A method for automatically monitoring indoor air quality

Abstract

The present invention discloses a method for automatic monitoring of indoor air quality, which belongs to the field of indoor air quality detection technology. In this method, a first data set and a second data set are obtained by monitoring cells, air inlets and air outlets of connected areas and indoor areas, an air state model of the connected area and each indoor area is constructed, the predicted state transition probability and the state input probability are substituted into the air state model, and then a state vector of the next state transition step is generated to reduce the data processing volume of multi-area air quality monitoring and achieve the effect of early prediction. Furthermore, the present invention monitors multiple components through the air state model, generates a negative parameter set and evaluates the negative air quality index from the aspects of pathogen infection, lung damage, explosion hazard, etc., generates a positive parameter set and evaluates the positive air quality index from the aspects of suitable temperature and humidity, suitable O2 concentration, suitable CO2 concentration, etc., to achieve comprehensive monitoring of air quality.

Description

Technical Field

[0001] The present invention relates to the technical field of air quality time series data processing, and in particular to a method for automatically monitoring indoor air quality. Background Art

[0002] There are multiple types of monitoring objects for indoor air quality. For example, fine particles such as PM2.5 harm the lungs and respiratory tract, bacteria and fungi in the air are infectious substances, and carbon monoxide in coal gas is a combustible gas. These components can be included in the monitoring range as harmful substances. For example, the concentration of oxygen, carbon dioxide, etc. is directly related to the suitability of body sensation. The Chinese patent application with publication number CN110410922A discloses a natural ventilator and a real-time indoor air quality monitoring method. The method uses an air quality monitoring sensor arranged indoors to send the monitored electrical signal to a data receiving module using a data sending module, and then obtains various indoor environmental parameters and the overall fresh air volume of the room after digital-to-analog conversion and analysis processing by the data analysis and processing module, and displays them in real time through a display panel. The display output of the current air quality is a basic issue in air quality monitoring. To further improve the ability to predict air quality, a Chinese patent application with publication number CN107145737A discloses a Markov chain-based inverse identification algorithm for pollution sources in unsteady flow fields. This method uses a Markov chain model to estimate the transfer of air pollutants and predict changes in pollutant concentrations and the location of pollution sources. For buildings with large areas and high traffic volume, such as office buildings, if the entire building is used as a prediction area, the amount of data required by the Markov chain is very large. Buildings usually contain multiple relatively independent spaces, and pollutants are exchanged between relatively independent spaces through areas such as corridors. Therefore, it is necessary to propose a new automatic indoor air quality monitoring method to solve the problem of monitoring air quality in multiple indoor areas. Summary of the Invention

[0003] To address the shortcomings of the aforementioned prior art, the present invention proposes an automatic indoor air quality monitoring method. This method obtains a first data set and a second data set by monitoring cells, air inlets, and air outlets in connected and indoor areas. Air state models are then constructed for the connected areas and each indoor area. The predicted state transition probabilities and state input probabilities are substituted into the air state models, and the state vector for the next state transition step is generated. This reduces the data processing workload for multi-area air quality monitoring and achieves advance prediction. Furthermore, the air state model monitors multiple air components and generates negative and positive air quality indices based on the concentrations of different components, enabling comprehensive monitoring of indoor air quality.

[0004] The technical solution of the present invention is achieved as follows:

[0005] A method for automatically monitoring indoor air quality comprises the following steps:

[0006] Step 1: Divide the target space into a connected area and at least two indoor areas, and set environmental monitoring nodes at the air inlet and outlet of the connected area and each indoor area;

[0007] Step 2: Divide the connected area and indoor area into multiple cells, arrange concentration monitoring nodes in multiple cells, and create air state models for the connected area and each indoor area;

[0008] Step 3: Collect the first data set of each concentration monitoring node and the second data set of the environmental monitoring node. If the sampling data of the first data set exceeds the threshold range of the corresponding component, issue a warning notification and end the task. Otherwise, proceed to step 4.

