Indoor environment detection method and system

Abstract

The present invention belongs to the field of environmental monitoring and provides an indoor environment monitoring method and system. Indoor environment data is collected through a network of multiple sensors distributed at different locations. The sensors are used to detect multiple environmental factors, including temperature, humidity, air quality, and light intensity. The environmental data is preprocessed, including data cleaning, data normalization, and data synchronization. Correlations between the environmental data are determined and an undirected graph is established based on the correlations. Based on the undirected graph, real-time indoor environment monitoring data is collaboratively corrected, data fused and inferred, and environmental status assessed. This scheme can efficiently perform indoor environment monitoring.

Description

Technical Field

[0001] The present invention belongs to the field of environmental detection, and in particular relates to an indoor environment detection method and system. Background Art

[0002] The widespread use of sensor networks in indoor environmental monitoring enables real-time collection and monitoring of key indoor environmental factors, such as temperature, humidity, air quality, and light intensity. However, existing sensor networks often face challenges in practical applications, particularly sensor drift, data loss, and measurement errors, which impact the accuracy of environmental data and system stability.

[0003] Over time, sensors' measurement accuracy often changes, a phenomenon known as sensor drift. Drift is particularly common in sensors that measure physical quantities such as temperature and humidity. Sensor drift can be affected by a variety of factors, including sensor component aging, increased electromagnetic interference in the environment, and sensor circuit loss. Drift can result in significant deviations between the sensor's output and the actual value, impacting the overall accuracy of the environmental monitoring system. For example, if a temperature sensor consistently measures temperatures that are too high or too low due to drift, it can trigger erroneous environmental control measures, impacting the comfort and safety of the indoor environment.

[0004] In real-world sensor networks, data loss often occurs due to unstable network transmission, sensor failures, low battery levels, or other unpredictable external factors. Data loss can occur when a sensor fails to transmit data for a short period of time or stops functioning for an extended period. In indoor environmental monitoring, data loss can prevent the system from acquiring continuous environmental data, hindering accurate assessments of environmental conditions. For example, in air quality monitoring, if an air quality sensor fails, preventing the acquisition of key pollutant concentration data for a period of time, the system will be unable to promptly determine whether the air is at a healthy level, potentially impacting user health.

[0005] Existing sensor deployments are typically based on fixed location planning and are unable to dynamically adapt to changes in the indoor environment. This fixed layout can lead to significant errors in environmental data measurement in certain areas, especially when sensors are subject to external interference (such as direct sunlight or cold air), causing the measured data to deviate significantly from the true value. Furthermore, sensor accuracy is limited by cost and performance. While high-precision sensors can improve measurement accuracy, they are often expensive in actual deployments and may be unstable in harsh environments.

[0006] To solve the above problems, existing technologies usually use calibration, data filling and other methods for processing, but these methods rely on fixed rules and models and are difficult to cope with complex and changing indoor environments. Summary of the Invention

[0007] In order to solve the problems in the prior art, the present invention provides an indoor environment detection method, which includes the following steps:

[0008] Collecting indoor environmental data through a network of multiple sensors distributed in different locations. The sensors are used to detect multiple environmental factors, including temperature, humidity, air quality, and light intensity;

[0009] Preprocessing the environmental data, including data cleaning, data normalization, and data synchronization;

[0010] Determining the correlation between the environmental data, and establishing an undirected graph based on the correlation;

[0011] According to the undirected graph, collaborative correction, data fusion and inference, and environmental status evaluation are performed on real-time indoor environment detection data.

[0012] Furthermore, establishing an undirected graph according to the association includes:

[0013] Use historical data to analyze the correlation between different types of sensor data through statistical correlation;

[0014] According to the correlation analysis, the correlation coefficients between different environmental factors are calculated to generate a correlation matrix. If the absolute value of the correlation coefficient of two environmental factors exceeds the set threshold, it is considered that there is a significant correlation between them;

[0015] According to the determined correlation, an undirected graph is constructed, where each node in the undirected graph represents an environmental element, and the edges represent the correlation between environmental elements;

[0016] If there is a significant correlation between two environmental factors, an undirected edge is added to the undirected graph to connect the corresponding nodes.

