Indoor environment data analysis system based on machine learning
Through the indoor environmental data analysis system based on machine learning, the problem of insufficient data quality and integrity in the existing technology is solved, efficient data screening and analysis is realized, the accuracy and monitoring capabilities of environmental data are improved, and early warning is provided Function.
Patent Information
- Application Number
- CN202411298352.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-18
- Publication Date
- 2024-12-31
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, there are shortcomings in the quality and integrity processing of indoor environmental data, which affects the accuracy of environmental data analysis.
Design an indoor environment data analysis system based on machine learning, including data acquisition module, data processing and transmission module, data display module and alarm module. It is connected through serial mode and uses machine learning module to perform data screening, analysis and prediction to ensure that Data integrity and accuracy.
It improves the integrity and accuracy of environmental data, simplifies the data processing process, enhances the ability to perceive and respond in advance to indoor environmental changes, and provides comprehensive environmental monitoring and early warning functions.
Smart Images
Figure CN119223352A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of indoor environment data analysis, and in particular to an indoor environment data analysis system based on machine learning. Background Art
[0002] In the existing technology, indoor environmental data analysis provides a scientific basis for solving environmental problems by collecting and analyzing a large amount of environmental data. It helps to protect the ecological environment and provide guidance for improving the indoor environment. For example, by analyzing the indoor temperature, humidity, ozone concentration and organic volatile compound concentration, the comfort and health impact of the indoor environment can be evaluated, and corresponding improvement measures can be taken to reduce the harm of these harmful substances to the human body in a timely manner;
[0003] At present, there are still some deficiencies in the processing of the quality and integrity of indoor environmental data. Since the quality and integrity of environmental data are crucial to the accuracy of environmental data analysis applications, more attention needs to be paid to the quality and integrity of data, and measures need to be taken to improve the reliability of data. At present, the method of establishing a machine learning model based on machine learning can effectively alleviate the deficiencies in the processing of indoor environmental data.
[0004] Therefore, this application proposes an indoor environment data analysis system based on machine learning, which solves problems existing in the indoor environment by systematically analyzing environmental data, thereby improving the ability to solve indoor environmental problems. Summary of the invention
[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an indoor environment data analysis system based on machine learning.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions:
[0007] A machine learning-based indoor environment data analysis system includes a processor, the processor is electrically connected to a data acquisition module, a data processing and transmission module, a data display module and an alarm module, the data acquisition module includes a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, a formaldehyde concentration sensor, an ozone concentration sensor and a multidimensional acquisition unit, the multidimensional acquisition unit acquires and collects meteorological data and ecological data to improve the type and dimension of the data, and the source of data acquisition is based on meteorological station data and research institution data.
[0008] Preferably: the data display module is integrated into the LED display screen, the processor and the data acquisition module are connected in serial mode, and after the data acquisition module, the data processing and transmission module, the data display module and the alarm module are initialized, a while loop is entered to call the data reading interface of each module in turn.
[0009] Preferably, the alarm module includes real-time monitoring of environmental changes, and detects abnormal changes in data, such as abnormal temperature and excessive pollutants.
[0010] Preferably: the data transmission module uploads the collected data to the cloud for backup, and uses the machine learning module to perform deep learning and analysis on the collected data. The machine learning module includes a data screening unit, an analysis unit, a data fusion unit and a prediction unit.
[0011] Preferably: the data screening unit screens the data based on the automatic encoder, eliminates misplaced data and redundant data, and converts the data into a useful format, thereby improving the quality of data imported into the analysis unit, extracts key features of the collected data based on a neural network, reduces the complexity of the data, simplifies the processing flow of high-dimensional data, and quickly finds features that are important for the indoor environment.
[0012] Preferably: the analysis unit predicts the changes in air quality and ambient temperature within a certain time range based on a recurrent neural network, thereby improving personnel's perception of environmental changes and their ability to respond in advance; and predicts the impact range and intensity of indoor particulate matter based on a deep learning model to achieve control of indoor particulate matter, and the deep learning model is constructed using a linear model, a nonlinear model and a neural network.
[0013] Preferably, the data fusion unit fuses various data in the indoor environment to provide more comprehensive environmental monitoring.
[0014] Based on the above scheme: after the prediction unit collects and summarizes the existing data, it provides early warning of changes in environmental data based on the machine learning module.
[0015] Preferably, based on the above solution, when the data acquisition module acquires environmental data, the following acquisition principles need to be ensured:
[0016] 1: Environmental data needs to be representative, that is, to ensure that each environmental data has good sample collection capabilities;
[0017] 2: The data collection method needs to be reasonable. When collecting data, it should be ensured that its adoption strategy, source and collection frequency meet the needs;
[0018] 3: Label and classify the type of each collected data to truly reflect the category of the data, so that the machine learning model can ensure the accuracy of the pattern when learning and analyzing;
[0019] 4: Adjust data collection strategies and processes according to changes in data parameters to improve the performance and applicability of the model.