[0009] Step 4: Generate the state transition probability and state transition step length of the air state model according to the first data set and the second data set of the connected area and the indoor area respectively;

[0010] Step 5: Generate gas mobility based on the first and second data sets of different indoor areas and connected areas at the same time and the state transition step, and then generate the state input probability of the air state model based on the gas mobility;

[0011] Step 6: Update the air state model according to the state transition probability and the state input probability, and predict the third data set of the next state transition step length according to the air state model and the first data set;

[0012] Step 7: Construct response curves of different components, and generate negative parameter sets and positive parameter sets for connected areas and indoor areas respectively according to the third data set and the response curves;

[0013] Step 8: Calculate the negative air quality index based on the negative parameter sets of the connected area and the indoor area, and calculate the positive air quality index based on the positive parameter sets of the connected area and the indoor area, report at least one set of negative air quality index and positive air quality index, and return to step 3.

[0014] In the present invention, in step 2, the number of cells in the connected area is I, the number of indoor areas is N, the number of cells in each indoor area is J, and the air state model of the connected area is S t+Δt 0 =S t 0 P t 0 , where S t+Δt 0 is the state vector of the connected region at time t+Δt, S t 0 is the state vector of the connected region at time t, P t0 is the state transfer matrix of the connected area at time t, and the air state model of indoor area n is S t+Δt n = S t n P t n , where S t+Δt n is the state vector of indoor area n at time t+Δt, S t n is the state vector of indoor area n at time t, P t n is the state transfer matrix of indoor area n at time t, and Δt is the state transfer step size.

[0015] In the present invention, in step 3, the first data set includes at least pathogen concentration, dust concentration, hazardous gas concentration, oxygen concentration, carbon dioxide concentration, and water vapor concentration, and the second data set includes at least inlet ventilation volume, outlet ventilation volume, inlet air flow rate, and import air flow rate.

[0016] In the present invention, in step 4, the state vector S of the connected area at time t t 0 The matrix C consists of 1×I groups of sample data 1×I and the matrix D consisting of 1×N groups of output states 1×N , S t 0 =[C 1×I ,D 1×N ], the state transfer matrix P of the connected area at time t t 0 The matrix P' consists of I × I groups of state transition probabilities I×I and the matrix P'' consisting of I×N groups of state output probabilities I×N , P t 0 =[P' I×I ,P'' I×N ] T , the state vector S of the indoor area n at time t t n The matrix C consists of 1×J sets of sample data 1×J and 1×1 group input state, S t n =[C 1×J ,D 1×1 ], the state transition matrix P of indoor area n at time t t n The matrix P' consists of J×J groups of state transition probabilities J×Jand the matrix P'' consisting of J × 1 sets of state input probabilities J×1 , P t 0 =[P' J×J ,P'' J×1 ] T .

[0017] In the present invention, in step 4, , when cell i is adjacent to cell m in the connected region, , p' 1im is the diffusion probability of the material in the i-th cell in the connected area transferred to the m-th cell, p' 2im is the convection probability of the material in the i-th cell in the connected area being transferred to the m-th cell, p' 3im is the random perturbation probability of the substance in the i-th cell in the connected area being transferred to the m-th cell, α0 is the diffusion weight, β0 is the convection weight, and γ0 is the random perturbation weight. When the cell i and the cell m in the connected area are not adjacent, p' im =0, p' im P' I×I The state transition probability of the group in the i-th row and m-th column.

[0018] In the present invention, in step 4, , , , δ is the diffusion coefficient, Δx is the distance between cells, c i C 1×I The i-th group of sample data, |v im | is the modulus of the wind speed from the ith cell toward the mth cell in the connected area, I0 is the number of cells adjacent to the ith cell in the connected area, Δt=Δx / (|v1|+|v2|), |v1| is the modulus of the inlet air flow velocity of the connected area, and |v2| is the modulus of the inlet air flow velocity of the connected area.