[0017] Furthermore, the collaborative correction includes:

[0018] During real-time monitoring, the system automatically detects whether a sensor's data is abnormal or drifting based on the changing trend of historical data or the deviation between the current measurement value and the data of other nodes in the undirected graph;

[0019] Determine the sensor nodes associated with abnormal or drifting data through an undirected graph and extract the data of these associated sensors;

[0020] According to the weights of the associated sensor data and the edges in the undirected graph, a collaborative correction model is used to correct the abnormal data. The correction formula is a weighted sum based on the contribution of the associated data.

[0021] Furthermore, data fusion and inference include:

[0022] During real-time monitoring, the system detects missing data or sensor failure from a certain sensor and identifies the environmental factors with missing data and their corresponding nodes.

[0023] According to the edges in the undirected graph, the data of other sensors associated with the missing data are extracted. The weight of the edge represents the influence of the associated sensor on the missing data.

[0024] Utilize the associated sensor data and fuse the multi-source data according to the edge weights in the undirected graph;

[0025] The inferred value of missing data is calculated through the data fusion model, and the result is adjusted according to the weight of each sensor in the undirected graph to ensure that the inferred missing data is reasonable and consistent with the current environmental status.

[0026] Furthermore, the environmental status assessment includes:

[0027] During the environmental monitoring process, the system collects data from each sensor in real time, monitors the status and change trend of each node in the undirected graph, and also monitors the weights and data relevance of the edges in the undirected graph.

[0028] The system detects whether there are significant changes in the edges of undirected graphs and whether the strength of the association between environmental factors deviates from the normal range. When the weight of a certain association relationship increases or decreases abnormally, it indicates that there may be abnormal or sudden changes in the related environmental factors.

[0029] According to the detected edge weight changes, the system analyzes which environmental elements have abnormal correlations and outputs them. Another aspect of the present invention also provides an indoor environment detection system, which includes the following modules:

[0030] a prediction module, configured to predict environmental parameters using a first machine learning model, the environmental parameters including temperature, humidity, and air duct pressure;

[0031] A collection module is used to collect indoor environmental data through a network of multiple sensors distributed in different locations. The sensors are used to detect multiple environmental factors, including temperature, humidity, air quality, and light intensity;

[0032] A preprocessing module, configured to preprocess the environmental data, including data cleaning, data normalization, and data synchronization;

[0033] An association module, configured to determine the association between the environmental data and establish an undirected graph based on the association;

[0034] The processing module is used to perform collaborative correction, data fusion and inference, and environmental status evaluation on real-time indoor environment detection data according to the undirected graph.

[0035] By constructing a sensor data association model based on an undirected graph, the present invention realizes dynamic collaborative correction, data fusion, and missing data inference of various environmental data in sensor networks, effectively solving problems such as sensor drift, data loss, and measurement errors in the prior art, and has the following beneficial effects:

[0036] Through the correlation between sensor nodes in the undirected graph, the present invention can dynamically monitor the anomalies and drift of sensor data, and automatically extract the data of related sensors for collaborative correction, ensuring the accuracy and stability of the measurement data and avoiding long-term measurement deviations caused by sensor drift.

[0037] In the event of a sensor failure or data loss, the present invention uses data from other sensors associated with the missing sensor to make inferences and generate reasonable alternative data, thereby ensuring the continuity and integrity of the data stream and avoiding environmental monitoring blind spots caused by data loss.

[0038] The present invention dynamically fuses data from multiple sensors through an undirected graph model and distributes data contributions according to their relevance weights, ensuring that the fused data is more consistent with actual environmental conditions and improving the system's perception capabilities in complex environments.

