air quality control operations by using a control method for indoor air quality by a five-constant system, including a data acquisition module, a preprocessing module, a sample data processing module, a time period information acquisition module, a model calculation processing module, an air quality feedback adjustment module, a secondary monitoring processing module, and a human-computer interaction module;

[0034] The data acquisition module is used to pre-install multiple detection locations in the room and collect indoor air quality information in real time;

[0035] The preprocessing module is used to communicate with the data acquisition module in real time and collect the indoor air quality information sent by the data acquisition module; the preprocessing module also preprocesses the indoor air quality information to obtain preprocessed target data, and then sends the preprocessed target data to the sample data processing module;

[0036] The sample data processing module is used to obtain the above-mentioned historical air quality information and the corresponding overall sample data of the output frequency data of the purification equipment in different time periods every day in history; associate the output frequency data of the purification equipment in different time periods every day with the historical air quality information to obtain the associated sample data; train the associated sample data as the training set of the machine learning model to finally obtain the target machine learning model; the overall sample data divides each day into 48 time periods;

[0037] The time segment information collection module is used to collect the current time and determine the time segment to which the current time belongs;

[0038] The model calculation processing module is used to input the current time segment and the current historical air quality information into the target machine learning model for model calculation processing, calculate the set similarity between the current time segment and the current historical air quality information and the associated sample data, and then determine the matching result with the highest similarity; the matching result includes the purification equipment output frequency data corresponding to the current time segment as the target control data;

[0039] The air quality feedback adjustment module is used to directly control the purification equipment to perform intervention control processing according to the matching result output by the model calculation processing module;

[0040] The secondary monitoring processing module is used to detect the indoor air quality information within a preset time after the current moment, and determine whether the indoor air quality information within the preset time after the intervention meets the standard control quality. If it is determined to be compliant, the control ends; if it is determined not to be compliant, the human-computer interaction module is started;

[0041] The human-computer interaction module is used to directly send alarm information, and then receive the manual intervention control instruction input by the user; analyze the manual intervention control instruction, and start the control operation of adjusting the output frequency of the purification equipment.

[0042] Compared with the prior art, the embodiments of the present invention have at least the following technical advantages:

[0043] By analyzing the specific scheme of the method for controlling indoor air quality by the five-constant system provided by the present invention, it can be known that in specific application;

[0044] The data acquisition module is pre-installed at multiple detection positions in the room: the data acquisition module collects indoor air quality information in real time; the pre-processing module is combined with the data acquisition module in a one-to-one correspondence and installed at multiple detection positions: the pre-processing module communicates with the data acquisition module in real time and collects the indoor air quality information sent by the data acquisition module; the pre-processing module also pre-processes the indoor air quality information to obtain pre-processed target data, and then sends the pre-processed target data to the sample data processing module;

[0045] The sample data processing module obtains the above historical air quality information and the overall sample data of the corresponding daily time-divided purification equipment output frequency data; associates the daily time-divided purification equipment output frequency data with the historical air quality information to obtain the associated sample data; trains the associated sample data as a training set for the machine learning model to finally obtain the target machine learning model; the overall sample data divides the 24-hour range of each day into 48 time periods;

[0046] The time segment information collection module collects the current time and determines the time segment to which the current time belongs;

[0047] The model calculation processing module inputs the current time segment and the current historical air quality information into the target machine learning model for model calculation processing, calculates the set similarity of the current time segment and the current historical air quality information with the associated sample data, and then determines the matching result with the highest similarity; the matching result includes the purification equipment output frequency data corresponding to the current time segment as the target control data;

[0048] The air quality feedback adjustment module directly controls the purification equipment to perform intervention control processing according to the matching results output by the model calculation processing module;

[0049] The secondary monitoring and processing module detects the indoor air quality information within a preset time after the current moment, and determines whether the indoor air quality information within the preset time after the intervention meets the standard control quality. If it is judged to be compliant, the control ends; if it is judged to be compliant, the human-computer interaction module is started;

[0050] From the analysis of the above content, we can know that the above-mentioned five-constant system control method for indoor air quality can collect and preprocess indoor air quality data in real time by setting up data acquisition modules and preprocessing modules, providing basic data support for subsequent machine learning modeling and control optimization. By collecting historical air quality data and purification equipment operation data, an associated sample data set is established, and a machine learning model that can predict the optimal purification equipment output frequency in the current time period is trained. This data-driven intelligent control model can achieve more accurate air quality control.