[0019] In the present invention, in step 5, the state output probability is the internal influence probability, and the state input probability is the external influence probability. , V n is the volume of the indoor area n, Δc n is the average concentration difference of indoor area n in the previous state transfer step, is the average concentration of indoor area n at time t, η is the mechanical ventilation efficiency, ΔC n is the average concentration difference between the connected area and the indoor area n at time t, C t is the average concentration of the connected area at time t, Q nt is the inlet ventilation volume of indoor area n at time t, Q' tis the inlet ventilation volume of the connected area at time t, Q'' n is the gas exchange rate between the channel area and the indoor area n, and the gas mobility r in the indoor area n n =Q'' n / Q nt , the state input probability of indoor area n to the connected area p'''n=r n , P'' I×N =[p''1,p''2,…,p'' n ,…,p'' N ], p'' n is the state output probability of the connected area to the indoor area n, p'' n =1-p'''n,P'' J×1 =[p'' n1 ,p'' n2 ,…,p'' nj ,…,p'' nJ ] T , p'' nj Input probability for the state of the jth cell in the indoor area n, p'' nj =r n L' nj , L' nj is the distance ratio of the jth cell in the indoor area n.

[0020] In the present invention, in step 7, the negative parameter set U1={y 10 、y 20 、y 30}, where the pathogen infection response curve , respiratory hazard response curve , explosion hazard response curve , a1 is the infectivity constant, v0 is the respiratory rate, M1 is the bacterial infection threshold, is the average concentration of bacteria, a2 is the respiratory hazard coefficient, M2 is the respiratory hazard threshold, is the average dust concentration, M3 is the dangerous gas concentration threshold, is the average concentration of hazardous gases.

[0021] In the present invention, in step 7, the positive parameter set U2={y 40 、y 50 、y 60}, where the oxygen suitable response curve , CO2 suitable response curve , humidity suitability response curve y 60 =(T0-M6) / M6+(R0-M7) / M7, M4 is the oxygen concentration threshold, is the average oxygen concentration, M5 is the carbon dioxide concentration threshold, is the average concentration of carbon dioxide, M6 is the temperature threshold, T0 is the average temperature, R0 is the average concentration of water vapor, and M7 is the water vapor concentration threshold.

[0022] In the present invention, in step 8, the negative air quality index , positive air quality index .

[0023] The implementation of the automatic indoor air quality monitoring method of the present invention has the following beneficial effects: the present invention obtains concentration data of various gases in the air through concentration monitoring nodes, and while determining whether the current concentration data exceeds a threshold range, predicts the state transition probability within an independent region and the state input probability between regions to construct an air state model with a Markov chain structure, and then generates a state vector for the next state transition step, thereby reducing the data processing volume for multi-region air quality monitoring. Furthermore, the present invention monitors multiple components through the air state model, generates a negative parameter set and evaluates the negative air quality index based on factors such as pathogen infection, lung damage, and explosion hazard, and generates a positive parameter set and evaluates the positive air quality index based on factors such as suitable temperature and humidity, suitable O2 concentration, and suitable CO2 concentration. The air quality negative index, positive air quality index, negative parameter set, and positive parameter set are combined to achieve comprehensive air quality monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a flow chart of the indoor air quality monitoring method of the present invention;

[0025] Figure 2 Schematic diagram of the indoor area distribution of the building of the present invention;

[0026] Figure 3 Schematic diagram of air exchange in various indoor areas of a building according to the present invention;

[0027] Figure 4 This is a schematic diagram of the distribution of concentration detection nodes in a preferred indoor area of ​​the present invention;

[0028] Figure 5 A schematic diagram of air circulation in two groups of areas in a building room according to the present invention;

[0029] Figure 6 This is a schematic diagram of a preferred method of air circulation between an indoor area and the outdoors according to the present invention;

[0030] Figure 7 The block diagram of the system for realizing the automatic monitoring method of indoor air quality of the present invention. DETAILED DESCRIPTION

[0031] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. Example 1

[0032] like Figures 1 to 6 As shown, the automatic indoor air quality monitoring method of the present invention includes the following steps.