[0039] The present invention can dynamically adjust the weights of edges in an undirected graph according to historical trends in sensor data and current environmental changes, ensuring that the system can adapt to associations under different environmental conditions, thereby enhancing the robustness and adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0042] Below, the invention is preferably described with reference to the accompanying drawings and specific embodiments.

[0043] This embodiment solves the above problem through the following steps:

[0044] In one embodiment, reference Figure 1This invention provides a method for indoor environmental monitoring. This process uses a sensor network deployed within a building to collect and monitor real-time data on various physical and chemical factors that affect indoor health, comfort, and safety. The method then uses data processing, fusion, and analysis techniques to assess and regulate the indoor environmental status. This method primarily focuses on key factors affecting the indoor environment, such as temperature, humidity, air quality (e.g., concentrations of pollutants such as PM2.5, CO2, and VOCs), light intensity, and noise levels. It utilizes Internet of Things (IoT) and data fusion technologies to form a comprehensive dynamic indoor environmental monitoring and feedback mechanism.

[0045] In this invention, the indoor environment refers to the climate, air, light, sound and other factors inside a building, mainly including temperature, humidity, air quality, noise, light intensity, etc., which directly affect the comfort, health and work efficiency of building users.

[0046] Indoor environmental data is collected through a network of multiple sensors distributed in different locations. The sensors are used to detect multiple environmental factors, including temperature, humidity, air quality, and light intensity.

[0047] The sensor network in the present invention realizes comprehensive monitoring of environmental factors in various indoor areas through multi-point layout, ensuring the breadth of the monitoring range and the accuracy of data collection.

[0048] The temperature sensor of the present invention is used to collect temperature data at different locations. By deploying multiple temperature sensors to cover different indoor areas, including but not limited to the center of the room, corners, and areas near windows or doors, accurate perception of indoor temperature distribution is achieved. The humidity sensor is used to monitor the relative humidity of indoor air. By deploying multiple humidity sensors throughout the environment, the humidity level in each area can be accurately reflected, ensuring the reliability of humidity monitoring, particularly in areas with large humidity fluctuations, such as kitchens and bathrooms.

[0049] The air quality sensor in this invention is used to detect pollutant concentrations in indoor air. This air quality detection includes, but is not limited to, real-time monitoring of multiple pollutants such as PM2.5, carbon dioxide (CO2), volatile organic compounds (VOCs), and formaldehyde. Accurate detection of these pollutants effectively ensures healthy indoor air quality. The light intensity sensor is used to monitor indoor lighting conditions, including both natural and artificial light sources. By placing light sensors at different heights and locations, dynamic monitoring of light distribution and changes is achieved.

[0050] The environmental data is preprocessed, and the preprocessing includes data cleaning, data normalization, and data synchronization.

[0051] The preprocessing step of the present invention aims to improve the accuracy, consistency and timeliness of environmental data collected from multiple sensor networks distributed in different locations through a series of processing operations, ensuring the effectiveness and stability of subsequent data analysis, fusion and regulation processes.

[0052] Among them, data cleaning is an important preprocessing link in the present invention, and the data cleaning step is used to identify and remove abnormal data, invalid data or noise data generated during the acquisition process. Specifically, the abnormal data may include abnormal readings caused by sensor failure, data transmission interruption or extreme environmental conditions. For example, during the sensor acquisition process, the temperature sensor may produce extremely deviated temperature values ​​due to external interference in a short period of time. Such data will cause misjudgment of the overall system. Therefore, the present invention detects such extreme abnormal values ​​through a data cleaning algorithm and deletes or corrects them. In addition, if some sensors have missing data due to network failure or power consumption problems, the cleaning step of the present invention will fill in the data through an interpolation algorithm or adjacent sensor data to maintain the continuity of the data. Through this step, the impact of abnormal conditions on the overall data quality can be effectively reduced, ensuring the stability of the system input data.