[0020] Based on the above solution, it is further preferred that when the machine learning module analyzes the indoor environment data, its analysis process includes the following steps:
[0021] S1: De-duplicate the data after acquisition;
[0022] S2: Check the type of overall data and the data missing;
[0023] S3: Clear the data with high repetition rate and the variables that do not need to be included in the model;
[0024] S4: Repair the type of variables and divide the processing methods of variable data.
[0025] The beneficial effects of the present invention are:
[0026] 1. By setting up a data acquisition module, it can cover the collection of environmental data from all aspects, thereby ensuring the integrity of environmental data, thereby providing sufficient support for the accuracy of environmental data analysis. By setting up a data processing module, various environmental data can be screened, which is crucial to the accuracy of the data.
[0027] 2. By processing and deleting incomplete and inaccurate data from the original data, the authenticity of the data is ensured, and at the same time the process of data screening is simplified, so as to facilitate repeated recycling in subsequent data analysis processes.
[0028] 3. Using the data transmission module, in the process of learning using the machine learning module, it takes corresponding measures for various data, removes duplicate records, handles data type mismatches, and processes missing values of continuous variables, thereby ensuring that subsequent analysis can be carried out effectively after data is missing. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] Figure 1 This is a schematic diagram of the overall structure of an indoor environment data analysis system based on machine learning proposed by the present invention. DETAILED DESCRIPTION
[0030] The technical solution of the present invention is further described in detail below in conjunction with specific implementation methods.
[0031] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.
[0032] Embodiment 1:
[0033] An indoor environment data analysis system based on machine learning, such as Figure 1 As shown, it includes a processor, the processor is electrically connected to a data acquisition module, a data processing and transmission module, a data display module and an alarm module, the data acquisition module includes a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor and a formaldehyde concentration sensor, an ozone concentration sensor and a multi-dimensional acquisition unit, the multi-dimensional acquisition unit collects and collects meteorological data and ecological data to improve the type and dimension of the data, and the source of data collection is based on meteorological station data and research institution data;
[0034] The temperature sensor monitors the indoor environment temperature, the humidity sensor monitors the relative humidity in the air, and the carbon dioxide concentration sensor and the formaldehyde concentration sensor monitor the amount of carbon dioxide and formaldehyde in the air;
[0035] The data display module is integrated in the LED display screen, and the processor and the data acquisition module are connected in serial mode. After the data acquisition module, the data processing and transmission module, the data display module and the alarm module are initialized, a while loop is entered to call the data reading interface of each module in turn;
[0036] The data transmission module uploads the collected data to the cloud for backup, and uses the machine learning module to perform deep learning and analysis on the collected data;
[0037] The alarm module includes real-time monitoring of environmental changes, and when abnormal changes occur in the data, such as abnormal temperature and excessive pollutants;
[0038] The machine learning module includes a data screening unit, an analysis unit, a data fusion unit and a prediction unit;
[0039] The data screening unit screens the data based on the automatic encoder, removes the misplaced data and redundant data, and converts the data into a useful format, thereby improving the quality of the data imported into the analysis unit, extracts the key features of the collected data based on the neural network, reduces the complexity of the data, simplifies the processing flow of high-dimensional data, and quickly finds the features that are important to the indoor environment;
[0040] The analysis unit predicts the changes of air quality and ambient temperature within a certain time range based on a recurrent neural network, thereby improving the personnel's ability to perceive and respond to environmental changes in advance; the influence range and intensity of indoor particulate matter are predicted based on a deep learning model to achieve control of indoor particulate matter, and the deep learning model is constructed using a linear model, a nonlinear model and a neural network;
[0041] The data fusion unit fuses various data in the indoor environment, such as temperature, carbon dioxide concentration, and formaldehyde concentration, to provide more comprehensive environmental monitoring;
[0042] After the prediction unit collects and summarizes the existing data, it provides early warning of changes in environmental data based on the machine learning module;
[0043] By setting up a data acquisition module, various aspects of environmental data collection can be covered, thereby ensuring the integrity of environmental data, thereby providing sufficient support for the accuracy of environmental data analysis. By setting up a data processing module, various environmental data can be screened, which is crucial to the accuracy of the data.