[0033] Step 1: Divide the target space into a connected area and at least two indoor areas, and set up environmental monitoring nodes at the air inlet and outlet of the connected area and each indoor area. Figure 2 and Figure 3 As shown, the middle space is the connected area, and the surrounding space of the connected area is the indoor area. The connected area and each indoor area have an air inlet and an air exhaust outlet. Environmental monitoring nodes are set at the air inlet and the air exhaust outlet of the connected area and each indoor area. Each environmental monitoring node is installed with a wind speed sensor. The connected area exchanges gas with each indoor area, and the arrows represent the direction of gas inlet and outlet.

[0034] Step 2: Divide the connected area and indoor area into multiple cells, arrange concentration monitoring nodes in multiple cells, and create air state models for the connected area and each indoor area. The number of cells in the connected area is I, and the number of indoor areas is N, as shown in Figure 4 As shown in the figure, the number of cells in each indoor area is J, and concentration monitoring nodes are deployed in multiple cells. Each concentration monitoring node is equipped with a pathogen sensor, dust sensor, hazardous gas sensor, CO2 sensor, O2 sensor, temperature and humidity sensor. The distance between each concentration monitoring node is within the range of 5 to 10 meters to ensure that the monitoring range of the concentration monitoring node can cover every cell.

[0035] The air state model of the connected area is S t+Δt 0 =S t 0 P t 0 , where S t+Δt 0 is the state vector of the connected region at time t+Δt, S t 0 is the state vector of the connected region at time t, P t 0 is the state transfer matrix of the connected area at time t. The air state model of indoor area n is S t+Δt n = S t n P t n , where S t+Δt nis the state vector of indoor area n at time t+Δt, S t n is the state vector of indoor area n at time t, P t n is the state transfer matrix of indoor area n at time t, and Δt is the state transfer step size.

[0036] Step 3: Collect the first dataset for each concentration monitoring node and the second dataset for each environmental monitoring node. If the sampled data in the first dataset exceeds the threshold range for the corresponding component, issue a warning notification and terminate the task. Otherwise, proceed to Step 4. The first dataset includes at least pathogen concentration, dust concentration, hazardous gas concentration, oxygen concentration, carbon dioxide concentration, and water vapor concentration. The second dataset includes at least inlet ventilation volume, outlet ventilation volume, inlet air flow rate, and inlet air flow rate. The collection frequency can be set based on different real-time requirements or the number of people in the target space and its volume. A buzzer will issue a warning to ensure timely evacuation of people to avoid danger.

[0037] In this embodiment, the threshold interval of the pathogen concentration is set according to the type of pathogen. For example, the threshold interval of Salmonella is generally [0,10 5 ], in milligrams per cubic meter. The dust concentration threshold range is set based on the type of dust. For example, the threshold range for silicon dioxide is generally [0, 0.05], in milligrams per cubic meter. The hazardous gas concentration threshold range is set based on the type of hazardous gas. For example, the hazardous gas concentration threshold for methane is generally [0, 327], in milligrams per cubic meter. The oxygen concentration threshold range is generally [255000, 307000], in milligrams per cubic meter; the carbon dioxide concentration threshold range is generally [1800, 9000], in milligrams per cubic meter; and the water vapor concentration threshold range is generally [5000, 15000], in milligrams per cubic meter.

[0038] Step 4: Generate the state transition probability and state transition step of the air state model based on the first data set and the second data set of the connected area and the indoor area. The state vector S of the connected area at time t t 0 The matrix C consists of 1×I groups of sample data 1×I and the matrix D consisting of 1×N groups of output states 1×N , S t 0 =[C 1×I ,D 1×N ]. The state transfer matrix P of the connected area at time t t 0 The matrix P' consists of I × I groups of state transition probabilities I×Iand the matrix P'' consisting of I×N groups of state output probabilities I×N , P t 0 =[P' I×I ,P'' I×N ] T The state vector S of the indoor area n at time t is t n The matrix C consists of 1×J sets of sample data 1×J and 1×1 group input state, S t n =[C 1×J ,D 1×1 ]. The state transition matrix P of indoor area n at time t t n The matrix P' consists of J×J groups of state transition probabilities J×J and the matrix P'' consisting of J × 1 sets of state input probabilities J×1 , P t 0 =[P' J×J ,P'' J×1 ] T The state transition probability and state transition step length of the air state model are described in detail in the second embodiment.