[0053] Data normalization is another important preprocessing step in the present invention. This normalization process is used to address the differences in dimensions and ranges of heterogeneous data obtained from multiple different types of sensors. For example, the sensors in the present invention may simultaneously collect data such as temperature, humidity, air quality (such as PM2.5 concentration), and light intensity. The numerical ranges and units of this data vary. Temperature is expressed in degrees Celsius (°C), humidity is expressed as a relative humidity percentage (%), and PM2.5 concentration in air quality is typically expressed in micrograms per cubic meter (µg / m³). To ensure that subsequent fusion and analysis algorithms can process this data on a unified data scale, the present invention introduces data normalization technology to convert different types of data into dimensionless data within the same dimension. Specifically, the normalization method can use maximum-minimum normalization to normalize each data value to a range of 0 to 1.

[0054] Data synchronization is a crucial part of the preprocessing step of the present invention, and the data synchronization is used to solve the time deviation problem caused by different sampling frequencies or transmission delays of different sensors. The sensor network of the present invention may include multiple sensors with different sampling frequencies. For example, the temperature sensor collects data once a minute, while the air quality sensor may collect data every 5 seconds. This inconsistency in sampling frequency will result in the inability to directly compare the data in time. The present invention uses data synchronization technology to align and interpolate data at different time points to ensure that the data of all sensors are processed on the same time axis. For example, if the latest data of the air quality sensor at a certain moment is before or after the sampling data of the temperature sensor, the system will generate the air quality data of the corresponding time point through the time interpolation method to ensure that the environmental data is synchronized and consistent at each moment. The data synchronization process also includes unifying the time base of sensors from different sources through timestamp matching technology to eliminate data deviations caused by transmission delays or asynchronous sampling times.

[0055] The pre-processing step of the present invention can greatly improve the quality and consistency of environmental data through the above cleaning, normalization and synchronization processing. For example, for temperature and light data collected from different locations, if the light sensor has high-intensity interference caused by direct sunlight, the cleaning process of the present invention can detect this abnormality and correct the abnormal value through historical data or data from neighboring sensors. The normalization step uniformly processes the data of different environmental factors, making the temperature, humidity, air quality, etc. comparable, ensuring that the accuracy of the model will not be affected by the difference in different data dimensions in subsequent algorithms. Data synchronization ensures that the data of various sensors can be accurately matched in time, avoiding misjudgments caused by time lags.

[0056] The correlations between the environmental data are determined, and an undirected graph is constructed based on the correlations. Undirected graphs are used to represent the correlations between different types of environmental data. This data is analyzed and processed using graph theory models to optimize the coordinated calibration and data fusion of various sensors in the sensor network. Based on the inherent correlations between various environmental factors (such as temperature, humidity, light intensity, and air quality), this invention automatically constructs a correlation model between these factors through historical data and algorithmic analysis. This model is expressed in the form of an undirected graph, providing a structured foundation for subsequent environmental monitoring and data processing.

[0057] The present invention first automatically identifies the correlations between environmental data through historical data analysis and algorithmic calculations. This correlation is determined by analyzing statistical correlations, physical dependencies, or machine learning-based correlation models between environmental element data (such as temperature, humidity, light intensity, and air quality) collected by multiple sensors.

[0058] To this end, the present invention can analyze different types of sensor data using correlation coefficient calculations (such as the Pearson or Spearman correlation coefficients), multivariate regression analysis, and machine learning models (such as support vector machines and random forests) to extract dependencies between different environmental factors. For example, if a high correlation is found between temperature and humidity in historical data, or a strong physical relationship is found between light intensity and temperature, the system can identify these as strongly associated data pairs.

[0059] Specifically, assuming that the temperature data collected in the sensor network , humidity data , light intensity data , air quality data Etc., through correlation analysis, the following correlation matrix is ​​obtained:

[0060] ,

[0061] in, represents the correlation coefficient between temperature and humidity, Represents the correlation coefficient between temperature and light intensity, and so on. When the absolute value of exceeds a certain threshold, such as , it is believed that there is a significant correlation between the two, and an undirected graph will be established based on this correlation.