[0044] Embodiment 2:
[0045] An indoor environment data analysis system based on machine learning, such as Figure 1 As shown, when the data acquisition module collects environmental data, it is necessary to ensure the following collection principles:
[0046] 1: Environmental data needs to be representative, that is, to ensure that each environmental data has good sample collection capabilities;
[0047] 2: The data collection method needs to be reasonable. When collecting data, it should be ensured that its adoption strategy, source and collection frequency meet the needs;
[0048] 3: Label and classify the type of each collected data to truly reflect the category of the data, so that the machine learning model can ensure the accuracy of the pattern when learning and analyzing;
[0049] 4: Adjust data collection strategies and processes according to changes in data parameters to improve the performance and applicability of the model.
[0050] When the data transmission module uses the machine learning module for learning, it takes corresponding measures for various data, removes duplicate records, handles data type mismatches, and handles missing values of continuous variables, specifically including deleting records with missing values and filling them with mean values, so as to ensure that subsequent analysis can be effectively carried out after data is missing;
[0051] When the machine learning module analyzes indoor environment data, the analysis process includes the following steps:
[0052] S1: De-duplicate the data after acquisition;
[0053] S2: Check the type of overall data and the data missing;
[0054] S3: Clear the data with high repetition rate and the variables that do not need to be included in the model;
[0055] S4: Repair the type of variables and divide the processing methods of variable data.
[0056] By processing and deleting incomplete and inaccurate data from the original data, the authenticity of the data is ensured, and the process of data screening is simplified, making it easier to reuse it in subsequent data analysis processes.
[0057] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.
Claims
1. A machine learning-based indoor environment data analysis system, comprising a processor, characterized in that: The processor is electrically connected to a data acquisition module, a data processing and transmission module, a data display module and an alarm module. The data acquisition module includes a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, a formaldehyde concentration sensor, an ozone concentration sensor and a multidimensional acquisition unit. The multidimensional acquisition unit acquires and collects meteorological data and ecological data to improve the type and dimension of the data. The source of data acquisition is based on meteorological station data and research institution data.
2. The indoor environment data analysis system based on machine learning according to claim 1, characterized in that: The data display module is integrated into the LED display screen, and the processor and the data acquisition module are connected in serial mode. After the data acquisition module, the data processing and transmission module, the data display module and the alarm module are initialized, a while loop is entered to call the data reading interface of each module in turn.
3. The indoor environment data analysis system based on machine learning according to claim 2, characterized in that: The alarm module includes real-time monitoring of environmental changes, and detects abnormal changes in data, such as abnormal temperature and excessive pollutants.
4. The indoor environment data analysis system based on machine learning according to claim 3 is characterized in that: The data transmission module uploads the collected data to the cloud for backup, and uses the machine learning module to perform deep learning and analysis on the collected data. The machine learning module includes a data screening unit, an analysis unit, a data fusion unit and a prediction unit.
5. The indoor environment data analysis system based on machine learning according to claim 4 is characterized in that: The data screening unit screens the data based on the automatic encoder, eliminates misplaced data and redundant data, and converts the data into a useful format, thereby improving the quality of data imported into the analysis unit, extracts key features of the collected data based on a neural network, reduces the complexity of the data, simplifies the processing flow of high-dimensional data, and quickly finds features that are important for the indoor environment.
6. The indoor environment data analysis system based on machine learning according to claim 5, characterized in that: The analysis unit predicts the changes in air quality and ambient temperature within a certain time range based on a recurrent neural network, thereby improving personnel's perception of environmental changes and their ability to respond in advance; and predicts the impact range and intensity of indoor particulate matter based on a deep learning model to achieve control of indoor particulate matter. The deep learning model is constructed using a linear model, a nonlinear model and a neural network.
7. The indoor environment data analysis system based on machine learning according to claim 6, characterized in that: The data fusion unit fuses various data in the indoor environment to provide more comprehensive environmental monitoring.
8. The indoor environment data analysis system based on machine learning according to claim 7, characterized in that: After collecting and summarizing the existing data, the prediction unit provides early warning of changes in environmental data based on the machine learning module.
9. The indoor environment data analysis system based on machine learning according to claim 8, characterized in that: When the data acquisition module collects environmental data, it is necessary to ensure the following collection principles: 1: Environmental data needs to be representative, that is, to ensure that each environmental data has good sample collection capabilities; 2: The data collection method needs to be reasonable. When collecting data, it should be ensured that its adoption strategy, source and collection frequency meet the needs; 3: Label and classify the type of each collected data to truly reflect the category of the data, so that the machine learning model can ensure the accuracy of the pattern when learning and analyzing; 4: Adjust data collection strategies and processes according to changes in data parameters to improve the performance and applicability of the model.
10. The indoor environment data analysis system based on machine learning according to claim 9, characterized in that: When the machine learning module analyzes indoor environment data, the analysis process includes the following steps: S1: De-duplicate the data after acquisition; S2: Check the type of overall data and the data missing; S3: Clear the data with high repetition rate and the variables that do not need to be included in the model; S4: Repair the type of variables and divide the processing methods of variable data.