[0039] Step 5: Generate gas mobility based on the first and second data sets of different indoor areas and connected areas at the same time and the state transition step, and then generate the state input probability of the air state model based on the gas mobility. Figure 5 and Figure 6 As shown, the indoor area and the connected area exchange gas, and at the same time, the indoor area and the connected area exchange gas with the outside world through the air inlet and exhaust port respectively. A fan and an air purifier are installed between the air inlet and the exhaust port to realize the mechanical ventilation mode. The state output probability is the internal influence probability, and the state input probability is the external influence probability. The gas mobility is the ratio of the gas exchange volume to the inlet ventilation volume. The calculation of the gas mobility, the state input probability and the state output probability of the air state model is described in detail in Example 3.

[0040] Step 6: Update the air state model based on the state transition probabilities and the state input probabilities, and predict a third data set for the next state transition step based on the air state model and the first data set. The third data set includes the state vector for the next state transition step of the connected area and the state vector for the next state transition step of each indoor area.

[0041] Step 7: Construct the response curves of different components, and generate the negative parameter set and positive parameter set of the connected area and the indoor area respectively according to the third data set and the response curve.10 、y 20 、y 30}, where the pathogen infection response curve , respiratory hazard response curve , explosion hazard response curve a1 is the infectivity constant, unitless. v0 is the respiratory rate, cubic meters per hour. M1 is the pathogen infection threshold, which is set according to different pathogen types. For example, the pathogen infection threshold for Salmonella is 10 5 indivual. is the average concentration of pathogens, measured in units of per cubic meter. a2 is the respiratory hazard coefficient, unitless. M2 is the respiratory hazard threshold, set based on the type of dust. For example, the respiratory hazard threshold for silica is 0.05 mg. M3 is the hazardous gas concentration threshold, which is set according to the type of hazardous gas. For example, the hazardous gas concentration threshold for methane is generally 3571 mg. It is the average concentration of dangerous gases in milligrams per cubic meter.

[0042] Positive parameter set U2={y 40 、y 50 、y 60}, where the oxygen suitable response curve , CO2 suitable response curve , humidity suitability response curve y 60 =(T0-M6) / M6+(R0-M7) / M7. M4 is the oxygen concentration threshold, generally 273,000 mg / m³. M5 is the carbon dioxide concentration threshold, generally 1,800 mg / m³. M6 is the temperature threshold, generally 28°C. M7 is the water vapor concentration threshold, generally 8,650 mg / m³. T0 is the average temperature, and R0 is the average water vapor concentration. is the average oxygen concentration, is the average concentration of carbon dioxide, in milligrams per cubic meter. The average concentrations of various gases in the response curves of the negative parameter set and the positive parameter set can be obtained through the state vector of the third data set.

[0043] Step 8: Calculate the negative air quality index based on the negative parameter sets of the connected area and the indoor area, and calculate the positive air quality index based on the positive parameter sets of the connected area and the indoor area, report at least one set of negative air quality index and positive air quality index, and return to step 3. Negative air quality index , positive air quality index The smaller the negative air quality index, the worse the air quality, and the larger the positive air quality index, the worse the air quality. The negative and positive air quality indexes provide feedback on air quality, clearly reflecting the real-time air status of the target space. Example 2

[0044] This embodiment further discloses a preferred method for updating state transition probability and state transition step length.

[0045] The state transition probability is the probability that the material in a cell in the same area is transferred to another cell due to diffusion, convection, or random disturbance. The state transition step is the time it takes for the material in a cell in the same area to transfer to another cell.