[0051] According to the current time period and air quality information, the machine learning model is called for matching analysis to obtain the optimal control parameters of the purification equipment, and the operation of the purification equipment is directly controlled through the feedback adjustment module, realizing fully automatic intelligent control of air quality.

[0052] In summary, the above control method greatly improves the intelligence level of indoor air quality management and provides effective technical support for creating a comfortable and healthy indoor environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 A flow chart of a method for controlling indoor air quality by a five-constant system provided in an embodiment of the present invention;

[0055] Figure 2 A flowchart of a specific operation in the method for controlling indoor air quality by the five constant systems provided in an embodiment of the present invention;

[0056] Figure 3 A flowchart of another specific operation in the method for controlling indoor air quality by the five constant systems provided in an embodiment of the present invention;

[0057] Figure 4 A schematic diagram of the principle structure of a control system for indoor air quality provided by the five-constant system in an embodiment of the present invention.

[0058] Labels: data acquisition module 10, preprocessing module 20, sample data processing module 30, time period information acquisition module 40, model calculation processing module 50, air quality feedback adjustment module 60, secondary monitoring processing module 70, human-computer interaction module 80. DETAILED DESCRIPTION

[0059] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] The present invention is further described in detail below through specific implementation examples in conjunction with the accompanying drawings.

[0061] See also Figure 1 The present invention provides a method for controlling indoor air quality using a five-constant system, comprising the following steps:

[0062] S10: The data acquisition module 10 is pre-installed at multiple detection locations in the room: the data acquisition module collects indoor air quality information in real time;

[0063] S20: The preprocessing module 20 is combined with the data acquisition module 10 in a one-to-one correspondence and is installed at multiple detection locations: the preprocessing module 20 is used to communicate with the data acquisition module 10 in real time and collect the indoor air quality information sent by the data acquisition module 10; the preprocessing module 20 is also used to preprocess the indoor air quality information to obtain preprocessed target data, and the preprocessing includes removing abnormal values, smoothing curves, etc. Then the preprocessed target data is sent to the sample data processing module;

[0064] S30: The sample data processing module 30 obtains the above-mentioned historical air quality information and the overall sample data of the corresponding daily time-divided purification equipment output frequency data; associates the daily time-divided purification equipment output frequency data with the historical air quality information (i.e., the corresponding daily time-divided purification equipment output frequency data and the historical air quality information of the corresponding time period are obtained) to obtain the associated sample data; the associated sample data is used as a training set of the machine learning model to train and finally obtain the target machine learning model; the overall sample data divides each day into 48 time periods; that is, every half hour is divided into a time period, because the daytime and nighttime of each day have great control requirements and different actual environments; the same is the daytime time, 6-8 am every morning is the time to get up, the indoor air is very easy to be polluted in a short time, and a lot of smoke and dust appear, and the demand for air quality purification is often very large at this time;

[0065] S40: The time segment information collection module 40 collects the time of the current moment and determines the time segment corresponding to the current moment;

[0066] S50: The model calculation processing module 50 inputs the current time period and the current historical air quality information into the target machine learning model for model calculation processing, calculates the set similarity between the current time period and the current historical air quality information and the associated sample data, and then determines the matching result with the highest similarity; the matching result includes the purification equipment output frequency data corresponding to the current time period as the target control data; it should be noted that the above matching result can be used as the target control data to adjust the operating frequency and output of the corresponding air purification equipment to maintain the indoor air quality within an ideal range.

[0067] S60: the air quality feedback adjustment module 60 directly controls the purification equipment to perform intervention control processing according to the matching result output by the model calculation processing module;

[0068] S70: The secondary monitoring and processing module 70 detects the indoor air quality information within the preset time after the current moment, and determines whether the indoor air quality information at the preset time after the intervention meets the standard control quality. If it is judged to be compliant, the control ends; if it is judged to be non-compliant, the human-computer interaction module 80 is started; the preset time is 15 minutes to 30 minutes.