[0062] According to the determined environmental data relevance, the present invention establishes an undirected graph , where each node in the graph Represents the environmental factor data collected by a sensor (such as temperature, humidity, light intensity, etc.), while the edge It indicates the correlation between these environmental factors.

[0063] Specifically, if there is a significant correlation between two environmental factor data, then in the undirected graph, Connect the corresponding nodes and For example, if there is a strong correlation between the data collected by the temperature sensor and the humidity sensor, then in the undirected graph, the temperature node and humidity nodes There will be an undirected edge between .

[0064] Assume that the correlation between temperature, humidity, light and air quality is as follows:

[0065] Temperature and humidity have a strong positive correlation;

[0066] There is a significant positive correlation between light intensity and temperature;

[0067] The correlation between air quality, temperature and light intensity is weak; the corresponding undirected graph can be expressed as follows:

[0068] ,

[0069] in, Represent temperature, humidity, light intensity and air quality respectively. Indicates the correlation between temperature and humidity, and temperature and light.

[0070] Data collaborative correction, data fusion and inference, and environmental status assessment are performed based on the undirected graph. The undirected graph is constructed by the correlations between different types of environmental data collected by multiple sensors and is used to represent the intrinsic relationships between the data of various environmental elements. Based on the undirected graph, the present invention applies graph theory algorithms to perform data collaborative correction, data fusion and missing data inference, as well as environmental status assessment, to achieve high-precision monitoring and intelligent control of indoor environments.

[0071] Data collaborative correction based on undirected graphs:

[0072] The present invention leverages the connections between nodes in an undirected graph to collaboratively correct anomalies or drift in measurement data. Each node in the undirected graph represents environmental data collected by a sensor, and undirected edges between different nodes represent correlations or dependencies between environmental element data. When drift or anomalies are detected in the data of a particular sensor, the system collaboratively corrects them using the data from associated nodes to improve data accuracy.

[0073] In an indoor environment monitoring system, the relationship between the temperature sensor and the humidity sensor is expressed through the edge in the undirected graph. Assume that the current temperature sensor measurement value is is 35°C, and the temperature value predicted by the system based on the undirected graph association By humidity value The calculated value is 30°C. Due to the large difference between the two, the system determines that the temperature sensor may have drifted, so it corrects it using the following collaborative correction formula:

[0074] ,

[0075] in and are weight coefficients automatically assigned by the system, and their sizes are determined by the edges of the undirected graph. Assume , , then the corrected temperature is:

[0076] ,

[0077] Through this collaborative correction, the present invention effectively reduces the impact of sensor drift on system accuracy and ensures the reliability of environmental data.

[0078] Data fusion and missing data inference based on undirected graphs:

[0079] This paper uses an undirected graph data association model to fuse data from different sensors. Specifically, when a sensor is missing or faulty, the system uses data from other sensors to infer the missing data. The edges in the undirected graph represent the strength of the associations between sensors, and the system infers the missing data based on these associations.

[0080] Assume that the light intensity sensor at a certain moment When a failure occurs and the light data cannot be collected, the system infers the light data based on the correlation between light, temperature, and humidity in the undirected graph. The correlation model can be expressed by the multivariate regression formula as follows:

[0081] ,

[0082] in, is the current temperature, is the current humidity, 、 is the regression coefficient, and its size is determined by the edge of the undirected graph. Assume that the current temperature ,humidity , obtained through the association model:

[0083] ,

[0084] The system uses this formula to infer that the light intensity is 72 lux. Even if the light sensor fails, the system can still provide reliable light data.

[0085] Environmental status assessment based on undirected graph:

[0086] This invention uses an undirected graph to assess the status of the entire indoor environment. It can quickly detect anomalies and issue warnings, particularly when the correlations between multiple environmental elements change significantly. The edges of the undirected graph represent the normal correlations between environmental elements. When a correlation significantly deviates from historical patterns, the system identifies a possible anomaly in that environmental element.