[0046] , when cell i is adjacent to cell m in the connected region, , when cell i and cell m in the connected region are not adjacent, p' im =0. p' im is the probability that the material in the i-th cell in the connected area is transferred to the m-th cell, that is, P' I×I The state transition probability of the group in row i and column m is α0, which is the diffusion weight, β0, which is the convection weight, and γ0, which is the random perturbation weight. α0+β0+γ0=1.

[0047] , p' 1im is the diffusion probability of the substance in the i-th cell in the connected area transferring to the m-th cell, δ is the diffusion coefficient, which is related to the specific gas, such as the diffusion coefficient of carbon dioxide in the air is 16 square millimeters per second. Δx is the distance between cells, c i C 1×I The i-th group of sample data. , p' 2im is the convection probability of the material in the i-th cell in the connected area transferring to the m-th cell, |v im | is the wind speed modulus from the i-th cell toward the m-th cell in the connected area. , p' 3im is the random perturbation probability of the material in the i-th cell in the connected region being transferred to the m-th cell, and I0 is the number of cells adjacent to the i-th cell in the connected region.

[0048] Δt = Δx / (|v1| + |v2|), where |v1| is the modulus of the air velocity at the inlet of the connected area, and |v2| is the modulus of the air velocity at the outlet of the connected area. The state transition probability of the indoor area can be calculated using the above method. Example 3

[0049] This embodiment further discloses a preferred method for updating gas mobility, state input probability, and state output probability.

[0050] First, the gas exchange rate is calculated using the following formula, and then the gas mobility is calculated. , V n is the volume of the indoor area n, in cubic meters. n Q is the average concentration difference of indoor area n in the previous state transfer step, in milligrams per cubic meter. nt is the inlet ventilation volume of indoor area n at time t, in cubic meters per hour. η is the mechanical ventilation efficiency, unitless. C t is the average concentration of the connected area at time t, Q' t is the inlet ventilation volume of the connected area at time t, is the average concentration of indoor area n at time t, ΔC n is the average concentration difference between the connected area and the indoor area n at time t, Q'' n is the gas exchange volume between the channel area and the indoor area n. , is the average concentration of indoor area n at time t-Δt, , c nj C 1×J The jth group of sample data in , The gas mobility r in the indoor area n n =Q'' n / Q nt .

[0051] Then the state input probability is calculated according to the gas mobility, and finally the state output probability is calculated according to the state input probability. The state input probability of the indoor area n to the connected area p'''n=r n , P'' I×N =[p''1,p''2,…,p'' n ,…,p'' N ], p'' n is the state output probability of the connected area to the indoor area n, p'' n =1-p'''n,P'' J×1 =[p'' n1 ,p'' n2 ,…,p'' nj ,…,p'' nJ ] T , p'' nj Input probability for the state of the jth cell in the indoor area n, p'' nj =r n L' nj , L'nj is the distance ratio of the jth cell in the indoor area n, , L nj is the distance between the jth cell in the indoor area n and the connected area. Example 4

[0052] like Figure 7 As shown, the system for implementing the automatic indoor air quality monitoring method of the present invention includes: a pathogen sensor, a dust sensor, a hazardous gas sensor, a CO2 sensor, an O2 sensor, a temperature and humidity sensor, a wind speed sensor, a data acquisition unit, a data error correction unit, a data analysis unit, a control unit, and a buzzer. The pathogen sensor is used to measure pathogen concentration. The dust sensor is used to measure dust concentration. The hazardous gas sensor is used to measure flammable gas concentration. The CO2 sensor is used to measure CO2 concentration. The O2 sensor is used to measure O2 concentration. The temperature and humidity sensor is used to measure temperature and water vapor concentration. The wind speed sensor is used to measure inlet ventilation volume, outlet ventilation volume, inlet air flow rate, and inlet air flow rate. The data acquisition unit is used to collect a first data set from each concentration monitoring node and a second data set from each environmental monitoring node and transmit them to the data error correction unit and the data analysis unit. The data error correction unit is used to identify abnormal data in the first and second data sets. The data analysis unit is used to analyze whether the sampled data in the first data set exceeds the threshold range of the corresponding component and, if so, issue a warning notification. The control unit is used to receive the warning notification and control the activation of the buzzer and transmit the acquisition signal to the data acquisition unit. The buzzer is used to alarm and indicate danger.