[0069] S80: the human-computer interaction module 80 directly sends an alarm message, and then receives a manual intervention control instruction input by a user; analyzes the manual intervention control instruction, and starts a control operation of adjusting the output frequency of the purification device.

[0070] The above-mentioned five-constant system control method for indoor air quality can collect and preprocess indoor air quality data in real time by setting up data acquisition modules and preprocessing modules, providing basic data support for subsequent machine learning modeling and control optimization. By collecting historical air quality data and purification equipment operation data, an associated sample data set is established, and a machine learning model that can predict the optimal purification equipment output frequency in the current time period is trained. This data-driven intelligent control model can achieve more accurate air quality control.

[0071] According to the current time period and air quality information, the machine learning model is called for matching analysis to obtain the optimal control parameters of the purification equipment, and the operation of the purification equipment is directly controlled through the feedback adjustment module, realizing fully automatic intelligent control of air quality.

[0072] In summary, the above control method greatly improves the intelligence level of indoor air quality management and provides effective technical support for creating a comfortable and healthy indoor environment.

[0073] At the same time, this method also adds human-computer interaction and monitoring functions: a secondary monitoring and processing module and a human-computer interaction module are added to monitor and evaluate the automatic control effect. When an abnormal situation occurs, an alarm can be given in time and manual intervention control instructions can be accepted, which improves the flexibility of the system.

[0074] See also Figure 2 , the preprocessing comprises:

[0075] S21: performing outlier detection on the collected indoor air quality information, and removing the current outlier if an outlier is detected;

[0076] S22: Smoothing the indoor air quality information data after removing abnormal values ​​using a preset smoothing method;

[0077] S23: Extract the feature of the time-period average value of the smoothed indoor air quality data to obtain the pre-processed target data.

[0078] It should be noted that the target data after the above preprocessing is statistical features such as average values. These features can better describe the state of indoor air quality and provide more effective input features for subsequent machine learning model training.

[0079] The specific operation logic steps of the preprocessing module 20 are as follows:

[0080] Removing outliers: Detecting and removing outliers from the indoor air quality information collected from the data acquisition module 10. Statistical methods such as the Z-score method and the IQR method can be used to identify and remove abnormal data that are obviously deviated from the normal range.

[0081] Smoothing curve: Smoothing the indoor air quality information data after removing outliers to reduce noise interference and improve the smoothness of the data. Generally speaking, the smoothing method includes using common smoothing algorithms such as moving average method, exponential smoothing method, Savitzky-Golay filtering, etc. This embodiment preferably uses Savitzky-Golay filtering method to better reflect the overall change trend of indoor air quality through smoothing.

[0082] Feature extraction: extract features from the smoothed indoor air quality data, such as statistical features such as mean value, variance, peak value, etc. This embodiment preferably uses the average value within the time period, which can better describe the state of indoor air quality and provide more effective input features for subsequent machine learning model training.

[0083] In general, the main function of the preprocessing module 20 is to clean, smooth and extract features from the original indoor air quality data to improve the quality and usability of the data and prepare for the subsequent machine learning model training. These specific preprocessing logic steps help to ensure the effectiveness and reliability of the air quality control method of the entire five-constant system.

[0084] Preferably, as an implementable solution, the preset smoothing method is a Savitzky-Golay filtering method.

[0085] Preferably, as an implementable solution, the data acquisition module includes a PM2.5 sensor, a CO2 sensor, and a formaldehyde detection sensor; the indoor air quality information includes PM2.5 concentration information, CO2 concentration information, and formaldehyde concentration information.

[0086] Preferably, as an implementable solution, the human-computer interaction module 80 includes an embedded touch screen; the embedded touch screen is used for manual intervention control instructions input by the user.

[0087] See also Figure 3 , input the current time period and the current historical air quality information into the target machine learning model for model calculation processing, and calculate the set similarity between the current time period and the current historical air quality information and the associated sample data, specifically including:

[0088] S51: inputting the current time period information and the current historical air quality information into the target machine learning model as query samples;

[0089] S52: Acquire previously trained associated sample data from the sample data processing module 30, including the air quality information of each historical time period and the corresponding purification equipment output frequency;

[0090] S53: Calculate the similarity between the current query sample and the historical associated samples using a cosine similarity algorithm;

[0091] S54: Selecting, from the historical associated samples, an associated sample having the highest similarity to the current query sample;

[0092] S55: extracting the purification equipment output frequency and the corresponding time period corresponding to the best matching associated sample as target control data.