[0087] In an office environment, there is usually a strong positive correlation between temperature and humidity. Through an undirected graph, the system records the historical correlation between temperature and humidity. When a temperature rise is detected, Humidity drops abnormally Below the normal range, the system uses the edges in the undirected graph Determine if there is an abnormal deviation in the correlation between temperature and humidity, and trigger the abnormal environmental status assessment mechanism.

[0088] Based on the overall state of each node and associated edges in the undirected graph, the system can determine whether indoor conditions include excessively dry air or air conditioning failure. It can then issue a warning signal to alert the user or automatically adjust equipment. For example, it might automatically activate a humidifier or reduce the cooling power of the air conditioner to restore normal environmental conditions.

[0089] On the other hand, the present invention also provides an indoor environment detection system, comprising:

[0090] A collection module is used to collect indoor environmental data through a network of multiple sensors distributed in different locations. The sensors are used to detect multiple environmental factors, including temperature, humidity, air quality, and light intensity;

[0091] A preprocessing module, configured to preprocess the environmental data, including data cleaning, data normalization, and data synchronization;

[0092] An association module, configured to determine the association between the environmental data and establish an undirected graph based on the association;

[0093] The processing module is used to perform collaborative correction, data fusion and inference, and environmental status evaluation on real-time indoor environment detection data according to the undirected graph.

[0094] The prior art mentioned in the above background technology section and specific embodiments section of the present invention can be regarded as part of the present invention and used to understand the meaning of some technical features or parameters.

Claims

1. A method for detecting an indoor environment, characterized in that: The method comprises the following steps: collecting indoor environmental data through a plurality of sensor networks distributed at different locations, wherein the sensors are used to detect a plurality of environmental factors, including temperature, humidity, air quality, and light intensity; Preprocessing the environmental data, including data cleaning, data normalization, and data synchronization; Determining the correlation between the environmental data, and establishing an undirected graph based on the correlation; Perform collaborative correction, data fusion and inference, and environmental status assessment on real-time indoor environment detection data according to the undirected graph; During real-time monitoring, the system automatically detects whether a sensor's data is abnormal or drifting based on the changing trend of historical data or the deviation between the current measurement value and the data of other nodes in the undirected graph; Determine the sensor nodes associated with abnormal or drifting data through an undirected graph and extract the data of these associated sensors; Based on the weights of the associated sensor data and the edges in the undirected graph, a collaborative correction model is used to correct abnormal data. The correction formula is a weighted sum based on the contribution of the associated data. The correction formula is: T 校正 (t) = w1T 测量 (t)+w2T 预测 (t), where w1 and w2 are weight coefficients automatically assigned by the system, and their sizes are determined by the edges of the undirected graph.

2. The indoor environment detection method according to claim 1, wherein: The determining the correlation between the environmental data and establishing an undirected graph based on the correlation includes: using historical data to analyze the correlation between different types of sensor data through statistical correlation; According to the correlation analysis, the correlation coefficients between different environmental factors are calculated to generate a correlation matrix. If the absolute value of the correlation coefficient of two environmental factors exceeds the set threshold, it is considered that there is a significant correlation between them; According to the determined correlation, an undirected graph is constructed, where each node in the undirected graph represents an environmental element, and the edges represent the correlation between environmental elements; If there is a significant correlation between two environmental factors, an undirected edge is added to the undirected graph to connect the corresponding nodes.

3. The indoor environment detection method according to claim 1, characterized in that: The data fusion and inference include: During real-time monitoring, the system detects missing data or sensor failure from a certain sensor and identifies the environmental factors with missing data and their corresponding nodes. According to the edges in the undirected graph, the data of other sensors associated with the missing data are extracted. The weight of the edge represents the influence of the associated sensor on the missing data. Utilize the associated sensor data and fuse the multi-source data according to the edge weights in the undirected graph; The inferred value of missing data is calculated through the data fusion model, and the result is adjusted according to the weight of each sensor in the undirected graph to ensure that the inferred missing data is reasonable and consistent with the current environmental status.