[0053] In addition, the system also includes a data processing unit, which is used to update and generate the state transition probability and state transition step of the air state model based on the first data set and the second data set, calculate the gas mobility, and then generate the state input probability of the air state model based on the gas mobility, update the air state model based on the state transition probability and the state input probability, predict the third data set of the next state transition step based on the air state model and the first data set, generate negative parameter sets and positive parameter sets for connected areas and indoor areas respectively based on the third data set and the response curve, calculate the negative air quality index based on the negative parameter set, and calculate the positive air quality index based on the positive parameter set.

[0054] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for automatically monitoring indoor air quality, characterized in that: The following steps are involved: Step 1: Divide the target space into a connected area and at least two indoor areas, and set environmental monitoring nodes at the air inlet and outlet of the connected area and each indoor area; Step 2: Divide the connected area and indoor area into multiple cells, arrange concentration monitoring nodes in multiple cells, and create air state models for the connected area and each indoor area; Step 3: Collect the first data set of each concentration monitoring node and the second data set of the environmental monitoring node. If the sampling data of the first data set exceeds the threshold range of the corresponding component, issue a warning notification and end the task. Otherwise, proceed to step 4. Step 4: Generate the state transition probability and state transition step length of the air state model according to the first data set and the second data set of the connected area and the indoor area respectively; Step 5: Generate gas mobility based on the first and second data sets of different indoor areas and connected areas at the same time and the state transition step, and then generate the state input probability of the air state model based on the gas mobility; Step 6: Update the air state model according to the state transition probability and the state input probability, and predict the third data set of the next state transition step length according to the air state model and the first data set; Step 7: Construct response curves of different components, and generate negative parameter sets and positive parameter sets for connected areas and indoor areas respectively according to the third data set and the response curves; Step 8: Calculate the negative air quality index based on the negative parameter sets of the connected area and the indoor area, and calculate the positive air quality index based on the positive parameter sets of the connected area and the indoor area, report at least one set of negative air quality index and positive air quality index, and return to step 3. Among them, the first data set includes at least pathogen concentration, dust concentration, hazardous gas concentration, oxygen concentration, carbon dioxide concentration, and water vapor concentration; the second data set includes at least inlet ventilation volume, outlet ventilation volume, inlet air flow rate, and outlet air flow rate.

2. The method for automatically monitoring indoor air quality according to claim 1, wherein: In step 2, the number of cells in the connected area is I, the number of indoor areas is N, the number of cells in each indoor area is J, and the air state model of the connected area is S t+Δt 0 =S t 0 P t 0 , where S t+Δt 0 is the state vector of the connected region at time t+Δt, S t 0 is the state vector of the connected region at time t, P t 0 is the state transfer matrix of the connected area at time t, and the air state model of indoor area n is S t+Δt n = S t n P t n , where S t+Δt n is the state vector of indoor area n at time t+Δt, S t n is the state vector of indoor area n at time t, P t n is the state transfer matrix of indoor area n at time t, and Δt is the state transfer step size.

3. The method for automatically monitoring indoor air quality according to claim 2, wherein: In step 4, the state vector S of the connected region at time t t 0 The matrix C consists of 1×I groups of sample data 1×I and the matrix D consisting of 1×N groups of output states 1×N , S t 0 =[C 1×I ,D 1×N ], the state transfer matrix P of the connected area at time t t 0 The matrix P' consists of I × I groups of state transition probabilities I×I and the matrix P'' consisting of I×N groups of state output probabilities I×N , P t 0 =[P' I×I ,P'' I×N ] T , the state vector S of the indoor area n at time t t n The matrix C consists of 1×J sets of sample data 1×J and 1×1 group input state, S t n =[C 1×J ,D 1×1 ], the state transition matrix P of indoor area n at time t t n The matrix P' consists of J×J groups of state transition probabilities J×J and the matrix P'' consisting of J × 1 sets of state input probabilities J×1 , P t 0 =[P' J×J ,P'' J×1 ] T .