[0093] This solution makes full use of the associated sample data previously established in the sample data processing module, which includes the air quality information of each historical time period and the corresponding purification equipment output frequency. Through similarity matching, this historical knowledge can be effectively applied to the current air quality control to improve the control effect. The intelligent control solution based on similarity matching makes full use of the experience knowledge accumulated from historical data to achieve more accurate and intelligent air quality control.

[0094] Preferably, as an implementable embodiment; the human-computer interaction module 80 also receives the purification equipment control satisfaction information sent by the current user, and determines the time period corresponding to the purification equipment control satisfaction information currently sent; the purification equipment control satisfaction information includes "good", "satisfied" and "unsatisfied";

[0095] The indoor air quality information corresponding to the "good" result reflected in the purification equipment control satisfaction information and the corresponding purification equipment output frequency data in the current time period are taken as typical samples and stored in the overall sample data.

[0096] The human-computer interaction module 80 also receives the purification equipment control satisfaction information sent by the current user, and determines the time period corresponding to the purification equipment control satisfaction information currently sent; the purification equipment control satisfaction information includes "good", "satisfied" and "unsatisfied";

[0097] However, only when the most ideal state, that is, the purification equipment control satisfaction information includes "good", is the relevant information extracted, it is most qualified to serve as the overall sample data; that is, the indoor air quality information corresponding to the "good" result reflected in the purification equipment control satisfaction information and the corresponding purification equipment output frequency data in the current time period are taken as typical samples and stored in the overall sample data.

[0098] Preferably, as an implementable solution, the data acquisition module is communicatively connected with the preprocessing module via a Zigbee module.

[0099] By adopting the Zigbee wireless communication module, wireless communication between the data acquisition module and the preprocessing module is realized. Zigbee is a low-power, low-cost short-range wireless communication technology with high stability and strong anti-interference ability, which is very suitable for data transmission needs in such distributed environments.

[0100] The Zigbee communication module belongs to the wireless communication architecture, which makes the data acquisition module and the pre-processing module not restricted by spatial location, and can be flexibly deployed according to the actual environment, which improves the convenience of system installation and expansion. At the same time, it also avoids the wiring cost and maintenance difficulty of traditional wired connection methods.

[0101] Embodiment 2

[0102] See also Figure 4 The present invention provides a control system for indoor air quality using a five-constant system, which uses a method for controlling indoor air quality using a five-constant system to implement indoor air quality control operations, including a data acquisition module 10, a preprocessing module 20, a sample data processing module 30, a time period information acquisition module 40, a model calculation processing module 50, an air quality feedback adjustment module 60, a secondary monitoring processing module 70, and a human-computer interaction module 80;

[0103] The data acquisition module 10 is used to pre-install multiple detection locations in the room and collect indoor air quality information in real time;

[0104] The preprocessing module 20 is used to communicate with the data acquisition module 10 in real time and collect the indoor air quality information sent by the data acquisition module 10; the preprocessing module 20 also preprocesses the indoor air quality information to obtain preprocessed target data, and then sends the preprocessed target data to the sample data processing module;

[0105] The sample data processing module 30 is used to obtain the above-mentioned historical air quality information and the corresponding overall sample data of the output frequency data of the purification equipment in different time periods every day in history; associate the output frequency data of the purification equipment in different time periods every day with the historical air quality information to obtain the associated sample data; train the associated sample data as a training set of the machine learning model to finally obtain the target machine learning model; the overall sample data divides each day into 48 time periods;

[0106] The time segment information collection module 40 is used to collect the current time and determine the time segment to which the current time belongs;

[0107] The model calculation processing module 50 is used to input the current time segment and the current historical air quality information into the target machine learning model for model calculation processing, calculate the set similarity between the current time segment and the current historical air quality information and the associated sample data, and then determine the matching result with the highest similarity; the matching result includes the purification equipment output frequency data corresponding to the current time segment as the target control data;