4. The indoor environment detection method according to claim 1, wherein: The environmental status assessment includes: During the environmental monitoring process, the system collects data from each sensor in real time, monitors the status and change trend of each node in the undirected graph, and also monitors the weights and data relevance of the edges in the undirected graph. The system detects whether there are significant changes in the edges of undirected graphs and whether the strength of the association between environmental factors deviates from the normal range. When the weight of a certain association increases or decreases abnormally, it indicates that there are abnormalities or sudden changes in the relevant environmental factors. Based on the detected changes in edge weights, the system analyzes which environmental elements have abnormal correlations and outputs them.

5. An indoor environment detection system, characterized in that: The system includes the following modules: a collection module for collecting indoor environmental data through a plurality of sensor networks distributed at different locations, wherein the sensors are used to detect a plurality of environmental factors, including temperature, humidity, air quality, and light intensity; A preprocessing module, configured to preprocess the environmental data, including data cleaning, data normalization, and data synchronization; An association module, configured to determine the association between the environmental data and establish an undirected graph based on the association; A processing module, configured to perform collaborative correction, data fusion and inference, and environmental status assessment on the real-time indoor environment detection data according to the undirected graph; During real-time monitoring, the system automatically detects whether a sensor's data is abnormal or drifting based on the changing trend of historical data or the deviation between the current measurement value and the data of other nodes in the undirected graph; Determine the sensor nodes associated with abnormal or drifting data through an undirected graph and extract the data of these associated sensors; Based on the weights of the associated sensor data and the edges in the undirected graph, a collaborative correction model is used to correct abnormal data. The correction formula is a weighted sum based on the contribution of the associated data. The correction formula is: T 校正 (t) = w1T 测量 (t)+w2T 预测 (t), where w1 and w2 are weight coefficients automatically assigned by the system, and their sizes are determined by the edges of the undirected graph.

6. An indoor environment detection system according to claim 5, characterized in that: The determining the correlation between the environmental data and establishing an undirected graph based on the correlation includes: using historical data to analyze the correlation between different types of sensor data through statistical correlation; According to the correlation analysis, the correlation coefficients between different environmental factors are calculated to generate a correlation matrix. If the absolute value of the correlation coefficient of two environmental factors exceeds the set threshold, it is considered that there is a significant correlation between them; According to the determined correlation, an undirected graph is constructed, where each node in the undirected graph represents an environmental element, and the edges represent the correlation between environmental elements; If there is a significant correlation between two environmental factors, an undirected edge is added to the undirected graph to connect the corresponding nodes.

7. An indoor environment detection system according to claim 5, characterized in that: The data fusion and inference include: During real-time monitoring, the system detects missing data or sensor failure from a certain sensor and identifies the environmental factors with missing data and their corresponding nodes. According to the edges in the undirected graph, the data of other sensors associated with the missing data are extracted. The weight of the edge represents the influence of the associated sensor on the missing data. Utilize the associated sensor data and fuse the multi-source data according to the edge weights in the undirected graph; The inferred value of missing data is calculated through the data fusion model, and the result is adjusted according to the weight of each sensor in the undirected graph to ensure that the inferred missing data is reasonable and consistent with the current environmental status.

8. An indoor environment detection system according to claim 5, characterized in that: The environmental status assessment includes: During the environmental monitoring process, the system collects data from each sensor in real time, monitors the status and change trend of each node in the undirected graph, and also monitors the weights and data relevance of the edges in the undirected graph. The system detects whether there are significant changes in the edges of undirected graphs and whether the strength of the association between environmental factors deviates from the normal range. When the weight of a certain association increases or decreases abnormally, it indicates that there are abnormalities or sudden changes in the relevant environmental factors. Based on the detected changes in edge weights, the system analyzes which environmental elements have abnormal correlations and outputs them.

Images