4. The method for automatically monitoring indoor air quality according to claim 3, characterized in that: In step 4, , when cell i is adjacent to cell m in the connected region, , p' 1im is the diffusion probability of the material in the i-th cell in the connected area transferred to the m-th cell, p' 2im is the convection probability of the material in the i-th cell in the connected area being transferred to the m-th cell, p' 3im is the random perturbation probability of the substance in the i-th cell in the connected area being transferred to the m-th cell, α0 is the diffusion weight, β0 is the convection weight, and γ0 is the random perturbation weight. When the cell i and the cell m in the connected area are not adjacent, p' im =0, p' im P' I×I The state transition probability of the group in the i-th row and m-th column.

5. The method for automatically monitoring indoor air quality according to claim 4, characterized in that: In step 4, , , , δ is the diffusion coefficient, Δx is the distance between cells, c i C 1×I The i-th group of sample data, |v im | is the modulus of the wind speed from the i-th cell toward the m-th cell in the connected area, I0 is the number of cells adjacent to the i-th cell in the connected area, Δt=Δx / (|v1|+|v2|), |v1| is the modulus of the inlet air flow velocity of the connected area, and |v2| is the modulus of the outlet air flow velocity of the connected area.

6. The method for automatically monitoring indoor air quality according to claim 5, characterized in that: In step 5, the state output probability is the internal influence probability, and the state input probability is the external influence probability. , V n is the volume of the indoor area n, Δc n is the average concentration difference of indoor area n in the previous state transfer step, is the average concentration of indoor area n at time t, η is the mechanical ventilation efficiency, ΔC n is the average concentration difference between the connected area and the indoor area n at time t, C t is the average concentration of the connected area at time t, Q nt is the inlet ventilation volume of indoor area n at time t, Q' t is the inlet ventilation volume of the connected area at time t, Q'' n is the gas exchange rate between the channel area and the indoor area n, and the gas mobility r in the indoor area n n =Q'' n / Q nt , the state input probability of indoor area n to the connected area p'''n=r n , P'' I×N =[p''1,p''2,…,p'' n ,…,p'' N ], p'' n is the state output probability of the connected area to the indoor area n, p'' n =1-p'''n,P'' J×1 =[p'' n1 ,p'' n2 ,…,p'' nj ,…,p'' nJ ] T , p'' nj Input probability for the state of the jth cell in the indoor area n, p'' nj =r n L' nj , L' nj is the distance ratio of the jth cell in the indoor area n.

7. The method for automatically monitoring indoor air quality according to claim 6, characterized in that: In step 7, the negative parameter set U1={y 10 、y 20 、y 30 }, where the pathogen infection response curve , respiratory hazard response curve , explosion hazard response curve , a1 is the infectivity constant, v0 is the respiratory rate, M1 is the bacterial infection threshold, is the average concentration of bacteria, a2 is the respiratory hazard coefficient, M2 is the respiratory hazard threshold, is the average dust concentration, M3 is the dangerous gas concentration threshold, is the average concentration of hazardous gases.

8. The method for automatically monitoring indoor air quality according to claim 7, characterized in that: In step 7, the positive parameter set U2={y 40 、y 50 、y 60 }, where oxygen suitable response curve , CO2 suitable response curve , humidity suitability response curve y 60 =(T0-M6) / M6+(R0-M7) / M7, M4 is the oxygen concentration threshold, is the average oxygen concentration, M5 is the carbon dioxide concentration threshold, is the average concentration of carbon dioxide, M6 is the temperature threshold, T0 is the average temperature, R0 is the average concentration of water vapor, and M7 is the water vapor concentration threshold.

9. The method for automatically monitoring indoor air quality according to claim 8, characterized in that: In step 8, the air quality index is negative , positive air quality index .

Images