[0108] The air quality feedback adjustment module 60 is used to directly control the purification equipment to perform intervention control processing according to the matching result output by the model calculation processing module;

[0109] The secondary monitoring processing module 70 is used to detect the indoor air quality information within a preset time after the current moment, and determine whether the indoor air quality information within the preset time after the intervention meets the standard control quality. If it is determined to be compliant, the control ends; if it is determined not to be compliant, the human-computer interaction module 80 is started;

[0110] The human-computer interaction module 80 is used to directly send alarm information, and then receive the manual intervention control instruction input by the user; analyze the manual intervention control instruction, and start the control operation of adjusting the output frequency of the purification equipment.

[0111] The above-mentioned five-constant system control method for indoor air quality can collect and preprocess indoor air quality data in real time by setting up data acquisition modules and preprocessing modules, providing basic data support for subsequent machine learning modeling and control optimization. By collecting historical air quality data and purification equipment operation data, an associated sample data set is established, and a machine learning model that can predict the optimal purification equipment output frequency in the current time period is trained. This data-driven intelligent control model can achieve more accurate air quality control.

[0112] According to the current time period and air quality information, the machine learning model is called for matching analysis to obtain the optimal control parameters of the purification equipment, and the operation of the purification equipment is directly controlled through the feedback adjustment module, realizing fully automatic intelligent control of air quality.

[0113] In summary, the above control method greatly improves the intelligent level of indoor air quality management. By accurately controlling the output frequency of the purification equipment, excessive or insufficient purification is avoided, thereby achieving a balance between air quality control and equipment energy consumption. This method realizes the intelligent linkage between air quality information and purification equipment, greatly improving the automation level of air quality control.

[0114] 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. A person skilled in the art may modify the technical solutions described in the above embodiments, or replace part 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 controlling indoor air quality using a five-constant system, characterized in that: The following steps are included: The data acquisition module is pre-installed at multiple detection locations in the room: the data acquisition module collects indoor air quality information in real time; The pre-processing module and the data acquisition module are combined in a one-to-one correspondence and installed at multiple detection locations: the pre-processing module is connected to the data acquisition module in real time communication and collects the indoor air quality information sent by the data acquisition module; The preprocessing module also preprocesses the indoor air quality information to obtain preprocessed target data, and then sends the preprocessed target data to the sample data processing module; The sample data processing module obtains the above historical air quality information and the overall sample data of the corresponding daily time-based purification equipment output frequency data; associates the daily time-based purification equipment output frequency data with the historical air quality information to obtain associated sample data; trains the associated sample data as a training set for a machine learning model to finally obtain a target machine learning model; In the overall sample data, the 24-hour range of each day is equally divided into 48 time periods; The time segment information collection module collects the current time and determines the time segment to which the current time belongs; The model calculation processing module inputs the current time segment and the current historical air quality information into the target machine learning model for model calculation processing, calculates the set similarity of the current time segment and the current historical air quality information with the associated sample data, and then determines the matching result with the highest similarity; the matching result includes the purification equipment output frequency data corresponding to the current time segment as the target control data; The air quality feedback adjustment module directly controls the purification equipment to perform intervention control processing according to the matching results output by the model calculation processing module; The secondary monitoring and processing module detects the indoor air quality information within a preset time after the current moment, and determines whether the indoor air quality information within the preset time after the intervention meets the standard control quality. If it is judged to be compliant, the control ends; if it is judged to be compliant, the human-computer interaction module is started; The human-computer interaction module directly sends an alarm message, and then receives a manual intervention control instruction input by the user; analyzes the manual intervention control instruction, and starts a control operation to adjust the output frequency of the purification equipment.

2. A method for controlling indoor air quality by a five-constant system as claimed in claim 1, characterized in that: The pre-processing comprises: The collected indoor air quality information is used for outlier detection, and if an outlier is detected, the current outlier is removed; Use a preset smoothing method to smooth the indoor air quality information data after removing abnormal values; The smoothed indoor air quality data is subjected to feature extraction of the time period average value to obtain the preprocessed target data.

3. A method for controlling indoor air quality by a five-constant system as claimed in claim 2, characterized in that: The preset smoothing method is a Savitzky-Golay filtering method.

4. A method for controlling indoor air quality by a five-constant system as claimed in claim 1, characterized in that: The data acquisition module includes a PM2.5 sensor, a CO2 sensor, and a formaldehyde detection sensor; the indoor air quality information includes PM2.5 concentration information, CO2 concentration information, and formaldehyde concentration information.

5. The method for controlling indoor air quality by a five-constant system as claimed in claim 1, characterized in that: The human-computer interaction module 80 includes an embedded touch screen; the embedded touch screen is used for manual intervention control instructions input by the user.

6. A method for controlling indoor air quality using a five-constant system as claimed in claim 1, characterized in that: The current time period and the current historical air quality information are input into the target machine learning model for model calculation processing, and the set similarity between the current time period and the current historical air quality information and the associated sample data is calculated, specifically including: Input the current time period information and the current historical air quality information as query samples into the target machine learning model; Obtain previously trained associated sample data from the sample data processing module, including the air quality information of each historical time period and the corresponding purification equipment output frequency; Use the cosine similarity algorithm to calculate the similarity between the current query sample and the historical associated samples; From the historical associated samples, select the associated sample with the highest similarity to the current query sample; The output frequency of the purification equipment and the corresponding time period corresponding to the best matching associated sample are extracted as target control data.

7. A method for controlling indoor air quality using a five-constant system as claimed in claim 1, characterized in that: The human-computer interaction module also receives the purification equipment control satisfaction information sent by the current user, and determines the time period corresponding to the purification equipment control satisfaction information currently sent; the purification equipment control satisfaction information includes "good", "satisfied" and "unsatisfied"; The indoor air quality information corresponding to the "good" result reflected in the purification equipment control satisfaction information and the corresponding purification equipment output frequency data in the current time period are taken as typical samples and stored in the overall sample data.

8. A method for controlling indoor air quality by a five-constant system as claimed in claim 1, characterized in that: The preset time is 15 minutes to 30 minutes.

9. A method for controlling indoor air quality using a five-constant system as claimed in claim 1, characterized in that: The data acquisition module is communicatively connected with the preprocessing module via the Zigbee module.

10. A five-constant system for indoor air quality control, characterized in that: The method for controlling indoor air quality by using the five-constant system as claimed in any one of claims 1 to 9 implements indoor air quality control operation, including a data acquisition module, a preprocessing module, a sample data processing module, a time period information acquisition module, a model calculation processing module, an air quality feedback adjustment module, a secondary monitoring processing module, and a human-computer interaction module; The data acquisition module is used to pre-install multiple detection locations in the room and collect indoor air quality information in real time; The preprocessing module is used to communicate with the data acquisition module in real time and collect the indoor air quality information sent by the data acquisition module; The preprocessing module also preprocesses the indoor air quality information to obtain preprocessed target data, and then sends the preprocessed target data to the sample data processing module; The sample data processing module is used to obtain the above-mentioned historical air quality information and the overall sample data of the corresponding daily time-based purification equipment output frequency data; associate the daily time-based purification equipment output frequency data with the historical air quality information to obtain the associated sample data; train the associated sample data as a training set for the machine learning model to finally obtain the target machine learning model; The overall sample data will be divided into 48 time periods per day; The time segment information collection module is used to collect the current time and determine the time segment to which the current time belongs; The model calculation processing module is used to input the current time segment and the current historical air quality information into the target machine learning model for model calculation processing, calculate the set similarity between the current time segment and the current historical air quality information and the associated sample data, and then determine the matching result with the highest similarity; the matching result includes the purification equipment output frequency data corresponding to the current time segment as the target control data; The air quality feedback adjustment module is used to directly control the purification equipment to perform intervention control processing according to the matching result output by the model calculation processing module; The secondary monitoring processing module is used to detect the indoor air quality information within a preset time after the current moment, and determine whether the indoor air quality information within the preset time after the intervention meets the standard control quality. If it is determined to be compliant, the control ends; if it is determined not to be compliant, the human-computer interaction module is started; The human-computer interaction module is used to directly send alarm information, and then receive the manual intervention control instruction input by the user; analyze the manual intervention control instruction, and start the control operation of adjusting the output frequency of the purification equipment.

Images